USF Libraries
USF Digital Collections

The quality of life among lymphedema patients due to lymphatic filariasis in three rural towns in Haiti

MISSING IMAGE

Material Information

Title:
The quality of life among lymphedema patients due to lymphatic filariasis in three rural towns in Haiti
Physical Description:
Book
Language:
English
Creator:
Kanda, Koji
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
reliability and validity
health behavior
gender
morbidity control
global health
Dissertations, Academic -- Public Health -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The worldwide eradication of lymphatic filariasis has recently started with two strategies, interruption of transmission and morbidity control. One of the most endemic countries, Haiti has experienced successful interventions through national and international efforts, but the morbidity control is still hindered by a lack of adequate information on quality of life (QOL) issues among those suffering from the chronic manifestations of the disease such as lymphedema. In addition, previous interventions have been focused primarily in a single community where an established lymphedema treatment clinic serves as a national reference center, so it is critical to expand programs to other areas in Haiti. The purpose of the study was to understand the issues of morbidity control and QOL among lymphedema patients due to lymphatic filariasis in three rural Haitian towns. Secondary data (n = 316) collected in an ongoing filariasis support group project was analyzed in terms of socio-demographic characteristics, including gender age, and regional perspectives. Also, two different commercial QOL instruments (EuroQol, CDC Healthy Days) and a subjective well-being assessment tool (CES-D) were introduced to describe their QOL and mental health status, respectively. The reliability and validity of the measurements were established at the same time. Regional differences were evident in patients illness history, knowledge of the illness, self-care and self-efficacy for legs, and major QOL indicators related to physical and mental health. Age of patients also influenced foot size, illness stage, and the QOL scores. However, other socio-demographic factors were poorly associated with filariasis related variables, including gender. The commercial QOL instruments and a standardized mental health tool satisfied a reasonable level of reliability and validity. Though additional discussion is needed regarding the validation of the mental health scales between EuroQol and the other instruments, they nevertheless offer utility for enhancing the quality of morbidity control programs. These findings offer a significant contribution for the development of filariasis prevention programs such as community-based morbidity control and support group activities in Haiti, as well as other areas of the filariasis-endemic world.
Thesis:
Thesis (M.S.P.H.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Koji Kanda.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 173 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001478765
oclc - 56389920
notis - AJS2455
usfldc doi - E14-SFE0000407
usfldc handle - e14.407
System ID:
SFS0025099:00001


This item is only available as the following downloads:


Full Text

PAGE 1

The Quality of Life Among Lymphedema Patie nts Due to Lymphatic Filariasis in Three Rural Towns in Haiti by Koji Kanda A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Community and Family Health College of Public Health University of South Florida Major Professor: Jea nnine Coreil, Ph.D. Melinda Forthofer, Ph.D. Eknath Naik, M.D., Ph.D. Date of Approval: June 23, 2004 Keywords: global health, morbid ity control, gender, health behavior, reliability and validity Copyright 2004 Koji Kanda

PAGE 2

Acknowledgement I have to mention that this thesis w ouldnt be completed without a great support from my advisor, Dr. Jeannine Coreil. She ga ve me a wonderful opportu nity to participate in the lymphatic filariasis project in Haiti, wh ich was my first step to involve in the field of global health. In addition, as an academi c advisor and department chair, she fully supported my two-year study in USF and in the United States. I really appreciate. I also want to thank my committees, Drs. Melinda Forthofer and Eknath Naik. They gave me precious advice to make the thesis better through in persons and in class. Also, I appreciate all faculties who advi sed me in person or in class. Finally, I have to say thank you to all of my friends who concerned about my thesis.

PAGE 3

i Table of Contents List of Tables................................................................................................................. .....iii List of Figures......................................................................................................................v List of Abbreviations.......................................................................................................... vi Abstract....................................................................................................................... .....viii Chapter One: Introduction...................................................................................................1 Chapter Two: Review of Literatures....................................................................................5 General Background Epidemiology, Etiology, Treatment, and Prevention...........5 Lymphedema..........................................................................................................11 Four Aspects of Lymphatic Filariasis....................................................................13 Haiti........................................................................................................................16 Factors Associated with Lympha tic Filariasis and Lymphedema..........................19 Gender Perspective ...............................................................................................22 Quality of Life........................................................................................................23 Reliability and Validity of the Quality of Life Measurements...............................28 Chapter Three: Methods....................................................................................................33 Study Design..........................................................................................................33 Objectives..............................................................................................................34 Population and Sample Size...................................................................................37 Sampling Method...................................................................................................39 Measurement..........................................................................................................40 Reliability and Validity..........................................................................................41 Data Analysis.........................................................................................................43 Chapter Four: Results........................................................................................................45 General Information...............................................................................................45 Demographics............................................................................................45 Illness History............................................................................................49 Knowledge.................................................................................................61 Foot Size and Illness Stage....................................................................................62 Self-care Practice and Self-efficacy.......................................................................74 Quality of Life........................................................................................................83 EuroQol......................................................................................................83 CES-D........................................................................................................94

PAGE 4

ii CDC Healthy Days....................................................................................97 Reliability and Validity............................................................................114 Chapter Five: Discussion and Conclusion.......................................................................117 Regional Differences...........................................................................................117 Gender Perspective..............................................................................................119 Lymphedema Condition and Its Related Variables..............................................119 Health-related Behavior.......................................................................................120 Quality of Life and Subjective Well-being Scales...............................................122 Reliability and Validity............................................................................122 Outcome of the Scales vs. So cio-demographic Variables........................124 Limitation.............................................................................................................126 Conclusion...........................................................................................................128 References........................................................................................................................130 Bibliography................................................................................................................... .135 Appendices.......................................................................................................................136 Appendix A: Lymphedema Stage........................................................................137 Appendix B: Filariasis Baseline Evaluation Survey Arcahaie..........................138 Appendix C: Map of Port-au-Prince Area...........................................................162

PAGE 5

iii List of Tables Table 1. Lymphedema Stage and Its Characteristics..................................................11 Table 2. Demographic Profile in Haiti.......................................................................17 Table 3. Brief Summary of the Survey Categories and Questionnaires.....................40 Table 4. Demographic Characteristics of Lymphedema Patients...............................47 Table 5. The First Impression of the Illness...............................................................51 Table 6. The First Symptom Noticed.........................................................................51 Table 7. Treatment Choice.........................................................................................52 Table 8. Precaution for Legs......................................................................................52 Table 9. History of Acute Attacks..............................................................................56 Table 10. Materials Purchased.....................................................................................60 Table 11. Daily Activities.............................................................................................60 Table 12. Cause of Illness............................................................................................61 Table 13. Foot Exam....................................................................................................63 Table 14. Stage of Illness.............................................................................................64 Table 15. Other Socio-demographic Variab les vs. Stage of Illness and Foot Sizes.............................................................................................................68 Table 16. Gender, Town vs. Self-care Practice for Legs..............................................75 Table 17. Other Socio-demographic Variab les vs. Major Self-care Practices for Legs..............................................................................................................76 Table 18. Gender, Town vs. Possible Leg Care in the Future......................................79

PAGE 6

iv Table 19. Other Socio-demographic Vari ables vs. Possible Leg Care in the Future...........................................................................................................80 Table 20. Degree of Confidence for Leg Care.............................................................82 Table 21. Gender, Town vs. EuroQol...........................................................................86 Table 22. Filariasis Related Variables vs. People Answered No Problem in EuroQol Questionnaires and EuroQol Overall Health Status.......................88 Table 23. Gender, Town vs. Total CES-D Score and the Proportion of the Score Indicating Depressive Cases ........................................................................95 Table 24. Other Socio-demographic Va riables vs. Total CES-D Score and the Proportion of the Score I ndicating Depressive Cases.............................96 Table 25. Gender, Town vs. Self-rated Health Status by CDC Healthy Days..............99 Table 26. Gender, Town vs. Healthy and Unhealthy Days.........................................100 Table 27. Gender, Town vs. Major Cause of Impairment or Health Problem............101 Table 28. Other Socio-demographic Variab les vs. Self-rated Health Condition and Unhealthy Days...........................................................................................104 Table 29. Other Socio-demographic Va riables vs. Unhealthy/healthy Days..............109 Table 30. Other Socio-demographic Variable s vs. Major Cause of Impairment or Health Problem...........................................................................................111 Table 31. Internal Consistency Reliab ility in QOL Scales and CES-D......................114 Table 32. Correlations between Represen tative Questions in QOL Instruments and CES-D..................................................................................................116

PAGE 7

v List of Figures Figure 1. Lymphatic Filariasis Endemic Countries.......................................................6 Figure 2. A Condition of Lymphedema.......................................................................12 Figure 3. Stage of Illness.............................................................................................65 Figure 4. Stage of Illness by Socio-demographic Variables........................................71 Figure 5. Gender, Town vs. Overall Health Status by EuroQol...................................87 Figure 6. Other Socio-demographic Vari ables vs. Overall Health Status by EuroQol........................................................................................................91 Figure 7. Gender, Town vs. Self-rated Health Status by CDC Healthy Days..............99 Figure 8. Other Socio-demographic Vari ables vs. Self-rated Health Status by CDC Healthy Days.....................................................................................106

PAGE 8

vi Lists of Abbreviations ADL Acute Adenolymphangitis BCG Bacillus of Calmette and Guerin CDC Center for Disease Control and Prevention CES-D Center for Epidemiologic Studies Depression Scale DEC Diethylcarbamazine DLQI Dermatology Life Quality Index DTP3 Diphtheria, Tetanus T oxoids, and Pertussis Vaccine EQ-5D EuroQol Five-dimension Health Scale EQ-VAS EuroQol Visual Analogue Scale EuroQol EuroQol Health Questionnaire GDP Gross Domestic Product GPELF Global Programme to Eliminate Lymphatic Filariasis HIV/AIDS Human Immunode ficiency Virus / Acquired Immuno-Deficiency Syndrome ICC Intraclass Correlation Coefficient LF Lymphatic Filariasis MPCE Ministere de la Planificat ion et de la Cooperation Externe NHP Nottingham Health Profile PAHO Pan American Health Organization QOL Quality of Life SES Socioeconomic Status

PAGE 9

vii SF-36 MOS 36-item Short-form Health Survey STDs Sexually Transmitted Diseases WHO World Health Organization

PAGE 10

viii The Quality of Life Among Lymphedema Patients Due to Lymphatic Filariasis in Three Rural Towns in Haiti Koji Kanda ABSTRACT The worldwide eradication of lymphatic fi lariasis has recently started with two strategies, interruption of transmission and morbidity control. One of the most endemic countries, Haiti has experienced succe ssful interventions through national and international efforts, but the morbidity contro l is still hindered by a lack of adequate information on quality of life (QOL) issues among those suffering from the chronic manifestations of the disease such as lym phedema. In addition, previous interventions have been focused primarily in a single community where an established lymphedema treatment clinic serves as a national reference center, so it is critical to expand programs to other areas in Haiti. The purpose of the study was to understand the issues of morbidity control and QOL among lymphedema patients due to lympha tic filariasis in three rural Haitian towns. Secondary data (n = 316) collected in an ongoing filariasis suppor t group project was analyzed in terms of socio-demographic ch aracteristics, including gender age, and regional perspectives. Also, two different commercial QOL instruments (EuroQol, CDC Healthy Days) and a subjective well-being asse ssment tool (CES-D) were introduced to describe their QOL and mental health status, respectively. The reliability and validity of

PAGE 11

ix the measurements were established at the same time. Regional differences were evident in patie nts illness history, knowledge of the illness, self-care and self-efficacy for legs, and major QOL indicators related to physical and mental health. Age of patients also infl uenced foot size, illness stage, and the QOL scores. However, other socio-demographic fact ors were poorly associated with filariasisrelated variables, including gender. The comm ercial QOL instruments and a standardized mental health tool satisfied a reasonable level of reliability and validity. Though additional discussion is needed regarding th e validation of the mental health scales between EuroQol and the other instruments, th ey nevertheless offer utility for enhancing the quality of morbidity control programs. These findings offer a significant contribu tion for the development of filariasis prevention programs such as community-based morbidity contro l and support group activities in Haiti, as well as other areas of the filariasis-endemic world. 330 words

PAGE 12

1 Chapter One Introduction Lymphatic filariasis (LF) is among the most common vector-borne diseases in tropical regions. Above a billion people ar e at risk and over 120 million people are infected with parasites in more than 80 c ountries in Africa, Asia, Central and South Americas, and the Pacific Isla nds (World Health Organization [WHO], 2000a). Of those, 44 million people are suffering from symptoma tic conditions such as lymphedema and hydrocele. Lymphatic filariasis is one of six eradi cable tropical diseases worldwide (Center for Disease Control and Prevention [CDC], 1993). In 1998, WHO started the Global Programme to Eliminate Lymphatic Filariasis (GPEFL), an initiative aimed at complete eradication of filariasis by 2020. The program is based on two strategies: interruption of transmission and morbidity control. The inte rruption of transmission is designed for atrisk and asymptomatic populations, and fairly achievable by several substrategies such as mass drug distribution and diethylcarbamazine (DEC)-fortifie d salt intake in endemic areas. The other approach, morbidity control, is targeted for those who live with chronic manifestations due to lymphatic filariasis. The control stra tegy includes adequate health education, compliance with hygiene regimens, a nd related efforts. However, it is still hindered by a lack of adequate information on quality of life (QOL ) issues, which could lead to more suitable, effective guidelines for morbidity control. Particularly, since

