xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004554
Help-seeking and utilization patterns among african american and caucasian mothers and fathers :
b an examination of parental problem recognition, barriers, and beliefs
h [electronic resource] /
by Idia Thurston.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Dissertation (PHD)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: The underutilization of mental health services is a pervasive problem that persists despite efforts by researchers and interventionists to make treatment accessible. Several factors have been hypothesized to contribute to these underutilization rates including sociopolitical factors (financial and structural barriers), and cultural/familial factors (race, ethnicity, socioeconomic status, gender, age, marital status, attitudes, beliefs, and stigma). The current study set out to explore patterns of child mental health service utilization based on parents' perceptions. Guided by "The Youth Help-Seeking and Service Utilization Model," the relationship between parental problem recognition and willingness to seek formal and informal help as influenced by parents' demographic variables, sociocultural beliefs, experience, perceived need, family characteristics, and barriers were examined. Parental perceptions of problem behaviors in children were examined through the use of 3 vignettes (internalizing, externalizing, and no diagnosis conditions) varying only by child gender. A total of 251 Black and White parents from the community participated in this study. Data analyses involved correlations, t-tests, general linear modeling procedures (including ANOVA, ANCOVA, and multiple regressions), non-parametric tests, and logistic regression analyses. As hypothesized, results revealed that more parents recognized the internalizing and externalizing vignettes as problematic, parents reported stronger intentions to seek help when they recognized a mental health problem, and they were more willing to seek help for a boy with an internalizing problem than a girl. Additionally, perceived severity was related to recognition of both internalizing and externalizing problems. Gender, race, and previous experience were related to parents' recognition and willingness to seek help; with mothers, white parents, and those with more experience recognizing problems and expressing willingness to seek help for an internalizing problem. Finally, perception of barriers and certain beliefs impacted parents' willingness to seek help. The implications of this study with respect to help-seeking patterns for youth will be discussed. In addition, results will be discussed with an eye toward service providers' and intervention researchers' shaping the referral process, keeping families in treatment, and developing strategies aimed at improving problem recognition and help-seeking with eventual goals of increasing actual utilization of mental health services for mothers, fathers, and their children.
Advisor: Vicky Phares, Ph.D.
t USF Electronic Theses and Dissertations.
Help Seeking and Utilization Patterns among African American and Caucasian Mothers and Fathers : An Examination of Parental Problem Recognition, Barriers, and Beliefs b y Idia B initie Thurston A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Vicky Phares, Ph D Michael Brannick, Ph.D. James Epps Ph D Ellis Gesten, Ph.D. Joseph Vandello, Ph.D. Date of Approval: June 18 20 10 Keywords: Children, Mental Health, Disparities Vignettes, Ethnic Identity Copyright 2010 Idia Binitie Thurston
Dedication I would like to dedicate this manuscript to my amazing husband, Rashard Thurston, who has been there for me every step of the way from undergrad, to graduate school, and internship. Thank you for talking me through my ideas, sitting next to me in the library, Special thanks also goes to my best friend, Dwan Samuel, who always asked me questions that made me think about what and how I wrote, who I wanted to be a s a I will always remain thankful for my family; mom, dad, Ehi, Eboni, EJ, Esigie, and Odion, who have supported me emotionally and financially along my Ph.D. journey. I love you all so much and could not have gotten this far without your love and support.
Acknowledgments There are so many people who have made this journey a reality for me, beginning with my major professor, advisor, mentor, and friend, Vicky Phares, who always knew exactly when to push and when to let me go. I am truly honored to have worked with such an i nspirational teacher and hope I continue to make her proud. This project would not have been possible without the expert guidance of my dissertation committee members, the dedication of my amazing research assistants Jill and Takiyah, the support of the Boys and Girls club of Tampa Bay, and all the amazing mothers and fathers who were willing to share their opinions with me in this research project. I would also like to acknowledge my lab mates for their support, especially Ariz Rojas ; my internship clas smates who ; my other mentors along the way, Jon Bailey, Trevor Stokes, and Celia Lescano. I am sincerely grateful to the psychology department and the graduate school, especially Rod Hale, for all of their dedication and support without which I could never have succeeded in my graduate school journey. Thank you to countless others who always helped me stay on the path!
i Table of Contents List of Tabl es ................................ ................................ ................................ ..................... iii List of Figure s ................................ ................................ ................................ .................... iv Abstrac t ................................ ................................ ................................ ................................ v Help Seeking and Utilization Patterns ................................ ................................ ................. 1 Background ................................ ................................ ................................ .............. 1 Underutilization of Mental Health Services ................................ ............................ 2 Sociopolitical Factors ................................ ................................ ............................... 4 Financial Barriers ................................ ................................ ................................ ..... 5 Structural Barriers ................................ ................................ ........................ 6 Cultural and Familial Factors ................................ ................................ ....... 7 Race and Ethnicity ................................ ................................ ................................ ... 8 Soc ioeconomic Status ................................ ................................ ............................ 10 Gender ................................ ................................ ................................ .................... 12 Age and Marital Status ................................ ................................ ........................... 14 A ttitudes, Beliefs, and Stigma ................................ ................................ ............... 15 Help Seeking and Service Utilization ................................ ................................ .... 19 Race, E thnicity, and Ethnic Identity ................................ ................................ ...... 20 Theories and Models ................................ ................................ .............................. 21 Problem Recognition ................................ ................................ ............................. 26 Clinical Assessment of Need ................................ ................................ ................. 27 Perceived (Subjective) Need ................................ ................................ .................. 28 Family Characteristics ................................ ................................ ........................... 32 Decision to Seek Help ................................ ................................ ............................ 32 Knowledge of Problem ................................ ................................ .......................... 33 The Current Study ................................ ................................ ................................ .. 34 Pilot Study ................................ ................................ ................................ .............. 39 Participants Pilot Study ................................ ................................ ........... 39 Measures Pilot Study ................................ ................................ .............. 39 Procedures Pilot Study ................................ ................................ ............ 4 0 Results Pilot Study ................................ ................................ .................. 4 0
ii Method ................................ ................................ ................................ ............................... 42 Participants ................................ ................................ ................................ ............. 42 Procedures ................................ ................................ ................................ .............. 44 Measures ................................ ................................ ................................ ................ 47 Vignettes ................................ ................................ ................................ .... 47 Demographics ................................ ................................ ............................ 48 Utilization ................................ ................................ ................................ .. 48 Multigroup Ethnic Identity Measure Revised (MEIM R) .................... 49 Barriers to Utilization ................................ ................................ ................ 50 Beliefs A bout Causes Revised (BAC R) ................................ ................ 51 Data Analy ses ................................ ................................ ................................ ........ 52 Results ................................ ................................ ................................ ................................ 54 Descriptives ................................ ................................ ................................ ............ 54 Group Differences ................................ ................................ ................................ .. 60 Hypothesis Testing ................................ ................................ ................................ 62 Post hoc Analyses for Hypothesis 4 ................................ .......................... 67 Post Hoc Analyses ................................ ................................ ................................ 82 Parental Confidence in R ating s of V ignettes ................................ ............. 82 ealth E xperienc e ................................ ... 83 Belief s a bout Causes of Mental Health Problems ................................ ...... 84 Discussion ................................ ................................ ................................ .......................... 88 Limitations and Future Research ................................ ................................ ......... 105 C linical Significance and Conclusion s ................................ ................................ 109 References ................................ ................................ ................................ ........................ 11 4 Appendices ................................ ................................ ................................ ....................... 1 31 App endix A: Letter of Invitation ................................ ................................ ......... 132 Appendix B: Referral Letter ................................ ................................ ................ 133 Appendix C: Boy Internalizing Vignette with follow up questions .................... 134 Appendix D: Externalizing and Control Vignettes ................................ .............. 136 Appendix E: Demographics ................................ ................................ ................. 137 Appendix F: Utilization ................................ ................................ ....................... 139 Appendix G: Multigroup Ethnic Identity Measure Revised (MEIM R) .......... 142 Appendix H: Barriers to Utilization ................................ ................................ ..... 144 Ap pendix I: Beliefs About Causes Revised (BAC R) ................................ ...... 146
iii List of Tables Table 1 Demographic Variables for Parents by Race and Gender ......................... 44 Table 2 Logistic Regression for Perceived Need Variables and Internalizing Problem Recognition ................................ ................................ ................. 6 6 Table 3 Logistic Regression for Selected Perceived Need Variables and Externalizing Problem Recognition ................................ ........................... 68 Table 4 Logistic Regression for Selected Perceived Need Variables and Willingness to Seek Help for an Externalizing Problem ........................... 68 Table 5 Logistic Regression for Parent Demographic Variables and Internalizing Problem Recognition ................................ ............................ 70 Table 6 Logistic Regression f or Parent Demographic Variables and Willingness to Seek Help for an Internalizing Problem ............................ 71 Table 7 Gender ................................ ................................ ................................ ........ 7 6 Table 8 Logistic Regression for Religious Beliefs and Parent Race and Gender ................................ ................................ ................................ ........ 77 Table 9 Intercorrelations between Barriers, Ethnic Identity, Formal and Informal Help Seeking for Internalizing and Externalizing Problems ...... 81 Table 10 Multiple Regression for Beliefs about Causes of Mental Health Problems and Willingness t o Seek Formal Help for an Internalizing Problem ................................ ................................ ................................ ...... 85 Table 11 Multiple Regression for Beliefs about Causes of Mental Health Problems and Willingness to Seek Formal Help for an Externalizing Problem ................................ ................................ ................................ ...... 86
iv List of Figures Figure 1 The Behavioral Model of Health Services Use ................................ ......... 2 2 Figure 2 The Youth Help Seeking and Service Utilization Model .......................... 25 Figure 3 Proposed research model ................................ ................................ ........... 35 Figure 4 Parental Experience with Various Disorders by Race and Gender ............ 56 Figure 5 Parental Beliefs of the Most Import a Health Problems ................................ ................................ ......................... 58 Figure 6 First Source of Help Seeking for Internalizing V ignettes by Race and Gender ................................ ................................ ................................ 59 Figure 7 Fi rst Source of Help Seeking for Ex ternalizing V ignettes by Race and Gender ................................ ................................ ................................ 60
v Help Seeking and Utilization Patterns among African American and Caucasian Mothers and Fathers: An Examination of Parental Problem Recognition, Barriers, and Beliefs Idia Binitie Thurston A bstract The underutilization of mental health services is a pervasive problem that persists despite effort s by researchers and interventionists to make treatment accessible Several factors have been hypothesized to contribut e to th ese underutilization rates incl uding sociopolitical factors (financial and structural barriers), and c ultural/ f amilial factors (race, ethnicity, socioeconomic status, gender, age, marital status, attitudes, beliefs, and stigma). The current study set out to explore patterns of child men tal health service utilization b ased on parents Seeking and Service willingness to seek formal and informal help as influenced by parents variables, sociocultural beliefs, experience, perceived need, family characteristics, and barriers were examined. Parental perceptions of problem behaviors in children were examined through the use of 3 vignettes ( internalizing, externalizin g, and no diagnosis condition s) var ying only by child gender. A total of 251 Black 1 and White 2 parents from the community participated in this study. Data analyses involved correlations, t tests, 1 The terms Black and African American will be used interchangeably throughout this paper. 2 The terms White and Caucasian will be used interchangeably throughout this paper.
vi general linear modeling procedure s (including ANOVA, ANCOV A, and m ultiple regression s ), non parametric tests, and logistic regression analyses. As hypothesized, r esults revealed that more parents recognized the internalizing and externalizing vignettes as problem atic parents report ed stronger intentions to seek help when they recognized a mental health problem, and they were more willing to seek help for a boy with an internalizing problem than a girl. Additionally, perceived severity was related to recognition of both internalizing and externalizing problems. Gender, race, and previous experience were related to parents recognition and willingness to seek help; with mother s white parents, and those with more experience recognizing problems and expressing willing ness to seek help for an internalizing problem. Finally, perception of The implications of this study with respect to help seeking patterns for youth will be discussed. In addition, re sults will be discussed with an eye toward service providers and intervention researchers shaping the referral process, keep ing f amilies in treatment, and develop ing strategies aimed at improving problem recognition and help seeking with eventual goals o f increasing actual utilization of mental health services for mothers fathers and their children
1 Help Seeking and Utilization Patterns Background Mental health services have been sorely underutilized in our society by adults and children alike Researchers have investigated various factors that may be contribut ing to underutilization of these services Some of these factors include demographic variables like race and gender; societal structure facto rs like income and socioeconomic status; individual attitudes and beliefs about mental health; various structural barriers to using mental health services ; and perceiv ing an actual need for mental health care. However, there is a dearth of systematic res earch on the patterns of utilization and help seeking and how these various factors may interact to increase or decrease actual utilization of mental health services in adults and children. Lack of understanding of the patterns of utilization and help see king may result in delayed progress in efforts to increase utilization in all individuals as well as targeted groups (such as racial and ethnic minorities) who have especially low rates of utilization. The purpose of this study is to conduct a preliminar y investigation of patterns related to underutilization of mental health services in a systematic fashion. Various researchers in the field have theorized patterns that may be contribut ing to utilization of mental health services. However, little has bee n done to establish these patterns in a systematic way and to determine if these theorized patterns are indeed paths to utilization. A thorough understanding of these patterns will provide intervention researchers and
2 mental health service providers with f ocus ed areas to target in efforts to improve utilization in children and their families Underutilization of Mental Health Services The seminal report of the Surgeon General based on major epidemiological surveys stated that over 20% of adult s in the U.S. population are affected by mental health disorders in a given year (U.S. DHHS, 1999). Re cent epidemiological studies estimate th at 26.2% of adult s will have a mental or addictive disorder s in a given year (Kessler, Chiu, Demler, & Walters, 2005) with lifetime prevalence estimates at 46.4% ( Kessler, Berglund, Demler, Jin, Merikangas, & Walters, 2005 ) Despite th e s e high prevalence rate s underutilization of mental health services in our society is well documented. Specifically, Kessler and colleagues examined utilization rates in adults with mental health disorders and found that 32 9 % had used mental health services in the preceding year (Kessler, Demler, et al. 2005). This utilization rate is an increase from acros s the general adult population, current mental health utilization rates for emotional disorder s increased from 12.2% between 1990 and 1992 to 20.1% between 2001 and 2003 ; with only half of these individuals actually meeting criteria for a DSM IV diagnosis (Kessler, Demler, et al. 2005) Even though mental health utilization increased over 10 years, it is still notable and concerning that most adults with mental health disorders are not receiv ing treatment Th e disparity between those individuals who need mental health services and those who actually utilize mental health services is not limited to adults. Epidemiologic studies have also documented the underutilization of mental health services in children and adolescents. Estimates indicate that a median of 12 14 % of children and adolescents
3 me e t criteria for a serious emotional disturbance (i.e. they meet diagnostic criteria and have impairment in functioning) each year ( Costello, Egger, & Angold, 2005 ; Wadde ll, Offord, Shepherd, Hua, & McEwan, 2002 ) Additionally, l ifetime prevalence of child m ental health disorders have been estimated between 53 57%, with more comprehensive estimates b ased on the National Comorbidity Survey Replication Adolescent Supplemen t (NCS A) forth coming (Kessler et al. 2009). Despite high prevalence rates, the report of the Surgeon General indicated that 21% of children and adolescents in the U.S. use mental health services every year ; with only half meeting criteria for a diagnos able mental or addictive disorder (U.S. DHHS, 1999) Of note, when treatment is utilized by children and adolescents with a serious emotional disturbance, only 22 25% actually receive services from the specialized mental health sector (Waddell et al ., 200 2). Thus, approximately 75 80% of children and adolescents fail to receive the specialty mental health services they need, and the majority of children and adolescents fail to receive any services at all ( Costello et al., 2005 ). An important aspect of s ervice utilization therefore, is where and from whom mental health services are being received. Results from the Great Smoky Mountain Study indicated that the majority of children with a serious emotional disturbance who had used mental health services were seen by providers in the education sector which included guidance counselors and/or school psychologists ( Farmer, Burns, Phillips, Angold, & Costello 2003) Other researchers have found that mental health services are often sought out from primary care provider s due to the significant increase in psychotropic medication use (Glied & Cuellar, 2003). Overall, the utilization of mental
4 health services from the school sector and primary care setting has been well established (Power, Eiraldi, Clarke, Mazzuca, & Krain, 2005). There are several contributing factors to the underutilization of mental health services by children, adolescents, and adults. Some of th ese factors were described in the includ ing : problems with cost, fragmentation of services, lack of availability of services, and societal stigma toward mental illness ( on Mental Health, 2003 ). The various factors contributing to underutilization have b e e n categ orized into two groups : sociopolitical factors and cultural and familial factors (Power, Eiraldi, Clarke, Mazzuca, & Krain, 2005). These factors will be discusse d in detail below In general, however, s ociopolitical factors include variables related to availability and coverage of health insurance, location of mental health services, and ease of accessibility to these services (Power et al. 2005). Cultural and f amilial factors include demographic variables that have been shown to influence utiliza tion of mental health services including: race ethnicity, socioeconomic status, and marital status. Also included in this category are beliefs that vary across familie s and cultural groups such as beliefs about mental health illness, validity of treatments, and trust in professionals who offer the mental health services (Power et al., 2005). Sociopolitical f actors These factors include variables related to the availability of health insurance with coverage of mental health services location of mental health services, and accessibility to these services (Power at al., 2005). From a broader perspective sociopolitical factors can be described as barriers to the utilization of mental health services based on access variables rather than cultural or attitudinal factors. These
5 sociopolitical barriers include financial barriers (cost and insurance coverage), and s tructural barriers ( accessibility to services, avail ability of services, etc). It is important to note however, that historically, research on barriers to service utilization often includes financial and structural barriers as well as other more cultural/ attitudinal barriers such as stigma, attitudes, and perceived barriers that are more unique to racial minorities such as distrust due to differences in race and language barriers. This study however, will attempt to discuss these cultural /attitudinal factor s separately from the sociopolitical factors but some overlap in the review of literature is probable Financial b arriers One of the highly influential factors related to the underutilization of mental health services is the cost of mental health treatment and the r elated cost of health insurance cov erage. A study by Ringel and Sturm (2001) found that 7% of families with a child claimed financial barriers as the reason for not receiving mental health care. Unfortunately the researchers did not report the percentage for adult populations. Research on the effects of health insurance coverage on mental health service utilization by children and adolescents has been well established For instance, Busch an d Horwitz (2004) found that after controlling for reported mental health need u ninsured children were less likely to have used mental health services than children who were privately insured Similar results were reported by Kataoka, Zhang, and Wells (2002) who found that in comparison t o families who had public or private insurance, uninsured families were less likely to receive mental health services and ethnic minority families were overrepresented among these uninsured families In addition, Ringel and Sturm (2001) found that childre n with Medicaid were most likely to receive mental health services while children
6 without any insurance were least likely to receive mental health services. Thus, it appears having some type of resource to cover mental health needs (either private or publ ic insurance ) results in higher rates of utilization of mental health services. S tructural b arriers There are numerous structural barriers that contribute to the u nderu tilization of mental health services by parents and their children. Some of the se ba rriers ing fragmentation of services and lack of availability of services (U.S. DHHS, 1999). The location of mental health services and ease of accessibility to these services have also been well documented as barriers to utilization of mental health services. R ecently, Sareen and colleagues (2007) investigated perceived barriers to service utilization in adults across several countries (US, Canada, and the Netherlands) and found higher rates of perceived attitudinal barriers than structural barriers overall More specific to structural barriers, however, they found that individuals who had a mood or substance disorder in the past year were mo st likely to endorse financial barriers Fur thermore, respondents in the U.S. were more likely to report that they were unsure of w h ere to go for services and reported that services would take too much time or be inconvenient (Sareen et al., 2007). Owens and colleagues (2002) investigated parent report s of barriers to their health utilization. They found that 20.7% of parents endorsed structural barriers, which included issues related to cost, accessibility, and availability of mental health services. Furthermore, 23.3% of paren ts endorsed barriers related to perceptions of mental health problems (such as deciding to handle the problem on their own), wh ereas 25.9% endorsed barriers related to perceptions of mental health services (such as being afraid of what family/friends might say; Owens et al., 2002). In terms of barriers to future service
7 utilization, Kekorian, McKay, and Bannon (2006) examined to s eeking mental health services for their children a second time and found that those parents who endo rsed more barriers to service utilization were more like ly to doubt the Some i ntervention pro grams have attempted to reduce these structural barriers to treatment util ization with little t o no impact on overall service utilization For instance, Bickman, Heflinger, Northrup, Sonnichsen, and Schilling (1998) implemented an intervention program that increase d parents knowledge of mental health services and their self efficacy but t he program did not increase treatment involvement, which was More recently, Stevens, Klima, Chisolm, and Kelleher ( 2009 ) utilized structured telephone support se rvices in attempt to reduce barriers to service utilization but were also unsuccessful. These findings suggest that without addressing some of the other variables influencing the under utilization of mental health services in addition to structural and fi nancial barriers to treatment interventions to increase utilization will likely be unsuccessful Some of t hese variable s are categorized as cultural and familial factors and are often more difficult or impossible to change Cultural and f amilial f actors These factors include demographic variables that have been shown to influence utilization of mental health services (such as race/ethnicity, socioeconomic status, gender, age, and marital status), and beliefs that vary across families and cultural groups such as beliefs about mental health illness, validity of treatments, and trust in professionals who offer the mental health services (Power et al.,
8 2005). Thus cultural/familial factors are those variables related to mental health utilization th at are based on specific demographic variables and attitudinal barriers. Race and Ethnicity. Racial and ethnic minorities make up approximately 30 percent of the United States population based on census data collected in 2000 (U.S. Census Bureau, 2001). By the year 2050, it is projected that racial and ethnic minorities will make up approximately 50 percent of the United States population (U.S. Census Bureau, 2004). Needless to say the United States has seen and will continue to see a tremendous rate of growth in racial and ethnic groups that will ultimately influence the way of life of all Amer icans. These statistics make the importance of racial and ethnic disparities an even more salient factor, as over half the population in the United States will be impacted eventually Despite its detrimental effects, t he underutilization of mental heal th services by individuals from minority cultures ha s been well documented and the research is fairly consistent across various studies prevalence of mental disorders for racial and ethnic minorities are equal to those for Whites, yet utilization of mental health services is extremely low for racial and ethnic minorities (U.S. DHHS, 2001). Th is s e minal report stated that ethnic minorities have n eeded mental health services; often receive poorer quality of mental health care; and are underrepresented in mental health rese These factors were reiterated in the p Resea rch indicated that racial and ethnic minorities experience a heavier disability burden from mental health illness than Caucasians, because they receive fewer services
9 and poorer quality care, and not because their illnesses differ in severity or prevalence (U.S. DHHS, 2001). However, Sue and Chu (2003) revisited this issue of the prevalence of mental health disorders in ethnic minorities, and reviewed data collected after the concluded that racial and ethnic di fferences are seen in the prevalence of mental disorders in adults Specifically, they reported that African Americans tend to have the lowest rates of mental health disorders; Mexican Americans, and Asian Americans and Pacific Islanders have slightly low er or similar rates to Non Hispanic Whites, while American Indians and Alaskan Natives have the highest rates of mental health disorders (Sue & Chu, 2003). Overall, however, regardless of the prevalence rates, research has consistently shown that racial a nd ethnic minorities underutilize mental health services more than Caucasians (Sue & Chu, 2003). Furthermore, t here are additional barriers that are unique to individuals of racial and ethnic minority groups including: mistrust and fear of treatment, raci sm/discrimination, differences in language and communication and cultural barriers in general (Thompson, Bazile, & Akbar, 2004). Dobalian and River s (2008) examined racial and ethnic differences in mental health service utilization and found that even af ter controlling for socioeconomic status and insurance racial and ethnic di sparities in service utilization persisted with African Americans and Hispanics being less likely to visit a mental health professional Kataoka, Zhang, and Wells (2002) also found that uninsured families were disproportionately overrepresented by ethnic minorities and both Hispanics and African American children had higher levels of unmet need than Whites However, as previously noted, even when they are insured ethnic mino rities still underutilize mental health services. This
10 utilization pattern is consistent in both the adult ( Dobalian & Rivers, 2008 ) and child (Ringel & Sturm, 2001) populations Furthermore, research has shown that even when minorities overcome barriers and actually utilize mental health services, they are at high risk of dropping out of treatment (Snowden & Yamada, 2005). Sue and Sue (2007 ) found that 50 percent of African American clients stop treatment after the first session compared with 30 percent of Caucasian clients. Additionally, African American s and Latino s are less likely to seek help from mental health professionals and agencies compared to Caucasian s (Dobalian & Rivers, 2008; Snowden, 2001) Furthermore, research found that African Americ an parents endorsed more barriers to mental health service utilization for their childre n than Caucasian parents (Thurston & Phares 200 8 ). Although no race differences were found in attitudes toward mental health treatment for their children. Socioecono mic s tatus. It is almost impossible to review the influence of socioeconomic status ( SES ) on mental health service utilization without also discussing race and ethnicity. It has been well documented th at a large proportion of minorities are represented i n lower SES classes. In truth, the finding that b arriers to mental health service utilization influence minorities more severely (U.S. DHHS, 2001) is likely because a large proportion of racial and ethnic minorities are in lower socioeconomic classes and most of the barriers to mental health service utilization have a more detrimental effect on individuals of lower SES. The U. S. Census Bureau (200 9 ) reported on th e 200 8 poverty levels of children under age 18 by race and ethnicity. Specifically, the report stated that approximately 3 4 percent of African American children, 30 percent of Hispanic children, 1 5 percent of
11 Caucasian children, and 1 4 percent of Asia n ch ildren are living below the poverty level. Another aspect of poverty that has a significant effect on racial and ethnic minorities is single parent status. In her review of socioeconomic disadvantage and child development, McLoyd (1998) noted that financ ial problems and risk of poverty are some of the burdens associated with raising children in single parent (usually single mother) households. She stated that these financial burdens are often due to low wages from just one parent, low education attainmen t, unfavorable economic conditions, and low rates and levels of child support (McLoyd, 1998). Data on children being raised by single mothers indicated that 50.2 percent of African American children and 23.3 percent of Hispanic children are being raised b y single mothers compared with 13.4 percent of Caucasian children (Hofferth, Stueve, Pleck, Bianchi, & Sayer, 2002). Data from the U. S. Census Bureau (2006) reported that 28.7 percent of children being raised by single mothers are below the poverty level compared to 5.1 percent of children with married parents. As stated earlier, m ost of the factors related to underutilization of mental health services have more detrimental effects on individuals of lower SES. Specifically, individuals of low SES perceiv ed more barriers to mental health service utilization than higher SES individuals due to individuals in low SES communities having fewer financial resources and the least educational attainment ( U.S. DHHS, 1999) Mo re recently, in a study of poor neighbor hoods, Chow and colleagues (2003) found higher rates of utilization of public mental health services in low poverty areas (where <20% of households had incomes below the poverty level) compared to high poverty areas (where 20% or more of the households had incomes below the poverty level) Similar
12 underutilization patterns are found in adolescents, Pumariega and colleagues (1998) found that socioeconomic status was consistently related to utilization compared to all the other non clinical factors they exami ned In contrast, when Kataoka and colleagues (2002) examined utilization and poverty they found that children from families who were at or below the poverty level were more likely to have used mental health services than children from non poor families. This finding is likely due to the fact that Medicaid and other public insurance s are most often utilized by poor families (Power, Eiraldi, C larke, Mazzuca, & Krain, 2005). Overall socioeconomic status appears to have major influences on utilization of services due to the s trong relationship between race, ethnicity and SES and the influence of financial barriers on utilization of mental health services. Gender. It is well known that parents have a tremendous impact on the lives of their children both genetically and environmentally. In comparison to mothers, however, less information is kno majority of studies that include parents tend to include only mothers rather than mothers and fathers. R ecent review s of studies of both developmental psychopathology and pediatric psychol ogy have shown that fathers are highly underrepresented in research ( Phares, Fields, Kamboukos, & Lopez, 2005; Phares, Lopez, Fields, Kamboukos, & Duhig 2005 ) This pattern is problematic given that when mothers and fathers are studied, both appear to ha 10 ). Overall, parental gender and child gender are differentially re lated to mental health service utilization by chi ldren and their families. Research has established that women utilize mental health services more often than men (Mahalik, Good, & Englar
13 Carlson, 2003). The influence of pare mental health service utilization, however, has not b een investigated thoroughly. Fals Stewart, Fincham, and Kelley (2004) found that substance abusing fathers were less willing than substance abusing mothers to allow their children to receive treatment. There is also evidence that fathers are less incline d to participate in treatment than mothers (Duhig, Phares, & Birkeland, 2002). A review by Phares and colleagues concluded that men often believe that problematic behaviors are normal for children or that therapy is a waste of time (Phares, Fields, & Bini tie, 2006). These beliefs might also influence fathers when deciding to seek help for their children. Overall, since women utilize mental health services for themselves more often than men, it is reasonable to assume that mothers would be more inclined t han fathers to seek mental health services for their children. Previous r esearch indicated that mothers had more positive attitudes toward mental health service utilization for their children than fathers ( Thurston & Phares 200 8 ). However, no parental gender differences were found in perceived barriers to seeking services for their children. Research on youth gender and service utilization is somewhat inconsistent. I n their s tudy of tri ethnic adolescents, Pumariega and colleagues found that females had more mental health visits than males (Pumariega, Glover, Holzer, & Nguyen, 1998). However, in their review of epidemiologic data of children ages 3 17 years, Kataoka and colleagues reported that boys were more likely to have receive d mental health ser vices than girls (K ataoka, Zhang, & Wells, 2002). With regard to factors associated with utilization, it has been shown that girls and adolescent females tend to have more positive help seeking attitudes perceive fewer barriers to help seeking ( Raviv, Ra viv, Vago
14 Gefen, & Fink, 2009) and report less stigma to utilizing mental health services (Chandra & Minkovitz, 2006) than boys and adolescent males The refore, although most studies consistently find that adult females utilize services more often than m ales and that female youth seek help more often than male youth; epidemiologic data show that male youth actually utiliz e mental health services more often than female youth. Furthermore, a review by Zwaanswijk and colleagues reported that gender differen ces in help seeking sought for boys in childhood and early adolescence wh ereas more help is sought for girls in late adolescence (Zwaanswijk et al., 2003). The research ers attributed these findings to the fact that externalizing problems (which are more prevalent in boys) tend to decrease with age while internalizing problems (more common in girls) increase with age. Age and m arital s tatus. Kessler et al. (2008) reported monotonic increases in mental health utilization rates in adults over age 18 until about age 44, at which time a decline in mental health utilization occurred with some of the lowest utilization rates in adults over age 60. With regard to youth utilization, older children use mental health services more frequently than y ounger children. Ringel and Sturm (2001) reviewed several epidemiologic studies to determine national estimates of mental health service utilization in youth They found that approximately 5 7% of children and adolescents ages 1 17 years use specialty mental health services in a year. Th ese data further br oke down to 1% to 2% of preschoolers, 6% to 8% of 6 11 years old, and 7% to 9% of youth ages 12 17 years (Ringel & Sturm, 2001). Similar data w ere also reported by Kataoka, Zhang, and Wells (2002) who reviewed three national surveys of mental health service
15 use among children ages 3 17 years old. Th e findings from these studies indicate that school age chi ldren a re 3 4 times more likely to receive mental health services than preschoolers. Mental health s ervice utilization appears to be higher in single parent homes compared to homes with two parents (Teagle, 2002) Specifically, Puma riega and colleagues found more lifetime mental health vi sits in adolescents from father absent households; however, the reverse was true for the Hispanic sample alone (Pumariega, Glover, Holzer, & Nguyen, 1998). Researchers also reported that unmarried individuals in both p rivate and public health sectors had greater odds of utilizing services than married individuals (Swartz et al., 1998). Overall, it appears that the odds of utilizing mental health services are greater for younger, unmarried adults; and older youth from si ngle parent home s Attitudes beliefs, and stigma the validity of treatment, their trust in mental health professionals, fear of being stigmatized by friends and family, and attitudes toward treatment ; all influence their help seeking intentions and utilization of mental health services. Attitudinal barriers (perceived barriers to service utilization based on attitudinal factors rather than structural/access factors) have been shown to have a strong relationship wit h mental health service utilization (Sareen et al., 2007). In this study of service utilization across three countries, Sareen and colleagues (2007) also found that attitudinal barriers were more prevalent than structural barriers. At the individual adul t level, studies have found strong associations between attitudes and help seeking intentions ( Mackenzie, Gekoski, & Knox, 2006 ). Kessler and colleagues (2001) found that the most commonly reported
16 reason by adults for failing to seek treatment and droppi ng out of treatment was their wanting to solve the problem on their own. They also noted that a major attitudinal emotional problem requiring treatment. s also appear to influence their willingness to seek mental health services for their children. Stiffman, Pescosolido, and Cabassa (2004) describe d parents as gatekeepers to child mental health service utilization toward m ental health services had a significant impact on their wilingness to seek help for their children. Other beliefs also influence mental health service utilization. Yeh and colleagues found relationships between type of parental belief about causes of mental health problems and service utilization (Yeh et al., 2005). Specifically, they reported that parents who believed that mental health problems were due to physical or trauma causes (biopsychosocial beliefs) were more likely to have u tilized mental health services 2 years later. Furthermore, they found that parents who believed that mental health problems were due to the influence of friends (sociological beliefs) were less likely to have utilized services. Finally they reported that certain beliefs (physical, trauma, relational issues, and prejudice) were partial mediators in the relationship between race/ethnicity and utilization ( Yeh et al., 2005). It is notable, however, that this study focused on parents of youth who were active in the public service sectors (i.e. juvenile courts, mental health, child welfare, etc). Alvidrez (1999) found that service utilization by family and friends and beliefs about causes of mental illness were significant predictors of utilization in African American, Hispanic, and Caucasian women Pumariega, Rogers, and Rothe
17 (2005) related that minority families are often suspicious of the mental health care system, which results in their discomfort when seeking care in su ch a system. Specifically, Africa n Americans history with discrimination, oppression, and involuntary treatment; Hispanics concern about language and cultural barriers; and Asian Americans experience of shame around mental illness, are likely contributing to their underutilization of mental health services (Pumariega, Rogers, & Rothe, 2005). Stigma has also been identified as a major factor related to mental health service utilization in children ( Chandra & Minkovitz, 2006 ) and adults ( Gary, 2005) Both studies report that fear of being stigmatized prevents adults and children from utilizing mental health services. Owens and colleagues (2002) found that the most common type of perceived parental barrier when seeking services for their children was related to perceptions of mental health services ( including thoughts that treatment will not help stigma, and not knowing who to trust ). G ary (2005) suggested that minorities are ofte n concerned about prejudice and discrimination and feel that they might suffer problem and so are reluctant to seek mental health services In summary, t he research reviewed above in dicates that overall mental health services are significantly underutilized by adults and youth alike. Several factors including financial, structural barriers, race and ethnicity, socioeconomic status, gender, age, marital status, attitudes, beliefs, and stigma influence help seeking intentions and service utilization at varying levels. Overall, in terms of youth utilization, it appears that young children, girls, minorities, and the uninsured are least likely to receive mental
18 health services and most y outh receive initial mental health care from schools or their primary care doctors. It is also notable that parents have a significant role in youth service utilization because i n most cases, before children and adolescents can receive mental health care parental consent must be obtained for their treatment. Thus, if parents are the unavoidable mediator between youth and psychological services, it is important for seek help for the ir children. R ecent epidemiological data on the lifetime prevalence of DSM IV disorders in adults ages 18 and older indicated that about half of the U.S. population will meet criteria for a DSM IV disorder in their lifetime with the first onset usually o ccurring in childhood or adolescence (Kessler, Berglund, Demler, Jin, Merikangas, & Walters, 2005). Thus it is essential that research focus on understanding how children and adolescents utilize mental health services and investigate ways to promote and i ncrease utilization in children and adolescents. Furthermore, r esearch has shown that parental barriers and attitudes for themselves differ from their barriers and attitudes toward treatment for their children, whereby parents are more willing to seek car e for their children than themselves ( Thurston & Phares 200 8 ). Thus, the influence parents have on youth mental health imperative. of a few related factors is necessary : help seeking service utilization, race, ethnicity, ethnic identify, and problem recognition
19 Help Seeking and Service Utilization Acro ss the literature distinctions have been made between help seeking and actual utilization of mental health services. Child related h elp seeking has been de scribed as the act of seeking assistance (formal or informal) for an emotional or behavioral proble m ( Srebnik, Cauce, & Baydar, 1996 ). Se rvice utilization however, can be described as the actua l utiliz ation of specific services or programs to help re solve a perceived emotional or behavioral problem. Historically, research on help seeking has focused solely on estimations of actual mental health service utilization ( Cauce et al., 2002 ). More recently, the utility and importance of understanding how families seek help and the factors that contribute to their decisions to seek help have been underscored A f ocus on help seeking pathways will allow researchers to assess the underutilization dilemma at the beginning of the process when the problem is first recognized (Cauce et al., 2002). Help seeking typically precedes service utilization meaning that parents often seek help for their children and ultimately utilize mental health services as a result of their help seeking. However, this is more of an ideal situation and is not always or e ven often the case. Sometimes parents are mandated by the juvenile justice system or child welfare to enroll their children in mental health care, thus in these situations services are utilized without ever actually seeking help voluntarily On the other hand, parents may seek help for their children and never actual ly utilize professional services due to any number of barriers to service utilization. It is notable that several theoretical models on factors related to service utilization and help seeking tend to use the terms interchangeably. The general theorized pathway to help seeking has been largely agreed
20 upon by several researchers ; however, there is very little research on how this model varies by race/ethnicity and other cultural factors. These models wil l be discussed in detail later in this review Race, Ethnicity, and Ethnic Identity Defining race and ethnicity is common practice in research on minority issues. As Cauce and colleagues (2002) summarized, race has been viewed historically as b iologically based while ethnicit y is often viewed as culturally based; however there is no scientific basis for these definitional constructs. Despite the lack of scientific backing, race is still a convenient but significant way to examine important cult ural differences experiences, and therefore their culture. However, the use of race as a marker of ethnicity only hints at th e cial norms (Cauce, Coronado, & Watson, 1998). Thus, measures of ethnic identity or cultural values should be used to provide information above and beyond what race classifications provide. loseness to their ethnic group, however, there is often limited consensus on what values should be included in these scales and values differ among ethnic groups, therefore between group comparison s of cultural values is not possible (Phiney & Ong, 2007). Ethnic identity, on the other a committed sense of belonging to a group, culture, and particular setting (Phiney & Ong, 2007). Thus, this concept can be compared across groups and may be more benefic ial in this regard. No studies to date have specifically examined the relationship between ethnic identity and help seeking or mental health service utilization.
21 Several theories have been established to assist researchers in understanding the utilizati on patterns of individuals. Two major theories guiding the current study are discussed below. Theories and Models The Behavioral Model of Health Services Use was initially developed in the (Andersen & Newman, 1973) Since its conception Andersen has made several revisions to the model with the primary concepts however, remaining constant (Andersen, 1995). This model helps guide the investigati on of the various factors that influence utilization of health services. It is notable that th e model w as initially developed to investigate health service utilization but has since been used extensively in the mental health field to guide understanding of mental health service utilization (Snowden & Yamada, 2005) Furthermore, a lthough this model focuses on service utilization in individual s it is relevant to assessing the individual parental factors related to seeking mental health services for children. The basic framework of the model identifies how environmental factors (health care system and extern al environment), population characteristics (predisposing characteristics, enabling resources, and need), and health behavior (personal health practices and use of health services) influence several outcome variable s (perceived health status, evaluated hea lth status, and consumer satisfaction ; Andersen, 1995 ). The model, however, is far from linear but rather includes feedback loops that show how the various outcome variables influence subsequent population characteristic variables and health behavior ( se e Figure 1)
22 Figure 1 The Behavioral M odel of Health Service s Use (Andersen, 1995) The two factors that influence utilizati on directly according to this model are population characteristics and o utcome s Population characteristics consist of predisposing characteristics, enabling resources, and need. Predisposing characteristics can be described as the factors that predispose an individual to use mental health services. These factors have been grouped into three categories: demograp hic characteristics, social structure, and beliefs. Some examples of demographic characteristics that can be measured include : age, gender, marital status, and past illness. Aspects of social structure include : education, race/ethnicity, occupation, family size, religion, and residential mobility. Finally, aspects of beliefs include : values concerning health and illness, attitudes toward health services, and knowledge about disease. Enabling resources are described as the conditions that allow a family to act on a value or satisfy a need Environment Population Characteristics Health Behavior Outcomes
23 concerning health service use. Enabling resources have been grouped into two categories: personal/ family and community resources. Some ex amples of personal/ family resources that can be measured include : income, health insurance, and other source of third party payment. Community resources include : ratios of health personnel and facilities to population, price of health services, region of country, and urban rural character of the communit y. The final influence on service utilization, according to this model, is the illness/need level. This level can be described as the ess. Illness level is grouped into two categories: perceived and evaluated need levels. Some examples of perceived need include: disability, perceived symptoms, diagnoses, and general state. Examples of evaluated need are: eva luated symptoms and diagnos es. Outcomes consist of perceived health status, evaluated health status, and consumer satisfaction. Perceived and evaluated health status are consistent with the descriptions of perceived and evaluated need levels above, with the former representing an perception of their health status and the latter representing a Consumer satisfaction is also relevant given the knowledge that outcome of services influence future use o f that service. Although this model has been useful in understanding service utilization patterns, more recent theoretical developments appear to better account for both service utilization and help seeking patterns. The second model guiding this study wa s described by Cauce et al. (2002) based on Andersen and Newman model described above The Youth H elp S eeking and S ervice U tilization M odel delineates the stages in the help
24 seeking pathway for youth and describes factors affecting movement through each stage ( Cauce et al., 2002 ). In addition, this model indicates the influence of social networks as a barrier or facilitator of service utilization. Three major stages along the path to help seeking and utilization are descr ibed including: Problem recognition, decision to seek help, and support network and service utilization patterns According to the model movement along the stages is determined by illness profile, predisposing characteristics, and barriers and facilitato rs to care These factors are show n in squares with dashed lines to indicate that they influence the major factors (see Figure 2). The illness profile variables include: clinical assessment of need, perceived (subjective) need, and family characteristics such as parental education, family size, marital conflict, and parental psychopathology. The predisposing characteristics defined as stable factors that influence readiness to seek help include: demographic character istics such as age, gender, and ethnic ity, and sociocultural values and beliefs such as attitudes, knowledge, and acculturation Finally, the barriers and facilitators of help seeking variables include: community and social network influences, economic factors such as income and health insura nce, service characteristics such as access and availability, and policy variables.
25 Figure 2 The Youth Help S eeking and Service Utilization Model (Srebnik et al., 1996; Cauce et al., 2002) Since its inception, the Y outh H elp S eeking and S ervice U tilization M odel has been modified slightly to illustrate how the three major stages (problem recognition, decision to seek help, and service selection and utilization ) are theorized to be interre lated (Cauce et al., 2002). Th e s e update s to the model are illustrated using dashed double headed arrow lines (see Figure 2). In addition, Cauce and colleagues (2002) also argue d that culture and context have a pervasive influence on the entire help seeking model and as such cannot be pinpointed to any specific s tage in the help seeking process. Furthermore, since culture and context are so pervasive Cauce and colleagues argue d that minority and non minority help seeking pathways should be studied separately. The current study use d these theoretical models to g uide the investigation of parental
26 influences o n youth utilization of mental health services. The study of the entire Y outh H elp S eeking and S ervice U tilization M odel is beyond the scope of this study. Therefore, the first and arguably, most important, s tep of the help seeking process w as investigated (i.e. problem recognition) In addition, factors that tie d this first step (problem recognition) to the next step (decision to seek help) w ere also examined in the current study Problem Recognition The rec ognition of a problem is the necessary first step before help seeking can occur (Cauce et al., 2002) Thus, problem recogn it ion is likely one of the most important step s in the help seeking and service utilization process. Once services have been utilized, other factors may influence continued use of services. However, before services are utilized the second, third, fourth or more time, individuals must recognize the re turn of the previous problem or the emergence of a new problem, thus problem recognition must occur again. Consequently each time help is sought by an individual or family, that individual, family, or a third entity ( such as teachers or the court system) must recognize a problem. As reviewed by Sayal (2006), fewer than half of the parents with a child with a disorder recognize a problem in their child. This finding suggests a significant barrier in the pathway to care and eventual service utilization (S ayal, 2006). Problem recognition indicates a need for care which can be defined in two ways: epidemiologically defined need and perceived need (Cauce et al., 2002). Epidemiologically defined need involves detailed clinical assessment by a professional, typically using measures based on the Diagnostic and Statistical Manual of Mental Disorders (DSM) or International Classification of Diseases (ICD) Perceived need,
27 however, involves the subjective perception of distress by the individual or caregiver. This subjective perception of need may be a stronger predictor of problem recognition than objective need due to the famil s own understanding of problematic behaviors (Srebnik et al., 1996) Thus, having a better understanding of the factors influencing parents According to the Y outh H elp S eeking and S ervice U tilization M odel problem recognition is influenced by several factors: clinical assessment of need, perceived need, and family characteristics. Clinical a ssessment of n eed Factors influencing the clinical assessment of need include symptoms / behaviors, diagnoses, and functional impairment (Srebnik et al., 1996). In their review, Srebnik and colleagues reported that symptom severity often distinguished between those yo uth who utilize formal mental health services (due to greater severity of problems) versus those who use informal sources of help (due to lower severity) Given that problem severity can be assessed objectively (by the professional) and assessed subjectiv ely (by the family) i t is possible that problem severity is only related to utilization of services if parents/caregivers also perceive the problem as severe. Functional impairment is another factor that can be assessed objectively and subjectively. Objectively, functional impairment is an important aspect of assessing specific mental health problems, in addition to symptoms and behaviors, using the diagnostic criteria of the DSM IV All mental health disorders in the D SM IV require that the symptoms experienced by the individual cause clinically significant distress and/or impairment in several areas of functioning including social, occupational, or other important areas of functioning (American Psychiatric Association, 2000). In children and adolescents, functional impairme nt typically includes problems with academic
28 performance, and relationships with peers and adults at home, school, and in the co mmunity (Powers et al., 2005). In general, clinical assessment of need involves a contrast to perceived need discussed below. Perceived (s ubjective) n eed This concept well being, symptoms, behaviors, and impairment and their definition of a problem or set of problems as mental health concern s As reviewed above subjective perception of need may be a stronger predictor of problem recognition than objective need (Srebni k et al., 1996). Thus subjective/perceived need may be one of the most important factor s in understanding parental problem recognition. Power and colleagues hypothesize that problem recognition by the family may serve as a mediator between problem sever ity as assessed by the professional and service utilization (Power et al., 2005). This relationship has not yet been investigated. Factors influencing perceived need include: symptoms, behaviors, functional impairment, caregiver burden, and problem thres hold. The first three factors ( i.e. symptoms, behaviors, and functional impairment ) are the same factors described above, except they are now being interpreted by the fa mily rather than professionals. In a review Broadhurst (2003) reported significant d ivergence between lay and professional definitions and rating of severity of problem behaviors Furthermore, an in vestigation of parental problem recognition revealed that parents were more likely to recognize that their child had a problem when they desc ribed more child symptoms or a greater impact on their family (Teagle, 2002) A study of Latino families also showed that perceived impairment and having disruptive behavior problems were the strongest predictors of service
29 utilization (Alegria et al., 2004). I n a review of studies published from 1992 2001, researcher s found that while child psychopathology was not sufficient to predict problem recognition, help seeking was enhance d with increased severity, comorbidity, and p ersistence of p roblems over time (Zwaanswijk, Verhaak, Bensing, Ende, & Verhulst, 2003). In addition, while presence of school and medical problems increased help seeking for child psychopathology, presence of school problems did not increase parental pro blem recognition (Zwaanswijk et al., 2003). T he researcher s also reported that th e effect s of type of problem (externalizing versus internalizing) on help seeking patterns w ere still inconclusive. However, Zimmerman (2005) reported significant gender and race differences in service utilization based on problem type in 7 14 year olds with symptom severity controlled They found that girls were less likely to receive treatment overall. More specifically they reported that girls were much less likely (1:5 ) to obtain treatment for externalizing problems than boys G irls were also somewhat less likely (1:2) to obtain treatment for depression than boys Problem type also influenced race differences in utilization. Specifically, Zimmerman reported that Afri can American children were much less likely to obtain treatment for depression but no less likely to obtain treatment for behavior disorders. Interestingly, in the entire sample, they also found that the presence of the father in the home reduced the like lihood that children would obtain treatment, especially for depression (Zimmerman, 2005) Caregiver burden is another factor that influence s parental perce ption of need. Angold and colleagues (1998) found that the strongest predictor of specialty mental health service use was perceived parental burden, followed by child symptom scores. Similar results were found by Teagle (2002) who reported that the i
30 mental health problems on the family wa s the strongest predictor of parent perception problem (i.e. problem recognition) These results were also confirmed by Zwaanswijk and colleagues (Zwaanswijk et al., 2003). In a separate study of elementary school aged children at high risk for ADHD, h igher levels of pare ntal/caregiver strain predicted formal service utilization (Bussing et al., 2003). It is notable, however, that caregiver strain /burden was specific to the impact that child ren ha d on the family /caregiver It is likely that high e r levels of overall stress may actually have a negative influence on problem recognition, as parents under significant stress may Accordingly, it is likely that related to increased utilization of formal services while overall caregiver stress may be of services. Conseque parental stressors is important. Additional stressors impacting parental problem recognition will be discussed in the next section. The fin al factor documented as influencing parental perception of need is problem threshold. This concept can be described as the classification of a set of behavior s as a mental health concern by the c aregiver/ family (Srebnik et al., 1996). This factor is impo for their children from a mental health professional versus other professionals (such as a physician if the problem is perceived as a health concern) or laypersons (such as friends for general parenting advice). Th u s th e
31 i dentification of behavior s or emotional problems as mental health difficultie s will not only influence the decision to seek help but also influences actual service utilization. Problem threshold has been show n to vary across various cultural grou ps (Weisz & Eastman, 1995) Specifically, Lambert and colleagues (1992) found that Jamaican and American adults had varying thresholds as to what problems they considered abnormal More specifically, Jamaican teachers unusual than American parents and teachers (Lambert et al., 1992). More recently, a study by Roberts and colleagues found differences in European African and Latino American parents ents, despite similar prevalence rates of diagnosis within the past year (Ro berts et al., 2005) Specifically, Roberts and colleagues found that European American parents were more likely to report that their adolescent had a mental health problem in the past year and rate d worse than both Latino and African American parents. Problem threshold has also been s hown to vary across g ender. In preliminary research Singh (2003) reported that fathers and mothers differed in their categorization of Attention Deficit/Hyperactivity Disorder (ADHD) symptoms as problematic, whereby fathers were less likely to classify symptoms of inattention and hyperactivity as needing medical attention. In a study of adolescent behavior, resear chers indicated that fathers reported significantly fewer problem behaviors than mothers reported (Seiffge Krenke & Kollmar, 1998). Furthermore, the authors found that significantly recognition have been described as family characteristics.