PAGE 13

2 filariasis is greatly associ ated with different socio-demographic variables (King & Freedman, 2000) and its clinical conditions are diverse by region or even within the same community (Dreyer, Figueredo-Silva, Neaf ie, & Addiss, 1998), providing appropriate morbidity control programs is a challenge in endemic areas including Haiti, which has one of the highest preval ence rates worldwide. One of the most common symptomatic manifestations due to LF is lymphedema. The non-fatal body disfigurement causes a huge burden of disability among infected people over decades. Currently about 15 m illion people are living with lymphedema worldwide. In one filariasis endemic area of Ha iti, it is estimated that more than 20% of the population are carriers of microfilaremia (Pan Amer ican Health Organization [PAHO], 2001), and approximately 5% of women suffer from a severe form of lymphedema called elephantiasis (Ebe rhard, Walker, Addiss, & Lammie, 1996). Lymphedema is also often associated with the comorbidity of hydrocele and/or acute attacks. Such conditions execerbate daily life, but little re search has been conducted on QOL among lymphedema patients. In Haiti, a morbidity control program has been in operation for almost a decade, at Ste. Croix Hospital in Leogane, and new treatment programs have recently been initiated in a fe w other areas of high endemicity. Therefore, understanding the issues of disabilities and the QOL among lymphedema patients due to LF would provide timely insights for improvi ng morbidity control in other filariasis endemic areas in Haiti. There are several objectives in this study. The first obj ective is to observe the association of filariasis-related variables among filariasis -related lymphedema in three rural Haitian towns. The association between the disease and socio-demographic

PAGE 14

3 indicators is complicated because it is known that the prevalence and characteristics of filariasis varies by region or even within the same community (D reyer et al., 1998) and that different populations infected with the same parasite can have very different clinical manifestations of the infection (Ottesen, 1987). Also, gender poses an important but complicated issue in LF and that there are no consistent results from an epidemiological standpoint. Above all, since it has been repor ted that there is higher prevalence of LF among women in Haiti (Lammie, Addiss, Leonard, Hightower, & Eberhard, 1993), attention to male Haitian patients has been ve ry limited. Thus, careful consideration of socio-demographic variables would be signifi cant for understanding of regional filariasis problems and the development of future morbidity control plans in the new communities. The other objective is to introdu ce and evaluate QOL measures among lymphedema patients due to LF. The QOL a ssessment procedure is now well established in chronic diseases. Though LF is categorized as an infectious disease, the symptomatic manifestations are chronic. Particularly, lym phedema due to LF is considered one of the most severe chronic disabilities among infectio us diseases. Therefore, the application of the common generic QOL instruments such as the CDC Healthy Days Scales and EuroQol Instrument (EuroQol) would be an innovative approach to assess the QOL among the filariasis-related lymphedema patients. Also, a subjective well-being assessment tool for mental health, Center for Epidemiologic Studies Depression Scale (CES-D) would provide supplemental informati on for morbidity control. In order to fit these tools into Haitian cultures, it is importa nt to establish the reli ability and validity of the measurements prior to the assessment. Since there is no gold standard for QOL assessment for LF, the findings would be a si gnificant contribution to future guidelines

PAGE 15

4 for QOL assessment of LF as well as provi de background information for the morbidity control programs.

PAGE 16

5 Chapter Two Review of Literatures General Background Epidemiology, Etiology, Treatment, and Prevention Lymphatic filariasis is one of the most common vector-borne diseases in tropical regions. More than a billion people are at ri sk and over 120 million people are infected in more than 80 countries in Africa, Asia, Central and South Americas, and the Pacific Islands, (WHO, 2000a; WHO, 2002a). In particul ar, more than 40% of infected people live in India, and one-third in Africa. Filariasis is cause d by eight different parasitic nematode worms: Brugia malayi, Brugia timori, M ansonella ozzardi, Mansonella perstan, Mansonella streptocerca, Loa loa, Onchocera volvulus, and Wuchereria bancrofti (King & Freedman, 2000). Of those, W. bancrofti B. malayi and B. timori are responsible for the infection in the lymphatic system of humans. W. bancrofti accounts for 90% of all infections of LF. It occurs in most tropical and subtropical filariasis endemic regions. On the other hand, the prevalen ce of brugian filariasis is limited to Asia. B. malayi is found from India in the west to South Korea in the northeast and Indonesia in the southeast (King & Freedman, 2000). B. timori is located only in eastern Indonesia. Brugian filariasis accounts for 13 million among all filari al-infected people. Figure 1 shows endemic countries of LF worldwide.

PAGE 17

Figure 1. Lymphatic Filariasis Endemic Countries (WHO, 2002a) 6

PAGE 18

7 The disease is transmitted by several kinds of mosquitoes. W. bancrofti is transmitted by Culex quinquefasciatus Anopheles and Adeles Anopheles is also responsible for B. malayi transmission as well as Coquilletidia and Mansonia. B. timori is transmitted by Anopheles barbirostris (King & Freedman, 2000). The life cycle of the filarial parasite is identical to other vector-borne diseases. When the female mosquitoes ingest microfila riae with blood from infected humans, the microfilariae develop into infective filariworm larvae in the thoratic muscles of the mosquitoes, and travel to the lymphatics th rough the bloodstream. Then the larvae enter the human body when the mosquitoes feed on blood. In the lymphatic system, the larvae develop into adult worms and the female worms yield microfilariae, which reach the bloodstream again after the 6 12 months incubation period. Infected persons show one or more c onditions after the incubation period. The conditions are divided into three stages: ay mptomatic damage to the lymphatic system and kidneys, acute attacks of filarial fever, and chronic conditions such as elephantiasis and hydrocele. Asymptomatic infection is id entified by the observation of millions of microfilariae in the blood vessel or adult worms in the lymphatic system without the onset of symptoms. The damage to the lym phatic system may enhance a risk of acute attacks such as acute adenolymphangitis (ADL). Kidney damage may cause blood and protein loss in the urine. Clinically asympt omatic microfilaremia is the most common manifestation of bancroftian filariasis, but a large proportion of infected people do not show any symptomatic sign (King & Freedma n, 2000). It is estimated that 120 million people have asymptomatic condi tions worldwide (WHO, 2000b). An acute episode is characterized by seve re pain and inflammation of skin, lymph

PAGE 19

8 nodes, and lymphatic vessels, often accompanie d by fever, nausea, and vomiting. It is triggered by bacterial infections, which ente r through breaks in the skin. ADL appears not only during the asymptomatic stage but also during the onset of chronic conditions. The attack usually lasts 5 7 days and usually occurs several times each year. Though some people have only a few attacks in their lifetime, they are likely to have experienced asymptomatic or subclinical conditions for years prior to the acu te attack (King & Freedman, 2000). The progression of elephantia sis and fibrosis is typically observed during the attack. In the wo rld, 15 million people are currently suffering from acute attacks (WHO, 2000b). Because the disease is rarely fatal, chr onic symptoms are the severe burdens of infected people. The major symptoms in clude adenopathy, gen ital manifestations, lymphedema, and tropical pulmonary eosinoph ilia. Adenopathy is a painless enlargement of a lymph node due to the presence of adu lt worms in the lymphatic vessel, but lymph node enlargement may be the onl y clinical symptom in infect ed people, regardless of the presense of microfilaremia (Dreyer et al ., 1998). Genital manifestations include hematuria, hydrocele, chylocel e, chyluria, lymphedema and elephantiasis of the scrotal wall, penis, and lymph scrotum. In partic ular, hydrocele is th e most common genital impairment among males. Hydrocele is a symp tom that the sac around the testes becomes inflated like a baloon with a volume of fluid inside. It is estimated that 25 million men suffer from hydrocele worldwide (WHO, 2000b). Chylocele also shows the same disfigurement of genitalia, but intestinal chyle are included in side of the sac. It is less frequent than hydrocele. Chyluria and hema turia may happen among infected people. In some cases, chyle and blood are present in the urine. Lymphedema is a lymphatic

PAGE 20

9 dysfunction which dilates the lymphatic vessels du e to filarial infections. It appears most frequently on the lower extremities of the legs, and the severe condition called elephantiasis includes dermatosclerosis and papillomatous lesions as well as the swelling of a part of or an entire body. Currently 15 million people worldwide are impaired by lymphedema/elephantiasis (WHO, 2000b). Tr opical pulmonary eosinophilia is an asthma-like symptom with paroxymal nocturnal cough and anorexia, but its frequency is relatively unusual (D reyer et al., 1998). The strategy of the treatment and prev ention of LF is well established. WHO (2000a) recommends annual, single-dose, tw o-drug regimens of diethylcarbamazine (DEC) with albendazole or ivermectin to ge t rid of microfilariae from the blood. For the infected surface areas, mainta ining rigorous simple hygiene by careful cleansing is extremely helpful in managing the chronic c ondition and preventing acute attacks. Also, protecting the surface from mosquito bites by the use of bednets, insecticides, and repellents is effective for interruption of larvae transmission. Likewise, the control of mosquito-favorable environments such as larg e areas of water is also helpful. Surgical operation is another option for treatment of chronic conditions. However, though it is technically feasible and great improvement of the conditi ons can be expected, most endemic areas usually face financial limitations and inadequate medical resouces so that the operation is often not practical. DEC-fortified salt intake is also useful and costeffective for future prevention. Lymphatic filariasis is also called one of six eradicable diseases in the world (CDC, 1993). WHO has recently initiated the Global Programme to Eliminate Lymphatic Filariasis (GPEFL) to achive complete erad ication of filariasis by 2020. The strategy of

PAGE 21

10 the global elimination program includes tw o cost-effective and socially-responsible components: interruption of transmission, a nd morbidity control. The interruption of transmission is fairly achievable because humans are the only reservoir of W. Bancrofti which is responsible for 90% of all pa rasites causing LF (Ottesen, 2000). Also, the transmission of the filarial worm is less fr equent than with othe r parasite diseases (Ottesen, 2000). Thus, WHO designed the strate gic procedures to in terrupt transmission in two steps; districts in which lymphatic filariasis is endemic must be identified, and then community-wide (mass treatment) programs implemented to treat the entire at-risk population (WHO, 2000a). In most countries, one time annual simultaneous single dose administration of two drugs, 400 mg albendazole plus 6 mg/kg DEC, will be effective. The areas where either onchocerciasis or lo iasis may also be endemic require the same regimen but use 200 mcg/kg ivermectin instead of DEC for 4-6 years. An alternative community-wide regimen is to provide common table/cooking salt fortified with DEC in the endemic region for a period of one year (WHO, 2000a). This mass treatment gives equal effectiveness to at risk populations. On the other hand, morbidity control is targeted for those suffering from chronic conditions. Since the non-fatal disease ofte n causes both suffering and disability among patients for a long time, it is necessary to alleviate their conditi ons. There are several approaches to morbidity control, but WHO ( 2000a) states that it will be necessary to implement community education programs to raise awareness in affected patients. This would promote the benefits of intensive local hygiene and the possible improvement, both in the damage already occurred, and in preventing painful, acute episodes of inflammation and future infection (WHO, 2000a). Also, the creation of hope and

PAGE 22

11 understanding among the patients and their co mmunities are additiona l important factors for morbidity control. Lymphedema Lymphedema is one of the commom chronic symptoms among LF patients (Figure 2). It is currenly estimated that 15 million people are living with lymphedema due to lymphatic filariasis worldwide (WHO, 2000b) The disease is caused by the presence of adult worms in the lymphatic vessels. It is sometimes triggered by physical events such as injury or pregnancy. The worms dilate the lymphatic vessels and make the lymphatic system malfunction. Lymphedema us ually occurs in the lower extremities, but it also appears in the arms, breast, and urogeni tal organs. In the pres ence of the symptom on lower legs, Dreyer, Addiss, Dreyer, and No roes (2002) categorized its symptom into seven stages (Table 1, Appendix A). Most c onditions are either of stage 1, 2, or 3, but sometimes more extreme cases can be detected. Table 1. Lymphedema Stage and Its Ch aracteristics (Dreyer et al., 2002). Lymphedema Stage Characteristic Stage 1 Swelling goes away overnight. Stage 2 Swelling does not go away overnight. Stage 3 Shallow skin folds. Stage 4 Knobs. Stage 5 Deep skins folds. Stage 6 Mossy lesions. Stage 7 Unable to care for self or perform daily activities.

PAGE 23

Figure 2. A Condition of Lymphedema. Like other filariasis symptoms, the prevalence of lymphedema increases as infected people get older. Shriram, Murheker, Ramaiah, and Sehgel (2002) found that the increase was stable between 20 and 30 years of age, but that it became significant afterwards. The prevalence of the disease is also associated with gender. However, the effects are not consistent. Lammie et al. (1993) found that the incidence of lymphedema was 5 10 times greater among females than males in Haiti. On the other hand, in India, more males experienced lymphedema than females (Shriram et al., 2002). Acute attacks frequently occurred among lymphedema patients. In Tanzania, 61.3% of lymphedema patients developed ADL (Gasarasi, Premji, Mujinja, & Mpembeni, 2000). Also, ADL was more common among patients with lymphedema than hydrocele in Ghana (Gyapong, Gyapong, & Adjei, 1996b). In addition, recurrent bacterial infections facilitated the progression of lymphedema to elephantiasis (Dreyer et al., 1998). Comorbidity of 12

PAGE 24

13 lymphedema and other chronic symptoms, es pecially hydrocele in male, is another concern. In eastern India, Shriram et al. ( 2002) reported that 15.9 % of hydrocele patients had lymphedema, and 90 out of 565 (12.4 %) of male filariasis patients had both hydrocele and lymphedema in the southern st ate of Tamil Nadu (Ramaiah et al., 2000). Regular, intensive hygiene is a critical part of the prevention and treatment of lymphedema. Though drug therapy such as albendazole and DEC is effective as an antiparasitic treatment, daily washing and dryi ng is extremely important for minimizing the chance of future acute attacks and maximizi ng the degree of elepha ntiasis improvement. Elevation of the legs is also an important exercise for alleviating the condition. In addition, patient education is critical for cha nging fatalistic beliefs about the disease and maintaining their motivation to follow the regimen (WHO, 2000a). Surgical procedures are limited in resource poor, filarial endemic settings. Four Aspects of Lymphatic Filariasis As mentioned above, morbidity control is one of two pillars of eliminating LF globally. WHO (2000b) categorized suffering and disability of LF into four different aspects: physical, social, psychological, and economic. Physical components of disability include asymptomatic or symptomatic body conditions such as acute inflammatory attacks, disfigurement of the body, decreased mobility and function of limbs, obesity, a nd hidden disease (WHO, 2000b). The acute attack is a painful bacterial infection of th e skin and superficial tissues. The incidence increases as they get older, with a peak in the 50s (Gasarasi et al., 2000; Gyapong et al.,