32 Family Characteristics According to the Y outh H elp S eeking M odel (Cauce et al., 2002) the factors influencing problem recognition in this area consist of structural and relational characteristics. These variables include: family size, parental education, family/marital conflict, warmth/cohesion, disorganization, abus e/neglect, parental psychopathology, and criminality. Zwaanswijk and colleagues found that education level and family stress influenced help seeking while parental psychopathology increased problem recognition but not utilization (Zwaanswijk et al., 2003) With regard to family size, presence of siblings reduced parental recognition of child symptoms as problematic but did not decrease the likelihood of help seeking. Finally, examination of type of abuse indicated that physical and sexual abuse resulted in increased service utilization while neglect was related to decrease d help seeking. The relational factors (including family size, family/marital conflict, warmth/cohesion, and disorganization) are important because they may influence the extent to whic identifiable by the care givers (Srebnik et al., 1996). The other factors (abuse/neglect, parental psychopathology, criminality, and low parent education) are important because they may influence the extent to which c problematic (Srebnik et al., 1996). Decision to Seek Help When a parent has recognized a problematic behavior, the next step involves deciding what to do. This decision usually falls in to a dichotomous decisi on : deciding to seek help or deciding not to seek help. Furthermore, the decision to seek help can often range from formal sources (psychologist, psychiatrist, etc) to informal sources (self help books, friends and family, etc) Several factors that may influence the help seeking
33 decision process include: knowledge of the problem, health locus of control, self efficacy, and acculturation (Power et al., 2005). The current study examine d one of these factors (knowledge) which is hypothesized to have a dire ct impact on problem recognition. As previously discussed, knowledge of a problem has been hypothesized to ed the influence of knowledge of a behavior problem on parental recog nition of symptoms. This study also examine d various factors related to problem recognition that may impact Knowledge of Problem Preliminary evidence has shown that parental knowledge of specific problematic behaviors has an influence on service utilization and help seeking patterns in youth (Power et al., 2005). Based on the model proposed, understanding a particular problem behavior and having knowledge of a how to treat the problem may have an ( Power et al., 2005). However, few studies have examined this concept. Research by Bu ssing and colleagues found ethnic differences in knowledge of ADHD among African American and Caucasian pa rents (mostly mothers). They found that fewer African American parents had heard of ADHD, knew some or a lot about ADHD, and were more parent s (Bussing, Schoenberg, & Perwien, 199 8). Although knowledge of a problem has been (Power et al., 2005), it is likely that knowledge of a problem is initially related to problem recognition as well. Th u s knowledge of a problem w a s investigated further in th e current study.
34 The Current Study The current study aim ed to investigate the factors related to youth problem recognition in African American and Caucasian parents through the use of vignettes Vignettes have been used in va rious community based studies to determine parent (usually mothers ) help seeking patterns and intentions to utilize mental health services in the future ( Raviv et al., 200 3; Shah et al., 2004 ). Vignettes can be used to control variables in such a way t hat participants are exposed to the same stimuli (rather than reporting on their reactions to their own children with varying degrees of behavior problems). Previous r esearch indicates decreased help seeking patterns and mental health service utilization by ethnic minorities. Some studies show th at th is difference in race and ethnic groups disappear s after socioeconomic status has been accounted for while other s tudies do not. In addition, although several studies have hypothesized differences in help se eking and ut ilization, very few studies attempt to explain these differences empirically This study also aim ed to identify factors affecting the decision to seek formal versus informal help when a mental health problem has been identified. The f act ors that w ere examined in this study are illustrated in the model below and are consistent with the first portion of the Youth Help Seeking & Service Utilization Model (Cauce et al., 2002)
35 Figure 3 Proposed research model The specific aims and hypotheses of this study are as follows: 1) To investigate the relationship between parental problem recognition and type of problem ( anxiety internalizing attention deficit/hyperactivity disorder externalizing, and control). I t is hypothesized that parents will be more likely to recognize the externalizing and internalizing vignettes as problems than the control vignette. 2 ) To predict the relationship between parental problem recognition and willingness to seek help. Based on previous research (Cauce et al., 2002), it is hypothesized that parents who recognize a mental health problem will be willing to seek help, irrespective of the type of problem. 3) To explain the effect of youth gender on parental problem recognition and willingness to seek help. Based on previous research (Zimmerman, 2005; Zwaanswijk
36 et al., 2003), it is hypothesized that parents will be more likely to recognize problems and more willing to seek help for boys than for girls (i.e. for male vignettes versus female vignettes ) 4) To explain the effect s of perceived need (severity, problem threshold, functional impairment, and family impact) on problem recognition and willingness to seek help. Based on previous research (Alegria et al., 2004; Power et al., 2005; Teagle, 2002), it is hypothesized that parents who endorse higher levels of perceived severity, problem threshold, functional impairment, and family impact would be more likely to recog nize problems and more willing to seek help 5) To explain the effects of family characteristics (family size, parent education, and parent psychopathology) on parental recognition of youth symptoms and willingness to seek help. Based on previous resea rch (Srebnik et al., 1996; Zwaanswijk et al., 2003), it is hypothesized that parents who report smaller family size, higher levels of education, and current psychopathology would be more likely to recognize problem s and more willing to seek help. 6 ) To d etermine which demographic variables (parent race, ethnic identity, gender, and SES) are most predictive of parental recognition and willingness to seek help Based on previous research ( Dobalian & Rivers, 2008; Mahalik et al., 2003; Pumariega et al., 199 8; Roberts et al., 2005), it is hypothesized that African Americans, fathers, and parents of lower SES would be less likely to recognize problems and less willing to seek help Although this has not been previously investigated, it is hypothesized that Af rican Americans with higher ratings of ethnic identity and Caucasians with lower ratings of ethnic identity will be less likely to recognize problems and less willing to seek help.
3 7 This hypothesis is based on previous research indicating that African Amer icans report higher levels of stigma and decreased levels of voluntary mental health service utilization than Caucasians ( U.S. DHHS, 2001). Thus, if findings are indeed due to ethnicity, it is likely that those African Americans who show higher levels of ethnic identity and Caucasians with lower levels of ethnic identity may be less likely to recognize problems, and less willing to seek help. 7) To determine which sociocultural beliefs (biopsychosocial, sociological, and spiritual/nature disharmony) are m ost predictive of parental problem recognition and willingness to seek help. Based on previous research (Yeh et al., 2005), it is hypothesize d that parents who endorse biopsychosocial beliefs ( p hysical c auses, p ersonality, r elational i ssues, f amilial i ssu es, and t rauma) would be more likely to recognize problems in youth and more willing to seek formal help. On the other hand parents who endorse sociological beliefs (friends, American culture, prejudice, and economic problems) and spiritual/nature disharmony beliefs (spiritual causes, and nature disharmony) would be less likely to recognize problems and more willing to seek informal help. 8) To determine the influence of previous experience on parental re cognition of youth problems Based on prev ious research (Power et al., 2005), it is hypothesized that parents who have had previous experience with anxiety and ADHD will be more likely to recognize the anxiety and ADHD vignettes as problematic. 9) To determine the influence of parental race and gender on type of beliefs endorsed, strength of beliefs endorsed, and previous experience with mental health disorders. Based on pr ior research (Yeh et al., 2005), race and gender differences in
38 type and strength of beliefs endorsed are hypothesized whe re, fathers and African Americans will endorse more sociological and spiritual/nature disharmony beliefs and less biopsychosocial beliefs than mothers and Caucasians. Furthermore, based on previous research (Bussing et al., 1998), it is hypothesized that race and gender differences will be found in previous experience; where fathers and African Americans w ould have less experience with mental health disorders than mothers and Caucasians. 10) To investigate the influence of parental race and gender on perce ption of barriers (financial, structural, and attitudinal). Based on previous research (Sareen et al., 2007), it is hypothesized that parents will endorse more attitudinal barriers than structural and financial barriers. Also based on previous research ( Thurston & Phares 200 8 ), fathers are hypothesized to endorse more attitudinal barriers than mothers and African Americans will endorse more structural barriers than Caucasians. 11) To investigate the relationship between various barriers and parental he lp seeking from formal sources, informal sources and parental ethnic identity Based on previous research (Mackenzie et al., 2006), it is hypothesized that parents who perceive more barriers overall, will be less willing to seek formal help irrespective of problem type 12) To investigate the influence of ethnic identity, barriers and beliefs on intentions to utilize mental health services in the future. Based on previous research ( Thurston & Phares, 200 8 ; Yeh et al., 2005), it is hypothesized that e ndorse ment of fewer barriers stronger biopsychosocial beliefs weaker sociological beliefs, and weaker spiritual/nature disharmony beliefs will predict higher intentions to utilize mental health services in the future.
39 In order to answer these research questions appropriately the vignettes that were designed to be used in this study were piloted with psychologists and psychology doctoral students The purpose of this pilot study was to ensure that the vignettes, which were created by the researcher, ac tually described the types of behaviors that were intended (i.e. validity of the vignettes). This pilot study focused on clinically assessed need, meaning that the vignettes were examined to determine if they met criteria for various types of disorders as assessed by clinicians. The actual study, however, focus ed on subjective need Thus, vignettes that were clinically assessed as problematic w ere examined by parents, to determine if parents subjectively perceive the vignettes as problematic. Below is a description of the pilot study. Pilot Study Participants p ilot s tudy A total of 16 participants ( 7 psychologists and 9 psychology doctoral students ) completed the pilot study. All psychologists were licensed professionals who work with children, adole scents, and their families and practice in academic or private settings. The doct oral students had a minimum of two years completed in their psychology program and over one year of clinical experience in working with youth and their families. Measures p ilot s tudy Vignettes were created based on criteria from the DSM IV (American Psychiatric Association, 2000) and the ICD 10 (World Health Organization, 1993 ) The names of children used in the vignettes were selected from the most popular name s for boys (Michael Chris, Joshua ) and girls (Ashley Brittany, and Jessica ) in Florida from 1990 1999 ( Popular baby names, 200 7 ) Five vignettes were created initially that describ ed one internalizing (anxiety) one externalizing (ADHD)
40 one pediatric he alth disorder (Juvenile Rheumatoid Arthiritis) one comorbid with health and internalizing disorder s (Diabetes and Depression ), and one control with no diagnosis After each vignette, three follow up questions were asked about ratings of severity of problems abnormality of problems, and categorization of the vignettes into one of five groups (internalizing, externalizing, pediatric, comorbid, and no diagnosis). The severity and abnormality questions were rated on a likert scale ranging from 1 5, wit h 1 being not at all serious/abnormal to 5 being extremely serious/abnormal. Procedures p ilot s tudy. A total of 23 psychologists and psychology graduate students who work with youth and their families were identified and asked to participate in this stu dy. A total of 16 of these individuals completed the pilot study, resulting in a participation rate of 70%. Participants were asked to read each vignette and answer the questions immediately following the vignette Initial pilot study testing revealed t hat including all five vignettes and the study questionnaires would increase overall participant involvement time by about 10 minutes, thus two of the five vignettes were dropped from the actual study. These deleted vignettes included the comorbid and the pediatric health vignettes. Although it is important to determine how parents utilize services for mental health and physical health the focus of this study i s on service utilization and help seeking for mental health problems. Thus dropping these two v ignettes did not affect the core purpose of this study. Results p ilot s tudy All participants correctly matched the three vignettes to the appropriate categories (i.e. the externalizing vignette (ADHD) was correctly identified as externalizing, etc). A paired samples T test indicated that the severity ratings of the anxiety vignette ( M = 3.25, SD = 0.45) w as not significantly different from
41 severity ratings of the ADHD vignette ( M = 3.12, SD = 0.34), t (15) = 1.00, p = .33. Another paired samples T test indicated that the abnormality rating of the anxiety vignette ( M = 3.06, SD = 0.68) w as significantly higher, albeit slightly, than the abnormality rating of the ADHD vignettes ( M = 2.81, SD = 0.54), t (15) = 2.24, p = .04. However, since these findings were marginal and actual study participants will be asked to rate the severity and not the abnormality of the vignettes ; both vignettes were retained for use in the actual study. As intended t he severity ( M = 1.31, SD = 0.48) and abnormality ( M = 1.19, SD = 0.40) ratings of the control vignette were significantly different from the severity and abnormality ratings of both the anxiety and ADHD vignettes. All analyses were significant at the .000 level. These analyses in dicate that the anxiety and ADHD vignettes meet criteria for clinically assessed need as illustrated by of greater severity and abnormality than on the control vignette. Thus, the control vignette is indeed a control, as it was rated a s not meeting criteria for any diagnosis. These vignettes w ere perception of need.
42 Method Participants A total of 2 87 3 parents with at least one child between the ages of 2 and 21 years old were recruited to participate in this study. A total of 46% of the participants were male and 54% were female. The sample comprised of parents of various racia l groups with oversampling for A frican American and Caucasian parents including 0.3% Asian, 4 3 9 % Black/ African American, 51.6% Caucasian, 0.3% Native Hawaiian/Pacific Islander, an d 3 8 % Biracial/multiracial Additionally, 9.1% of the sample described their ethnicity as Hispanic. No pa rents were denied participation in the study, however, only the data from the African American and Caucasian participants were used in the current analyses This procedure was because the focus of th e study was a more concise comparison of the unique facto rs impacting these minority and majority groups Thus the final sample used in the following analyses was 251 parents; 49 % African American and 51 % Caucasian. Par ents ranged in age from 20 to 66 ye ars of age, with a mean age of 40 years (SD = 8.1 5 ). The sample consisted of approximately equal numbers of fathers (49 %) and mothers (51 %), and the majority of the parents ( 7 3 3 % ) were married, 11.2 % were single with /without a partner, 11.2% were divorced, 3 2 % were separated and 1.2% were 3 There were 289 questionnaires returned, but two participants were left out of all the analyses because one participant on did not meet study criteria.
43 widowed The mean education of fathers was 16. 36 years (SD = 2 .2 0 ) and mothers had a mean education of 15.4 7 years (SD = 2.26 ). A total of 8 .8% of parents were receiving some kind of public assistance B ased on Hollingshead criteria for socioeconomic status (Hollingshead, 1975), the social strata for the average participating parent represented medium businesses, minor professionals, and te chnical jobs (M = 47 76 ; SD = 10. 18 ). Additionally, the average household income for the sam ple ranged from $50,001 $65,000 a year. Pa rents had an average of 2. 48 children (SD = 1.31 ). A total of 52% of their children were female and 48% were male. In addition, 45.8% of the children were African American, 45.8% Caucasia n, and 8.4% biracial/multiracial with 1.8% describ ing their children ethnicity as Hispanic. Th e children described in the current sample ranged in age including, 8.5% infants/toddlers (0 2 years old), 14.3% preschoolers (3 5 years old), 35.8% school age (6 12 years old), 16.3% adolescents (13 17 years old), and 25.1% adult children (18 years and above). Note that all parents had at least one child between the age of 2 and 21. A majority of parents had physical health care insurance for themselves (9 0 8%) and their children (94.8 %). When asked about mental he alth care insurance there was m or e variability in responses. A total of 72 .1% of parents had insurance for themselves, 14 7 % did not, and 1 1 6 % did not know if their insurance covered me ntal health needs. Similarly, 75 .3% of parents had mental health insuranc e c overage for their children, 9 6% did not, and 12 7 % did not know if their insurance covered mental health care needs for their children.
44 Based on a power analysis with power set at .80, alpha set at .05, and expecting a medium effect size, 30 participant s per group were required to t est the hypotheses adequately via regressions and analyses of variance ( Cohen, 2003 ) Since the focus of this study was to examine similarities and differences between African American and Caucasian mothers and fathers who re ad male and female vignettes (i.e. 2 X 2 X 2 model) a minimum of 240 parents ( 60 per rac e /gender group) was necessary to test for a medium effect size. Thus, the final sample was sufficient to test the hypotheses adequately. The demographic information for parents by gender and race is presented below (see Table 1). Similarities and differences in these groups are discussed further in the Results section. Table1 Demographic Variables for Parents by Race and Gender Variable Statistic Black White Black White Fathers Fathers Mothers Mothers (n = 6 0) (n = 63) (n = 63 ) (n = 65 ) Age Mean 3 9 36 42 05 37 84 40 7 3 SD 9 89 6.69 8.21 7. 13 Soci oeconomic Status Mean 49.03 51 22 4 4 92 4 6 02 SD 9 38 9.1 9 10 .9 6 10 .0 8 Year s of Education Mean 16 03 1 6 67 1 5 .6 9 15.26 SD 2.34 2 02 2.22 2.29 Number of Children Mean 2 .88 2.27 2. 52 2.26 SD 1.53 0.97 1.58 0.97 Note. SD = Standard deviation Procedures Parents were recruited to participate in the study through the use of flyers, advertisements, and direct invitation (snow balling and direct approach). Flyers were
45 clinics, grocery stores, community centers, libraries, an d housing/apartment complexes. As recruiting minorities in research is often a difficult task and recruiting low SES minorities even more taxing (Sadler et al., 2005) extra effort was made to recruit a wide variety of individuals from varying SES ranges by placing flyers in well off, moderate, and impoverished areas. Similarly advertisements were placed on several internet listserves that targeted both African American and Caucasian parents ( FAMU alumni listserv e 100 black men of Tampa bay, and USF Alu mni listserv e ) A total of 66% of participants were recruited through direct invitation, 13% through the Boys and Girls Clubs, 9% through community organizations, 6% through online ads, 3% had participated in previous research & agreed to be conducted for future studies, and 3% saw the flyer for this study and expressed interest. In addition to flyers and advertisements, two direct approach methods were used to recruit parents for this study. The snow ball method was used, where parents who participated i n the study and non parents who were approached to participate in the study were asked to invite other parents to participate. Leaders of various organizations ( fraternity and sorority groups, churches, community organizations and parent groups) were als o approached and the parents at these organizations were invited by the researcher or the leader of the organization to participate in the study. Additionally, parents whose children attended after school programs at several Boys and Girls Clubs in the lo cal area were approached and invited to participate. Effort was made to recruit all parents ( biological, step, and/or adoptive parents ) who had a t least one child between the ag es of 2 21 and at least monthly face to face
46 contact with their child within this age range These criteria were selected to ensure that chosen because empirically supported treatments such as Behavioral Parent Training usually begin at age two (Barkley, 1997). Parents were not required to be dyads. This decision was important in order to ensure that proportion s of African American families would not be excluded, given that 52.5% of African American children are being raised by single mothers (Hofferth, Stueve, Pleck, Bianchi, & Sayer, 2002). Given that this study was based on a community s ample and because participants we re only asked to complete questionnaires, data collection took place largely through the mail and drop off boxes a t community sites. Although no order effects we re expected, questionnaires and vignettes w ere presented to participants in randomized order so that possible biases due to ord er effects w e re minimized. Those parents who call ed or sent an inquiry via email had the requirements of the study reiterated to them and i nterested participants were asked to provide their mailing address. The questionnaires along with a cover letter explaining the study (Appendix A), a business reply envelope and business reply p ostcard (for the drawing) w ere then mailed to them. Interested participants who wer e approached directly ha d the option of completing the questionnaires immediately dropping off the questionnaires in the boxes at various sites, or mailing the completed qu estionnaires back in the reply envelope provided If participants chose to complete the questionnaires immediately, their post cards w ere immediately separated from their survey responses to protect their privacy. Q uestion naires took approximately 20 30 m inutes to complete.
47 A total of 806 surveys were distributed and 2 89 were returned. Thus, there was a 3 6 % participation rate. This number is consistent with other survey research in the community with adults (Kropf & Blair, 2005; Shumway, Unick, McConnell Catalano, & Forster, 2004). After the participa nt target number was reached (30 parents from each vignette gender, parent gender and racial group), t he data were entered into SPSS for analysis A ll participants who provided a mailing address via the index cards w ere entered into a drawing for one of six prizes including : one $100 monetary prize, two $50 monetary prizes and three $25 gift certificates from merchants in the surrounding community. Prizes were mailed out to winners via standard mail. Up on completion of the study (when the results have been defended) a synopsis of the results and a referral list of mental health facilities (Appendix B) will be sent to all participants who provided physical mail or email addresses in case they wish to see k mental health treatment for themselves or for their children in the future. Measures Vignettes. Please r efer to pilot study section above for a description of the creation and piloting of the vignettes. T hre e vignettes were us e d in the final study. Th ese vignettes include d an internalizing (anxiety) an externalizing (ADHD) and no diagnosis (i.e. control) vignette. The three vign ettes w ere i dentical for all study participants, with the exception of the which w as randomized among all the pare nts by group. Thus, approximately equal numbers of black fathers, white fathers, black mothers, and white mothers complete d the three vignettes with either all boy or all girl descriptions (for example: 30 black fathers compl ete d all boy vignettes and another 30 black fathers complete d all girl vignettes) This procedure was used in order to allow
48 for ease of data collection and to ensure that parents w ere not motive to examine gender differences. F ollow up questions w ere included after each vignette. P articipants were asked to rate the severity of problems, concern about problems, impact of problems on the hypothetical problems on the hypothetical ily. Participants rate d their responses on a likert scale ranging from 1 5, with 1 being not at all to 5 being extreme. Participants we re then ask ed to determine if the child had a mental health problem, if they would seek help for the child, and the lik elihood of seeking help from several different sources. Appendix C shows the finaliz ed boy anxiety vignette with follow up questions The additional vignettes ( ADHD and no diagnosis) are shown in Appendix D Demographics. The demographics measure include d basic background questions about gender, age, marital status, annual income, zip code, insurance, public services received, education (for self and spouse/partner) occupation (for self and spouse/partner) age gender and race/eth nicity of children, and number of individuals in the ir household (Appendix E ). The mothers and fathers who participate d in this study we re not necessarily dyads; therefore demographic information on both the paren t and his/her spouse/partner wer e collecte d from the parent who complete d the questionnaires. Socioeconomic s tatus (SES) w as derived based on household gender, marital status, education, and occupation al information obtained from this measure based on the Hollingshead four factor index of social status ( Hollingshead, 1975) Utilization Participants complete d a measure about the ir history of mental health service utilization for themselves and their children This measure is based on Burns et al., (1995). Parents w ere asked if either they or their children ha d ever seen or been
49 referred to see a mental health professional for emotional, behavioral, alcohol, or drug related problems. Participants w ere a lso asked about satisfaction with previous services, current need for psychological care, a nd their likelihood of seeking care for themselves or their children. In addition, parents w ere service utilization from specific sources including health, mental health, education, religious, child welfare, and ju venile justice sectors Finally, parents w ere asked if they had any personal or professional experience with five psychological disorders : ADHD, anxiety disorder, depression, ODD, or alcohol/drug abuse (Appendix F ). Multigroup Ethnic Identity Measure Revised (MEIM R ; Phinney & Ong, 2007 ) Th e original version of this measure was developed by Phinney (1992) As reviewed by Avery and colleagues (2007) th e original measure was valid, internally consistent (with alphas ranging from .71 to .92) and show ed evidence of equivalence across racial and ethnic groups A revised version of the MEIM was recently developed (Phinney & Ong, 2007). This new version more accurately measure s ethnic identity, (2007). The previous version not only included items measuring ethnic identity but also assessed for participation in ethnic activities and orientation to other ethnic groups. However, as Phinney and Ong described, individuals may have a strong sense of belonging to their ethnic group but choose not to participate in the Thus, i n this version some of the previous items were dropped other items were reworded two factors (exploration and commitment) were identified with subscales measuring each one and based on factor analysis a 6 item scale was finalized with good reliability of .76 for the explor ation subscale, .78 for the commitment
50 subscale and .81 for the combined 6 item scale. The exploration subscale (items 1, 4, and 5) ethnic group while the commitment subscale (items 2, 3, and 6 ) and sense of commitment to their ethnic group (Phinney & Ong, 2007) T he measure beg an with an open label and conclude d with a list of race and ethnic gro ups for the participants to identify their own (Appendix G ) Participants respond ed to questions on a five po int likert scale ranging from strongly disagree (1) to strongly agree (5) with higher scores indicating a s tronger committed sense of belonging to their ethnic group The mean of the scores for the full measure and the individual subscales w ere used in the final analyses. This measure was used with permission from the author. In the current sample, internal very s Barriers to Utilization This measure inquire d about possible barriers that pare nts may experience when seeking or deciding to seek mental health care for their children. Th e measure wa s adapted from the National Institute of Mental Health (NIMH) M ethods for the Epidemiology of Child and Adolescent Mental Disorders (M ECA ) study, where it was used in an interview format to assess potential access barriers to mental health services (Flisher et al., 1997) Although specific data on reliability is not available, the MECA study was an extensive study of youth mental health disorders with over 128 5 parent/youth dyads at four sites in the USA and Puerto Rico The measure was modified to reflect a self report format and was used in this study with permission from the authors P articipants respond ed to questions on a likert scale ranging from strong ly
51 disagree (1) to strongly agree (4) where higher scores indicate d stronger perception of barriers to service utilization (Appendix H ). Items 1, 3, 5, 8, 11, 15, and 18 are reverse scored items. The barriers were separated into three groups: financial barriers (items 1 and 4), attitudinal barriers (items 2, 3, 5, 6, 7, 9, 10, 11, 12, and 13), and structural barriers (items 8, 14, 15, 17, and 18). One additional question (item 16) assess ed if services have been used in the past and did not help. In the current sample, internal for the overall barriers measures .73). Internal consistencies for the subscales were varied; significantly low for the However, given that subscale analyses were proposed, mean scores for the full measure and subscale s were use d in the analyses but interpreted with caution B eliefs About Causes Revised (BAC R ; Yeh & Hough, 2005 ) This questionnaire examined via a semi structured interview format The measure was crea ted based on previous research, literature reviews, and consultation with cultural experts (Yeh et al., 2005). A revised version of the original measure was recently developed and was modified for use in the current study with permission from the author The BAC R is comprised of 11 scales which fall into three broad categories as were determined by confirmatory factor analysis. They include : Biopsychosocial ( 5 scales: Physical Causes, Personality, Relational Issues, Familial Issues, and Trauma), S ociol ogical ( 4 scales: Friends, American Culture, Prejudice, and Economic Problems), a nd S piritual/nature disharmony ( 2 scales: Spiritual Causes, and Nature Disharmony). Each of the 11 scales is comprised
52 of several individual items which are used to assess pa categories ( Appendix I ) On the BAC R, e ndorsement of an individual item on one of the 11 scales indicates endorsement of that entire scale and as such belief in that specific cause of mental health problems. For example, if a parent endorsed d Immediately following the 11 scales (A K) wa s an open ended item which asked participants to identify the most important beli ef from all those endorsed. In addition to converting the measure to a self report format, an additional item was added to the BAC R measure. Participants were now asked to rate how strongly they believe in each of the 11 categories on a likert scale fro m not at all (1) to a great deal (5). Since the BAC R was a fairly new unpublished measure, reliability data we re not yet available on this version However, the original measure (which varies only slightly from the updated measure) showed excellent re p roducibility on seven scales (kappas > 85%), good reproducibility on two scales, and two scales (personality and friends) had marginal reproducibility ( Yeh et al., 2005 ) Construct validity was also supported (2005). As discussed above endorsement of an item indicate d endorsement of that scale. Scale endorsement w as w ere used to compare beliefs across ethnicities and gender. In the current sample, Data Analyses Most of the analyses in this study w ere conducted using a series of logistic regression analyses. Logistic regression analysis has several advantages over linear regression analy s is Specifically, logistic regressions do not make assumptions about the
53 distribution of the outcome variables as is the case in linear regressions (Pedhazur, 1997). Furthermore, the assumption of homoscedasticity is invalid and errors are not normally di stributed (1997). Logistic regressions are therefore especially desirable for use when the dependent variable is dichotomous (such as with the problem recognition variable) as undesirable results will be found if linear regressions are used. Logistic re gression analysis allows for the predict ion or explanation of a discrete outcome from dichotomous, discrete, continuous, and mixed independent variables. Given that this research paper is focused on hypothesis testing, the forced entry method was used fo r all regression analyses (Field, 2009). Other analyses used in this study were Chi square tests repeated measures logistic regression, Mann Whitney test s paired samples t tests, correlations, and multiple regression analysis Bon ferroni c orrected p values w ere used for each research question based on the number of analyses being run per question. SES was calculated based on the four factor index of socioeconomic status (Hollingshead, 1975), which uses gender, marital status, educa tion, and occupation to calculate the SES of study participants. In order to assess race differences, SES was controlled statistically so as not to confound race and SES. This procedure was achieved by covarying out the influence of SES on the participan this procedure is sometimes seen as a methodological limitation, as it arbitrarily creates a are confounded in reality Thus, although the data are reported with SES controlled, t hey were also analyzed without SES controlled to ensure that the data w ere examined thoroughly.
54 Results Descriptives As previously d iscussed a total of 251 Africa n American and Caucasian mothers and fathers particip ated in this study. Of these parents, 41% had used mental health services for themselves in the past These parents who had used mental health services ranged in age from 2 0 6 4 years ( M = 40 39 SD = 8.25), 52.4% were female and 47.6% were male, and 65 % were Caucasian and 35 % were African American. Based on a scale of 1 5, t he average satisfaction rating of mental health services received was 3.19 ( SD = 1.23) indicating that on average, most parents were at least moderately satisfied with the services th ey had received. With respect to future utilization, of the 28 parents ( 11.2 % of the total sample) who self disclosed having a current mental health problem, 50 % were quite or extremely likely to seek treatment from a mental health professional. that a total of 30.3% of their children had used mental health services in the past. Based on a scale of 1 5, t he average satisfaction rating of mental health services their children had received was 3.05 ( SD = 1.26), con satisfaction with their own services A dditionally, 59.2% of the parents of children who had utilized mental health servi ces had also used services themselves in the past. Most of the children had utilized mental health services from the mental health sector (82.9%). However, some of them had also received mental health treatment from the education sector (51.3%), health
55 sector (35.5%), religious sector (21.1%), Juvenile Justice (10.5% ), and Child Welfare ( 9.2%). These percentages were not mutually exclusive as the majority of these children had utilized services from more than one s ector; in fact only 3 8.2% had received mental health treatment from only 1 sector. Wi th respect to futu re mental health utilization for their children 45 parents in the current sample ( 17.9% of the total participants ) reported that at least one of their children had a current mental health problem. Of these parents, 68.9 % were quite or extremely likely to seek treatment for this child from a mental health professional. Parents indicated their previous personal and/ or professional experience with some pre identified me ntal health problems including Attention Deficit/Hyperactivity Disorder (ADHD), Anxiety, D epression, Oppositional Defiant Disorder (ODD), and Alcohol/Drug Abuse. Within the entire sample, 47% of parents had professional or personal experience with ADHD, 47% with depression, 38% with anxiety, 37% with Alcohol/drug abuse, and 22% with ODD. These percentage s varied by parental race and gender (see Figure 4 ) Parents scores on the Multigroup Ethnic Identity Measure Revised (MEIM R) have a potential range from 6 30 with higher scores indicating a stronger commit ted sense of belonging to ethnic group. This sample had mean scores ranging from 1 5 ( M = 3.64, SD = 1.00) indicating a positively skewed distribution of scores. A paired samples T test indicated significant differences between the MEIM R subscale scores t (250) = 7.38, p < .001 Specifically, parents endorsed a stronger sense of Ethnic I dentity Commitment ( M = 3.81, SD = 1.00) than Ethnic Identity E xploration ( M = 3.46, SD = 1.14). Further examination of MEIM R scores using a one way ANOVA indicated
56 that Black parents ( M = 4. 20, SD = .76) endorsed a stronger sense of belonging to their ethnic group than White parents to their own ethnic group ( M = 3.10, SD = .90), F (1, 249) = 107.84, p < .001. Figure 4. Parental Experience with Various Disorders by Race and Gender The scores for the Barriers to Utilization measure ranged from 0 to 3, with higher scores indicating more perceived barriers to treatment utilization children The current sample had scores ranging from 0 .11 to 1.83 ( M = .94, SD = .36). The ANCOVA used to examine the relationship between race, gender, and barriers (covarying for SES) was significant F (4, 237) = 1.47, p < .05. There was a main effect of gender with fathers ( M = 1.01) reporting significantly more barriers to service util ization for their children than mothers ( M = .87). Neither race nor the interaction effect was 0 10 20 30 40 50 60 70 80 90 100 ADHD Anxiety Depression ODD Substance abuse Black Dads Black Moms White Dads White Moms 100%
57 significant. The most frequently endorsed barriers re 36.3%, 31.1%, and 30.3%, respectively). Further examination of barriers based on parental race and gender indicated similarities and differences in the most frequently endorsed barrier. Specifically, both mo thers (33.6%) and father ould be a Overall, 45.4% of parents endorsed financial barriers, 60.6% endorsed structural barriers, and 77.7% endorsed attitudinal barriers. Parents reported their b eliefs about causes of mental health problems from a list of 11 differ ent areas. Overall, t he most frequently endorsed cause s of mental health problems in children were physical causes, family or parenting issues, and trauma with 98.4%, 97.6%, and 96.4% of parents endorsing items from each of these areas, respectively. It is notable that all three of these subscales fall under the broad category of biopsychosocial cause s of child mental health problems. In fact, 99.2% of parents endorsed a biopsychosocial cause, 93.6% endorsed a sociological cause, and 51.8% endorsed a spi ritual/nature disharmony cause of mental health problems in children. After parents endorsed their beliefs about the causes of mental health problem s in children overall they were then asked to indicate what they felt was the most important cause of ment al health pr oblems in children ; th is y ield ed smaller percentages and some va riability by race and gender (see Figure 5 ) Three areas were not identified by any
58 parent as the most important cause of mental health problems in children including: American cu lture, Discrimination or Prejudice, and Disharmony with Nature. Parents responses to the vignettes w ere examined in greater detail in the hypothesis testing section; however, examination of parents responses about their first source of help seeking for the internalizing and externalizing problem vignettes indicated 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Black Dads Black Moms White Dads White Moms
59 that most parents would seek help from the for both internalizing (see Figure 6) and externalizing problems (see Figure 7) Figure 6 First Source of Help Seeking for Internalizing Vignettes by Race and Gender 0% 10% 20% 30% 40% 50% 60% Black Dads Black Moms White Dads White Moms
60 Figure 7. First Source of Help Seeking for Externalizin g Vignettes by Race and Gender Group Differences Analyses of Variance (ANOVAs) were conducted to determine if the groups (Black fathers, Black mothers, White fathers, and White mothers) were significantly different from each other on any of the major demo graphic variables including: age, number of children, household size, education, socioeconomic status, and income. Assumptions of normality, equality of variance, and independence were met for most of the analyses. For a few cases, equality of variance was not met, however ANOVA is fairly robust to this violation. In cases where outliers resulted in a skewed distribution, the analyses were run with and without outliers. When significant diff erences occurred, the results a re described separately A Bon ferroni corrected p value of .01 was used given that four ANOVAs were run. 0% 10% 20% 30% 40% 50% 60% Black Dads Black Moms White Dads White Moms
61 Examination of parental age indicated a significant main effect for race, F (1, 238) = 7.28, p < .01, where Black parents were significantly younger ( M = 38.60) than White parents ( M = 41.39). There was no main effect for gender and the interaction effect was not significant. There was no difference between groups on the number of children parents had However, given that the data was positive ly skewed and not normal ly distributed final analyses involving group differences were run with and without outliers removed There was also no significant difference between groups on household size. An ANOVA examining parental education indicated a significant main effect for gender, F (1, 244) = 9.61, p < .01, with fathers reporting higher levels of education ( M = 16.35) than m others ( M = 15.48). Neither race nor the interaction effect w as significant. Given t hat t he average number of years of education was 15.9 ( SD = 2.27) this sam ple is notable for having a highly educated group of both Black and White parents, as there were no significant differences in race. For socioeconomic status (SES), there was a significant main effect for gender F (1, 2 43 ) = 13 57 p < .01 Specifically, fathers reported higher SES ( M = 50.12) than mothers ( M = 45.47). There was no main effect for race and the interaction effect was not significant. When the data w ere re examined with outliers controlled, results remained consistent. With respect to inco me, there was a significant main effect for gender, F (1, 240) = 11.62, p < .01, with fathers ( M = 13.92) reporting higher income than mothers ( M = 12.47). Neither the main effect for race nor the interaction effect were significant. As the income variabl e was negatively ske wed and not normally distributed outliers were controlled but results remained consistent Given
62 the known confound between SES and race it is notable that this sample did not differ significantl y on SES across racial groups. As init ially proposed SES was to be statistically controlled, thus the gender difference in the SES variable will be captured by this statistical control of SES Education differences will already be captured by controlling SES as well, as education is one of the four variables that make up the SES variable. Differences in income will also be captured by controlling SES, given the high correlation between these two variables ( r =.57, p < .001 in this sample). Relevant analyses were re examined with parental a ge statistically controlled and no differences were found Hypothesis Testing The first and most important analysis was the investigation of the relationship between parental problem recognition and problem type (hypothesis 1) Given that the same set of parents rated all 3 vignettes, hypothesis 1 was examined using non parametric nature of the categorical variables which were repeatedly measured and there fore dependent. This test was used to examine the relationship between problem recognition and problem type including internalizing ( item 5 on anxiety vignette), externalizing ( item 5 on ADHD vignette), and control problem (item 5 on the no diagnosis vign ette). Additionally, odds ratio hand calculations were conducted to determine the likelihood of parents recognizing internalizing and externalizing problems in comparison to control vignettes. In the overall sample, the percentage of parents who correctly identified the vignettes (i.e. correctly recognized internalizing and externalizing vignettes as problematic, and the control vignette as not problematic) were 51.8% for the internalizing
63 vignette, 60.6% for the externalizing vignette, and 88.8% for the c ontrol vignette. As hypothesized, p airwise comparisons using a corrected Bonferroni of p = .01 7 indicated that parents were more likely to identify the internalizing vignette as proble matic than the control vignette Q (1) =89.63, p < .00 1 and the externalizing vignette as problematic than the control vignette Q (1) =116.74, p < .00 1 No statistically significant difference was found between parents recognition of externalizing versus internalizing problems. Further odds ratio calculation s revealed that, compared to the control vignette, parents were 9.7 times more likely to recognize an internalizing problem and 14. 5 times more likely to recognize an externalizing problem as problematic. However, it is important to emphasize that althoug h parents were often able to distinguish between internalizing and externalizing vignettes from the control vignette, 48.2% and 39.4% of parents did not recognize that there was an internalizing or externalizing problem at all. Additionally, p arents were 1.5 times more likely to recognize an externalizing than an internalizing problem as problematic; however, this difference w hile not statistically significant after Bonferroni correction reveal ed a trend ( p = .023). Hypothesis 2 stated that parents who rec ognize a mental health problem w ould be willing to seek help, irrespective of the type of problem. Two Chi square tests were used to examine th is relationship between problem rec ognition (item 5 on the anxiety and ADHD vignettes) and willingness to se ek help (item 6 on the anxiety and ADHD vignettes) respectively. Additionally, a repeated measures logistic regression was used to examine the overall relationship between problem recognition (item 5 on the anxiety and ADHD vigne ttes) and willingness to s eek help (item 6 on the anxiety and ADHD vignettes) The first c hi square tes t revealed a significant relationship between parental
64 recognition of an internalizing problem and their decision to s eek help (1) = 65.66 p < .001. Specifically, odd s ratio ca lculations indicated that the odds of parents seeking help for an internalizing problem were 35.28 times higher when they recognized a problem than when they did not recognize a problem A significant relationship was also found between parental recognition of an externalizing problem and their decision to seek help (1) = 50.31, p < .001 Similarly, the odd s ratio indicated that parents were 69.12 times more likely to seek help for an externalizing problem when they recognized a pro blem than when they did not As expected, the repeated measures logistic regression was significant (1) = 53.97, p < .001 and revealed that overall parents were more likely to report that they would seek help when they recognized a problem regardless of the problem type B (1) = 3.81 (0 .52 ) p < .001 Thus hypothesis 2 was fully supported. Hypothesis 3 was analyzed with a series of chi square tests and stated that parents w ould be more likely to recognize problems and more willing to seek help for boys than for girls. One set of chi squares examined the relationship between child vignette gender and problem recognition for a) an internalizing problem (item 5), and b) an externalizing problem (item 5). The second set of chi square s examined the relationship between child vignette gender and willingness to seek help for a) an internalizing problem (item 6), and b) an externalizing problem (item 6). The first set of chi square analyse s used to examine Hypothesis 3 was no n significant. The se analyses examin ed the relationship s between child vignette gender and recognit ion of internalizing (1) = 2.13 p = .14 and externalizing (1) = .25, p = .62 problems The second set of chi squares examin ing the relationship betwe en child vignette gender and willingness to seek help for an
65 ex ternalizing problem (1) = .31 p = 58 was non significant T he chi square examining the relationship between child vignette gender and willingness to seek help for an internalizing proble m was significan t ( using a corrected alpha of .025 ) (1) = 6.72, p = .0 1 Specifically, based on odds ratio calculations, parents were 2.2 times more likely to seek help for a boy with internalizing symptoms than a girl with the same symptoms V = .16). Thus, there was partial support for hypothesis 3. Hypothesis 4 w as examined with two sets of logistic regressions and stated that parents who endorse higher levels of perceived severity, problem threshold, functional impairment, and family impac t would be more likely to recognize problems and more willing to seek help. The first set of logistic regressions w ere used to determine what aspects of perceived need [severity (item 1), problem threshold (item 2), functional impairment (item 3), and fam ily impact (item 4)] were most predictive of parental recognition of a) an internalizing problem, and b) an externalizing problem. While t he secon d set of logistic regressions examined w hat aspects of perceived need [severity (item 1), problem threshold ( item 2), functional impairment (item 3), and family impact (item 4)] were most predictive of parental willingness to seek help for a) an internalizing problem, and b) an externalizing problem. The s eries of logistic regressions (using a corrected alpha of .025 per question) beg an with predicting the relationship between perceived need ( severity, problem threshold, functional impairment, and family impact ) and recognition of an internalizing problem The overall logistic regression was significant with sev erity being the only significant predictor of internalizing problem recognition (see Table 2) Specifically, for every unit increase in severity rating s parents were 2.8 times more likely to recognize an internalizing problem.
66 Table 2 Logistic Regression for Perceived Need Variables and Internalizing Problem Recognition 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 5.10 *** (.82) Severity 1.03 ** (.32) 1.50 2.81 5. 26 Problem Threshold 0.02 (.35) 0.52 1.02 2.02 Functional Impairment 0.34 (.29) 0.80 1.41 2.48 Family Impact 0.09 (.23) 0.69 1.09 1.72 Note : R 2 = .19 (Hosmer & Lemeshow), 23 (C ox & Snell), .31 (Nagelkerke). Model (4 ) = 66.74 p < 001. ** p < .0 0 1 ** p < .0 1 Another logistic regression was used to predict the relationship between the perceived need variables and externalizing problem recognition. Although the overall model was significant, (4) = 55.92, p < .001, none of the predictor variables were significant. Logistic regressions were also used to examine the second part of hypothesis 4 (i.e. the relationship between the perceived need variables and willingness to seek help for internalizing and externalizing problems). The overall model for the perceived need variables and willingness to seek help for an internalizing problem was significant, (4) = 49.20, p < .001; however none of the predictor variables were significant. Finally, the overall logistic regression model for the perceived need variables and willingness to seek help for an externalizing problem was significant, p < .001 ; however again none of the predictor variables were significant, although there was a trend for severity, B (1) = 1.06 (0. 50 ), p = .0 3.
67 Post hoc analyses for Hypothesis 4 Given consistently null findings despite significant overall models, hypothesis 4 was re examined with post hoc analyses in order to understand the data more clearly The relationship among perceived need variables ( severity, problem thresh old, functional impairment, and family impact ) for internalizing and externalizing vignettes were explored with correlation al analyses. For the internalizing vignette, a very strong correlation was found between severity and problem threshold ( r = .80, p < .001). For the externalizing vignette very strong correlations were found between severity and problem threshold ( r = .8 3 p < .001) and functional impairment and family impact ( r = .82, p < .001). Thus, the data w ere re examined with only one of thes e variables given that they appeared to be measuring similar underlying factors. Severity was selected over problem threshold and functional impairment over family impact based on the consistent relationship between these selected factors and help seeking as previously established in the literature (Zwaanswijk et al., 2003). Examination of the internalizing vignette, with problem threshold dropped from the model, revealed no differences with prob lem recognition or willingness to seek help. Examination of the externalizing vignette however, revealed significant changes for both problem recognition and willingness to seek help using a corrected alpha of 0.025 Specifically, the overall logistic r egression model was significant with severity and functional impairment as significant predictor s of ex ternalizing problem recognition (see Table 3 ). Results indicated that for every unit increase in perceived severity ratings parents were 2.3 times more likely to recognize an ex ternalizing problem. Similarly, for every unit increase in perceived functional impairment, parents were 1.7 times more likely to recognize an externalizing problem.
68 Table 3 Logistic Regression for Selected Perceived Need Variable s and Ex ternalizing Problem Recognition 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 4 37 *** (.8 1 ) Severity 0 82 ** (.2 3 ) 1.44 2.27 3 59 Functional Impairment 0.54* (.24 ) 1.08 1.72 2. 74 Note : R 2 = .16 (Hosmer & Lemeshow), 19 (Cox & Snell), 26 (Nagelkerke). Model ( 2 ) = 53 19 p < 001. ** p < .0 0 1 p < .0 5 With respect to willingness to seek help for an externalizing problem, the overall logistic regression model was significant with severity and functional impairment as significant predictors (see Table 4). Table 4 Logistic Regression for Selected Perceiv ed Need Variables and Willingness to Seek Help for an Ex ternalizing Problem 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 4. 55 *** ( 1 06 ) Severity 1 11 ** (.3 8 ) 1.44 2.27 3.59 Funct ional Impairment 0. 89 (. 3 4) 1.08 1.72 2.74 Note : R 2 = 28 (Hosmer & Lemeshow), .19 (Cox & Snell), 37 (Nagelkerke). Model (2 ) = 5 2 .9 1 p < 001. ** p < .0 0 1 * p < .0 1, p < .0 5 Results revealed that for every unit increase in perceived severity ratings, parents were 3.0 times more likely to report willingness to seek help for an externalizing problem. Similarly, for every unit increase in perceived functional impairment, parents were 2.4 times more likely to report willingness to seek help for an externalizing problem. Overall, there was partial support for Hypothesis 4.
69 The fifth hypothesis was explored with logistic regressions and stated that parents who report smaller family size, higher levels of education, and current psychopathology would be more likely to recognize problems and more willing to seek help. The first set of logistic regressions w ere used to determine the aspects of family characteristics [family size (demographics item 4), parent education (dem ographics item 10 self), and parent self reported psychopathology (utilization item 4)] that were most predictive of recognition of a) an internalizing problem, and b) an externalizing problem. The second set of logist ic regressions helped to determine as p ects of family characteristics (same as those listed above) that we re most predictive of willingness to seek help for a) an internalizing problem, and b) an externalizing problem. A c orrected alpha of .025 was used for each l ogistic regression With res pect to the relationship between the family characteristic variables and parental problem recognition, neither the internalizing problem (3) = 7.37, p = .06 nor the externalizing problem ( 3 ) = 0 70 p = 89 logistic models were significant. Similarly, examination of the relationship between the family characteristic variables and willingness to seek help were non significant for internalizing (3) = 2.87, p = .41 and externalizing (3) = 1.78, p = .62 problems Thus, hypothesis 5 was n ot supported. Two sets of logistic regressions w ere used to examine the s ixth hypothesis which stated that African Americans, fathers, and parents of lower SES would be less likely to recognize problems and less willing to seek help It was also hypothe sized that African American parents with higher ratings of ethnic identity and Caucasian parents with lower ratings of ethnic identity will be less likely to recognize problems and less willing to seek help. The first set of logistic regressions were used to predict the relationship between
70 demographic characteristics (parent race, ethnic identity (mean score), parent gender, and SES) and problem recognition of a) an internalizing problem, and b) an externalizing problem. The second set of logistic regres sions w ere used to predict the relationship between demographic characteristics (parent race, ethnic identity (mean score), parent gender, and SES) and willingness to seek help for a) an internalizing problem, and b) an externalizing problem. Logistic reg ressions were explored using a corrected alpha of .025 With respect to the relationship between parental demographic variables and internalizing problem recognition, the overall model was significant with race and gender as significant predictors (see Ta ble 5 ) Table 5 Logistic Regression for Parent Demographic Variables and In ternalizing Problem Recognition 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 2 77 ** ( 1 00 ) SESscore 0 01 (.0 1 ) 0.99 1.01 1.04 Ethnic Identity 0.28 (. 16 ) 0.97 1.33 1 82 Race (1) 1.60 *** (. 44 ) 2.10 4.94 11 .65 Gender (1) 1 22 ** (. 39 ) 1.56 3.37 7 .2 9 Race (1) by Gender (1) 1 .1 5 (. 53 ) 0.11 0.32 0 90 Note: R 2 = 06 (Hosmer & Lemeshow), .08 (Cox & Snell), ) = 20 11 p < 0 1. ** p < .0 0 1, ** p < .01 Specifically, White parents were 4.9 times more likely than Black parents and mothers were 3.4 times more likely than fathers to recognize an internalizing problem. Of note, there was a trend ( p =.03) for White mothers to recognize internalizing problems more often than other parents.
71 Examination of the relationship between parental demographic variables and externalizing problem recognition revealed a non significant model (5) = 8.48, p = .13. The second part of hypothesis 6 examined the relationship between parent demographic characteristics and willingness t o seek help. For willingness to seek help for an internalizing problem the overall model was significant with gender as the only significant predictor (see Table 6 ) Spec ifically, m others were 4.5 times more willing to seek help for an internalizing problem than fathers. Of note, th ere was a trend for increased willingness to seek help for an internalizing problem for white parents ( p = .03) and white mothers, specifically ( p = .03). Table 6 Logistic Regression for Parent Demographic Variables and Willingness to Seek Help for an I n ternalizing Problem 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 0.38 (1. 17 ) SESscore 0.01 (.0 2 ) 0.98 1.01 1.0 5 Ethnic Identity 0. 17 (. 21 ) 0.56 0 84 1.2 6 Race (1) 1.0 2 (.4 8 ) 1 .0 8 2 79 7 16 Gender (1) 1. 51 ** (. 45 ) 1. 87 4.54 11 07 Race (1) by Gender (1) 1. 38 (. 65 ) 0. 07 0. 2 5 0.90 Note: R 2 = 07 (Hosmer & Lemeshow), 07 (Cox & 18 33 p < .01. ** p < .01 Examination of the relationship between parental demographic variables and willingness to seek help for an externalizing problem revealed a non significant relationship, p = .0 3. Thus, overall, there was only partial support for hypothesis 6.