PAGE 25

14 1996b). The impairment of the body influences daily activities, esp ecially among female patients. Ramaiah, Vijay Kumer, Ramu, Pani, and Das (1997) found that women suffering from chronic conditions had signifi cant difficulties in domestic chores. Also, there is a tendency for women to refrain fr om traveling due to the physical impairment (Bandyopodhyay, 1996; Coreil, Mayard, Louis-Char les, & Addiss, 1998; Ramaiah et al., 1997). In addition to the physical burdens, th ere are huge negative impacts on social, psychological, and economic conditions am ong chronic patients. From a social perspective, malfunction of the body increases the difficulty of self-care. Although it is unusual to have a self-untreatable condition, the limitation of the body function enhances the stress of care in daily lives. Patients also suffer from shame, stigma, and discrimination due to disfigurement of th e body. In Haiti, patients reported negative effects of their illness on family relations, and they experienced at least some sort of discrimination in the community (Coreil et al., 1998). Also, women experienced embarrassment, shame, cultural constraints, and social taboos preventing them from seeking help (Bandyopodhyay, 1996). People from the community often refuse to marry, sit beside, or eat with LF affected pers ons (Rauyajin, Kamthornwachara, & Yablo, 1995). Those suffering from genital impairment expe rience sexual disability. Above all, those with hydrocele experien ce severely impaired sexual func tion as well as decreased work capacity, and it has apparently negative effects on the QOL for the patients, families, and communities (Ahorlu, Dunyo, Asamoah, & Simonsen, 2001). Also, unmarried men with hydrocele found it difficult to find a spouse due to the condition, and married men experienced various degree of sexual dys function (Gyapong, Gyapong, Weiss, & Tanner,

PAGE 26

15 2000). Related to the social aspects, psychological burdens also influence patients lives. It is significant to observe the negative consequence in mental health. Typically, people suffer from depression, passivity, hopelessness, and fatalism, in some cases even leading to suicide (WHO, 2000b). Such phenome na might happen among both patients and family members. However, low-cost psychos ocial interventions su ch as support groups, offer significant benefits and satisfaction in allevieting the psychological burden of disease as well as improvement in QOL for people with LF in developing countries (Coreil, Mayard, & Addiss, 2003). Finally, economic limitation is the other ne gative aspect of LF. Because of the nature of the disease, it significantly aff ects poor people. The impairment minimizes both quantity and quality of work among infected pe ople. A large number of acute patients are completely absent from their jobs during th e onset of attacks, us ually 3 4 full days (Gasarasi et al., 2000; Gyapong, Gyapong, Evans, Aikins, & Adjei, 1996a; Ramaiah et al., 2000). Though the coping mechanism works among chronic patients, they also fail to pursue their works at the same level as befo re (Babu et al., 2002; Gyapong et al., 1996a; Ramaiah et al., 2000). In India, for example, 32 % of the total days suffering from attacks are considered as a loss of any economic wo rk among chronic patients due to lymphatic filariasis, and a loss of work is significant in those with lymphedema only, those with hydrocele only, and those with both lymphede ma and hydrocele (Ramaiah et al., 2000). Womens productivities are also impaired by the disease in endemic regions (Bandyopodhyay, 1996; Coreil et al., 1998). Leve l of absenteeism is influenced by disease conditions as well as other personal characteristics such as age, gender, and

PAGE 27

16 family type (Babu et al., 2002). Since the disease manifestations are more prevalent among householders, the family suffers from in come shortage. In India, about 7 % of household income goes for the treatment of LF, and the cost of hydrocele surgery exceeds more than one-third of the aver age household income (Babu et al., 2002). Therefore, economic loss eliminates the possibi lity of recovery and threatens daily life. Then the economic difficulty indirectly aff ects negatively on social and psychological impairment among both patients and family (B abu et al., 2002). As a result, a single disease causes a vicious cycle of physical, social, psychological, and economic burdens. Haiti The Republic of Haiti is located on the island of Hispaniola, surrounded by the islands of Cuba and Jamaica on the west and Puerto Rico on the east. Haiti occupies a land area of 27,700 km2 on the western-third of the island, and the rest is governed by the Dominican Republic. The nation consists of nine departments, 133 minicipalities, and 561 districts, with a total of 8.2 million people (PAHO, 1998; WHO, 2002b). Because of the high fertility rate (4.4 per woman), ch ildren under 15 years old accounts for 40 % of the total population (PAHO, 2003). On the other hand, people over 65 only account for 4 % due to low life expectancy at birth (52.8 years in males and 56 years in females) (PAHO, 2003). Haiti is also one of the most densely populated count ries in Central and Latin Americas. Owing to the small territory with rapid increase of the population, the population density is 260 inhabitants per km2 nationwide and 885 inhabitants per km2 of cultivated land, as of 1995 (PAHO, 1998). More than one-third of the total population

PAGE 28

17 (34.7%) are living in the capital Port-au-Prince, and there is a tendency toward rapid increase of the urban population in recent years (PAHO, 1998). This basic demographic information is summarized in Table 2. Table 2. Demographic Profile in Hai ti. (PAHO, 1998; WHO, 2002b; PAHO, 2003) Country name The Republic of Haiti Location N 19, W 72, western one-third of Hispaniola Land area 27,700 km2 Capital Port-au-Prince Administrative divisions 9 department s, 133 minicipalities, 561 districts Population 8.2 millions (1997) 15 years or younger 42 % 16-64 years 54 % 65 years or older 4 % % living in capital 34.7 % Total fertility rate (TFR) 4.4 Population density (nationwide) 260 people per km2 (1995) (cultivated land) 885 people per km2 (1995) Life expectancy at birth 52.8 years in male, 56 years in female Haiti is categorized as one of the least developed countries in the world. Social service infrastructure is poorly established or nonexistent. All types of infrastructure such as water, sewage and sanitation systems, e nviromental pollution due to the rapid increase of motor vehicles, and road networks are definitely inadequate (PAHO, 2003). Moreover, due to the chronic political instability and economic cris is, the actual gross domestic product showed decline to 1980s level in late 1990s, and the unemployment rate exploded instead (PAHO, 1998; World Bank, 2002) Currently, more than 80% of the total population is below the poverty line (World Bank, 2002). The gross domestic product (GDP) per capita was US$460 in 1999 (World Bank, 2002), or 1,094 international dollars in 2000 (WHO, 2002b), both of which are the least amount in the Americas.

PAGE 29

18 There is no systematic method to collect, process, and disseminate information on mortality in Haiti (PAHO, 1998). However, it is estimated that nearly one-half of all deaths happen during the first fi ve years of life due to diarrh eal diseases, acute respiratory infections, and malnutrition (PAHO, 1998). V accine-preventable diseases such as measles, diphtheria, and neonatal tetanus are still prevalent due to unsatisfactory vaccination rates. In 2000, vaccination covera ge of measles, DTP3, and BCG was 75%, 45%, and 55%, respectively (PAHO, 2003). As a result, a certain number of vaccinepreventable diseases occur intermittantly. For adolescents and adults, both communicable and noncommunicable diseases are common. Tropical vector-borne diseases such as malaria and dengue fever as well as lymphatic filariasis are endemic. Malaria is found throughout the country but its occurrence is more frequent in rural coastal areas, varying year to year in relation to season and the amount of rainfall (PAHO, 1998). In 1999, 973 cases and 59 deaths attributed to Plasmodium falciparun were reported (PAHO, 2003). Likewise, dengue fever has been prevalent particularly in urban areas, though the epidemiological data are insufficient to estimate its magnitude (PAHO, 2003). The outbreak of more severe fo rms of dengue, including dengue hemorrhagic fever and dengue toxic shock symdrome, is a constant threat throughout the country (PAHO, 1998). Other communicable diseases such as t uberculosis, HIV/AIDS, sexually transmitted diseases (STDs), and zoonoses are also found all over the country. However, poor surveilance systems often cause und erestimation in the entire nation. Haiti is one of the most endemic countries with lymphatic filariasis. The presence of bancroftian filariasis in Haiti originated from the history of African slave trade (King & Freedman, 2000). Although the exact nati onwide prevalence is unknown, it is

PAGE 30

19 estimated that more than 20 % of the populat ion in most coastal cities are carriers of microfilaremia (PAHO, 2001). In the city a nd surrounding areas of Leogane, for example, the prevalence of micr ofilaremia is about 25 %, approximately 5 % of women have a chronic symptom of elephantiasis, and th e prevalence of hydrocele among men is up to 30% (Eberhard et al., 1996). Recently a fila riasis mapping for children has completed (Beau de Rochears et al., in press). It shows that two major areas, Leogane and areas surrounding of Port-au-Prince, and the North De partment, record the highest prevalence of microfilaremia. Factors Associated with Lympha tic Filariasis and Lymphedema There are complicated issues in sociodemographic variables associated with lymphedema due to LF. Age is greatly associated with the onset of symptoms. The identification of both microfila ria and symptomatic conditions is rare in early childhood, but the incidence increases as people get ol der (Gasarasi et al., 2000; Gyapong et al., 1996b; Hyma, Ramesh, & Gunasekaran, 1989; King & Freedman, 2000; Shriram et al., 2002; Weerasooriya, Weerasooriya, Gunaward ena, & Samarawickrema, 2001). Likewise, the prevalence of lymphedema also increases as infected people get older. Shriram et al. (2002) found that, although the increase was st able between 20 and 30 years of age, it became significant afterwards. On the other hand, the gender profile remain s unclear. It is generally held that more females suffer from lymphedema than males. In Haiti, Lammie et al. (1993) found that the incidence of lymphedema was 5 10 times higher among females than males.

PAGE 31

20 However, the tendency seems inconsistent in other areas of the world. Shriram et al. (2002) showed that more males experienced lymphedema than females in India. No significant gender differences were observed among elephantiasis patients with filarial fever in Sri Lanka (Weerasoriya et al., 2001). The unclear gender perspective also tends to occur in other filariasis conditions. Seve ral studies mentioned that men generally had higher microfilaremia levels and some clini cal features such as hydrocele than women (King & Freedman, 2000; Weerasooriya et al ., 2001). Gasarasi et al. (2000) found that males experienced acute attacks more often than females in Tanzania; however, females had higher prevalence of acute attacks than males in Ghana (Gyapong et al., 1996b). Thus, the gender perspective is one of the important factors for consideration in LF elimination. More issues related to gender will be discussed in later chapters. Socioeconomic status (SES) is an anothe r important factor in LF. Filariasis principally affects persons of the lowest SES, especially those who are unable to protect themselves from mosquitoes sufficiently and/or who live in the mosquito-favorable environments (King & Freedman, 2000). Educational status as well as employment status of the household-head were positively associat ed with protection agai nst mosquito vector contacts (Mwobobia & Mitsui, 1999). Place of residence is also a source of variation. It is known that the prevalence and characteristics of LF varies by region or even within the same community (Dreyer et al., 1998). Also, different populations infected with the same parasite can have very different clinical ma nifestations of the infection (Ottesen, 1987). In Haiti, most research has been conducte d in the city and surr ounding areas of Leogane only. Thus, different aspects of lymphatic fila riasis would be observed within the country. In terms of knowledge of the illness, the la ck of current knowledge about filariasis

PAGE 32

21 is a significant risk factor for higher rates of morbidity among those affected. Reports from several countries indicate that poor knowledge about LF significantly execerbates conditions and contributes to high risk beha vior (Ahorlu et al., 1999; Eberhard et al., 1996; Gyapong et al., 1996a; Rauyajin et al., 19 95). Particularly, in spite of the obvious recognition of both acute and chronic sympto ms, there is almost no understanding of the etiology of these conditions and the role of mosquitoes that transmit the disease in Haiti (Coreil et al., 1998; Eberhard et al., 1996) In addition, the lack of knowledge is less likely to promote self-care practice at home. Si nce the simple daily care is important for the alleviation of the leg condition, the l ack of knowledge would lead to the poor compliance and the lower self-efficacy of th e self-care regimen at home, and therefore the QOL among patients would remain lower in th eir later life. However, there is little research about this issue and a more in-depth understanding is needed. Certain occupations contribute to th e high prevalence of LF. Because the distribution of the disease is in rural areas of the deve loping world, those engaging in agricultural-related work using water have higher chances of exposure. Pa rticularly, it was reported that the rainy season was associat ed with more frequent episodes of acute attacks (Gasarasi et al., 2000; Gyapong et al., 1996b). T hus, the type of occupation and the prevalence of filari asis seems to be related. However, there is no information about such relationships in Haiti. As mentioned in the section on lymphede ma, comorbid conditions related to filariasis are likely to affect lymphedema pa tients. An acute attack is one of the most common complications. In Tanzania, 61.3 % of lymphedema patients developed ADL (Gasarasi et al., 2000). Also, ADL is more frequent among patients with lymphedema

PAGE 33

22 than hydrocele in Ghana (Gyapong et al., 1996b). In addition, recurrent bacterial infections facilitate the progr ession of lymphedema to elephantiasis (Dreyer et al., 1998). Gender Perspective The gender perspective is an important aspe ct of LF. There is a lot of controversy regarding gender differences in LF. From an epidemiological standpoint, the prevalence of infection is often higher among men than women during the childbearing years (Dreyer et al., 1998). Also, males have hi gher microfilaremia levels and clinical symptoms because of a frequent conse quence of hydrocele (King & Freedman, 2000; Weerasooriya et al., 2001); in contrast, females experien ce a higher incidence of lymphedema and ADL (Gyapong et al., 1996b; King & Freedman, 2000). However, the opposite results have also been reported in different regions (Gasarasi et al., 2000). These outcomes indicate that the risk factors of infection by gende r vary by location. Though several studies have focused on females, there has been little attention to the disease among males in Haiti. At the same time, QOL among chronic patients is also a gender sensitive issue. Because the nature of the body disfigurement differs by gender, it affects their QOL in a different manner. A bove all, because of higher prevalence of chronic conditions and social ignorance about the diseas e among females, research among males is quite limited. The dominan t symptomatic condition among males, hydrocele, also makes it difficult to intervene in some cultures because of the location of the disfigurement. Therefore, this study al so aims at a gender comparison of filarial infection on risk factors and QOL.