72 Hypothesis 7 stated that parents who endorse biopsychosocial beliefs would be more likely to recognize problems in youth and more willing to seek formal help while p arents who endorse sociological and spiritual/nature disharmony beliefs would be less likely to recognize problems and more willing to seek informal help. This hypothesis w as examined with c hi square analyses 2 tailed was used when m ore than 20% of the expected frequencies were less than 5) and Mann Whitney tests. Non parametric tests were used given the violations of normality (Field, 2009). The c hi e xact tests were used to explain the relationship between each of t he biopsychosocial beliefs (physical causes, personality, relational issues, familial issues, and trauma) sociological beliefs (friends, American culture, prejudice, and economic problems), and spiritual/nature disharmony beliefs (spiritual causes, and na ture disharmony) and a) internalizing problem recognition (item 5), and b) externalizing p roblem recognition (item 5) The Mann Whitney tests were used to explain the relationship between biopsychosocial, sociological, and spiritual/nature disharmony beli efs and a ) willingness to seek formal help for internalizing and externalizing problem s (mean s of items 8 1, 8 2, and 8 4 for internalizing and externalizing vignettes), and b ) willingness to seek informal help for internalizing and externalizing problem s (mean s of items 8 3, 8 5, and 8 6 for internalizing and externalizing vignettes ). Chi s revealed non significant relationships between internalizing problem recognition and B iopsychosocial B eliefs in cluding Physical C auses ( p = .36; ), P ersonality ( ( 1 ) = 1 .4 0 p = 24), Relational I ssues ( (1) = 0.35, p = .55), P arenting/ F amilial issues ( p = .11; FET ), and T rauma ( p = 1.00; FET ). T here were significant relationships between internalizing problem recognition
73 and some of the Sociological Beliefs (using corrected alpha of 0.013 ) including Friends ( (1) = 11 .66, p < .01 ) and Economic Problems ( (1) = 7.51 p < .0 1 ) Specifically, the odds of parents recognizing an internalizing problem were 4.6 times higher when they believed mental health problems w ere at least partially caused by Friends ) and 2.2 times higher when they believed mental health problems were at least partially .17) The relationship between internalizing problem recognition and the Sociological Belief s of American Culture ( (1) = 5.21, p = .02) and Prejudice ( (1) = 2.66, p = .10) were non significant; although as illustrated there was a trend for American Culture beliefs Similarly, there were non significant relationships between internalizing problem recognition and Spiritual/Nature Disharmony Beliefs including Spiritual Causes ( (1) = 0.08, p = .78) and Nature Disharmony ( (1) = 1.02, p = .31) With respect to ex ternalizing problem recognition and Biopsychosocial Beliefs non significant relationships were found with Physical Causes ( p = 6 4; FET ), Personality 0 .0 2 p = 90 10 p = 75 ), Parenting/Familia l issues ( p = .1 0 ; FET ), and Trauma ( p = 1.00; FET ). Similarly, non significant rel ationships were found between ex ternalizing problem recognition and the Sociological Beliefs including Friends ( 0 09 p = 76) American Culture ( 1 47, p = 23) 1 81 p = 18) and Economic Problems ( 1 .0 2 p = 31) Additionally, there were non significant relationships between ex ternalizing problem recognition and Spiritual/Nature Disharmony Beliefs including Spiritual Cause 1 23 p = 27 0 .0 03 p = 96 ).
74 The sec ond portion of hypothesis 7 examined the relationship between the various beliefs and willingness to seek formal and informal help using Mann Whitney test s and corr ected al pha of 0.0125 for each b elief category examined Analyses revealed non significant relationships between Biopsychosocial beliefs and willingness to seek formal help for internalizing ( U = 241, p = .94) and externalizing ( U = 157.50, p = .39) problems Si milarly, non significant relationships were also found between Biopsychosocial beliefs and willingness to seek informal help for internalizing ( U = 161, p = .39) and externalizing ( U = 174.50, p = .49) problems Examination of the relationship between Sociological Beliefs and willingness to seek in formal help for internalizing ( U = 1739.50, p = .62) and externalizing ( U = 1835, p = .99) problem s were non s ignificant T he relationship s between Sociological beliefs and willingness to seek formal help for internalizing ( U = 1354, p = .06) and externalizing ( U = 1247, p = .03) problems were also non significant but a trend was revealed. A s hypothesized, a significant relationship was revealed between Spiritual/Dish armony Beliefs and willingness to seek informal help for internalizing ( U = 6167, p < .01) and externalizing ( U = 5265.50, p <.001) problems. Further examination of the data revealed that parents who believed that mental health problems were at least part ially caused by spiritual/religious reasons were more willing to seek informal help for internalizing ( U = 6491.50, p = .02, r = .15) and externalizing ( U = 5666.50, p < .01, r = .22) problem s. Non significant relationships were revealed between Spiritual/Disharmony Beliefs and willingness to seek formal help for internalizing ( U = 7556.50, p = .59) and externalizing ( U = 7400.50, p = .78) problems. Thus, overall, there was only partial support for hypothesis 7.
75 Two chi square tests were used to examine the eighth hypothesis which stated that parents who have had previous experience with anxiety and ADHD would be more likely to recognize the anxiety and ADHD vignettes as problematic One chi square w as used to explain the relationship between p revious professional or personal experience with anxiety (utilization item 12) and recognition of an internalizing problem (item 5). The second chi square w as used to explain the relationship between previous professional or personal experience with ADHD (utilization item 12) and recognition of an externalizing problem (item 5). Results revealed a significant relationship between previous experience with anxiety and recognition of an internalizing (anxiety) problem vignette (1) = 19 01 p < .001. Spec ifically, parents were 3.3 times more likely to recognize an internalizing (anxiety) problem when they had previous experience with ). There was a non significant relationship, however, between previous experience with ADHD and re cognition of an externalizing (ADHD) problem vignette (1) = 1 09 p = 30 Further analysis using an independent samples t test revealed a significant relationship between previous mental health experience in general and recognition of an internalizing (anxiety) problem, t (247) = 4.05, p < .001. Specifically, parents who accurately recognized an internalizing problem vignette had more pr ior mental health experience ( M = 0.49) than those who di d not ( M = 0.30). However, p revious mental health experience did not have a significant relationship with externalizing (ADHD) problem recognition, t (243) = 0 15 p = 88. Specifically, parents who accurately recognized an externalizing problem vignett e did not differ in previous mental health experience ( M = 0.40) from parents who did not recognize the externalizing problem vignette ( M = 0.40). Thus, hypothesis 8 was partially supported.
76 The ninth hypothesis stated that fathers and African American pa rents would endorse more sociological and spiritual/nature disharmony beliefs and fewer biopsychosocial beliefs than mothers and Caucasian parents. Additionally, it was hypothesized that fathers and African American parents would have less experience with mental health disorders than mothers and Caucasian parents. This hypothesis w as examined using logistic regressions and Analyses of covariance (ANCOVA). Logistic regressions w ere used to examine the relationship s between each specific belief (relational issues, friends, American culture, prejudice, economic problems, spiritual causes, and nature disharmony) and parental race and gender. However, four of the beliefs (physical causes, personality, parenting/familia l issues, and trauma) were not examined because they were endorsed by 93% 98% of all parents resulting in more than 20% of their cells with expected counts below 5 Using a corrected alpha of .007 analyses proceeded with predicting the relationship betw een Relational belief s parent race and gender, which was non significant ( 2 ) = 5. 5 8 p = 06 However, t he relationship between Friends belief s parent race and gender was significant with gender as the only significant predictor (see Table 7 ) T able 7 Logistic Regression for Parent Race and Gender 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 1 25*** (. 29 ) Gender (1) 1.27** (.4 6 ) 1 .45 3.56 8 77 Race (1) 0 74 (. 4 2 ) 0 91 2.09 4 7 9 Note: R 2 = 07 (Hosmer & Lemeshow), 05 (Cox & Snell), 0 9 (Nagelkerke). Model p < .01 *** p < .001, ** p < .01
77 Specifically, mothers were 3.6 times more likely than fathers to endorse that mental elief in American Culture as a potential cause of mental health problems and their race and gende r were examined next and revealed a non significant relationship ( 2 ) = 0 17 p = 92 T he relationship between parent race gender and Discrimination belief s ( 2 ) = 3 0 2 p = 22 and Economic belief s ( 2 ) = 6 17 p = 0 5 were also non significant after Bonferroni correction T he relationship between parent race gender and Religious belief s was significant with race as the only si gnificant predictor (see Table 8 ). Specifically, B lack parents were 2 3 times (1/.43, give n negative beta weight) more likely than White parents to endorse that mental health problems a re due at least in part to religious or spiritual reasons Tab le 8 Logistic Regression for Religious Beliefs and Parent Race and Gender 95% CI for Odds Ratio Variable B (SE) Lower Odds Ratio Upper Constant 0 .25 (. 23 ) Gender (1) 0 12 (. 26 ) 0 68 1.13 1 87 Race (1) 0 86** (. 26 ) 0 26 0 43 0 7 1 Note: R 2 = 0 3 (Hosmer & Lemeshow), 04 (Cox & Snell), 0 6 (Nagelkerke). Model p < .01 ** p < .01 Finally, the relationship between parent race gender and Nature Disharmony belief s was non significant after Bonferroni correction 2 ) = 7 83 p = .0 2; alt hough, a trend was revealed
78 T he second and third part s of hypothesis 9 w ere examined using two way ANCOVA s Due to the strong relationship between race and SES in previous research the influence of SES w as covaried to en sure that the results we re not due to the influence of SES but rather to actual rac e differences. There w ere two between subject factors gender (mother vs. father) and race (Black vs. White). These ANCOVAs w ere used to investigate the relationships between parental race and g ender and a) biopsychosocial beliefs (mean score), b) sociological beliefs (mean score), c) spiritual/nature disharmony beliefs (mean score), and d) prior experience with mental health disorders (mean experience score). The second part of hypothesis 9 wa s examined usin g three two way ANCOVAs (corrected alpha of 0.017) The first ANCOVA was between B iopsychosocial beliefs, race (Black and White), and parent gender (mother and father ), covarying for the influence of SES. While the overall statistic was si gnificant F (4, 241) = 3 11 p = .0 16 partial 2 = .0 5 none of the main effects ( parent gender, F (1, 241) = 4.60, p = .03; race, F (1, 241) = 4.29 p = .0 4 ) or the interaction effect F (1, 241 ) = 2 8 9, p = .09 w ere significant The second ANCOVA exa mining the relationship between S oci ological beliefs, race and parent gender, while covarying for the influence of SES was no n significant, F (4, 241) = 1.00, p = .41. The third ANCOVA explored the relationship between Spirit ual/ Disharmony beliefs race and pa rent gender, covarying for the influence of SES Results indicated a significant main effect of race F (1, 24 0) = 13 90 p < .0 01 partial 2 = 06 indicating that Black parents ( M = 2 03 ) endorsed stronger Spiritual/Disharmony beliefs than White parents ( M = 1 59 ). However, n either the main
79 effect of parent gender, F (1, 240) = 0 07 p = .80, nor the interaction effect, F (1, 240) = 0.07, p = 79 were significant. The third part of Hypothesis 9 used an ANCOVA to examine the relationship between prior experience with mental health disorders, race, and parent gender, while covarying for the influence of SES. There was a significant main effect of parent gender, F (1, 240) = 7.10, p < .01, partial 2 = .03, indicating that mothers ( M = .46) reported more prior mental health experience than fathers ( M = .33). The main effect for race was also significant, F (1, 24 0 ) = 25 3 9, p < .0 01 partial 2 = 10 indicating that White parents ( M = .51 ) reported more prio r experience with mental health disorders than Black parents ( M = 28 ). However, the interaction effect was non significant, F (1, 240 ) = 0 01 p = 93 Thus overall, t here was partial support for portions of hypothesis 9. Hypothesis 10 stated that paren ts would endorse more attitudinal barriers than structural and financial barriers. Additionally, it was hypothesized that fathers would endorse more attitudinal barriers than mothers and African American parents would endorse more structural barriers than Caucasian parents. P aired samples t test s and ANCOVA s were used to examine this hypothesis The initial paired samples t tests were used to determine differences among the types of barriers [attitudinal (mean subtest score), structural (mean subtest score), and financial (mean subtest score)] The ANCOVA s (covarying SES) were used to investigate the relationship between race (black and wh ite) and parent gender (mothers and fathers) and attitudinal, structural, and financial barriers. The first part of Hypothesis 10 using paired samples t tests revealed medium sized correlations between financial and attitudinal barriers ( r = .31 p < .001 ), financial and structural barriers ( r = .39 p < .001 ), and attitudinal and structural barriers
80 ( r = .46 p < .001 ). However, t tests comparing each of the following pairs were non significant; financial and attitudinal barriers, t (245) = 0.85, p = .40 financial and structural barriers, t (245) = 1.79, p = .07, and attitudinal and structural barriers, t (245) = 1.19, p = .24. The second part of Hypothesis 10 used ANCOVAs to explore the relationship between barriers, parent gender, and race, covarying for SES (corrected alpha 0.017) The first ANCOVA exploring the relationship between financial barrier s parent gender, and race, covarying for SES was non significant, F (4, 237) = 2.19, p = .07. The second ANCOVA examined the relationship between attitu dinal barriers, parent gender, and race, covarying for SES was also non significant, F (4, 237 ) = 1. 97 p = 10. The third ANCOVA used to analyze the relationship between structural barriers, parent gender, and race, while covarying for SES was not significant either, F (4 237) = 2 64 p = .0 4. Thus, hypothesis 10 was not supported. The eleventh hypothesis stated that parents who perceive more barriers would be less willing to seek formal help, irrespective of problem type Correlations w ere used to ex plore the relationship s between barriers (mean score) formal help seeking (mean of items 8 1, 8 2, and 8 4) for internalizing and externalizing problems, inform al help seeking ( mean of items 8 3, 8 5, and 8 6) for internalizing and externali zing problems and ethnic identity (mean score ; see Table 9 ). There was a weak inverse relationship between barriers and formal help seeking for an externalizing problem ( i.e. parents were less willing to seek formal help for an externalizing problem when they perceived more barriers ) Additionally, help seeking regardless of source (formal vs. informal) or problem type (internalizing vs. externalizing) had moderate to strong inter relationships. Of note, there were very strong relationships between simi lar sources of help seeking
81 regardless of problem type (see Table 9 ). Ethnic identity was not significantly correlated with any of the other variables. Thus, there was partial support for hypothesis 11. Table 9 Intercorrelations between Barriers, Ethnic Identity, Formal and Informal Help Seeking for Internalizing and Externalizing Problems Variable Mean SD Barriers Formal HS Forrnal HS Informal HS Informal HS Ethnic Internalizi ng Exte rnal i zing Internalizing Externalizing Identity Barriers .94 .36 1 3 14 09 .11 .0 6 Formal HS 3.14 1.03 .5 5 * 40*** .21** 05 Internalizing Formal HS 3.39 .99 24*** 3 5* * 0 5 Externalizing Informal HS 2.59 .94 74 *** 04 Internalizing Formal HS 2.46 .99 02 Externalizing Ethnic 3.64 1.00 Identity Note. HS = Help seeking p < .05, ** p < .01, *** p < .001 Hypothesis 12 stated that endorsement of fewer barriers, stronger biopsychosocial beliefs, weaker sociological beliefs, and weaker spiritual/nature disharmony beliefs would predict higher intentions to utilize mental health services in the future. M ultiple regression analysis w as services for their child in the future based on their perceived barriers (mean score), biopsychosocial beliefs (mea n score), sociologic al beliefs (mean scale score), spiritual/nature disharmony beliefs (mean score), and ethnic identity (mean score). Only parents who endorsed that at least one of their children had a current mental health problem were used in this anal ysis. Of the se 45 parents (17.9% of t he t otal sample)
82 68.9% of them were quite or extremely likely to seek help for their child. However, t he overall regression equation was non significant, F (5, 39) = 1.28, p = .29 Thus, the hypothesized variables di d not predict intentions to utilize mental health services; hypothesis 12 was not supported. Post Hoc Analyses Given that some of the proposed hypotheses were not supported and revealed unexpected null results, further analyses were conducted to explore t he variables in a more thorough manner These results are presented in this section because they were not proposed with the original hypotheses. Parental c onfidence in rating s of vignettes. P arents were asked to rate their c onfidence in their ratings of each vignette as problematic or not. Overall parents were confident in their own ratings (measured on a scale from 1 5); internalizing ( M = 3.60, SD = .89), externalizing ( M = 3.65, SD = .97), and control ( M = 3.89, SD = .96). Independent samples t tests were used to explore differences in confidence ratings among parents who did or did not identify vignettes correctly. With respect to the internalizing vignette, a significant relationship was revealed, with parents who correctly recognized the internali zing vignette expressing more confidence in their rating ( M = 3.73) than parents who did not recognize the internalizing vignette ( M = 3.46; t (248) = 2.45, p = .02 ) Similarly with the externalizing vignette, parents who correctly recognized this vignet te expressed more confidence in their rating ( M = 3.83) than parents who did not recognize the externalizing vignette ( M = 3.35; t (243) = 3.88, p < .001). There were no significant differences for the control vignette, t (244) = 0.96, p = .34.
83 ANOVAs were used to explore differences in confidence ratings by race and parent gender covarying for the influence of SES and a ge Results indicated non significant relationships with internalizing, F (5, 232) = 1.98, p = .08 and control, F (5, 228) = 1.21, p = .31. However, for the externalizing vignette, there was a main effect for race, F (1, 228) = 36.94, p < .001 partial 2 = .14 Specifically, Black parents ( M = 4.03) were more confident in their rating of externalizing symptoms than were White parents ( M = 3.31). Neither the main effect for parent gen der, F (1, 228) = 0.97, p = .33 nor the interaction effect were significant, F (1, 228) = 2.55, p = .11. mental health experience Examination of parents who currently had a child with a mental health problem revealed that parents whose children had utilized mental health services in the past reported higher intentions ( M = 4.17, SD = 1.16) to utilize service s in the future than parents whose children had not used services ( M = 3.11, SD = 1.36) in the past, t (43) = 2.36, p < .05 r = .34 Furthermore, there was a very strong correlation between satisfaction with past mental health treatment and intentions to utilize services in the future, r = .42, p < .001. Chi square tests were us mental health experience with their own children and recognition of internalizing and externalizing problems as presented in vignettes A significant relationship was found between having a child with a mental health problem and internalizing problem recognition, 1 ) = 12 20 p < .0 01. Specifically, parents who reported having a child with a current mental health problem were 3.5 ti mes more likely to recognize the However, this relationship was non significant for the externalizing vignette, 1 ) = 1 86 p = 17. Independent samples
84 t tests were used to examine the relation health services for their children and willingness to seek formal help for internalizing and externalizing problems as presented in vignettes. Parents were more willing to seek formal help for the internali zing vignette if their own children had used mental health services in the past, t (249) = 3.85, p < .001, r = .24. Similarly, parents were also more willing to seek formal help for the externalizing vignette if their own children had used mental health services in the past, t (244) = 2.85, p < .01, r = .18. Belief s about causes of m ental health p roblems Given poor variability and significant skew in the BAC R measure, logistic and multivariate regressions were used to re examine the relationships between beliefs about causes of mental health problems and recogniti on and willingness to seek formal and in formal help. First, correlations between endorsement of each belief category (mean of items A K on the BAC R) and strength rating of each belief category (item L: A K) revealed moderate (.23 for physical causes) to very strong (.60 for discrimination) c orrelations. The overall logistic regression model predicting the relationship between beliefs and internalizing problem recognition was significant, 11 ) = 26 01 p < 01. However, none of the predictors were significant. Additionally, t he relations hip between beliefs and externalizing problem recognition was non significant, 11 ) = 10 05 p = 53. The first multiple regression examining the relationship between beliefs and willingness to seek formal help for an internalizing problem was signific ant F (11, 227) = 4.16, p < .001, R 2 = .17 with Personality, Trauma, and Economic problems as the only predictors accounting for a significant amount of variance in the model (see Table 10). These results suggest that as d economic problems increase d so did their willingness
85 to seek formal help for an internalizing problem. Also, as their beliefs in trauma decreased, they were more willing to seek formal help for an internalizing problem. Table 10 Multiple Regression f or Beliefs about Causes of Mental Health Problems and Willingness to Seek Formal Help for an Internalizing Problem Variable B SE B Constant 2.29 0.37 Physical Causes 0.01 0.07 0.01 Personality/Emotional Struggles 0.31 0.08 0.29*** Getting Along with Others 0.11 0.09 0.13 Trauma 0.24 0 08 0.22** Family/Parenting Issues 0.06 0.08 0.06 Friends 0.13 0.10 0.14 American Culture 0.08 0.08 0.09 Discrim ination/Prejudice 0.17 0.09 0.21 Economic Problems 0.23 0.09 0.27** Spiritual/Religious Reasons 0.01 0.07 0.01 Disharmony with Nature 0.04 0.08 0.04 Note. R 2 = .17 *** p < .001, ** p < .01 The second regression revealed a significant relationship between beliefs and willingness to seek formal help for an externalizing problem, F (11, 222) = 2.27, p < .01, R 2 = .10 with Economic problems as the only predictor accounting for a significant amoun t of variance in the model (see Table 11). These results suggest that the stronger parents believed that economic issues cause mental health problems the more willing they were to seek formal help for an externalizing problem.
86 Table 11 Multiple Regr ession for Beliefs about Causes of Mental Health Problems and Willingne ss to Seek Formal Help for an Ex ternalizing Problem Variable B SE B Constant 2.60 0.37 Physical Causes 0.0 5 0.07 0.0 5 Personality/Emotional Struggl es 0.1 0 0.08 0. 10 Getting Along with Others 0. 04 0.09 0. 05 Trauma 0. 06 0.08 0. 06 Family/Parenting Issues 0. 1 0 0.08 0. 11 Friends 0.10 0.10 0.1 2 American Culture 0. 12 0.08 0. 14 Discrimination/Prejudice 0.1 1 0.09 0. 13 Economic Problems 0.2 5 0.09 0. 31 ** Spiritual/Religious Reasons 0.0 5 0.07 0.0 5 Disharmony with Nature 0.06 0.08 0.0 6 Note. R 2 = .1 0 ** p < .01 The third regression revealed a non significant relationship between beliefs and willingness to seek in formal help for an in ternalizing problem, F (11, 22 7 ) = 1 33 p = 21. The fourth regression also revealed a non significant relationship between beliefs and willingness to seek informal help for an externalizing problem, F (11, 222) = 1.03, p = .42. A MANCOVA (Multivariate Analyses of Covariance) was used to explore the relationship between strength rating of each belief category, race and parent gender, covarying for SES and age given their known relationships with the factors. There was a significant effect of age F (11, 211) = 3 29 p < 001 parent g ender, F (11, 2 11 ) = 2 .33,
87 p < 01 and race, F (11, 211 ) = 4 .3 6 p < .001 on the 11 beliefs. Separate univariate ANOVAs (using corrected alpha of 0.01) revealed significant relationships between age and Disharmony beliefs, F (1, 221) = 7.67, p < .01. Additionally, significant relationships were found between race and P hysical Causes, F (1, 2 2 1) = 11 51 p < .0 0 1 partial 2 = .05; Spiritual Problems, F (1, 221) = 9.39, p < .01, partial 2 = .04, and Disharmony, F (1, 221) = 10.67, p < .001, partial 2 = .05. Specifically, Black parents had stronger beliefs ( M = 2.35) in Spiritual Causes than White parents ( M = 1.87) ; Black parents had stronger beliefs ( M = 1.73) in Disharmony beliefs than White parents ( M = 1.32) ; and White parents had stronger beliefs ( M = 4.18) in Physical Causes than Black parents ( M = 3.72) The relationship betwee n barriers (mean score) and strength rating of each belief category was examined with correlations. There w ere small negative correlation s between barriers and Physical Causes ( r = .14, p < .05), barriers and Personality/Emotional Struggles ( r = .17, p < .01), barriers and Trauma belief ( r = .18, p < .01), and barriers and Family/Parenting Issues ( r = .18, p < .01). These results suggest that parents who perceived more barriers endorsed weaker beliefs in physical, personality/emotional, trauma, and fam ily/parenting causes of mental health problems in children.