PAGE 34

23 Quality of Life Lymphatic filariasis is know n as the disease which is the second leading cause of disability worldwide (WHO, 1995). Since mortality is rarely an issue in filariasis, acute attacks and several chronic conditions greatly affect their QOL. One of the obviously negative aspects is that acute attacks and clinical manifest ations limit physical activity. Reports in endemic countries indicate that both acute and clinical manifestations significantly hamper daily activities of inf ected people, especially women engaging in housework such as cooking, cleaning, washing and bringing up children (Ramaiah et al., 1997; Ramaiah et al., 2000). Male patients al so go through the impairment of sexual function due to hydrocele (Ahorlu et al., 2001). In the Haitian case, more than 70 % of patients with lymphedema in Haiti experien ced limited physical activ ities due to acute attacks (Dahl, 2001). In relation to physical activities, economic loss is another burden among infected persons. Many studies show that symptomatic conditions forced people to reduce their economic productivity or even stay at home all the time. In particular, those with ADL are affected much more se verely. Although chronic patients are usually capable of maintaining their economic activitie s at some level (Gyapong et al., 1996a), it is almost impossible for those with ADL to have any income during the period of symptoms (Ahorlu et al., 1999; Babu et al., 2002; Gyapong et al., 1996a; Ramaiah et al., 1997; Ramaiah et al., 2000). However, it is still significant that there is a regional difference in loss of work based on gender (Babu et al., 2002; Ramaiah et al., 2000). In addition, since the economic loss not only reduces their income but directly increases the

PAGE 35

24 proportion of treatment cost in a family, pati ents health-seek ing behavior is restricted (Gyapong et al., 1996a). Hence, they are more likely to endure their painful conditions. Psychosocial burden and QOL are strongly re lated in filariasis patients. Social stigma and discrimination prevent them from participating in the community and society as well as induce mental and psychological stress among them due to their abnormal physical features (Bandyopadhyay, 1996; Coreil et al., 1998). Patients with hydrocele also face a severe psychosocial burden (Ahorlu et al., 2001; Gyapong et al., 2000). However, surgical repair of hydrocele greatly improves QOL both physically and mentally, even though its cost is prohibitiv e in resource poor count ries (Ahorlu et al., 2001). Physical disability also makes a negative psychosocial impact on not only patients per se but also their family and community. Co reil et al. (1998) found that patients experienced difficulties in maintaining a good re lationship with their family, and that the family underwent social discrimination and ostracism from the community. Therefore, QOL among lymphedema patients is likely to be considered poor. On the other hand, effective morbidity c ontrol programs have a great impact on communities. Coreil et al. (2003) reported th at successful support group activities in a filariasis endemic area of Haiti showed significant improvement on patients QOL. However, even if the treatment regimen of lymphedema was well introduced in the community, patients often felt unconfortable fo llowing some practices such as bandaging and elevation of the legs due to discomfo rts, pain, itching, and tightness of bandages (Coreil et al., 1998). Therefore, it is importa nt to make an additional control strategy by careful understanding of QOL among them. There is no gold standard in QOL assessment for chronic patients due to

PAGE 36

25 lymphatic filariasis. Several different QOL measurement forms for chronic diseases have been established and widely available over the last few decades. Yet, there are challenges for QOL assessment of lymphedema patient s due to LF by using health-related QOL measurements. The latest try attempted QOL assessment of the introduction of hygiene and skin care regimens in Guyana, using the Dermatology Life Quality Index (DLQI) (McPherson, 2003). It was concluded that th e improved DLQI scores after the regimen indicated that the morbidity management by trained nurses was an effective intervention to enhance their QOL. However, the sample size was small (n=15), and the validity and reliability of the measurement was established only when lymphedema was considered as a skin disease. Another QOL study among lymphedema patients was conducted by the Nottingham Health Profile (NHP) to evaluate conservative treatment for general chronic lymphedema patients rather than lymphe dema due to LF (Sitzia & Sobrido, 1997). However, the reliability and validity of the questionnaires was not established, and the study concluded that the firs t part of NHP was less useful for QOL assessment among them so that the authors recommended ot her assessment tools such as the SF-36, the MOS 36-item short-form health survey (Ware & Sherbourne, 1992). Pereira de Godoy, Braile, de Fatima Godoy, and Longo (2003) recently examined the QOL among lymphedema patients with the SF-36. Though the objective of the assessment was designed not only for those due to LF but for general lymphedema patients, the researchers concl uded that both physical and mental health as well as social interaction among the lymphedema group showed a statistically significant lower QOL. However, in addition to the small sample si ze (n=23), neither does Ha itian Creole version

PAGE 37

26 of SF-36 exist in QualityMetr ic, Co. Ltd, the organization which retains the right to access SF-36, nor have translation efforts am ong researchers been accepted. Therefore, it is currently impossible to conduct the QOL assessment among Haitian lymphedema patients with the SF-36. Therefore, it would be beneficial to introduce other QOL assessment tools among lymphedema patients. Considering the characte ristics of the disease manifestation, there are several generic instruments which could apply to Haitian lymphedema cases. One of the most commonly used instruments is the EuroQol Instrument, developed by the EuroQol Group, a consortium of five European countries in 1987. It is designed to examine the feasibility of jointly developing a standardized no n-disease-specific instrument for describing and valuing health-related QOL (Brooks, 1996). The test consists of two parts: EuroQol five-demensi on health scale (EQ-5D) and EuroQol Visual Analogue Scale (EQ-VAS). EQ-5D consists of fi ve questions from five different healthrelated concepts each: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each question has three di fferent answers, and respondents have to choose one of them, on the basis of the curr ent days health status. Thus, there are 35 = 243 possible combinations to describe ones health. These outcomes are weighted by EuroQol group guidelines. On the other hand, the EQ-VAS measures ones overall health status by pointing a 20cm vertical visual analogue scale, which ranges from zero as a worst imaginable health state to 100 as a best imaginable health state. Like EQ-5D, EQVAS also assesses current health status. A nu mber of researchers have used the EuroQol in QOL assessment for various chronic dis eases (e.g. Fransen & Edmonds, 1999; Hurst, Kind, Hunter, & Stubbing, 1997; Konig et al., 2002; Myers & Wilkes, 1999).

PAGE 38

27 Another widely used form is the CDC Healthy Days questionnaire, developed by CDC (2000). It is a compact set of measures about recent perceived physical and mental health status and activity limita tion. The latest version of th e questionnaire consists of three parts: four general ques tions of self-determined health condition, five measures of acitvity limitation, and five additional statem ents regarding QOL. The first four core questions assess self-rated general health status, the duration of the unhealthy condition regarding physical and mental health, and the length of activity limitation due to both poor physical and mental health. Each questi on except the first one requires the answer during the past 30 days. The s econd part, five measures of acitvity limitation, assesses the major cause and duration of limitation and how it affects people's routine and personal care only if they have any activity limitation. The last pa rt evaluates the length of unhealthy days due to pain, depression, anxiety, sleeplessness, and vitality within last 30 days. CDC (2000) indicates that the Healthy Days measures are useful for identifying health disparities, tracki ng population trends, and building broad conditions around a measure of population health compatible with the definition of health by WHO. Particularly, the questionnaires are more focused on activity liminations so that the context would be suitable for the evaluation of lymphedema-oriented QOL. For assessing QOL in me ntal health, Center for Epidemiologic Studies Depression Scale (CES-D) developed by Radlof f (1977) in National Institute of Mental Health is widely used. It is a 20-question self-repor ting scale to evaluate ones depressive condition in the past week by four simple sc ores. Most of the a bove questionnaires ask about negative experiences but some a ssess positive atittudes of respondents.

PAGE 39

28 Reliability and Validity of the Quality of Life Measurements Since the reliability and validity of the generic QOL instruments are not yet established in filariasis-endemic Haitian communiti es, it is important to assure the quality of the measurements prior to a survey administration. Reliabili ty is the extent to which a measure yields consistent results In other words, it looks at th e extent to which scores are free of random error. Ary, Jacobs, and Razav ieh (2002) explains that three different consistency measures or reliability coefficient (rxy) are widely used for reliability measurement: 1) test-retest coefficient; 2) al ternate-form coefficient; and 3) internalconsistency coefficient. The test-retest co efficient is one derived from correlating individuals scores on the sa me test in two different administrations with a certain interval. The alternate-form coefficient co mes from correlated individuals scores on different sets of equivalent items. The cal culations of these two coefficients require laborious works due to at least two administrati ons of the same test or two equivalent test forms. On the other hand, the internal-consis tency coefficient is based on the relationship among scores derived from individual items or subsets of items within a single test so that only one administration of the test is necessary. Common measurements for internalconsistency coefficients are coefficient alpha developed by Cronbach (1951) and the Spearman-Brown formula for spli t-half reliability methods. Although reliability is relate d to the consistency of a scale as a measure of a specific variable, validity is associated w ith its adequacy. DeVellis (2003) summarizes that validity mainly consists of three categories: content validity, criterion-related

PAGE 40

29 validity, and construct validity. Content validity is the extent to which a specific set of measurements reflects a content domain. It is often easier to assess it when the domain is clearly defined, but more difficult when measuri ng attributes such as be liefs, attitudes, or disposition due to their ambiguous conceptu alization (DeVellis, 2003). Criterion-related validity is the extent to which a measure is empirically associated with another measure or procedure, such as a gold standard in th at field. Thus, it is used to demonstrate the accuracy of a measure used in a study. Construc t validity is the extent to which a question correlates with other measures that it shoul d correlate with. It indicates how much a theoretical construct or hypothesis agrees with a sp ecific measurement. It can be subcategorized into convergent validity and discriminate validity. Convergent validity looks at how much theoretically related meas ures agree with one another. Discriminant validity, on the other hand, indi cates how much theoretically unrelated measures show a lack of the relationship with one another. For the EuroQol, Brooks (1996) summarized the following issues. He notes that the test-retest reliability for the standard EuroQol questionnaires for the general Dutch population could be established by the ge neralizability study, a study focused on determining to what extent scores are comparable across different levels of a facet (van Agt, Essink-Bot, Krabbe, & Bonsel, 1994). They concluded that four different versions of questionnaires generated stable valuations over time. Good test -retest reliability (Intraclass Correlation Coefficient [ICC] = 0.78) was also obtained for the visual analogue scale of EuroQol instrument among healthy British people (Gudex, Dolan, Kind, & Williams, 1996). For those suffering from chronic conditions such as rheumatoid diseases and inflammatory bowel disease, ICCs of 0.70 0.85 indicated that there was

PAGE 41

30 moderate to high level reliab ility of the EuroQol questi onnaire (Fransen & Edmonds, 1999; Hurst et al., 1997; Konig et al., 2002). On the other hand, Brazier, Jones, and Kind (1993) made a comparison of the validity between the EuroQol instrument and the SF-36. They concluded that there was reasonable evidence for construct validity in the EuroQol dimension responses and the total scores, by comparing the percentages of each health problem with sociodemographic variables such as age. Also, th e authors figured out that convergent and discriminant validity had a reasonable agreement between EuroQol and SF-36 in the general population. Although the outcome of EuroQol was less variable than SF-36 due to the limited number of questionnaires and answers, they suggested that it was more applicable for a general population and suitable for those with major morbidity. The validity studies were performed for severa l chronic disease conditions, and similar performances of construct and criterion-related validity were observed (Fransen & Edmonds, 1999; Hurst et al., 1997; Konig et al., 2002; Myer s & Wilks, 1999). However, the results were not consistent by the type of disease, possibly because of ceiling effects due to the simplicity of the questionnaires in EuroQol. My ers and Wilks (1999) suggested that the EuroQol was a useful rapid-assessment means for chronic fatigue syndrome patients, and Coons, Rao, Keininger, and Hays (2000) indicated that EQ-5D was generally a preference-based measure de signed to summarize the QOL. However, no investigation has been conducted in lympha tic filariasis or chronic lymphedema. There is much less discussion of reliabil ity and validity of the CDC Healthy Days instrument, but CDC has documented releva nt studies in the publication Measuring Healthy Days. It reported that good internal consistency reliability was established

PAGE 42

31 among 2400 Norwegian adults (CDC, 2000, p.17). Also, acceptable test-r etest reliability was found among Americans who suffered from known disabilities (CDC, 2000, p.17). Validity of the measurements was more careful ly examined in various situations (CDC, 2000, pp.15 19). Good construct validity was establ ished in studies of statewide normal adults, low-income elderly, those with disabi lities, and low-income older male minorities, especially in terms of socio-demographic status and disease conditions. Also, acceptable or good correlations with SF-36 and CES-D que stionnaires were found in several healthrelated domains. However, like the EuroQol, the application of the CDC Healthy Days is a challenge in the Haitian LF cases. In contrast, such arguments for CES-D have been active for more than two decades. Radloff (1977) reported in his article that very high internal consistency was established by different types of reliabil ity measurements among general population and psychiatric patients. He f ound that reliability coefficients of from 0.77 to 0.92 were obtained by coefficient alpha, split-halves r, and spearman-brown methods. Also, moderate but stable test-retest reliability wa s observed in different intervals of retest administrations. All but one gained th e range of 0.45 0.70 on 2, 4, 6, and 8 weeks intervals of mail back administrations and 3, 6, and 12 months of reinterviews. The author also showed substantial to excellent evidence of criterion-related validity and construct validity. Criterion -related validity was establishe d by looking at patterns of correlations with other self-re port measures such as th e Lubin and Bradburn Negative Affect scale. Though the result was at an acceptable level, comparisons with the variety of self-reported measures helped document concurrent and discriminant validity. In addition, overall construct validity was established by examining correlations with

PAGE 43

32 clinical ratings of de pression and by relationships with other variables such as sociodemographic indicators. Therefore, CES-D is a strong indicator for looking at current mental health status so that the questionnaires are expected to be applied well among the lymphedema patients in Haiti.