88 Discussion This study aimed to ex plore help seeking and service utilization patterns in Black and White mothers and fathers by examining their ability to recognize child mental health problems, exploring their beliefs about causes of mental health problems in children, and examining their perceived barriers to service utilization for their children Overall, over one third of parents in this sample had used mental health services in the past. Consistent with the literature, women and white parents had used services more often than men an d black parents (Mahalik, Good, & Englar Carlson, 2003; Sue & Chu, 2003). Similarly, a third of the children described in this study had used mental health services in the past 83% of whom utilized the mental health sector for these services. This menta l health utilization rate is much higher than expected given that previous research found higher rates of mental health service utilization from non mental health sectors including schools and primary care settings (Power et al. 2005). Of note, 62% of t he children who had utilized mental health services did so from more than one sector thus it is likely that these children may have utilized schools and/or primary care settings first or in tandem with the mental health sector. M ental health prevalence rates in this study were lower than national estimates given that 11% of parents reported having a current mental health problem and half of them report ed a strong likelihood of seeking help from a mental health professional for these problems Epidemiological r esearch (Kessler, Chiu, et al., 2005 ) described yearly
89 prevalence rates of mental health disorders at 2 6 % wi th only one third actually receiving services. Thus this study had fewer parents reporting mental health pr oblems and more parents indicating that they would seek help or were already utilizing services Given status was not assessed formally, it is likely that some parents would have met criteria for a mental health disorder but are unaware. Also as discussed in detail earlier in this paper, while help seeking intentions can be a strong predictor intentions rarely equate to actual service utilization due to numerous barriers Child mental health prevalence rates in this s tudy were reported at 18% and thus were slightly higher than the national estimates of 1 2 14 % ( Costello et al. 2005; Waddell et al. 2002 ) However, it is important to note that utilization estimates have varied significantly given that past research has estimated child service utilization rates as high as 21% (U.S. DHHS, 1999). In this study, a lmost 70% of parents who described having a child with a current mental health problem repor ted that they w ould seek help for thei r child. These reports are much higher than national estimates which show that less than half of children with a diagnosable mental health disorder receive services (U.S. DHHS, 1999) However, given the increase in service utilization rates for adults over the past 10 years (Kessle r, Chiu, et al., 2005) ; it is probable that child service utilization rates have also increased. Thus, more updated epidemiological research on child mental health utilization is needed and already underway (Kessler et al., 2009). The specific aims of thi s study were examined with 12 hypotheses. The first hypothesis accurately predicted that parents would be more likely to recognize the externalizing and internalizing vignettes as problem atic compared to the control vignette. This difference was quite la rge and indicated that our experimental process was
90 successful S pecifically in comparison to the control vignette, parents were 9.7 times more likely to recognize the internalizing vignette, and 14.5 times more likely to recognize the externalizing vign ette. Consistent with previous research ( e.g., Wu et al., 1999) problem recognition was slightly better with externalizing than internalizing problems although this difference was marginal This marginal difference is not surprising given inconclusive f inding s of previous reviews with respect to parental recognition of internalizing and externalizing problems (Zwaanswijk et al., 2003). It is also notable that parents who did recognize the vignettes were more confident in their ratings than parents who d id not recognize these problems. Th ese results mean that parents who recognize d the problem vignette s were confident that a problem was indeed present rather than just guessing about it Furthermore, Black parents were more confident in their ratings of externalizing symptoms than White parents. These results indicate that parents have a sense of their own abilities to recognize problem behaviors in children. The fact that Black parents were more confident in their ratings of externalizing problems indicates that they may need more guidance and specific strategies on how to recognize internalizing problems in their children. This information is relevant to intervention researchers; it wi ll be important to conduct qualitative research with parents who recognize problems in order to gain an understanding of their cognitive process in determining whether a set of symptoms is indeed problematic or not Although the majority of parent s in this study recognized the internalizing (52%) and externalizing (61%) vignettes, it is important to note that many parents still did not recognize these problems. Previous research ha s demonstrated s imilar patterns. In one study less than half of the parents with a child w ith a mental health disorder recognized a
91 problem (Sayal, 2006) Given that problem recognition is the gateway to help seeking (Cauce et al., 2002), it is crucial to understand the factors that contribut e to problem recognition in order to increas e help seeking intentions and mental health service utilization. The contribution of problem recognition to help seeking intentions was f urther e xplored in hypothesis 2 which accurately predicted that pa rents who recognized a problem would be mor e willing to seek help. Specifically, when parents recognized a n internalizing problem, they were 35 times more likely to seek help and when they recognized an externalizing problem, they were 69 times more likely to seek help. Therefore, as Cauce et al. (2002) described when parents recognize a problem, they tend to seek help. Furthermore, results from this study indicated that when parents recognized the vignette child as having an externalizing problem they almost always indicated that they would see k help which is also consistent with the literature (Wu et al., 1999). As the relationship between problem recognition and willingness to seek help was well established in this study, an understanding of the factors contribut ing to problem recognition and willingness to seek help was sought Interestingly, in several analyses the factors contributing to problem recognition were not always con gruent with the factors contributing to willingness to seek help. For instance, hypothesis 3, which predicted a higher likelihood of problem recogni tion and willing ness to seek help for the male rather than the female vignettes, was only partially supported. Speci fically, parents who recognized problems did so regardless of the gender of the child in t he vignette. This is consistent with Teagle (2002) who also found that child gender did not predict parental problem perception in children who met DSM IV criteria for psychiatric
92 diagnosis Given that the vignettes in th e current study were written to m eet criteria for DSM IV psychiatric diagnoses, it is probable that with more variability in psychiatric symptomatology like what occurs in the real world, child gender may actually have a significant relationship with problem recognition ; however examinati on of this relationship is beyond the scope of the current study In contrast child vignette gender did predict willingness to seek help but only for internalizing and not externalizing problems Specifically, parents were more willing to seek help for a male with an internalizing problem than a female with the same problem In their review, Zwaanswijk et al. (2003) reported that parents sought help more often for boys in young childhood and early adolescence; however, in late adolescence the reverse was found with parents seeking more help for girls Given that the child vignettes used in this study were all the same age of 10 years old th e current finding is consistent with previous research Thus, as hypothesized, we would expect that more parent s would be willing to seek help for the boy in the vignette than the girl in the vignette regardless of the problem type given the age of the child vignette T he fact th at th is relationship was not found for externalizing problems is likely attributable t o the perceived stronger impact of externalizing problems over internalizing problems Specifically, the previous hypotheses established that almost all parents who recognized a n externalizing problem were willi ng to seek help, th us there was less variab i l ity in responses G ender (as has been sho wn in hypothesis 3) did not affect this relationship. Additionally, internalizing problems are usually seen as typical problems for girls and more unusual for boys ( Rescorla et al., 2007; Zwaanswijk et al., 2003); this pattern
93 further increases the likelihood that more parents would be willing to seek help for a boy with internalizing problems than a girl with similar problems. Hypothesis 4 examined the relationship between perceived ne ed variables problem recognition and willingness to seek help Consistent with previous research (Power et al., 2005; Teagle, 2002) perceived severity was significantly related to i nternalizing problem recogni tion externalizing problem recognition, and willingness to seek help for an externalizing problem. Of note, perceived severity was not significantly related to internalizing help seeking intentions This finding is likely due to the s trength of the relation ship between internalizing problem recognition and willi ngness to seek help; such that when pa rents recognize d an internalizing problem (which in this study was more difficult to recognize than an externalizing problem), they we re willing to seek help rega rdless of how severe they perceived the problem to be Hypothesis 4 also revealed that perceived functional impairment was significantly related to externalizing problem recognition and willingness to seek help for an externalizing problem. Although pre vious research suggests that perceived functional impairment is related to both internalizing and externalizing problems (Power et al., 2005), t h e current finding further emphasizes the higher burden externalizing symptoms are perceived to have on parent al and child functioning Thus, even though both internalizing and externalizing vignettes in this study me t criteria for DSM IV diagnose s by clinician interpretation, parent own perce ptions of impairment apparently di f f ered, functioning than internalizing symptoms. This finding is especially relevant for intervention program development Specifically, parents will likely benefit from
94 psychoeducation on how, when, and wh at internalizing symptoms should be of concern to them Prevention should also be one of the main focus areas of these parental interventions, with researchers providing psychoeducation to parents about how internalizin g symptoms ten d to be less visible externally but how the symptoms can easily spiral downward with worsening impact if appropriate help is not sought early. O verall the finding s that parental perceived severity and functional impairment are significantly related to problem recognition and willingness to seek help (w hich in turn are related to mental health utilization ) emphasizes the importance of parental knowledge functioni ng. The n ull findings f rom hypothesis 5 between f amily characteristics problem recognition and willingness to seek help were surprising giv en previous research ( Cauce et al., 2002). T hese results are likely because the child vignettes in the current s tudy we re only simulated and not actually related to these parents lives or about their own children. Thus, family size, parent education, and parent psychopathology may be variables that impact problem recognition and help seeking intentions in real lif e situations but th ese relationship s could not be examined in the current study Additionally, this was a group of highly educated Black and White parents, with no differences found in years of education by race. Thus, it is probable that when differences in education are consistent across racial groups, some of the findings of dispar ities in mental health utilization and help seeking are neutralized. Results from h ypothesis 6 indicated that mothers recognized and were more willing to seek help for internalizing problems than fathers This finding is consistent
95 with previous resear ch that has established differing utilization rates for males and females with women utilizing services more often than men ( Koopmans & Lamers, 2007; Mahalik et al. 2003) These findings c ontribute to our understanding of why women and girls utilize men tal health service s more often than men and boys. Specifically, given the strong relationship s between service utilization problem recognition, and help seeking intentions, it is likely that mothers who are better at recognizing mental health problems t han fathers also use mental health services for themselves and their children more often seeking intentions and thus utilization. Hypothesis 6 also revealed that White parents were more likely to recognize internalizing problems than Black parents. Again this is consistent with past research that reported lower rates of service utilizat i on by Black individuals and children ( Roberts et al., 2005; Snowden & Yamada, 2005 ). This could be explained by differing problem thresholds that have been shown to vary across cultures/racial groups (Weisz & Eastman, 1995). Specifically, due to the more co vert nature of internalizing problems, it is likely that Black parents were less likely to recognize the internalizing vignette because they have different thresholds for classifying mental health problems In other words, if symptoms are not related to obvious impairment then the symptoms are not (Roberts et al., 2005). Furthermore, t he lack of finding s for race and gender differences with the externalizing vignette could be attributed to the strong relationship between problem recognition and willingness to seek help for an ext ernalizing problem. Specifically, the externalizing vignette in this study appeared to
96 have been perceived by m ost parents as severe and likely to cause significant functional impairment to the extent that race and parental gender did not impact this find ing. This explanation is reinforced by the fact that Black parents were more confident in their rating of externalizing symptoms but not internalizing symptoms (as shown in hypothesis one) Neither ethnic identity nor parental SES were significantly rela ted to problem recognition or willingness to seek help. Th ese findings are like ly attributable to the lack of variability among these variables. Specifically, even though Black parents scored higher on the ethnic id entity measure, within gr oup difference s were not found ( i.e. most Black parents score high on the measure ) Similarly SES was positively skewed with the majority of families fall ing in the minor business/ professional range and averaging a yearly income of $50,000 $65,000/year. Thus with mor e variability across participants differences may have been found. Another possibility that requires further exploration in future studies is that ethnic identity might not significantly impact problem recognition or help seeking intentions; however more variability in the measure will be needed to explore this question thoroughly. Other potential contributors to the relationship between problem recognition and willingness to seek help were investig ated in hypothese s 7 and 9 Specifically, analyses from hypothesis 7 showed that parents who attributed the cause of mental health problems to poor friendships and economic problems were more likely to recognize internalizing problems. Of note, hypothesis 9 revealed that mothers were more likely to attribute t he cause of mental health problems to poor friendships. These findings were contrary to what was expected given that friendships and economic problems are types of
97 sociological beliefs that were expected to decrease problem recognition especially given t hat they are less focused on individual /internal factors and more focused on environmental /external factors (Yeh et al., 2005) Thus, it appears that attribution of mental health problems to external causes actually increased the likelihood of internalizing problem recognition It is probable that parents (mostly mothers in this study) who attribute d the cause of mental health problems to these external factors ( poor friendship s and economic problems) are better able to recognize these problems because they themselves are removed from the situati on as potential causes. Alternatively, p arents who attribute mental health problems to internal factors, may engage in self blam e and may be unwilling to admit to being at least partially to blame for a mental health problems This idea is in line with previous vignette research by Raviv and professional mental health services than their own child (Raviv, Sharvit, R aviv, & Rosenblat Stein, 2009). Hypothesis 7 also revealed that parents who endorsed spiritual beliefs were more willing to seek informal help (religious leader s, family/friends, and self help) for both internalizing and externalizing problems. Of note, hypothesis 9 indicated that Black parents were more likely to attribute the cause of ment al health problems to spiritual reason s. These findings were as expected given previous research on spirituality beliefs and service utilization (Yeh et al., 2005). Additionally, research has frequently shown increased utilization of clergy by African Americans for mental health services (Ayalon & Young, 2005). P arental belief s in spiritual causes of mental health problems and increased willingness to seek help from a religious leader (among other informal
98 help sources) are inextricably linked Thus, i t makes sense that parents who attribute mental health problems to poor spiritual beliefs would seek help from their pastor. Additionally, it is likely that these parents (mostly Black in this study) are more open to alternate sources/providers of mental health services due to their own hesitation/perceived barriers to using formal mental health services. Alternatively parents who endorse spiritual beliefs ma y have a stronger affinity for informal help seeking because formal services do n o t usually address spiritual beliefs in standard care H owever there is emerging evidence that indicat es the importance of collaboration with religious leaders in order to increase service utilization (Milstein, Manierre, Susman, & Bruce, 2008; Wang, Berglund, & Kessler, 2003). The relationship s between professional/personal mental health experience and p arental p roblem recognition were examined in hypothesis 8. The f inding that previous experience s with anxiety and mental health ( in general ) w ere significantly related to internalizing problem recognition was not surprising; however, the null finding for externalizing problem recognition w as unexpected. Of note, hypot hesis 9 revealed that both mothers and White parents had more previous professional/personal mental health experience than fathers and Black parents. Previous research indicated that knowledge of mental health problems and associated treatments increase h elp seeking intentions and service utilization (Power et al. 2005). Thus, the fact that externalizing problem recognition was not significantly related to previous mental health experience in this study indicates that other factors such as severity and f unctional impairment (see hypothesis 4) are the primary factors that influence externalizing problem recognition. It appears that (at least as assessed in this study), externalizing problems were so overt and
99 easily recognizable by most parents that havin g previous experience did not i ncrease the odds of recognizing these problems However, given the covert nature of internalizing problems, prior knowledge d id contribute to recognition of these clusters of symptoms. These findings again emphasize the imp ortance of parental psychoeducation about mental health symptoms especially internalizing problems which appear to be harder to recognize. The finding that mothers and White parents were more likely than fathers and Black parents to report previous pro fessional/personal mental health experience s could help explain some of the variability in problem recognition and help seeking intentions especially with internalizing problems Alternatively this finding could also mean that White parents and mothers w ere simply more comfortable or less concerned about reporting their previous professional/personal mental health experiences in this study However, given that Black individuals have lower rates of mental health disorders than White individuals (Harris, E dlund, & Larson, 2005 ; Sue & Chu, 2003 ) it is more likely that less exposure to and experience with mental health illness is contributing to some of the differences in problem recognition and help seeking intentions found between Black and White parents. Post hoc analyses indicated that th e relationship s between pr evious mental health experience, problem recognition, and help seeking intentions w ere also found among parents whose own children had used mental health services in the past. Specifically, parents of children who had used mental health services in the past were more likely to recognize an internalizing problem, more willing to seek help f or both internalizing and externalizing problems, and had stronger intentions to use mental health services for their
100 child ren in the future. Thus prior knowledge/experience increases problem recognition and help seeking intentions which in turn sh ould i ncrease mental health service utilization and would be another prime area for intervention researchers to target (i.e. increasing parental knowledge/experience with child mental health ; Power et al., 2005 ). Examination of t he Barriers to Utilization measu re revealed significant their children with fathers endorsing significantly more barriers than mothers. This is consistent with previous research indicating lower ra tes of mental health service utilization in m en compared to women (Mahalik et al., 2003), decreased perception of the need for therapy in fathers compared to mothers ( Phares, Rojas, Thurston, & Hankinson, 2010 ), decreased interest in child related therapy in fathers compared to mothers (Duhig et al., 2002), children ( Phares et al., 2006) to service utilization for their children is likely underl ying some of the gender differences in problem recognition and willingness to seek help In other words, it is likely that because fathers perceive more barriers to service utilization, they have less experience and thus are less able to recognize interna lizing problems and less willing to seek help for these problems. There were no race differences in overall perception of barriers which was unexpected. Specifically, previous research using separate measures of barriers to service utilization and attitu des towa rd service utilization found race differences in perception of barriers but not attitudes; and gender differences in attitudes toward mental health treatment but not perception of barriers ( Thurston & Phares, 2008). The lack of
101 findings could be a ttributed to the measure s used in th ese stud ies Specifically, Hypothesis 10 explored the relationships among three subsets of barriers (financial, structural, and attitudinal) as assessed with the Barriers to Utilization measure. Of note, internal consistency reliabilities were poor for the financial and structural subscales and acceptable for the attitudinal subscale. Furthermore, strong correlations were found among the subscales. Th ese patterns lead to the possibility that th e Barriers to Utili zation measure may not have been a good assessment tool for both structural and attitudinal barriers but rather w as measuring the same underlying concept. Furthermore, given that the re are additional barriers that are unique to individuals of racial and e thnic minority groups ( including mistrust and fear of treatment, racism/discrimination, differences in language and communication a nd cultural barriers in general ; Thompson et al. 2004) it will be important to identify/develop measures that take into ac count the unique experiences of racial and ethnic minorities in their own help seeking process Thus examination of the impact of attitudinal and structural barriers would need to be reassessed in future studies with more stable measures. Hypothesis 11 revealed that parents who were more willing to seek help ( either from formal or informal sources) were willing to do so regardless of the type of problem being experienced by the child However, parents who sought informal help for internali zing problems almost always did so for externalizing problems as well and vice versa (i.e. parents who sought formal help were just as likely to do so for both internalizing and externalizing problems). This alludes to the possibility that source of help seeking is consistent across parents regardless of problem type. However, this study also showed a strong co nnec tion between formal and informal help seekers. Thus,
102 seeking informal help i s tied to the likelihood of seeking formal help. As researchers, our goal is to increase parental help seeking intentions, which should then increase utilization of mental health services for children. Given such a strong relationship between formal and informal help seeking intentions, intervention research ers should target their interventions on two levels. First, identify the help seekers from the non help seekers and creat e separate interventions for each of these groups of parents. Second, among the help seekers, it will be important to establish interventions fo r those who are choosing to seek informal help even in situations that require formal help such as a child meeting criteria for a DSM IV diagnosis. Psychoeducation will also be helpful to teach parents how to recognize when formal versus informal help sho uld be sought (Stiffman et al., 2004) Additionally, informal help sources such as religious leaders should be engaged in the process of increasing formal mental health service utilization either through increasing their own awareness of when formal versu s informal help should be sought, providing on site clinicians to provide treatment when needed and/or maintaining an updated referral list with community clinicians who can provide formal services to these families. The C.O.P.E (Clergy Outreach and Prof essional Engagement) program was developed to facilitate and increase the collaboratio n between clinicians and clergy (Milstein et al., 2008); this and other strategies (see Harris, Edlund, & Larson, 2006) can be utilized to promote and increase mental health utilization in families who would ordinarily prefer informal help sources Consistent with previous research, p arents who perceived more barriers to service utilization were less willing to seek formal help for externalizing problems (Kekorian et al ., 2006; Sareen et al., 2007). The fact that this relationship was not found for the
103 internalizing problem could be attributed to parents being mo re open to seek help from any available source for internalizing symptoms; t hus there was no significant imp act on their formal help seeking However, given that externalizing problems are viewed as more impactful, parents are more likely to seek formal help sources for these symptoms thus making them more susceptible to increased barrier perception. Hypothe sis 12 examined data from parents who had at least one child with a current mental health problem and revealed that neither perceived barriers, ethnic future for t heir children Given that previous research has found relationships among these factors (Thurston & Phares, 2008; Yeh et al., 2005) it is likely that the small sample size (45 parents) may have resulted in in sufficient power to predict this relationship. However, it was notable that most parents (69%) who reported that their c hild had a current mental health problem reported being quite or extremely likely to seek help. This finding is likely inflated by social desirability and the fact that intentions to use services were being assessed as opposed to actual service utilization. Additionally, consistent with previous research ( Garland Haine & Boxmeyer 2007; Hawley & Weisz, 2005) parental satisfaction with p revious mental health services w as related t o higher intentions to use services for children in the future Thus, clinician s should not only assess satisfaction with their own provided services but also examine satisfaction with previous services. Post hoc analyses examining relationships among strength s of beliefs revealed several patterns. Specifically, parents were more willing to seek formal help for internalizing problems as their beliefs in personality /emotional and economic causes of
104 child mental health problems increased and beliefs in t rauma causes decreased. Alternatively, parents were more willing to seek formal help for externalizing problems as their beliefs in economic causes of child mental health problems increased. Thus consistently, parents who believed that mental health prob lems were related to economic difficulties were more open to seeking formal help. This pattern is likely because economic causes are external factors, thus perceived by parents as outside of their immediate control which emphasizes the need for formal he lp. Although personality/emotional struggles are more internal factors (i.e. factors within the child), the specific items (s discipline, self control, or difficulties with ng through a certain age or developmental stage ) are more related to transient factors that are probably perceived by parents as malleable and tied to internal struggles, thus parents are able to see the potential b enefit of seeking help from a formal sour ce Results also indicated that B lack parents had stronger spiritual and disharmony with nature beliefs than White parents wh er e as White parents had stronger physical cause beliefs than Black parents. This finding helps further explain some of the race differences found in sources of help seeking. Specifically, given that Black parents tend to attribute child mental health problems to spiritual reasons their tendency to seek help from clergy is reasonable Additionally, White parents tendency to seek help from formal sources more often is also reasonable given that they most commonly attribute mental health problems to physical causes. This information will be relevant to intervention research ers and clinicians. Specifically, these findings encourage the use of
105 differing strategies to engage and maintain Black and White parents in treatment based on their beliefs (Yeh et al., 2005). Finally, there was a significant relationship between barrier endorsement and b eliefs in physical, personali ty/emotional, trauma, and family/parenting causes (biopsychosocial beliefs ) of mental health problems in children. Given that the Barriers to Utilization measure examined barriers to formal mental health service utilization, it is not surprising that thos e parents who perceived more barriers also had weaker beliefs in the biopsychosocial causes of mental health problems in children. Thus, it will be pertinent for clinicians and researchers attempting to increase service utilization to examine beliefs in a ddition to barriers to service utilization, as the type s of beliefs parents endorse can help guide the types of referrals that are given. Specifically, since the goal is to increase utilization and results from this study have shown that informal help see king is related to formal help seeking, one of the importan t aspects of increasing utilization appears to be getting parents to seek some kind of help first. Then their first help seeking source (i.e. pediatrician, teachers, pastor, et c) c ould assist the parents with getting into the specific type of service that would be most effective for their child Given emerging research on the adapt ation of evidence based treatments to minorities (Horrell 2008), th e findings about the varying beliefs o f parents in this study will also be useful in adapting treatments further to meet the needs of minorities and thus increase and maintain treatment utilization. Limitations and Future Research Like all studies, t his study had some limitations which can be used to inform future resea r ch First, as with all other cross sectional studies, this study only assessed a
106 (Field, 2009 ; Teagle, 2002 ) Future studies should consider longitudinal data collection in order to fully examine the impact of problem recognitio n and willingness to seek help i n relation to child mental health utilization. The g eneralizability of this study is also limited given that participants were not samp led nationally. More representative national sampling could be a goal of future research projects on this topic Additionally, the sole use of parental self report measures i s a study lim itation given that parents may have be e n influenced by social desir ability among other factors thus leading to the underreporting of information Future projects could strive to obtain data from multiple sources to further support study findings. This study utilized vignettes to provide some control of child characteri stics and to However, because the v ignette used in this study focus ed on a 10 year old child, we were unable to examine the impact of age on problem recognition and willingness to seek help F uture studies should use multiple age ranges such as early childhood (especially given the dearth of research in this area), school age, and adolescence. Additiona lly, v ignette descriptions used in this study all met criteria for DSM IV diagnoses which was necessary to allow for examination of d ifferences and similarities between clinically assess ed need and parental perceived need. However, this practice limited t he ability to assess parental threshold differences of the various levels at which parents recognize clinical symptoms and begin to consider informal and formal help seeking. More variability in psychiatric symptomatology would a lso a llow for a more detail ed exploration of the impact of child gender on problem recognition (Zwaanswijk et al., 2003) Thus, future research should
107 explore mild, moderate, and severe internalizing and externalizing vignettes in order to better understand how sub threshold sympto matology may impact problem recognition and help seeking intentions. As noted above, t here were some limitations with several of the measures. Specifically, the Beliefs About Causes measure did not focus on causes of mental health problems for specific mental health diagnoses (i.e. parents were asked about their beliefs about causes of mental health problems in general, rather than beliefs about causes of anxiety, ADHD, etc). It is possible that the type of mental illness may impact what parents percei ve to be caus ing these symptoms (Yeh et al., 2005) Thus it would be beneficial for future r esearch to target specific mental health problems and examine beliefs about causes of each specific mental health problem along with barriers to seeking services for that specific problem. Both the Multigroup Ethnic Identity Measure (MEIM) and the Hollingshead SES scores had poor variability. This limited variability likely contributed to several null findings in this study. Future research examining these facto rs should recruit a wide range of participants at various stages on the MEIM and at various levels of SES to better assess the impact of these variables on problem recognition, help seeking intentions, and mental health service utilization. Poor variabili ty in Ba rriers to Utilization measure also resulted in an inability to properly assess differ ences among the subscales of financial, structural, and attitudinal barriers. The Barriers to Utilization measure was the most comprehensive measure available at the beginning of the study but f uture studies should try to develop more comprehensive and established measures that separately examine attitudes and structural barriers.
108 On the questions following the vignette measure, the h igh correlations between sev erity and problem threshold, a s well as functional impairment and family impact had an effect of our analyses. Specifically, these patterns limited the extent to which the effects of all four variables on problem recognition and help seeking could be exam ined. Future studies should provide more clear and succinct operational definitions for each of these concepts to avoid overlap and assessment of the same underlying concepts (Power et al., 2005) There were some limitations with the s mall sample size o f parents who had children with mental health difficulties ; thus examining the various factors that were hypothesized to be associated with future mental health service utilization was difficult (Singh, 2003) Future studies should oversample for parents w ho have children w ith mental health problems in order to fully explore the relationship between ethnic identity, barriers, and beliefs on future intentions to utilize mental health services for children. Finally, although there are numerous advantages to quantitative research, there are also some limitations. Specifically, due to t he quantitative nature of this study, it was impossible to determine what those parents w ho recognize d symptoms did differently or similarly to parents who did not recognize sy mptoms. The process of problem recognition and help seeking intent was lost in the quantitative nature of this study. The richness of qualitative research might allow for exploration of some of these underlying factors that c ontribut e to the end goal of recognition and help seeking (Bussing, Zima, Gary, & Garvan, 2003; Singh, 2003) Future studies should explore qualitative research options with parents who recognize symptoms to help understand their decision making process in determining whether a set o f symptoms is indeed problematic or not Results
109 from that type of investigation could be used to create psychoeducational interventions to teach other parents these skills. C linical Significance and Conclusion s Guided by several theoretical models, this study set out to explore factors related to hel p seeking and service utilization in African American and Caucasian mothers and fathers based on their ability to recognize problems, their perception of barriers, and their beliefs. There are several implic ations of this study both for intervention researchers and practitioners alike. First, despite the finding that many parents are able to recognize severe problem behaviors when given a hypothetical situation, there were several parents who did not recogn ize these problems. Thus, there is a need for interventions aimed at increasing problems. This finding underscores the importance of psychoeducation and empowering parents with the skill set to be able to recognize problem atic behavior s (especially internalizing problems) and seek appropriate help when needed. The knowledge that problem recognition often leads to willingness to seek help, further underscores the imp ortance of increasing problem recognition skills in parents. Second, clinicians and researchers alike need to think more broadly than solely considering treatment in a typical outpatient setting where patients come to the clinics to receive services. Current and past research has shown that youth and parents are often reluctant to attend outpatient appointments and are receiving mental health treatment from non mental h ealth specific sites. Thus, rather than insisting parents and youth come to outpatient providers, we may need to develop strategies to bring the services to them.
110 Furthermore, given all the barriers and obstacles parents perceive around mental health car e, typical mental health settings are not sufficient to provide services to parents across race and gender groups. Thus, more innovative programs should be used, such as the for the Development of Attention and Readiness (CUIDAR) early intervention parent training program which begin s at an early age and is innovatively set up to be more easily accessible to minority families (Lakes et al., 2009). Additionally, providing the rapeutic services at primary care provider offices ( Brown, Wissow, Zachary, & Cook, 2007; Lieberman, Adalist Estrin, Erinle, & Sloan, 2006), providing high quality clinical services at school (Owens & Murphy, 2004), and utilizing clergy to refer and/or ini tiate therapeutic services ( Milstein et al., 2008 ) are promising ways to initiate and maintain service utilization. Given that parents often do not bring up primary care doctors, clergy or school staff (Briggs G owan, Horwitz, Schwab Stone, Levanthal, & Leaf, 2000), interventionists should also target these service providers and teach problem recognition skills and ways to assess and respond to mental health problems in children Research comparing adult mental h ealth services in primary care versus specialty mental health settings found that patients seen in primary care settings tend to have fewer visits and treatment is often less effective than in specialty mental health centers (Uebelacker, Wang, Berglund, & Kessler, 2006). Additionally, more severely mentally ill individuals tend to receive services from specialty mental health rather than primary care settings (Druss & Rosenheck, 2000; Uebelacker et al., 2006). However, an innovative mental health delivery program in Canada found effective results of treatment when they brought
111 clinicians and psychiatrists to primary care offices and delivered services there (Kates, Crustolo, Farrar & Nikolaou 2002). Similar programs can be used here in the U.S. by clerg y, primary care providers, and schools in order to maintain effectiveness of treatment while allowing easy access to services. Unfortunately, research comparing child mental health services across settings (primary care, specialty mental health, and schoo l) was not found. Thus, more research is needed on the quality and effectiveness of child mental health services across settings. Third, when attempting to address help seeking and service utilization disparities; all contributing factors need to be taken into account. The stronger and more lasting interventions will likely be those that target multiple contributing factors including perceived need variables, family characteristics, demographic variables, beliefs, and experience as show n in the study mode l. Targeting these interventions and providing psychoeducation on the availability of efficacious treatments for both internalizing and externalizing symptoms should work in tandem to increase service utilization. G iven th at study variables interact with externalizing and internalizing symptoms differently, mothers and fathers differently, and ethnic minorities and white parents differently, it is worth considering separate models and thus separate interventions aimed at addressing these disparities (Cauce et al., 2002; La Greca, Silverman, & Lockman, 2009). Specifically, parents appear to need more guidance around recognizing internalizing symptoms in particular, and although not assessed in this study, parents will also likely benefit from understa nding subclinical symptoms and when referral to formal versus informal help sources should occur. Additionally, with respect to gender parents need to be encouraged to seek help for girls even when they have internalizing symptoms rather
112 than ignor ing th girls being girls. Fathers are at a disadvantage when it comes to problem recognition and help seeking for internalizing symptoms, thus preventive interventions should target increasing skills in these areas and addressing barriers that fathers perceive to limit access to service utilization for their children. Similarly with respect to race differences Black parents need more guidance on how to recognize internalizing problems. Furthermore, given their li mited experience with child mental health problems, Black parents will need even more guidance on how to recognize problems an d where to seek help when problems occur in their children. As Black parents are more comfortable going to thei r spiritual adviso rs for help, educational efforts should target clergy to support and provide referrals to parents when specialized mental health services are needed. The clinical implications of having multiple targeted models and interventions based on type of problem b ehaviors, parental gender, and parental race/ethnicity are significant because the se models will guide the way we provide services to youth and their families, refer them to providers, and teach them how to seek out services for themselves. In conclusion given that many lifetime mental health problems emerge in childhood and adolescence (Kessler, Berglund, et al., 2005), addressing factors that contr ibut e to mental health utilization in childhood is crucial. Previous r esearch describes parents a s the ga te keepers to child mental health service utilization ( Stiffman et al., 2004) and th is study showed a significant difference between parent perceived need and clinically assessed need Thus, a s clinicians and researchers, we must take these two sides in to account when creating interventions and providing therapeutic services. Unless parents are able to perceive a need for services, they will neither initiate
113 nor maintain treatment participation. Given that problem recognition and help seeking intention s are strongly related and our goal is to increase service utilization, t he good news is that there are several places along the help seeking process where we can intervene. Although we should aim to address all contributing factors improving parental p r oblem recognition appears to be a good place to start
114 References Alegria, M., Canino, G., Lai, S., Ramirez, R. R., Chavez, L., Rusch, D., care use. Medical Care, 42, 447 455. doi:10.1097/01.mlr.0000124248.64190.56 Alvidrez, J. (1999). Ethnic variations in mental health attitudes and service use among low incom e African American, Latina, and European American young women. Community Mental Health Journal, 35, 515 530. doi:10.1023/A:1018759201290 American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders DSM IV TR (4 th ed.). Was hington, DC: Author. Andersen, R. M. (1995). Revisiting the behavioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36, 1 10. doi:10.2307/2137284 Andersen, R. M., & Newman, J. F. (1973). Societal and individual determinants of medical care utilization in the United States. Milbank Memorial Fund Quarterly Health and Society 51 (1) 95 124. Retrieved from http://www.jstor.org/stable/3349613 Angold, A., Messer, S. C., Stangl, D., Farmer, E. M., Costello, E. J., & Burns, B. J. (1998). Perceived parental burden and service use for child and adolescent psychiatric disorders. American Journal of Public Health, 88, 75 80. doi:10.2105/AJPH.88.1.75
115 Avery, D. R., Tonidandel, S ., Thomas, K. M., Johnson, C. D., & Mack, D. A. (2007). Assessing the multigroup ethnic identity measure for measurement equivalence across racial and ethnic groups. Education and Psychological Measurement, 67, 877 888. doi :10.1177/0013164406299105 Ayalon, L ., & Young, M. A. (2005). Racial Group Differences in Help Seeking Behaviors. Journal of Social Psychology 145, 391 403. doi: 10.3200/SOCP.145.4.391 404 Barkley, R. A. (1997). training (2 nd ed.). New York, NY: Guilford Press. Bickman, L., Heflinger, C. A., Northrup, D., Sonnichsen, S., and Schilling, S. (1998). Long term outcomes to family caregiver empowerment. Journal of Child and Family Studies, 7, 269 282. doi:10.1023/A:1022937327049 B riggs Gowan, M. J., Horwitz, S. M., Schwab Stone, M. E., Leventhal, J. M., & Leaf, P. J. (2000). Mental health in pediatric settings: Distribution of disorders and factors related to service use. Journal of the American Academy of Child & Adolescent Psychi atry 39 841 849. doi:10.1097/00004583 200007000 00012 Broadhurst, K. (2003). Engaging parents and carers with family support services: What can be learned from research on help seeking? Child and Family Social Work, 8, 341 350. doi:10.1046/j.1365 2206.2 003.00289.x Brown, J., Wissow, L., Zachary, C., & Cook, B. (2007). Receiving advice about child mental health from a primary care provider: African American and Hispanic parent attitudes. Medical Care 45 1076 1082. doi:10.1097/MLR.0b013e31812da7fd
116 Busch, S. H., & Horwitz, S. M. (2004). Access to mental health services: Are uninsured children falling behind? Mental Health Services Research, 6, 109 115. doi:10.1023/B:MHSR.0000024354.68062.e8 Bussing, R., Schoenberg, N. E., & Perwien, A. R. (1998). Knowledge and information about ADHD: evidence of cultural differences among a frican american and white parents. Social Science Medicine, 46, 919 928. doi:10.1016/S0277 9536(97)00219 0 Bussing, R., Zima, B. T., Gary, F. A., & Garvan, C. W. (2003). Barriers to detec tion, help seeking, and service use for children with ADHD symptoms. The Journal of Behavioral Health Services & Research 30 176 189. doi:10.1007/BF02289806 Bussing, R., Zima, B. T., Gary, F. A., Mason, D. M., Leon, C. E., Sinha, K., C. W. (2003). Social networks, caregiver strain, and utilization of mental health services among elementary school students at high risk for ADHD. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 842 850. doi:10.1097/01.CHI.0000046876.272 64.BF Cauce, A.M., Coronado, N., & Watson, J. (1998). Conceptual, methodological, and statistical issues in culturally competent research. In M. Hernandez & M. R. Isaacs (Eds.), (pp. 305 32 9). Baltimore, MD: Paul H. Brookes Publishing Co. Cauce, A. M., Domenech Rodriguez, M., Paradise, M., Cochran, B. N., Shea, J. M., Srebnik, D., (2002). Cultural and contextual influences in mental health help seeking: A focus on ethnic minorit y youth. Journal of Consulting and Clinical Psychology, 70, 44 55. doi:10.1037/0022 006X.70.1.44
117 Chandra, A., & Minkovitz, C. S. (2006). Stigma starts early: Gender differences in teen willingness to use mental health services. Journal of Adolescent Health 38 754.e1 754.e8. doi:10.1016/j.jadohealth.2005.08.011 Chow, J. C., Jaffee, K., & Snowden, L. (2003). Racial/ethnic disparities in the use of mental health services in poverty areas. American Journal of Public Health, 93, 792 797. doi:10.2105/AJPH.93.5. 792 Cohen, J. (2003). A power primer. In A. Kazdin (Ed.), Methodological Issues & Strategies in Clinical Research (3rd ed.) (pp. 427 436). Washington, DC: American Psychological Association. Costello, E. J., Egger, H., & Angold, A. (2005). 10 Year Researc h Update Review: The epidemiology of child and adolescent psychiatric disorders: I. Methods and public health burden. Journal of the American Academy of Child & Adolescent Psychiatry 44 972 986. doi:10.1097/01.chi.0000172552.41596.6f Dobalian, A. & Rive rs, P. A. (2008). Racial and ethnic disparities in the use of mental health services. The Journal of Behavioral Health Services & Research, 35, 128 141. doi:10.1007/s11414 007 9097 8 Duhig, A. M., Phares, V., & Birkeland, R. W. (2002). Involvement of fathers in therapy: A survey of clinicians. Professional Psychology: Research and Practice, 4, 389 395. doi:10.1037/0735 7028.33.4.389 Druss, B. G., & Rosenheck, R. A. (2000). Locus of mental health treatment in an integrated service system. Psychiatric Services 51 890 892. doi:10.1176/appi.ps.51.7.890
118 Fals Stewart, W., Fincham, F. D., & Kelley, M. L. (2004). Substance abusing parents A comparison of mothers versus fathers. Journal of Family Psychology, 18, 666 671. doi:10.1037/0893 3126.96.36.1996 Farmer, E. M., Burns, B. J., Phillips, S. D., Angold, A., & Costello E. J. (2003). Pathways into and through mental health s ervices for children and a dolescents Psychiatric Serv ices 54 60 66. doi:10.1176/appi.ps.54.1.60 Field, A. (2009). Discovering Statistics Using SPSS (3 r d ed.). Thousand Oaks, CA: Sage Flisher, A. J., Kramer, R. A., Grosser, R. C., Alegria, M., Bird, H. R., Bourdon, K. H., Hoven, C. W. (1997). Correlates of unmet need for mental health services by children and adolescents. Psychological Medicine, 27, 1145 1154. doi:10.1017/S0033291797005412 Gary, F. A. (2005). Stigma: Barrier to mental health care among ethnic minor ities. Issues in Mental Health Nursing, 26, 979 999. doi :10.1080/01612840500280638 Garland A.F., Haine R.A. & Boxmeyer C.L. (2007) Determ ina t e s of youth and parents satisfaction in usual care psychotherapy. Evaluation and Program Planning 30 45 55. doi:10.1016/j.evalprogplan.2006.10.003 Glied, S. & Cuellar, A. E. (2003). Trends a nd issues i n c hild a nd a dolescent mental h ealth Health Affairs, 22, 39 50. doi: 10.1377/hlthaff.22.5.39 Harris, K. M., Edlund, M. J., & Larson, S. (2005). Racial and ethnic differences in mental health problems and use of mental health care. Medical Care, 43, 775 784. doi:10.1097/01.mlr.0000170405.66264.23
119 Harris, K. M., Edlund, M. J., & Larson, S. L. (2006). Religious involvement and the use of mental health care. Health S ervices Research, 41, 395 410. doi: 10.1111/j.1475 6773.2006.00500.x Hawley, K. M., & Weisz, J. R. (2005). Youth versus parent working alliance in usual clinical care: Distinctive associations with retention, satisfaction and treatment outcome. Journal of Clinical Child and Adolescent Psychology, 34, 117 128. doi: 10.1207/s15374424jccp3401_11 Hofferth, S. L., Stueve, J. L., Pleck, J., Bianchi, S., & Sayer, L. (2002). The demography of fathers: What fathers do. In C.S. Tamis LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary Perspectives (pp. 63 90). Mahwah, NJ: Lawrence Erlbaum Associates. Hollingshead, A. B. (1975). Four Factor index of social status. New Haven, CT: Yale University, Department of Sociology. Horrell, S. C. (2 008). Effectiveness of Cognitive Behavioral therapy with adult ethnic minority clients: A review. Professional Psychology: Research and Practice 39, 160 168. doi: 10.1037/0735 7028.39.2.160 Kataoka, S. H., Zhang, L., & Wells, K. B. (2002). Unmet need for mental health care among U.S. children: Variation by ethnicity and insurance status. American Journal of Psychiatry, 159, 1548 1555. doi:10.1176/appi.ajp.159.9.1548 Kates, N., Crustolo, A., Farrar, S., & Nikolaou, L. (2002). Counsellors in primary care: b enefits and lessons learned. The Canadian Journal of Psychiatry / La Revue canadienne de psychiatrie 47 (9) 857 862. Retrieved from PsycINFO database.
120 Kerkorian, D., McKay, M., & Bannon, W. M. (2006). Seeking help a second time: cterizations of previous experiences with mental health services for their children and perceptions of barriers to future use. American Journal of Orthopsychiatry, 76, 161 166. doi :10.1037/0002 94188.8.131.52 Kessler, R. C. Avenevoli, S., Green, J., Gruber M. J. Guyer, M., He, Y., K. R. (2009). National Comorbidity Survey Replication Adolescent Supplement (NCS A): III. Concordance of DSM IV/CIDI diagnoses with clinical reassessments. Journal of the American Academy of Child & Adolescent Psychiatry 48 386 399. doi:10.1097/CHI.0b013e31819a1cbc Kessler, R. C., Berglund, P. A., Bruce, M. L., Koch, J. R., Laska, E. M., Leaf, P. J., Wang, P. S. (2001). The prevalence and correlates of untreated serious mental illness. Health Services Resear ch, 36 (6) 987 1007. Retrieved from Academic Search Premier database. Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walter s E. E. (2005). Lifetime prevalence and age of onset distributions of DSM IV disorders in the national comorbidity survey replication. Archives of General Psychiatry, 62, 593 602. doi:10.1001/archpsyc.62.6.593 Kessler, R. C., Chiu, W. T., Demler, O., & Wal ters, E. E. (2005). Prevalence, severity, and c omorbidity of 12 Month DSM IV d isorders in the national comorbidity survey r eplication Archives of General Psychiatry, 62, 617 627. doi:10.1001/archpsyc.62.6.617
121 Kessler, R. C. Dem l er, O., Frank, R. G. Olfs on, M., Pincus, H. A. Walters, E. Zaslavsky, A. M. (2005). Prevalence and t reatment of m ental d isorders, 1990 to 2003. The New England Journal of Medicine 352 2515 2523. doi:10.1056/NEJMsa043266 Kessler, R.C., & Wang, P.S. (2008). The descriptive epidemiology of commonly occurring mental disorders in the United States. Annual Review of Public Health, 29, 115 129. doi: 10.1146/annurev.publhealth.29.020907.090847 Koopmans, G. T., & Lamers, L. M. (2007). Gender an d health care utilization: The role of mental distress and help seeking propensity Social Science and Medicine, 64, 1216 1230. doi: 10.1016/j.socscimed.2006.11.018 Kropf, M. E., & Blair, J. (2005). Eliciting survey cooperation: Incentives, self interest, and norms of cooperation. Evaluation Review, 29, 559 575. doi :10.1177/0193841X05278770 La Greca, A. M. Silverman, W. K. & Lochman, J. E. (2009). Moving beyond efficacy and effectiveness in child and adolescent intervention research. Journal of Consulting and Clinical Psychology 77 373 382. doi:10.1037/a0015954 Lakes, K. D., Kettler, R. J., Schmidt, J., Haynes, M., Feeney Kettler, K., Kamptner, L., preschoolers at risk for beha vioral disorders: An innovative practice for reducing disparities in access to service. Journal of Early Intervention 31 167 178. doi:10.1177/1053815109331861. Lamb, M. E. (Ed.) (20 10 ). The role of the father in child development ( 5 th ed.). Hoboken, NJ : Wiley
122 Lambert, M. C., Weisz, J. R., Knight, F., Desrosiers, M., Overly, K., & Thesiger, C. (1992). Jamaican and American adult perspectives on child psychopathology: Further exploration of the threshold model. Journal of Consulting and Clinical Psychology 60 146 149. doi:10.1037/0022 006X.60.1.146 Lieberman, A., Adalist Estrin, A., Erinle, O., & Sloan, N. (2006). On site mental health care: A route to improving access to mental health services in an inner city, adolescent medicine clinic Child: Care, Health & Development 32 407 413 doi:10.1111/j.1365 2214.2006.00620.x Mackenzie, C. S., Gekoski, W. L., & Knox, V. J. (2006). Age, gender, and the underutilization of mental health services: The influence of help seeking attitudes. Aging and Mental Healt h, 10, 574 582. doi :10.1080/13607860600641200 Mahalik, J. R., Good, G. E, & Englar Carlson, M. (2003). Masculinity scripts, presenting concerns, and help seeking: Implications for practice and training. Professional Psychology: Research and Practice 34, 1 23 131. doi :10.1037/0735 7028.34.2.123 McLoyd. V. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185 204. doi:10.1037/0003 066X.53.2.185 McMiller, W. P. & Weisz, J. R. (1996). Help seeking preceding mental health clinic intake among African American, Latino, and Caucasian youths. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1086 1094. doi:10.1097/00004583 199608000 00020
123 Milstein, G., Manierre, A., Susman, V. L., & Bruce, M. L. (2008). Implementation of a program to i m prove the continuity of mental h ealth care t hrough Clergy Outreach and Professional Engagement (C.O.P.E.) Professional Psychology: Research and Practice 3 9, 218 228. doi: 10.1037/0735 7028.39.2.218 Owens, P. L., Hoagwood, K., Horwitz, S. M., Leaf, P. J., Poduska, J. M., Kellam, S. G., Journal of the American Academy of Child and Adolesc ent Psychiatry, 41, 731 738. doi:10.1097/00004583 200206000 00013 Owens, J. S., & Murphy, C. E. (2004). Effectiveness research in the context of school based mental health. Clinical Child and Family Psychology Review 7 195 209. doi:10.1007/s10567 004 608 5 x Pavuluri, M. N., Luk, S., & McGee, R. (1996). Help seeking for behavior problems by parents of preschool children: A community study. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 215 222. doi:10.1097/00004583 199602000 00015 Pedhazur, E. J. (1997). Multiple Regression in Behavioral Research: Explanation and Prediction (3 rd ed.). Orlando, FL: Harcourt Brace College Publishers. Phares, V., Fields, S., & Binitie, I. (2006). G etting fathers involved in c hild related t herapy Cognitive and Behavioral Practice, 13, 42 52. doi:10.1016/j.cbpra.2005.06.002 Phares, V., Fields, S., Kamboukos, D., & Lopez, E. (2005). Still looking for Poppa. American Psychologist 60 735 736. doi:10.1037/0003 066X.60.7.735
124 Phares, V., Lopez, E., Fie lds, S., Kamboukos, D., & Duhig, A. M. (2005). Are fathers involved in pediatric psychology research and treatment? Journal of Pediatric Psychology, 30, 631 643 doi:10.1093/jpepsy/jsi050 Phares, V., Rojas, A., Thurston, I. B., & Hankinson, J. C. (2010 ). I ncluding fathers in c linical interventions for children and a dolescents. In M. E. Lamb (Ed.), The Role of the Father in Child Development (5th ed.) (pp 459 485). Hoboken, NJ: Wiley. Phinney, J. S. (1992). The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7, 156 176. doi:10.1177/074355489272003 Phiney, J. S., and Ong, A. D. (2007). Conceptualization and measurement of ethni c identity: Current status and future directions. Journal of Counseling Psychology, 54, 271 281. doi:10.1037/0022 0184.108.40.2061 Popular baby names. (n.d.). Retrieved October 1, 200 7 from http://www.yeahbaby.com/popular baby names/ Power, T. J., Eiraldi, R B., Clarke, A. T., Mazzuca, L. B., & Krain, A. L. (2005). Improving mental health service utilization for children and adolescents. School Psychology Quarterly, 20, 187 205. doi:10.1521/scpq.220.127.116.11510 lth (2003). Achieving the promise: Transforming mental health care in America. Final report. DHHS Pub. No. SMA 03 3832 Rockville, MD. Pumariega, A. J., Glover, S., Holzer, C. E., & Nguyen, H. (1998). Utilization of mental health services in a tri ethnic sa mple of adolescents. Community Mental Health Journal, 34, 145 156. doi:10.1023/A:1018788901831
125 Pumariega, A. J., Rogers, K., & Rothe, E. (2005). Culturally competent systems of care Community Mental He alth Journal, 41, 539 555. doi:10.1007/s10597 005 6360 4 Raviv, A., Raviv, A., Vago Gefen, I., & Fink, A. S. (2009). The personal service gap: Factors affecting adolescents' willingness to seek help. Journal of Adolescence 32 483 499. doi:10.1016/j.adole scence.2008.07.004 seeking help for their children from school and private psychologists. Professional Psychology: Research and Practice, 34, 95 101. doi:10.1037/0735 7028.3 4.1.95 Raviv, A., Sharvit, K., Raviv, A., & Rosenblat Stein, S. (2009). f r eluctance to s eek p sychological h elp for t heir c hildren Journal of Child and Family Studies, 18, 151 162. doi: 10.1007/s10826 008 9215 0 Rescorla, L., Achenbac h, T. M., Ivanova, M. Y., Dumenci, L., Almqvist, F., Bilenberg, qualities reported by adolescents in 24 countries. Journal of Consulting and Clinical Psychology 75 351 358. doi:10.1037/0022 006X.75.2.351 Ringel, J. S., & Sturm, R. (2001). National estimates of mental health utilization and expenditures for children in 1998. Journal of Behavioral Health Services and Research, 28, 319 333. doi:10.1007/BF02287247 Roberts, R. E. Alegria, M., Roberts, C. R., & Chen, I. G. (2005). Mental health problems of adolescents as reported by their caregivers: A comparison of European, African, and Latino Americans. Journal of Behavioral Health Services and Research, 32, 1 13. doi:10.1007/B F02287324
126 Kolodner, R. D. (2005). Recruiting Research Participants at Community Education Sites. Journal of Cancer Education 20 235 239. doi:10.1207/s15430154jce2004_12 Sareen, J., Jagdeo, A., Cox, B. J., Clara, I., Have, M. T., Belik, S., (2007). Perceived barriers to mental health service utilization in the United States, Ontario, and the Netherlands. Psychiatric Services, 58, 357 364. doi:10.1176/appi.ps. 58.3.357 Sayal, K. (2006). Annotation: Pathway to care for children with mental health problems. Journal of Child Psychology and Psychiatry, 47, 649 659. doi:10.1111/j.1469 7610.2005.01543.x Seiffge Krenke, I., & Kollmar, F. (1998). Discrepancies between m parent adolescent agreement on internalising and externalising problem behaviour. Journal of Child Psychology and Psychiatry, 39, 687 697. doi:10.1017/S0 021963098002492 Shah, R., Draycott, S., Wolpert, M., Christie, D., & Stein, S. (2004). A comparison of problems. Emotional and Behavioral Difficulties, 9, 181 190. doi:10.11 77/1363275204047808 Shumway, M., Unick, G. J., McConnell, W. A., Catalano, R., & Forster, P. (2004). Measuring community preferences for public mental health services: Pilot test of a mail survey method. Community Mental Health Journal, 40, 281 295. doi:10 .1023/B:COMH.0000035225.19252.92
127 diagnosis, and drug treatment. Harvard Review of Psychiatry, 11, 308 316. doi:10.1080/714044393 Snowden, L. R. (2001). Barriers to effective m ental health services for African Americans. Mental Health Services Research 3 (4), 181 187. doi:10.1023/A:1013172913880 Snowden, L. R. & Yamada, A. (2005). Cultural differences in access to care. Annual Review of Clinical Psychology, 1, 143 166. doi:10.1146/annurev.clinpsy.1.102803.143846 Srebnik, D., Cauce, A. M., & Baydar, N. (1996). Help seeking pathways for children and adolescents. Journal of Emotional and Behavioral Disorders, 4, 210 220. doi:10.1177/106342669600400402 Stevens, J., Klima, J. Chisolm, D., & Kelleher, K. J. (2009). A trial of telephone services to increase adolescent utilization of health care for psychosocial problems. Journal of Adolescent Health 45 564 570. doi:1 0.1016/j.jadohealth.2009.04.003 Stiffman, A. R., Pescosolido B., & Cabassa, L. J. (2004). Building a model to understand youth service access: The Gateway Provider Model. Mental Health Services Research 6 189 198. doi:10.1023/B:MHSR.0000044745.09952.33 Sue, S., & Chu, J. Y. (2003). The mental health of ethnic m inority groups: Challenges Culture, Medicine, and Psychiatry, 27, 447 465. doi:10.1023/B:MEDI.0000005483.80655.15 Sue, D. W., & Sue, D. (2007 ). Counseling the c ulturally diverse : The ory and p ractice (5 th ed.) New York, NY: John Wiley & Sons.