PAGE 44

33 Chapter Three Methods Study Design The thesis is based on the analysis of secondary data obtained from a crosssectional, correlational study along with a de scriptive epidemiological survey. A crosssectional study is a field study that collect s data about activities, events, or other experiences at a single point of time. It is easier and less expens ive to perform than longitudinal studies, and allows us to genera te and test hypotheses. The disadvantages are that it cannot establish temporal relationships between variable s, there are potential biases in the selection of the population/sample and it cannot control potential confounders. However, the study also allows researchers to evaluate multiple risk factors and outcomes. A correlational study is a study that describes and postulat es the associations between variables of interests by using correlation coefficients. It is useful to obtain a first look at the population, but does not control for pot ential confounding factors. Finally, a descriptive st udy is a study that describes the existing distribution of variables without regard to causal associations. Ther efore, the thesis intends to investigate associations between socio-demographic vari ables and filariasis-related variables among Haitian patients and to describe the QOL among them at a certain point in time.

PAGE 45

34 Objectives The purpose of the study consists of tw o main objectives. Th e first objective is to observe the association of filariasis-related variables among lymphedema patients. The variables include demographic characteristics, illness history, knowledge of lymphedema, and self-care practice and behavior related to the illness. The risk factors for the filarial infection have been identified through various international studies; however, risk factors are sensitive to the environment and peopl es life style (King & Freedman, 2000). Also, different populations infected with the same parasite can have very different clinical manifestations of the infection (Ottese n, 1987). Moreover, Haiti is a country mixing western hemisphere with indigenous culture so that it is anticipated that any demographic variables, knowledge about the disease, or their unique daily custom and behavior may influence the disease. This cross-sectiona l correlational study also includes the consideration of regional differences and gender perspectives. The prevalence and characteristics of LF can vary by region or even within the same community (Dreyer et al., 1998). Above all, since there is no resear ch on the regional comparisons of LF in Haiti, the assessment of multiple communitie s is noteworthy for future filariasis management in Haiti. There are some differ ences known about the clin ical manifestations and social and behavioral factors by ge nder (Bandyopadhyay, 1996; Coreil et al., 1998), but they are still poorly understood. In Haiti, there has been no systematic comparison of gender perspectives, especially no representative information on men with LF in Haiti. Therefore, understanding the ge nder perspective of lymphatic filariasis is significant for

PAGE 46

35 morbidity management. It will be clarifie d by careful consideration of demographic, social, and behavioral factors associated with the disease. The other objective is to evaluate the QOL among lymphedema patients. As mentioned above, morbidity control is one of two main strategies for the global elimination of LF. However, since there are few systematic QOL assessments of lymphedema due to various limitations such as sample size and reliab ility and validity of the measurements, the obtained information has limited utility for morbidity control. Therefore, a general introduction of the QOL assessment is highly recommended. To fulfill the primary objective, secondary data collected for the ongoing project called Evaluation of Support Groups in the Management of Lymphedema Caused by Lymphatic Filariasis was used. The survey form was created on the basis of the questionnaires previously used in a survey in Leogane. In the project, the form was designed for the assessment of background information about LF in three rural Haitian towns. The data includes socio-demographi c information, illness history, foot exam, knowledge of the illness, self-care practice, and self-efficacy of the care. The form can be viewed in Appendix B. Regional assessments and gender comparisons were achieved by the evaluation of each questionnaire by regions and gender, respectively. The second objective, the QOL assessment is also a part of the ongoing project, but establishing the reliability and validity of th e instruments is an original work for this thesis. The generic QOL tools include the CDC Healthy Days Survey and EuroQol. CESD, Center for Epidemiologic Studies Depre ssion Scale was also applied for a future reference of mental health assessment of lym phedema patients. Due to the high illiteracy rate, interviews were conducted by trained pe rsonnel using Creole-language forms. Also,

PAGE 47

36 since there are few representative reports of re liability and validity of the questionnaires among lymphedema patients, a basic discussion of these issues was considered through application of commonly used validation procedures. Therefore, following research questions will be of interest in this research. Research questions 1. How is the condition of LF manifest ed in three rural Haitian towns? 1.1. To what extent does place of residence influence lymphedema conditions? 1.2. To what extent are socio-demographic variables such as age, gender, and SES associated with lymphedema conditions? 1.3. What is the knowledge of the illn ess and history of the illness? 1.4. How does the lymphedema condition inte ract with the frequency of comorbid condition such as acute attacks? 2. How do patients health-related behaviors aff ect morbidity control in their daily lives? 2.1. How do LF patients seek treatment for their legs? 2.2. How do place of residence, SES, and the cost of medication affect on treatmentseeking behavior? 2.3. To what extent do previous illness history and knowledge of the illness interact with help-seeking behavior? 2.4. What is the effect on self-efficacy? 3. How does the condition of lymphedema affect quality of life? 3.1. To what extent does the difference of the place of residence influence on QOL? 3.2. To what extent are differences of sociodemographic variables such as age, gender, and SES associated with the QOL among lymphedema patients? 3.3. To what extent are differences obser ved between lymphedema stage and QOL, such as morbidity, physical limitation, and anxiety/depression? 3.4. What is the association with knowledge of the illness and illness history? 3.5. To what extent do simultaneous symptoms exacerbate QOL?

PAGE 48

37 4. To what extent do the generic QOL asse ssment tools for chronic diseases have reliability and validity for measuring QOL among lymphedema patients due to LF in Haiti? Population and Sample Size The dataset came from an ongoing research project titled Evaluation of Support Groups in the Management of Lymphedema Caused by Lymphatic Filariasis. A collaboration between Haitian and US public he alth institutions expanded the successful support group intervention among lymphedema patients in Leogane, Haiti, which was funded by WHO (Coreil et al., 2003). The dataset included background information among LF patients living in one of the most LF endemic areas located north of the nations capital. Thus, the dataset was pr imarily aimed to provide the baseline information prior to support gr oup introduction in the new areas. The data were collected in three localit ies, Arcahaie, Cabaret, and La Plaine. Arcahaie is a coastal town situated 30 miles north of the capital, Po rt-au-Prince. Cabaret is another community located between Port -au-Prince and Arcahaie. Both towns are situated on or near the main norther n highway Route National No. 1 within L arrondissement dArcahaie (District of Arcahaie). Particularly, Cabaret is just right of the main highway so that it is more urbanized than Arcahaie. However, sampling in Cabaret included more residents from the rura l zones so that the effect of urbanization seems to be lower. On the other hand, La Pl aine is a dispersed ar ea that is a part of La Commune de Delmas, (Delmas County) on or near Route National No. 1. The community

PAGE 49

38 of Delmas is one of six communities in La rrondissement de Port-au-Prince (District of Port-au-Prince), which includes the metropolitan area. Therefore, residents in La Plaine have greater access to the capital due to its closeness to the metropolitan area. The locations of these localities except La Plaine are available in Appendix C. Though the exact population size in each town is unknown, it is estimated to be 100,000 for Arcahaie, 60,000 for Cabaret, and 10,000 for La Plaine. The target population was lymphedema patients of all ages. The prevalence of lym phedema is poorly understood because there is no case reporting on surveillance in these areas However, Beau de Rochars et al. (in press) figured out that these communities were in three of the most filariasis endemic areas in the country, based on the national mapping of infected school children. The towns are also located in one of two regions which have the highest prevalence of microfilaremia. Therefore, analysis of the da ta collected in these towns would be one of the most representative information about LF in Haiti. The sample size of the project was based mainly on the information collected on the preliminary census in early 2003. At that time, at least 60 and 25 adult lymphedema patients were found in Arcahaie and Cabare t, respectively. Since the census was conducted in a short period of time with a mi nimum effort, it was anticipated that many more people suffering from lymphedema were collectible as samp les. Also, population data were unavailable so that the statistical calculation of the sample size was impossible. Thus, the sample size was based on resour ces available for the study and patients available in the study site. Fortunately, a total sample of 316 were able to be collected including 123 in Arcahaie, 72 in Cabare t, 120 in La Plaine, and 1 unknown site.

PAGE 50

39 Sampling Method The sample was based on the availability of patients. Since the study was focused on a particular condition of a single disease and there was no list or information of the study population, non-probability sampling was conducted instead of probability sampling. Non-probability sampling is one of tw o sample selections that researchers can make a subjective decision for the characteristic s of samples. It is especially appropriate when resources for samples are limited, th e members of a population are difficult to identify, or the list of a population is unavail able. Because samples cannot be collected by random selection such as probability sampli ng, it is difficult to reduce or eliminate potential biases and confounders. There are several types of non-probability samplings. Henry (1990) lists six major designs of nonprobability sampling: convenience, most similar/dissimilar, typical cas es, critical cases, snowball, and quota. Of those, the approach in this project was more likely to be convenience sampling that a group of individuals who are available for a study, or quota sampling that interviewers select a member of samples until filling out quota. Interviews were conducted in the infec tious disease department of the local hospital in Arcahaie and the public clinic in Ca baret and La Plaine, or in people's homes. It was restricted by person-to-person intervie w only, due to the nearly 50% of nationwide illiteracy rate. A team of trained interviewers conducted the interviews. A small refreshment and transportation fee was provide d at the end of the interview. Informed consent was obtained prior to the data collection.

PAGE 51

40 Measurement The questionnaires used in this study ha d two different formats. The first form, Filariasis Baseline Evaluation Survey, was a set of questions adapted from a previous filariasis survey in Leogane, Haiti. It cons ists of six different cat egories: demographics, illness history, foot exam, knowledge about th e illness, self-care practices for leg, and self-efficacy. All the items were written in both E nglish and Haitian Creole, although the instruments were administered in Creole only. The contents of each category are summarized in Table 3. Also, the full quest ionnaires are available in Appendix B. Table 3. Brief Summary of the Survey Categories and Questionnaires. Demographic: address, gender, age, marital status, general information about their living children, religion, occupation, educational level, literacy, economic scale. Illness history: age of first awareness of the illness, first impression of the illness, recognition of the first symptom, treatment of the illness, precautions taken with ones foot, number of acute attacks in the past year, presence of comorbidity of lymphedema and acute attacks. Foot exam : the sizes of foot (10cm from toe), ankle (10cm from floor), and leg (20cm from floor) for both limbs, the stage of illness (stage 1 to 7, from mild to severe), location and condition of lesions. Knowledge about the illness : what the respondent thinks is the cause of the illness, which care options can help one's lymphedema, which care options can be done to help preven t acute attacks and what treatments are available, what kinds of care can be provided to ease one's acute attacks. Self-care practice for legs : what kinds of practices and how often one does each for self-care of the legs daily, once a week, once a month, or less often. Self-efficacy : how confident they feel in their ability to practice all the care techniques available to take care of their legs.

PAGE 52

41 The other form was a combination of tw o generic QOL assessment tools, the CDC Healthy Days Survey and EuroQol, a nd a subjective well-be ing assessment tool, CES-D. All were translated by researchers who engaged in research on LF in Haiti. As mentioned in the previous chapter, these form s were designed for assessing one's general health condition. Particularly, the CDC H ealthy Days is more focused on activity limitation, CES-D is a common mental health indicator, and the EuroQol evaluates five health dimensions of mobility, self-car e, usual activities, pain/discomfort, and anxiety/depression. The forms ar e available in Appendix B. Reliability and Validity In order to confirm to what extent the measurements have reliability, coefficient alpha or Cronbach alpha was calculated. One of the reasons that the Cronbach alpha was selected is that this is a cross-sectiona l study. Since the project was only a single time administration, test-retest relia bility could not be introduce d. The second reason is that, due to the limited period of survey ad ministration and no alternative equivalent assessment tools, alternate-form reliability was excluded for reliability measurement. In addition, it was impossible to use the KuderRichardson Procedure because there were no dichotomous measures in the questionnaires Therefore, the internal-consistency reliability formula was used to confirm the re liability of the instruments. The formula for coefficient alpha is as follows:

PAGE 53

42 = (K/(K-1))*((sx 2 si 2 ) / sx 2 ), where, K = number of items on the test Sx2 = variance of the test scores for all K-items Si2 = sum of variances of the item scores. There are some limitations for these test s. Coefficient alpha assumes that the items on the form are homogeneous. Since EuroQol and the CDC Healthy Days are designed for evaluating one's general health, the questionnaires include different domains of health status such as physical and mental health and disability. Thus, overall alpha might be lower than expected. On the other hand, establishing validity required more elaborate procedures. Criterion-related validity was observed by compar ing a single domain in one form with a related from in another, EuroQol and the CDC Healthy Days. Als o, CES-D is a mental health measurement only so that the questi onnaires in CES-D were used in comparison with comparable domains in the other two form s. The approach to construct validity was dependent on the previous findings or constructs that have al ready been established in the literature, because it is impossible to prove the validity of a measure. According to the literature review in the previ ous chapter, the relationships between certain characteristics of the symptoms and a dimension of the quest ionnaires would be expected. For example, since people with lymphedema in Haiti expe rienced limited physical activities due to acute attacks (Dahl, 2001), they would be e xpected to claim more activity limitations in the CDC Healthy Days. It has been reported that women with abnor mal physical features due to lymphedema experience mental a nd psychological stress (Bandyopadhyay, 1996; Coreil et al., 1998). This suggests that they would have a higher score on the

PAGE 54

43 anxiety/depression scale in EuroQol and CDC Healthy Days and in many of the questionnaires in CES-D. Also, SES would be expected to affect responses to the questionnaires. Thus, construct validity wa s established by analyses of the hypotheses shown above by using statistical tests such as correlational analys is. Lastly, convergent and discriminant validity was established by comparing the most similar and dissimilar health domains of the questionn aires of one form with anothe r. For example, the EQ-VAS scale in EuroQol and self-determined overall health condition in the CDC Healthy Days should be correlated each othe r because both domains address the general health of respondents. Likewise, mobility questions in EuroQol also should be correlated with activity limitation questions in the CDC Healthy Days. On the other hand, there should be no or weak relationships between anxiety and depression scales and physical health status. Content validity was not examined because it is beyond the purpose of this thesis. Data Analysis All the data were analyzed using SAS version 8.02 (SAS Institute Inc, Cary, NC). The Filariasis Baseline Evaluation Survey was described by simple descriptive statistics. Each category was summarized by variables of interest for a general profile of the sample. Some of the categories such as knowle dge about the illness, self-care practices for legs, and self-efficacy were utilized as QOL measurements so that simple bivariate statistical analysis such as ttest, correlation, chi-square te st, and others as appropriate were performed between variables of interest including socio-demogr aphic indicators and self-care practices for leg. On the other ha nd, scores from the QOL measurements were

PAGE 55

44 analyzed by following the guidelines for each QOL evaluation instrument, in addition to the establishment of reliability and validit y discussed in the previous section. The variables investigated in the Filariasis Base line Evaluation Survey were used for further analyses of QOL. For instance, the scores obtained in the QOL instruments were examined by variables of interest such as age and gender, and compared by simple statistical tests to observe th e differences between them.