128 Swartz, M. S., Wagner, H. R., Swanson, J. W., Burns, B. J., George, L. K., & Padgett, D. K. (1998). Comparing use of public and private mental health services: The enduring barriers of race and a ge. Community Mental Health Journal, 34, 133 144. doi:10.1023/A:1018736917761 Teagle, S. E. (2002). Parent problem recognition and child mental health service use. Mental Health Services Research, 4, 257 266. doi:10.1023/A:1020981019342 Thompson, V. L., Ba psychotherapy and psychotherapists. Professional Psychology: Research and Practice, 35, 19 26. doi :10.1037/0735 7028.35.1.19 Thurston, I.B., & Phares, V. (2008). Mental health service utilizat ion among African American and Caucasian mothers and fathers. Journal of Consulting and Clinical Psychology, 76, 1058 1067. doi : 10.1037/a0014007 Uebelacker, L. A. Wang, P. S. Berglund, P., & Kessler, R. C. (2006). Clinical differences among patients treated for mental health problems in general medical and specialty mental health settings in the National Comorbidity Survey Replication. General Hospital Psychiatry 28 387 395. doi:10. 1016/j.genhosppsych.2006 .05.001 U.S. Census Bureau. (2001). Census 2000 Redistricting Data (P.L. 94 171) Summary File for states, Tables PL1, PL2, PL3, and PL4. U.S. Census Bureau (2004). U.S. Interim Projections by Age, Sex, Race, and Hispanic Origin. http://www.census.gov/ipc/w ww/usinterimproj/ U.S. Census Bureau (200 9 ). Income, poverty, and health insurance coverage in the United States: 200 8 http://www.census.gov/prod/2009pubs/p60 236.pdf
129 U.S. Department of Health and Human Services (1999). Mental health: A report of the Surg eon General. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services, National Institute of Mental Health. U.S. Department of Health and Human Services (2001) Mental health: Culture, race, and ethnicity a supplement to mental health: A report of the Surgeon General. Rockville, MD: U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Center for Mental Health Services. Waddell, C., Offord, D. R. Shepherd, C. A. Hua, J. M. & McEwan, K. (2002). Child psychiatric epidemiology and Canadian public policy making: The state of the science and the art of the possible. The Canadian Journal of Psychiatry / La Revue canadienne de psychiatrie 47 (9) 825 832. Retrieved from PsycINFO database. Wang, P. S., Berglund, P. A., & Kessler, R. C. (2003). Patterns and correlates of contacting clergy for mental disorders in the United States. Health Services Research, 38, 647 6 73. doi: 10.1111/1475 6773.00138 Weisz, J. R., & Eastman, K. E. (1995). Cross national research on child and adolescent psychopathology. In F. C. Verhulst & H. M. Koot (E ds. ), The Epidemiology of Child and Adolescent Psychopathology (pp. 42 65). New York NY : Oxf ord University Press World Health Organization (1 993 ) The ICD 10 Classification of Mental and Behavioural Disorders Diagnostic Criteria for Research. WHO: Geneva.
130 Wu, P., Hoven, C. W., Bird, H. R., Moore, R. Roper, M. T. (1999). Depressive and disruptive disorders and mental health service utilization in children and adolescents. Jour nal of the American Academy of Child & Adolescent Psychiatry 38, 1081 1090. doi: 10.1097/00004583 199909000 00010 Yeh, M., & Hough, R. L. (2005). Beliefs about causes of child problems questionnaire. Unpublished measure. Yeh, M., McCabe, K., Hough, R. L., Lau, A., Fakhry, F., and Garland, A. (2005). Why bother with beliefs? Examining relationships between race/ethnicity, parental beliefs about causes of child problems, and mental health service use. Journal of Consulting and Clinical Psychology, 73, 800 80 7. doi :10.1037/0022 006X.73.5.800 Zimmerman, F. J. (2005). Social and economic determinants of disparities in professional help seeking for child mental health problems: Evidence from a national sample. Health Research and Educational Trust, 40, 1514 1533. doi :10.1111/j.1475 6773.2005.00411.x Zwaanswijk, M., Verhaak, P. F., Bensing, J. M., Ende, J. V., & Verhulst, F. C. (2003). Help seeking for emotional and behavioral problems in children and adolescents: A review of recent literature. European Child and A dolescent Psychiatry, 12, 153 161. doi:10.1007/s00787 003 0322 6
132 Appendix A : Letter of Invitation Dear Parent, I would like to invite you to participate in a study about seeking help for your children. The purpose of this research project is to better understand how typical parents and their children seek help for different things. You are being asked to read som e stories and answer the questions that follow. You are also asked to complete a series of services, your beliefs, and some background information. You do not need to hav e received mental health services in the past to participate in this study. The entire study should take about 30 40 minutes. Your participation in this survey is completely voluntary. You are free to participate in the study or withdraw at any time wit hout penalty. Your consent to participate is shown by your decision to complete the questionnaires. We will not need to contact your child to participate in this study. The potential benefits for participating in this study are raising your awareness ab out mental health services. There are no known risks for those who take part in this study. All participants who provide their contact information, using the business reply post cards, will be entered into a drawing for one of six prizes. All contact in formation provided will be kept separate from questionnaires which will be identified by subject code to protect your privacy. The prizes include: one $100 cash prize, two $50 cash prizes and three gift certificates from merchants in the surrounding commu nity. Your privacy and research records will be kept confidential to the extent of the law. Authorized research personnel, employees of the Department of Health and Human Services, and the USF Institutional Review Board, its staff and other individuals acting on behalf of USF may inspect the records from this research project. If you have any questions about your rights as a person who is taking part in a research study, you may contact the Division of Research Compliance of the University of South Flo rida at (813) 974 5638. If you have any questions about this research study, please contact Idia B. Thurston, Department of Psychology, University of South Florida, 4202 E. Fowler Avenue., PCD 4118G, Tampa, FL 33620, 813 974 9222, email@example.com Thank you. Sincerely, Idia B initie Thurston, M.A. Psychology Doctoral Candidate
133 Appendix B : Referral Letter Dear Parent Thank you for participating in our research study. As you recall, this study was about understanding how typical parents and their children use mental health services and the reasons why they use mental health services. Since some people wonder about where to receive services, this letter is being sent out to all participants regardle ss of their answers to the survey. In case you are interested in seeking mental health services, we wanted you to know that most health insurance companies cover some type of mental health service. Therefore, you should first check with your insurance company to see what type s of mental health services are covered. Many insurance companies also may seek treatment. Please get this information clarified with your insurance company if you are concerned a bout payment for mental health services. If you do not have health insurance or if your health insurance does not cover mental health services, you may want to consider one of the following facilities in the Tampa Bay area (all of which have either low fe es or fees on a sliding scale): USF Psychological Services Center (813) 974 2496 At USF in the Psychology Department Mental Health Care (813) 272 2244 5707 North 22 nd street, Tampa, FL www.mhcinc.org Northside Mental Health Center (813) 977 8700 12512 Bruce B. Downs Blvd, Tampa, FL www.northsidemhc.org C hild and Family Counseling (813) 744 5953 Hillsborough County Peace River Center (863) 519 0575 Locations in Polk, Hardee, & Highlands Counties www.peace river.com service that is run through the American Psychological Association. This service can be accessed on line ( http://loc ator.apahelpcenter.org ) or through a toll free phone call (1 800 964 2000). In addition, most communities throughout the United States and Canada have a Community Mental Health Center, so a quick call to the local information center might help you gain ac cess to the services in your community, if you are interested. If you have any questions about this research study, please contact Idia Binitie Thurston, Department of Psychology, University of South Florida, 4202 E. Fowler Avenue., PCD 4118G, Tampa, FL 33620, 813 974 9222, firstname.lastname@example.org Thank you. Sincerely, Idia Binitie Thurston M.A. Psychology Doctoral Candidate
134 Appendix C : Boy Internalizing Vignette with follow up questions Please read the desc riptions of children below and a nswer the questions that follow. year olds? 1 Not at all Serious 2 A little bit Serious 3 Moderately Serious 4 Quite Serious 5 Extremely Serious 1 Not at all Concerned 2 A little bit Concerned 3 Moderately Concerned 4 Quite Concerned 5 Extremely Concerned 3. How much do 1 Not at all 2 A little bit 3 Moderately 4 Quite 5 Extremely 1 Not at all 2 A little bit 3 Moderately 4 Quite 5 Extremely 5. Do you think Michael has a mental health problem? Yes No 6. How confident are you in your response to the question above? (i.e., if you answered NO, how confident are you that Mi chael does not have a mental health problem; if you answered YES, how confident are you that Michael has a mental health problem)? 1 Not at all Confident 2 A little bit Confident 3 Moderately Confident 4 Quite Confident 5 Extremely Con fident A.) Michael is a 10 year old boy who has been overly worried and nervous about various things at home and school for the past 6 months. Some of his worries include getting perfect grades at school, his performance on the soccer team, and keeping his room tidy. Michael has also been experiencing some difficulty concentrating at school due to his worrying and as a result his grades are beginning to fall. He is easily tired throughout the day and is unable to sleep at night. have been making fun of him. He realizes that he worries too much and wishes he could control it so he could be more like other children.
135 Appendix C (Continued) 7 If you were responsible for Michael, w ould you seek help for him? Yes No 8 How likely are you to seek help for Michael from each of the following sources ? (Be sure to circle an answer for each source; if you will not seek help from a particular source, please circle option 1 not at all likely) 1 Not at all Likely 2 A little bit Likely 3 Moderately Likely 4 Quite Likely 5 Extremely Likely 1.) Pediatrician/Family doctor 1 2 3 4 5 2.) Psychologist/Other mental health professional 1 2 3 4 5 3.) Religious leader (Pastor, Spiritual leader, Imam, Rabbi, etc) 1 2 3 4 5 4.) Teacher/School guidance counselor 1 2 3 4 5 5.) Family/Friends/Co workers 1 2 3 4 5 6.) Self help books/Internet websites 1 2 3 4 5 7.) None (I would not seek any help) 1 2 3 4 5 9 Which of the sources are you most likely to seek help from first (c heck one): Pediatrician/Family doctor Psychologist/Other mental health professional Religious leader (Pastor, Spiritual leader, Imam, Rabbi, etc) Teacher/School guidance counselor Family/Friends/Co workers Self help books/internet websites None (I would not seek any help)
136 Appendix D : Externalizing and Control Vignettes Girl Externalizing Vignette B.) Brittany is a 10 year old girl who has become distracted easily and forgetful at home and school over the past 6 months. She often fails to finish her chores and school wo rk and repeatedly makes careless mistakes on assignments. Brittany has difficulty paying attention for long periods of time and does not appear to listen when spoken to. She has a hard time waiting her turn, talks a lot, and often interrupts others when they are talking. Brittany usually has difficulty playing quietly. At school, she is out of her seat constantly and has become very fidgety. Brittany has always been an active child, but her recent behavior is now affecting her school work and ability t o keep and make new friends. Boy Control Vignette C .) Joshua is a 10 year old boy who has been receiving A and B grades in school over the past 6 months. He has several friends at home and school who he enjoys spending time with. Although he usuall y gets along with most children, Joshua sometimes gets into minor arguments with his friends when playing games or when he does not get his way. Occasionally, when Joshua gets angry or upset, he yells or slams his door; however, once he cools down he usua lly feels bad and apologizes for his behavior. Joshua participates in several activities after school such as soccer and reading club.
137 Appendix E : Demographics INSTRUCTIONS: Please respond to the following questions about you and your family. 1. What is your gender? Male Female 2. How old are you? ______ 3. Are you: Married L iving with partner In a relationship but living separately Single Separated Divorced Widowed Other (specify ____________ ________) 4. How many people, including yourself, live in your home? __________ 5. How many children (biological, stepchildren, adopted and other children) do you have? ____ 6. How many of these children are currently living in your home? _______ 7. List the ages, gender, and race of all your children (example: 10 year old white male): ________________________________________________________________________ ________________________________________________________________________ ____ ____________________________________________________________________ 8. Do you have monthly face to face contact with at least one of your children, ages 2 to 17? Yes No 9. Employment status: Self Your Spouse/Partner Employed as (list job)_________________ Employ ed as (list job)________________ Unemployed ________________________ Unemployed _______________________ Retired (list previous job)______________ Retired (list previous job)_____________ Other ______________________________ Other __________ __________________
138 Appendix E (Continued) 10. Highest level of education completed (please circle one response): Self Grade School Middle School High School College Graduate School 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 >20 Your Spouse/Partner Grade School Middle School High School College Graduate School 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 >20 11 H ome zip code : _____________ 12. Please select your total household income per year (This information is confidential) < $10,000 $20,501 $24,000 $34,501 $38,000 $50,001 $65,000 $10,001 $13,500 $24,001 $27,500 $38,001 $41,500 $65,001 $80,000 $13,501 $17,000 $27,501 $31,000 $41,501 $45,000 $80,001 $100,000 $17,001 $20,500 $31,001 $34,500 $45,001 $50,000 > $100,000 13. What types of public assistance do you receive? Food Stamps Welfare Help with Housing Temporary Assistance for Needy Families Other, please describe: _______________ ________________ None 14. Do you have insurance coverage (Private, Medicaid, Medicare) for physical health needs? For Self For your Children Yes Yes No No 15. Do you have insurance coverage (Private, Medicaid, Medicare) for mental health needs? For Self For your Children Yes Yes No No w
139 Appendix F : Utilization INSTRUCTIONS: Please respond to the following questions about you and your family. A. QUESTIONS ABOUT YOU 1. Have you ever been referred to see a mental health professional (such as: a psychologist, psychiatrist, social worker, or counselor) for emotional or behavior problems, alcohol or drug use? Yes No 2. Have you ever seen a mental health professional (such as: a psychologist, psychiatrist, social worker, or counselor) for emotional or behavior problems, alcohol or drug use? Yes No 3. Overall, how satisfied are you with the mental health services you received from mental health profess ionals? 0 Have never used services 1 Not at all satisfied 2 A little bit satisfied 3 Moderately satisfied 4 Quite satisfied 5 Extremely satisfied 4. Do you think you currently have a mental health problem (i.e. an emotional, behavioral, alcohol, or drug problem)? Yes No 5. How likely are you to seek treatment for the problems you are currently experiencing from a mental health professional? 0 Not currently experiencing any problems 1 Not at all likely 2 A little bit likely 3 Moderately likely 4 Quite likely 5 Extremely likely B. QUESTIONS ABOUT YOUR CHILDREN 6. Have any of your children ever been referred to see a mental health professional (such as: a psychologist, psychiatrist, social worker, or counselor) for emotional or behavior problems, alcohol or drug use? Yes No
140 Appendix F (Continued) 7. Have any of your children ever seen a mental health professional (such as: a psychologist, psychiatrist, social worker, or counselor) for emotional or behavior problems, alcohol or drug use? Yes No 8. Overall, how satisfied are you with the mental health servic es your children have received from mental health professionals? 0 My children have never used services 1 Not at all satisfied 2 A little bit satisfied 3 Moderately satisfied 4 Quite satisfied 5 Extremely satisfied 9. Have any of your children ever received mental health care from any of the following sectors (please check all that apply)? Health sector (pediatrician, family doctor, emergency room, etc) Mental health sector (psychologist, psychiatrist, counselor, social worker, residential, etc) Education sector (guidance counselor, school psychologist, special class, etc) Religious sector (clergy, pastor, spiritual leader, traditional healer, etc) Child welfare (social services counseling, etc) Juvenile Justice System (detention center/jail, court counselor, etc) Other (please describe): ________________________________________________ My children have never used mental health services 10. Do you think any of your children currently have a mental health problem (i.e. an emotional, behavioral, alcohol, or drug problem)? Yes No 11. How likely are you to seek treatment for the problems your child is currently experiencing from a mental health p rofessional? 0 My child is not experiencing any problems 1 Not at all likely 2 A little bit likely 3 Moderately likely 4 Quite likely 5 Extremely likely
141 Appendix F (Continued) 12. Please circle YES or NO, to indicate if you have had any personal or professional experience with any of the following disorders that are sometimes found in children and adults. Attention Deficit/Hyperactivity Disorder (ADHD) YES NO Anxiety Disorder YES NO Depression YES NO Oppositional Defiant Disorder YES NO Alcohol/Drug Abuse YES NO
142 Appendix G : Multigroup Ethnic Identity Measure Revised (MEIM R) People come from a lot of different cultures and there are many different words to describe the different backgrounds or ethnic groups that people come from. Some examples of the names of ethnic groups are M exican American, White, African American, American Indian, and Asian American. Every person is born into one or more ethnic groups, but people differ on how important their et hnicity is to them, how they feel about it, and how much their behavior is affected by it. INSTRUCTIONS: Please write and circle your answers to the questions below about your ethnic group (s) and how you feel about it. In terms of ethnic group (s) I c onsider myself to be ______________________________ 1. I have spent time trying to find out more about my ethnic group (s) such as its history, traditions, and customs. 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree 2. I have a strong sense of belonging to my own ethnic group (s) 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree 3. I understand pretty well what my ethnic group membership means to me 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree 4. I have often done things that will help me understa nd my ethnic background better. 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree 5. I have often talked to other people in order to le arn more about my ethnic group (s) 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree 6. I feel a strong attachment towards my own ethnic group (s) 1 Strongly Disagree 2 Disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree
14 3 Appendix G (Continued) 7. What is your race (check all that apply)? American Indian or Alaskan Native Asian, Asian American, or Oriental Black or African American Native Hawaiian or Other Pacific Islander White, Caucasian, or European Other (specify ________________________) 8. What is your ethnicity (check one)? Hispanic or Latino Not Hispanic or Latino 9. What is your race (check all that apply)? American Indian or Alaskan Native Asian, Asian American, or Oriental Black or African American Native Hawaiian or Other Pacific Islander White, Caucasian, or European, not Hispanic Other (specify ________________________) 10. What is your ethnicity (check one)? Hispanic or Latino Not Hispanic or Latino 11. What is your race (check all that apply)? American Indian or Alaskan Native Asian, Asian American, or Oriental Black or African American Native Hawaiian or Other Pacific Islander White, Caucasian, or European, not Hispanic Other (specify ________________________) 12. What is your ethnicity (check one)? Hispanic or Latino Not Hispanic or Latino
144 Appendix H : Barriers to Utilization INSTRUCTIONS: Below are some reasons parents consider when seeking professional mental health services for their children. Please circ le a response for each question that best represents how you currently feel about seeking mental health services for your children SD Strongly Disagree D Dis agree A Agree SA Strongly Agree 1. My health insurance would cover this type of treatment. SD D A SA 2. Help probably would not do any good. SD D A SA 3. I think the problem would not get better by itself. SD D A SA 4. Services would be too expensive. SD D A SA 5. I would not be concerned about what others might think. SD D A SA 6. It would take too much time or be inconvenient. SD D A SA 7. My child would want to solve the problem on his/her own. SD D A SA 8. There would not be a language problem. SD D A SA 9. I would be scared about my child being put into a hospital or taken away from me against my will. SD D A SA 10. My child would refuse to go. SD D A SA 11. I trust mental health professionals. SD D A SA 12. A family member/friend would object. SD D A SA 13. I think the staff would be unfriendly or disrespectful. SD D A SA 14. I would have transportation problems. SD D A SA 15. I am satisfied with available services. SD D A SA 16. My child went in the past but it did not help. SD D A SA 17. I would be unsure about where to go for help. SD D A SA 18. I would be able to get an appointment. SD D A SA
145 Appendix I : Beliefs About Causes Revised (BAC R) INSTRUCTIONS: When a child has mental health problems, parents can have many different explanations for why the problems began. The lists below show some of these any emotional, behavioral, or alcohol/drug problems that a child may have. Please circle A. PHYSICAL CAUSES: 1. Genetics or her edity? Yes No 2. Alcohol, drugs, or other substances taken during Yes No pregnancy? 3. Something else (not alcohol/drugs) taken by the father Yes No or mother during pregnancy? 4. Other pregnancy or birth related complications? Yes No (e.g., prematurity, no prenatal care) 5. A serious physical illness, injury, or condition? Yes No 6. Alcohol, drugs, or other substances taken by the child? Yes No 7. Something (not alcohol/drugs) eaten by the child? Yes No 8. A lack or imbalance of chemicals, proper vitamins, Yes No hormones, or other nutritional elements? B. PERSONALITY OR EMOTIONAL STRUGGLES: Yes No discipline, self control, or difficulties Yes No with anger? 3. Emotional struggles or inner conflicts? (e.g., self esteem Yes No issues) Yes No developmental stage in life? C. GETTING ALONG WITH OTHERS: Yes No
146 Appendix I (Continued) Yes No relationships? Yes No D. TRAUMA: 1. Trauma suffered by the child? (e.g., physical abuse, Yes No sexual abuse, seeing a violent act) 2. Trauma suffered by the family? (e.g., war, poverty, Yes No hardship) E. FAMILY OF PARENTING ISSUES: Yes No 2. The absence of, or being separated from, one or both Yes No parents? (e.g., due to divorce, foster care, incarceration, death) 3. Drug, alcohol, or mental health problem of a parent? Yes No 4. Something related to parenting skills? Yes No 5. Not having enough time with a parent or adult? Yes No 6. A child not following or disobeying the beliefs and Yes No teachings of the family or parents? 1. A child not having enough friends, being teased, or being Yes No bullied? Yes No or peers? Yes No G. AMERICAN CULTURE: 1. The influence of popular American culture (e.g. through Yes No television or movies)? 2. The influence of Ame rican culture, which is different Yes No 3. Difficulty adjusting to American cul ture? Yes No
147 Appendix I (Continued) H. DISCRIMINATION OR PREJUDICE: 1. Racial or ethnic discrimination or prejudice? Yes No I. ECONOMIC PROBLEMS: 1. Not having enough money for things like food, clothing, Yes No housing, etc.? 2. Not having the necessary books/school materials at home Yes No to help him/her succeed in school? 3. Not having enough money for extra things that the child Yes No wants? J. SPIRITUAL, COSMIC, OR RELIGIOUS REASONS: 1. Bad luck or chance? Yes No 2. Spirit possession, the influence of spirits or ghosts, Yes No someone casting a spell, magic, or witchcraft? 3. Punishment for the deeds of the child, his/her family, or Yes No ancestors? 4. Not having enough faith, spirituality, or involvement in Yes No religion? 5. The will of God, gods, deities, or some other supernatural Yes No beings? 6. Fate? Yes No 7. Being born during a particular day, year, or cosmic sign? Yes No K. IMBALANCE OR DISHARMONY WITH NATURE OR NATURAL ELEMENTS: 1. An arrangement of physical elements or objects (e.g., in Yes No the home) that is not in line with nature or natural forces? 2. A yin and yang imbalance? Yes No Yes No
148 Appe ndix I (Continued) L. This list shows all the parent explanations given above, please rate how strongly you believe in each of these possible causes of mental health problems. 1 Not at all 2 A little bit 3 Somewhat 4 Much 5 A great deal A. Physical Causes 1 2 3 4 5 B. Personality or Emotional struggles 1 2 3 4 5 C. Getting along with others 1 2 3 4 5 D. Trauma 1 2 3 4 5 E. Family or Parenting Issues 1 2 3 4 5 F. Friends 1 2 3 4 5 G. American Culture 1 2 3 4 5 H. Discrimination or Prejudice 1 2 3 4 5 I. Economic problems 1 2 3 4 5 J. Spiritual or Religious reasons 1 2 3 4 5 K. Disharmony with Nature 1 2 3 4 5 M. ONE of these do you believe is the most important cause _______________________________________________________________________ ______________________________________________ _________________________
About the Author Idia Binitie Thurston completed her first independent research project through the honors department and earned her Bachelor of Science degree in Ps ychology from Florida State University in 2003 Three years later, Idia received her Master of Arts in Clinical Psychology from the University of South Florida While working on her Doctor ate of Philosophy in Clinical Psychology at the University of South Flori da, Idia was the recipient of the Tom Tighe Graduate Student Research Award, USF Counseling Center Faculty St aff Scholarship, and Graduate Diversity and Access Fellowship. She has published t hree peer reviewed article s and two book chapters and participated in both oral and poster presentations at several national conferences. Idia was also involve d with the Socie ty o f Pediatric Psychology, where she wa s an active member on two committees In 2008, She completed an NIMH Predoctoral Fellowship in HIV biobehavioral research at Brown University. Sh e also served as the principal investigator on a clinical services g rant for the implementation of an HIV/AIDS therapeutic group for teenage girls. Most recently, Idia has been completing her clinical internship training at