PAGE 56

45 Chapter Four Results General Information The research was reviewed and approved on February 17, 2004, for ethics in human subject research by the Institutional Revi ew Board at University of South Florida. Demographics Table 4 shows the demographic character istics by gender and regions. Of 316 total respondents, there were 255 (80.7 %) females and 61 (19.3 %) males, and 120 people lived in Arcahaie, 72 in Cabaret, a nd 123 in La Plaine (1 missing). The gender proportion was slightly differe nt among towns. There were 2.78 times more females than males in Cabaret, but 6.23 times in La Plaine. The mean age was 44.52 years. No significant age differences were found for eith er gender or regions. There was almost no previous visit of Ste. Croix Hospital in Le ogane, where a lymphatic filariasis treatment program is currently available. The demographic characteristics were near ly identical for males and females as well as across towns. Statis tically significant differences were found only in marital status between gender ( 2=14.89, df=6, p=0.02), occupation by gender ( 2=90.07, df=5, p<0.01) and regions ( 2=44.26, df=10, p<0.01), and working days per week by gender

PAGE 57

46 (T=-3.11, df=81.7, p<0.01) and regions (F=8.76, df=2, 19, p<0.01). In marital status, plase was the most common status in both genders, but the proportion of single respondents was also the highest among males. Type of occupation varied by gender. More than 35 % of respondents were engage d in vending at home or market, but it was exclusively a female-dominated job. On th e other hand, agriculture-related work was more common among men. This fact may aff ect the difference on working days per week. However, there was no significant ge nder difference in average income. The regional differences were also obvious in occupation. There were more vendors in Arcahaie and La Plaine, but more farmers in Cabaret, probably due to the difference in gender distribution. Working days per week were also different by region (F=8.76, df=2, 194, p<0.01); however, no significant regional differences were found in average income.

PAGE 58

Table 4. Demographic Characteristics of Lymphedema Patients (#: 1 missing, *: % total, *: different between gender (p<0.05), **: different among towns (p<0.05), US $1 = 42.5 Haitian Groude (Gde) as of 2/18/04). Gender Town# Total Male Female Arcahaie Cabaret La Plaine Number of respondents (N (%*)) 316 (100) 61 (19.3) 255 (80.7) 120 (38.1) 72 (22.9) 123 (39.1) Age (mean SD (range)) 44.5 18.4 (9 95) 43.5 20.7 (9 95) 44.8 17.9 (9 90 ) 43.5 18.5 (10 90) 44.9 17.7 (12 86) 45.1 18.9 (9 95) Visit of Ste.Croix Hospital (n) 3 0 3 2 0 1 Married 47 (14.9) 10 (16.4) 37 (14.5) 18 (15.0) 8 (11.1) 21 (17.1) Plase 115 (36.4) 17 (27.9) 98 (38.4) 36 (30.0) 31 (43.1) 48 (39.0) Live together 9 (2.9) 2 (3.3) 7 (2.8) 4 (3.3) 3 (4.2) 2 (1.6) In relationship 21 (6.7) 5 (8.2) 16 (6.3) 10 (8.3) 4 (5.6) 7 (5.7) Single 44 (13.9) 17 (27.9) 27 (10.6) 17 (14.2) 7 (9.7) 20 (16.3) Separated/divorced 41 (13.0) 5 (8.2) 36 (14.1) 20 (16.7) 10 (13.9) 10 (8.1) Marital status (n (% N)) ** Other 36 (11.4) 5 (8.2) 34 (13.3) 15 (12.5) 9 (12.5) 15 (12.2) Those who have living children (n (%*)) 239 (75.6) 36 (59.0) 203 (79.6) 84 (70) 58 (80.6) 96 (78.1) Number of living children (mean SD (range)) 2.8 2.5 (0 11) 2.5 2.6 (0 8) 2.9 2.5 (0 11) 2.5 2.5 (0 9) 2.8 2.3 (0 8) 3.1 2.7 (0 11) Age of oldest child (mean SD (range)) 25.1 13.7 (0 65) 25.2 13.0 (2 55) 25.1 13.9 (0 65) 25.6 14.2 (1 65) 25.1 .6 (0 55) 24.6 13.5 (1 60) Age of youngest child (mean SD (range)) 15.8 11.0 (0 50) 13.4 10.0 (0 35) 16.3 11.2 (0 50) 16.8 10.6 (2 46) 14.8 12.1 (0 46) 15.6 10.9 (0 50) Number of children who go/went to school (mean SD (range)) 1.5 1.9 (0 11) 1.5 2.0 (0 7) 1.5 1.9 (0 11) 1.3 1.7 (0 6) 1.5 2.0 (0 7) 1.6 2.1 (0 11) (Continued on the next page) 47

PAGE 59

Table 4 (Continued). Gender Town Total Male Female Arcahaie Cabaret La Plaine Catholic 156 (49.4) 32 (52.5) 124 (48.6) 56 (46.7) 34 (47.2) 66 (53.7) Protestant 121 (38.3) 18 (29.5) 103 (40.4) 45 (37.5) 29 (40.3) 46 (37.4) Voodooist 11 (3.5) 1 (1.6) 10 (3.9) 5 (4.2) 2 (2.8) 4 (3.3) Religion (n (% N)) No religion or other 27 (8.5) 10 (16.4) 18 (7.1) 14 (11.7) 7 (9.7) 7 (5.7) Farmer 44 (13.9) 28 (45.9) 16 (6.3) 20 (16.7) 20 (27.8) 4 (3.3) Seller at home 66 (20.9) 1 (1.6) 65 (25.5) 18 (15.0) 16 (22.2) 32 (26.0) Seller at market 52 (16.5) 0 (0) 52 (20.4) 30 (25.0) 9 (12.5) 13 (10.6) Tailor / seamstress 17 (5.4) 2 (3.3) 15 (5.9) 4 (3.3) 1 (1.4) 11 (8.9) Unemployed 90 (28.5) 13 (21.3) 77 (30.2) 31 (25.8) 19 (26.4) 40 (32.5) Occupation (multiple answers) (n (% N)) ** Other 64 (20.3) 21 (34.4) 43 (16.9) 23 (19.2) 9 (12.5) 32 (26.0) Number of working days per week (mean SD) ** 5.2 2.0 5.9 1.5 5.0 2.1 4.4 2.2 5.6 .7 5.6 1.8 Income per day (Gde (US$)) 332.2 ($7.8) 415.1 ($9.8) 309.5 ($7.3) 416.5 ($9.8) 350.8 ($8.3) 250.5 ($5.9) Able to read and write (n (% N)) 192 (60.8) 41 (67.2) 151 (76.1) 74 (61.7) 38 (52.8) 79 (64.2) Number of school years completed (mean SD (range)) 3.9 4.3 (0 15) 4.1 4.3 (0 14) 3.8 4.3 (0 15) 4.4 4.5 (0 15) 2.9 4.0 (0 14) 3.9 4.3 (0 14) Radio 238 (75.3) 44 (72.1) 194 (76.1) 98 (81.7) 52 (72.2) 88 (71.5) Storage set 121 (38.3) 21 (34.4) 100 (39.2) 55 (45.8) 27 (37.5) 38 (30.9) Living room 69 (21.9) 11 (18.0) 58 (22.8) 18 (15.1) 16 (22.2) 35 (28.5) Number of people having: (n (% N)) Bicycle/motorcycle 141 (44.6) 31 (50.8) 110 (43.1) 76 (63.3) 35 (48.6) 30 (24.4) 48

PAGE 60

49 Illness History In order to describe the first experien ce of the illness and its related health behaviors that respondents had, the responses regarding illne ss history were described in Table 5 8. A chi-square test was conducted to look at gender and regi onal differences at the significance level of = 0.05. Table 5 shows the age at which the res pondent first noticed symptoms and what the perceived illness was. The mean age was 28.0 years old, and this was consistent for gender and regions. About a quarter (24.4 %) of respondents thought that the illness was a chill, followed by bad blood, gland, magi cal powder, and sprain. Also, 14.9 % of respondents couldnt identify th e illness. Surprisingly, onl y 2 people were able to recognize the illness as filariasis. A significant difference was observed both in gender ( 2=16.83, df=7, p=0.02) and towns ( 2=42.42, df=14, p<0.01). Those who answered bad blood were dominated by females, and mo re people in Cabaret thought it was gland rather than blood. Table 6 describes the first symptom no ticed. Swollen foot, swollen gland, and pain were common symptoms reported. There was a significant regional difference ( 2=55.61, df=12, p<0.01). More people in La Plaine experienced fever, headache, and hot foot than those in other towns. Howe ver, no significant gender differences were found. Table 7 shows how people treated the illness at that time. Nearly half of them relied on traditional medication such as an herbal remedy. Also, Western medicine was

PAGE 61

50 an alternative choice (95 or 30.1 % of them vi sited a hospital/clinic or health center and 83 or 26.3 % went to the pharmacies). Although no significant gender differences were found in their help-seeking behavior ( 2=13.35, df=7, p=0.06), females were more likely to follow traditional medicine. Regional differences were also evident ( 2=67.55, df=14, p<0.01). People living in Arcahaie preferred traditional healers, but those in La Plaine were dependent more on health professionals as well as herbal remedies and use of cupping/leeches. Table 8 describes what precautions people usually take with their affected legs. Keeping clean/hygiene, wash/soak/soap legs, herbal remedy, and wearing sandals/socks/stockings/shoes were the most common precautions taken. However, some of those who answered other practiced cont radictory precautions such as exposing legs to cold water or washing legs with cold water or urine, instead of avoiding cold water or washing legs with hot water. There was no significant gender difference in precautions taken, but regional differe nces were significant (2=50.74, df=20, p<0.01). People in Arcahaie preferred to use herbal remedies, pom ade, and leg covering such as sandals and stockings. Avoiding cold water and something cold or wet was a more common precaution in Cabaret. On the other hand, those in La Plaine tried to band, wrap, tie legs, not to walk on ground with bare feet, and to keep clean by washing legs more than residents of other areas.

PAGE 62

Table 5. The First Impression of the Illness (multiple responses, *:1 missing, **: different between gender (p<0.05), ***: different among towns (p<0.05)). Gender Town* Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Age of first awareness of the illness (mean SD (range)) 28.0 15.6 (1 95) 27.1 18.4 (1 95) 28.2 14.8 (1 78) 27.3 .6 (4 72) 27.1 15.3 (3 72) 29.4 16.8 (1 95) Chill 77 (24.4) 16 (26.2) 61 (23.9) 32 (26.7) 15 (20.8) 30 (24.4) Gland 62 (19.6) 15 (24.6) 47 (18.4) 13 (10.8) 23 (31.9) 26 (21.1) Magical powder 56 (17.7) 13 (21.3) 43 (16.9) 32 (26.7) 4 (5.6) 20 (16.3) Sprain 54 (17.1) 12 (19.7) 42 (16.5) 18 (15.0) 13 (18.1) 23 (18.7) Filariasis 2 (0.6) 1 (1.6) 1 (0.4) 0 (0) 1 (1.4) 1 (0.8) Dont know 47 (14.9) 7 (11.5) 40 (15.7) 17 (14.2) 15 (20.8) 14 (11.4) Blood 70 (22.2) 2 (3.3) 68 (26.7) 27 (22.5) 4 (5.6) 39 (31.7) First impression of the illness (n (% N)) ** *** Other 53 (16.8) 8 (13.1) 45 (17.7) 21 (17.5) 10 (13.9) 22 (17.9) Table 6. The First Symptom Noticed (n (% N), multiple responses, *:1 missing, **: different among towns (p<0.05)). Gender Town*** Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Swollen foot 266 (84.2) 53 (86.9) 213 (83.5) 100 (83.3) 55 (76.4) 110 (89.4) Pain 179 (56.7) 31 (50.8) 147 (57.6) 65 (54.2) 33 (45.8) 80 (65.0) Swollen gland 190 (60.1) 43 (70.5) 147 (57.6) 51 (42.5) 49 (68.1) 89 (72.4) Fever 141 (44.7) 29 (47.5) 112 (43.9) 39 (32.5) 23 (31.9) 78 (63.4) Headache 47 (14.9) 8 (13.1) 39 (15.3) 5 (4.2) 5 (7.0) 37 (30.1) Hot foot 115 (36.4) 18 (29.5) 97 (38.0) 30 (25.0) 14 (19.4) 71 (57.7) Other 70 (22.2) 12 (19.7) 58 (22.7) 36 (30.0) 14 (19.4) 20 (16.3) 51

PAGE 63

Table 7. Treatment Choice (n (% N), multiple responses, *:1 missing, **: different among towns (p<0.05)). Gender Town*** Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Traditional healer 33 (10.4) 7 (11.5) 26 (10.2) 22 (18.3) 3 (4.2) 8 (6.5) Herbal remedy 143 (45.3) 22 (36.1) 121 (47.5) 54 (45.0) 22 (30.6) 67 (54.5) Pomade 49 (15.5) 11 (18.0) 38 (14.9) 23 (19.2) 10 (13.9) 16 (13.0) Herbal leaves on legs 31 (9.8) 3 (4.9) 28 (11.0) 16 (13.3) 9 (12.5) 6 (4.9) Cupping / leeches 51 (16.1) 7 (11.5) 44 (17.3) 11 (9.2) 2 (2.8) 38 (30.9) Pharmaceutical medicine 83 (26.3) 21 (34.4) 62 (24.3) 31 (25.8) 24 (33.3) 28 (22.8) Health professional 95 (30.1) 18 (29.5) 77 (30.2) 18 (15.0) 20 (27.8) 57 (46.3) Other 49 (15.5) 2 (3.3) 47 (18.4) 15 (12.5) 14 (19.4) 19 (15.4) Table 8. Precaution for Legs (n (% N), multiple responses, *:1 missing, **: different among towns (p<0.05)). Gender Town*** Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Avoid cold water/something cold/being wet 34 (10.8) 5 (8.2) 29 (11.4) 9 (7.5) 14 (19.4) 11 (8.9) Band/bandage, wrap, tie legs 25 (7.9) 1 (1.6) 24 (9.4) 6 (5.0) 3 (4.2) 16 (13.0) Do not walk (put legs) on ground (mud) 33 (10.4) 5 (8.2) 28 (11.0) 8 (6.7) 3 (4.2) 22 (17.9) Keep clean/hygiene, wash/soak/soap legs 58 (18.4) 9 (14.8) 49 (19.2) 19 (15.8) 7 (9.7) 32 (26.0) Herbal remedy (leaves/herbs) 56 (17.7) 11 (18.0) 45 (17.6) 28 (23.3) 9 (12.5) 19 (15.4) Take medicine (komprime/flanax/dolex, etc,) 22 (7.0) 4 (6.6) 18 (7.1) 8 (6.7) 8 (11.1) 6 (4.9) Put pomade 47 (14.9) 10 (16.4) 37 (14.5) 23 (19.2) 13 (18.1) 11 (8.9) Raise legs 20 (6.3) 4 (6.6) 16 (6.3) 6 (5.0) 3 (4.2) 11 (8.9) Put sandal/socks/stocking/shoes 58 (18.4) 14 (23.0) 44 (17.3) 30 (25.0) 13 (18.1) 15 (12.2) Nothing 68 (21.5) 12 (19.7) 56 (22.0) 28 (23.3) 16 (22.2) 23 (18.7) Other 94 (29.7) 16 (26.2) 78 (30.6) 39 (32.5) 20 (27.8) 35 (28.5) 52

PAGE 64

53 Table 9 shows the history of acute attack s in the previous year. To determine gender and regional differences, the following st atistical tests were used as appropriate: T-test, F-test, and chi-square test The criterion of significance is = 0.05. Most people (94.6 %) experienced at least one attack in the previous year. Each attack lasted 10.6 days on average. Females had a significantly longer attack period than males (T=-2.59, df=298, p=0.01). People sought treatment in nearly half of the attacks. Of those, more than one-third of people went to a clinic, a nd 12.6 % visited an herb alist. The other sites mostly consisted of health centers or pha rmacies. Regional differences were significant ( 2=16.87, df=8, p=0.03). Particularly, people in Arcahaie preferred the herbalist, but those in Cabaret visited pharmacies more often. Approximately 5.1 days had elapsed between the onset of the illness and the visi t to health facilities, and there was a significant gender difference (T=2.99 df=213, p<0.01); interestingly, females stayed at home longer, though they had a longer attack period. In each attack, people visited health care facilities about 2.5 times, averaging 7.7 hours for a trip and 25.2 Gde (US $0.59) for each transportation fee. About 30 % of respondents received help from others during the visit to health facilities, and most helpers were family members. Besides accompanying the patient to the clinic, helpers mainly t ook over patients routin e household chores and prepared and administered medicine, he rbal remedies, and some other common treatments for lymphedema. The average co st of consultation per visit was 54.01 Gde (US $1.27), and for each attack the total average cost was 130.49 Gde (US $3.07). These fees include the cost of consultation only and exclude the cost of treatment. The consultation in Cabaret seemed to be less expensive.

PAGE 65

54 About three-quarters (74.4 %) of people w ho went to health care facilities obtained medicine. Also, 25.6 % of them receive d herbal remedies or pomade. Most of the tests/exams were blood exams. The cost wa s dependent on the type of treatment. The treatment related to Western medicine (test/e xam and shot/injection) was more expensive than other traditional procedures, except for herbal remedies. However, the average cost of treatment was inconsistent among its type, possibly due to the large number of missing responses. On the other hand, people selected herbal remedy more often than medicine when they made a treatment at home. More than two-thirds of people chose pharmacy medicines (flanax komprime ) or a common antibiotic (ampicillin). A significant difference was present both among gender ( 2=16.48, df=7, p=0.02) and towns ( 2=81.11, df=14, p<0.01). Particularly, females relied more on indigenous treatment such as herbal remedy or leg washing. Also, in comparison wi th people in Arcahaie and Cabaret, those in La Plaine were more concerned about th e treatment for legs generally, especially washing legs, taking medici ne, and raising legs. In more than 40 % of attacks, people r eceived some help from a third person during the illness. Mostly, family members or relatives took over respondents daily housework and assisted in their treatment, including herbal remedies, soaping legs, and use of pomades. No gender difference was found in the type of help sought. In contrast, regional differences were noted ( 2=13.32, df=2, p<0.01). Particularly, in comparison with the other cities, people in La Plaine sought more help in da ily activities than the treatment for legs. Also, in 58 % of attacks, the respondents werent able to engage in their work activities during the illness. The average number of missing days was 11.7

PAGE 66

55 days, which indicates that they were unable to work during the entire period of the attack. There were neither significant ge nder nor regional differences.

PAGE 67

Table 9. History of Acute Attacks (#: 1-3 missing, *: difference between gender (p<0.05), **: difference among towns (p<0.05), ***: The average cost of treatment obtained in health settings was calculated only for total due to a small number of the valid sample). Gender Town# Total (N = 316) Male (N = 61) Female (N = 255) Arcahaie (N = 120) Cabaret (N = 72) La Plaine (N = 123) People who had acute attacks during the past year (n (% N), T=total attacks) 299 (94.6) T = 462 58 (95.1) T = 93 241 (94.5) T = 369 113 (94.1) T = 187 69 (95.8) T = 106 116 (94.3) T = 166 Number of attacks (mean SD) 1.5 0.8 1.6 0.8 1.4 0.8 1.5 0.8 1.5 0.8 1.3 0.7 Duration of attack (mean days SD) 10.6 11.9 8.6 6.3 11.1 13.0 10.9 10.5 9.2 6.7 11.0 15.6 Clinic 165 (35.7) 44 (47.3) 121 (32.8) 66 (35.3) 39 (36.8) 60 (36.1) Herbalist 58 (12.6) 11 (11.8) 47 (12.7) 30 (16.0) 11 (10.4) 17 (10.2) Voodoo priest 7 (1.5) 1 (1.1) 6 (1.6) 5 (2.7) 1 (0.9) 1 (0.6) Other site 47 (10.2) 11 (11.8) 36 (9.8) 15 (8.0) 19 (17.9) 13 (7.8) Location of treatment sought (n (% T), multiple responses) ** Nowhere 212 (45.9) 32 (34.4) 180 (48.8) 79 (42.2) 42 (39.6) 88 (53.0) Days spent before Rx (mean SD) 5.1 8.1 3.2 2.7 5.6 9.1 5.6 9.3 3.2 4.2 5.7 8.5 No time 50 (10.8) 12 (12.9) 38 (10.3) 18 (9.6) 16 (15.1) 16 (9.6) No money 57 (12.3) 12 (12.9) 45 (12.2) 19 (10.2) 17 (16.0) 21 (12.6) Cant go out 31 (6.7) 10 (10.8) 21 (5.7) 15 (8.0) 1 (0.9) 15 (9.0) Treatment at home 23 (5.0) 2 (2.2) 21 (5.7) 8 (4.3) 5 (4.7) 10 (6.0) Other 44 (9.5) 7 (7.5) 37 (10.0) 14 (7.5) 11 (10.4) 16 (9.6) Reasons for days spent before treatment (n (% T), multiple responses) Missing, not applicable 278 (60.2) 24 (25.8) 133 (36.0) 122 (65.2) 60 (56.6) 96 (57.8) Number of visits for the same attack (mean SD) ** 2.5 1.7 2.3 1.7 2.6 1.8 2.4 1.5 2.1 1.5 2.9 2.2 Accompanied to health centers (n (% T)) 146 (31.6) 37 (39.8) 109 (29.5) 70 (37.4) 32 (30.2) 44 (26.5) Of those, by family / relative (n (% above)) 122 (83.6) 32 (86.5) 90 (82.6) 58 (82.9) 27 (84.4) 37 (84.1) Time to arrive at the place (hours SD) 7.7 23.7 7.6 10.5 7.7 26.7 6.7 10.1 9.6 11.6 7.6 39.5 Paid for transportation (n (% T)) 132 (28.6) 33 (35.5) 99 (26.8) 59 (31.6) 32 (30.2) 41 (24.7) (Continued on the next page) 56

PAGE 68

Table 9 (Continued). Gender Town# Total (T = 462) Male (T = 93) Female (T = 369) Arcahaie (T = 187) Cabaret (T = 106) La Plaine (T = 166) Sought care at health center (H (%)) 250 (54.1) 61 (65.6) 189 (51.2) 108 (57.8) 64 (60.4) 78 (47.0) Number of consultation visits per attack (mean SD)** 2.9 4.9 3.2 6.6 2.7 4.1 2.0 1.4 2.1 3.4 4.6 7.6 Cost per visit (mean Gde SD) ** 54.0 87.5 51.2 72.7 55.0 92.4 66.5 99.2 25.7 28.4 62.8 .0 Cost for total (mean Gde SD) 130.5 244.7 162.4.4 119.6.7 163.8.5 86.1.9 174.9.8 Shot / injection 55 (22.0), 163.6 232.8 18 (29.5) 37 (19.6) 15 (13.9) 24 (37.5) 16 (20.5) Medicine 186 (74.4), 44 (72.1) 174.5 271.1 142 (75.1) 82 (75.9) 42 (65.6) 62 (79.5) Herbal remedy 64 (25.6), 145.6 208.4 11 (18.0) 53 (28.0) 33 (30.6) 11 (17.2) 20 (25.6) Cupping / leech 30 (12.0), 65.9 85.7 2 (3.3) 28 (14.8) 10 (9.3) 0 (0) 20 (25.6) Pomade 64 (25.6), 76.2 148.9 12 (19.7) 52 (27.5) 21 (19.4) 20 (31.3) 23 (29.5) Massage 14 (5.6), 52.5 130.6 0 (0) 14 (7.4) 0 (0) 0 (0) 14 (17.9) Bandage 23 (9.2), 34.3 39.8 0 (0) 23 (12.2) 1 (0.9) 0 (0) 22 (28.2) Test / exam 20 (8.0), 314.2 434.8 3 (4.9) 17 (9.0) 5 (4.6) 1 (1.6) 14 (17.9) Other 11 (4.4), 380 0 1 (1.6) 10 (5.3) 5 (4.6) 1 (1.6) 5 (6.4) Treatment obtained in health setting (n (% H) **, cost (gde) SD, multiple responses) *** Missing 34 (13.6),N/A 8 (13.1) 26 (13.8) 0 (0) 10 (15.6) 24 (30.8) (Continued on the next page) 57

PAGE 69

Table 9 (Continued). Gender Town# Total (T = 462) Male (T = 93) Female (T = 369) Arcahaie (T = 187) Cabaret (T = 106) La Plaine (T = 166) Took medicine 125 (27.1) 27 (29.0) 98 (26.6) 37 (19.8) 12 (11.3) 76 (45.8) Herbal remedy 214 (46.3) 34 (36.6) 180 (48.8) 80 (42.8) 47 (44.3) 84 (50.6) Pomade 132 (28.6) 16 (17.2) 116 (31.4) 54 (28.9) 32 (30.2) 46 (27.7) Washed legs 183 (39.6) 33 (35.5) 150 (40.7) 52 (27.8) 27 (25.5) 101 (60.8) Raised legs 94 (20.3) 18 (19.4) 76 (20.6) 24 (12.8) 14 (13.2) 56 (33.7) Prayed 56 (12.1) 1 (1.1) 55 (14.9) 5 (2.7) 2 (1.9) 46 (27.7) None 20 (4.3) 4 (4.3) 16 (4.3) 6 (3.2) 10 (5.3) 4 (2.4) Other 63 (13.6) 6 (6.5) 57 (15.4) 25 (13.4) 10 (5.3) 28 (16.9) Self-care at home (n (% T)) ** Missing 76 (16.5) 22 (23.7) 54 (14.6) 33 (17.6) 29 (27.4) 14 (8.4) People received help (n (% T)) ** 193 (41.8) 41 (44.1) 152 (41.2) 76 (40.6) 31 (29.2) 86 (51.8) by family/relative (n (% above)) 166 (86.0) 35 (85.4) 131 (86.2) 64 (84.2) 28 (90.3) 74 (86.0) People who couldnt work during the attack (n (% T)) 270 (58.4) 59 (63.4) 211 (57.2) 105 (56.1) 71 (67.0) 91 (54.8) Days missing work (mean SD) ** 11.7 22.7 10.0 25.2 12.2 21.8 9.3 9.5 6.4 5.7 19.6 37.8 58

PAGE 70

59 Table 10 shows respondents tr eatment for legs in the previous attack. About 50 % of the respondents purchased goods for thei r treatment and prevention. Forty percent of those who purchased goods obtained sa ndals, and 36.4 % applied pomades. T-test indicated that the cost of sandals was more th an twice as expensive as that of pomades at the significance level of = 0.05 (T=4.48, df=166, p<0.01), but there was no significant difference in purchase history between gender. In contrast, regional preferences were apparent ( 2=20.51, df=6, p<0.01). Particularly, those in Cabaret preferred pomade over use of a bandage. Table 11 describes peoples daily activities. Among five different categories, more than half of them engaged in wash ing clothes, going shoppi ng, and going to church. Including go selling, however, these activ ities were exclusively womens roles (2>7.4, df=1, p<0.01, and RR>2.2). Unlike gender differences, regional differences were less obvious. Only the numbers of people going shopping and selling were significantly higher in Arcahaie than in the other towns. Th e average days of the activities were also calculated. Going selling occurred much more often than other activities. This is probably related to source of income.

PAGE 71

Table 10. Materials Purchased (n (% N), multiple answers, #: 1 missing, *: different among gender (p<0.05)). Gender Town# Total (N = 316) Male (N = 61) Female (N = 255) Arcahaie (N = 120) Cabaret (N = 72) La Plaine (N = 123) Cost (gde SD (range)) People who bought 160 (50.6) 31 (50.8) 129 (50.6) 66 (55.0) 40 (55.6) 54 (43.9) 123.2 131.7 (0 1000)* Shoes/sandals 90 16 74 44 10 36 159.5 119.9 (0 500)* Pomade 82 18 64 32 24 26 68.5 66.3 (0 350) Bandage 23 5 18 8 1 14 68.8 52.0 (0 175) Other 30 7 23 9 11 10 205.7 274.9 (0 1000)* Table 11. Daily Activities (n (% N), mean days per week SD), *: frequency different between gender (p<0.05), **: frequency different among towns (p<0.05), #: days different between gender (p<0.05), ##: days different among towns (p<0.05), &: 1 missing). Gender Town& Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Wash clothes ** 209 (66.3) 2.3 1.4 10 (16.4) 2.4 1.1 199 (78.4) 2.3 1.4 72 (60.0) 2.2 1.1 41 (57.8) 2.0 0.9 95 (77.2) 2.6 1.8 Go shopping ## 162 (51.4) 2.5 1.7 11 (18.0) 3.5 2.7 151 (59.5) 2.4 1.6 72 (60.0) 2.1 1.0 31 (43.7) 2.3 1.0 59 (48.0) 3.0 2.3 Go selling ** # ## 61 (19.5) 3.5 2.2 2 (3.3) 2.0 0 59 (23.4) 3.5 2.1 32 (26.9) 2.2 1.1 14 (19.7) 4.1 2.0 15 (12.3) 5.4 2.0 Go to church # 178 (56.5) 1.9 1.6 25 (41.0) 1.4 0.7 153 (60.2) 2.0 1.7 68 (56.7) 1.8 1.4 38 (53.5) 1.7 1.2 71 (57.7) 2.1 1.9 Take children to school ** 68 (21.9) 1.9 1.0 9 (14.8) 1.7 0.7 59 (23.7) 1.9 1.1 13 (10.9) 1.8 0.9 15 (21.1) 2.1 1.3 40 (33.6) 1.9 1.0 60

PAGE 72

61 Knowledge Table 12 describes respondents knowledge of the cause of the illness. More than half of them did not know a cause. Only 9.8 % of them could identify that the illness was due to an insect bite or worms. Twenty-one people thought that the illness was related to blood. Under the chi-square test, there wa s no gender difference in knowledge; however, regional differences were evident ( 2=31.52, df=10, p<0.01). People living in Arcahaie reported more knowledge about th e cause of the illness. Table 12. Cause of Illness (n (% N), multip le responses, *: 1 missing, **: different among towns (p<0.05)). Gender Town* ** Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Insect bite or worms 31 (9.8) 5 (8.2) 26 (10.2) 16 (13.3) 5 (6.9) 10 (8.1) Magic 33 (10.4) 8 (13.1) 25 (9.8) 23 (19.2) 3 (4.2) 7 (5.7) Sprain or foot injury 30 (9.5) 4 (6.6) 26 (10.2) 12 (10.0) 9 (12.5) 9 (7.3) Chill 43 (13.6) 13 (21.3) 30 (11.8) 22 (18.3) 8 (11.1) 13 (10.6) Other 41 (13.0) 6 (9.8) 35 (13.7) 22 (18.3) 9 (12.5) 10 (8.1) Do not know 177 (56.0) 31 (50.8) 146 (57.3) 50 (41.7) 43 (59.7) 83 (67.5)

PAGE 73

62 Foot Size and Illness Stage To observe the lymphedema condition and determine the relationship between this condition and the variables of interest, T-te st and F-test were used for foot size, and the chi-square test was conducted for illness stage. The level of significance is = 0.05 for all tests. Table 13 presents the results of a foot exam. The mean foot size was 24.6 cm for right and 24.8 cm for left. The ankle size was 26.2 cm for right and 27.0 cm for left. The size of the leg was 35.5 cm for right and 35.9 cm for left. Due to the nature of physical appearances, t-test revealed that there were significant differences in foot and leg sizes between genders (all p-values < 0.01); however interestingly, the sizes of legs, which measured 20 cm above the ground, were signifi cantly larger among females than males. There were no significant differe nces in ankles by gender. Also, there were no significant differences among towns, except left foot wa s larger in La Plaine (F=3.65, df=2, 309, p=0.03). Table 14 and Figure 3 show the distributi on of illness by stag e. More than twothird of peoples legs were categorized as st age 2 or lower, except for the left foot among those living in La Plaine. Some patients had lymphedema on one side only. No gender differences were observed for illness stage in both legs, but those who live in La Plaine experienced more severe symptoms of lym phedema, particularly of the left leg ( 2=13.95, df=2, p<0.01).

PAGE 74

63 Stages and sizes of feet are also summ arized in Table 15 and Figure 4, by current age (age group), age of onset, education, income, occupation, knowledge of the illness, and the number of acute attack s in the previous year. Only the number and/or proportion of stage 2 or lower was described in stages of illness. Current age was correlated with illness stage and leg size. In each foot location, the mean foot size increased with age (all p-values < 0.05). In particular people in their 50s experi enced more severe symptoms than younger cohorts. Also, the stage of the illness varied among the age groups but was more likely to be mild in younger groups. On the other hand, the other variables were less related to either stage or foot size. Significant variations we re observed only for size of right ankle by education and knowledge of the illness, and the size of left leg by occupation. Therefore, other va riables seem to be poorly a ssociated with foot size and stage. Table 13. Foot Exam (mean cm SD (range), *: different between gender (p < 0.05), **: different among towns (p < 0.05)). Gender Town Total Male Female Arcahaie Cabaret La Plaine Foot (right)* 24.6 3.0 (19 42) 25.8 3.5 (20 42) 24.4 2.8 (19 40) 24.6 3.4 (19.5 42) 24.2 2.7 (20 35) 24.9 2.7 (19 34) (left) ** 24.8 2.9 (18.5 38) 25.7 2.6 (20 32) 24.6 3.0 (18.5 38) 24.4 3.0 (18.5 38) 24.6 2.6 (20 32) 25.4 3.0 (20 34) Ankle (right) 26.2 4.9 (18 49) 27.1 6.3 (18 49) 26.0 4.5 (18 45) 26.0 4.2 (19 40) 25.7 4.6 (18 44) 26.8 5.6 (18 49) (left) 27.0 5.4 (17 57) 27.2 5.2 (21 57) 27.0 5.4 (17 51) 26.5 4.7 (17 48) 27.1 5.2 (18.5 48) 27.4 6.1 (19 57) Leg (right) 35.5 5.6 (23 54) 34.1 6.8 (23 54) 35.8 5.3 (24 54) 35.4 5.7 (23 49) 35.2 5.5 (25.5 53) 35.8 5.6 (24 54) (left) 35.8 6.4 (22.5 80) 33.9 5.1 (24.5 52) 36.3 6.6 (22.5 80) 35.1 6.0 (22.5 57) 36.1 7.3 (25.5 80) 36.5 6.2 (25 62)

PAGE 75

Table 14. Stage of Illness (n (% N), *: different among towns (p<0.05)). Gender Towns* Stage (left) Total (N = 314) Male (N = 59) Female (N = 255) Arcahaie (N = 119) Cabaret (N = 71) La Plaine (N = 123) 0 Normal 26 (8.3) 4 (6.8) 22 (8.6) 7 (5.9) 2 (2.8) 17 (13.8) 1 Swelling is reversible overnight 71 (22.6) 10 (16.9) 61 (23.9) 28 (23.5) 20 (28.2) 23 (18.7) 2 Swelling is not reversible overnight 125 (39.8) 32 (54.2) 93 (36.5) 61 (51.3) 31 (43.7) 33 (26.8) 3 Shallow skin fold 58 (18.5) 8 (13.6) 50 (19.6) 17 (14.3) 15 (21.1) 25 (20.3) 4 Knobs 21 (6.7) 2 (3.4) 19 (7.5) 3 (2.5) 1 (1.4) 17 (13.8) 5 Deep skin folds 6 (1.9) 1 (1.7) 5 (2.0) 0 (0.0) 2 (2.8) 4 (3.3) 6 Mossy lesions 7 (2.2) 2 (3.4) 5 (2.0) 3 (2.5) 0 (0.0) 4 (3.3) Stage 2 or lower 222 (70.7) 46 (67.9) 176 (69.0) 96 (80.7) 53 (74.7) 73 (59.3) Gender Towns Stage (right) Total (N = 314) Male (N = 60) Female (N = 254) Arcahaie (N = 120) Cabaret (N = 70) La Plaine (N = 123) 0 Normal 43 (13.7) 8 (13.3) 35 (13.8) 14 (11.7) 4 (5.7) 24 (19.5) 1 Swelling is reversible overnight 106 (33.8) 25 (41.7) 81 (31.9) 46 (38.3) 28 (40.0) 32 (26.0) 2 Swelling is not reversible overnight 84 (26.8) 14 (23.3) 70 (27.6) 33 (27.5) 24 (34.3) 27 (22.0) 3 Shallow skin fold 52 (16.6) 5 (8.3) 47 (18.5) 18 (15.0) 11 (15.7) 23 (18.7) 4 Knobs 21 (6.7) 3 (5.0) 18 (7.1) 8 (6.7) 1 (1.4) 12 (9.8) 5 Deep skin folds 7 (2.2) 4 (6.7) 3 (1.2) 1 (0.8) 2 (2.9) 4 (3.3) 6 Mossy lesions 1 (0.3) 1 (1.7) 0 (0.0) 0 (0.0) 0 (0.0) 1 (0.8) Stage 2 or lower 233 (74.3) 47 (78.3) 186 (73.3) 83 (77.5) 56 (80.0) 83 (67.5) 64

PAGE 76

Figure 3. Stage of Illness. illness stage (left) stage 123%stage 240%stage 62%stage 52%stage 47%stage 318%normal8% normal stage 1 stage 2 stage 3 stage 4 stage 5 stage 6 illness stage (right) stage 133%stage 227%normal14%stage 47%stage 317%stage 60%stage 52% normal stage 1 stage 2 stage 3 stage 4 stage 5 stage 6 (Continued on the next page) 65

PAGE 77

Figure 3 (Continued). illness stage (gender, left leg)0102030405060normalstage 1stage 2stage 3stage 4stage 5stage 6level of stage% respondents male (left leg) female (left leg) illness stage (gender, right leg)01020304050 60 normalstage 1stage 2stage 3stage 4stage 5stage 6level of stage% respondents male (right leg) female (right leg) (Continued on the next page) 66

PAGE 78

Figure 3 (Continued). illness stage (towns, left leg)0102030405060normalstage 1stage 2stage 3stage 4stage 5stage 6level of stage% respondents Arcahaie (left) Cabaret (left) La Plain (left) illness stage (towns, right leg)0102030405060normalstage 1stage 2stage 3stage 4stage 5stage 6level of stage% respondents Arcahaie (right) Cabaret (right) La Plaine (right) 67

PAGE 79

Table 15. Other Socio-demographic Variables vs. Stage of Illness (n (% N)) and Foot Sizes (mean cm SD) (*1: different total between left and right legs (N left / N right) due to the one side of leg missing among patients, *2: different among age group (p<0.05), *3: different among occupation (left only, p<0.05), *4: different among educational level (left only for stage, right only for ankle p<0.05), *5: different among knowledge of the illness (left only, p<0.05)). Stage 2 or lower *2 Foot*2 Ankle*2 Leg*2 N Left Right Left Right Left Right Left Right Age group (yrs) 19 or less 33 29 (87.9) 30 (90.9) 22.9 1.9 23.2 1.8 23.6 2.9 23.0 2.6 31.5 3.7 31.5 3.6 20 29 42 37 (88.1) 36 (85.7) 23.9 2.1 23.7 2.2 25.0 3.9 24.6 3.4 35.0 4.4 34.8 4.2 30 39 49 35 (71.4) 39 (79.6) 25.5 2.9 24.8 3.1 27.8 5.8 26.2 3.6 36.8 5.6 35.7 4.4 40 49 63 44 (69.8) 45 (71.4) 25.0 2.5 24.8 2.9 27.5 5.6 27.1 5.7 37.2 7.4 36.7 6.1 50 59 52 32 (61.5) 33 (63.5) 25.5 3.6 26.1 3.8 28.3 5.6 28.2 5.2 37.5 6.6 37.9 6.0 60 69 39 25 (64.1) 27 (69.2) 25.1 3.0 24.2 2.5 27.5 4.3 26.2 5.8 35.9 6.9 35.1 6.5 70 or more 33 17 (51.5) 21 (63.6) 25.3 3.4 24.9 2.8 28.4 6.6 27.1 4.7 35.0 7.2 34.1 5.3 Missing 3 3 (100.0) 2 (66.7) 23.3 1.2 24.0 1.0 22.0 2.0 25.0 2.6 32.3 1.5 34.0 4.4 Age of onset (yrs) 9 or less 15 9 (60.0) 12 (80.0) 24.6 3.3 23.9 2.6 27.0 6.5 24.7 4.6 35.0 6.7 34.1 6.0 10 19 92 64 (69.6) 68 (73.9) 24.7 2.7 24.7 3.3 26.8 5.8 26.2 5.2 35.4 5.8 35.0 5.5 20 29 57 44 (77.2) 41 (71.9) 24.6 2.9 24.2 2.6 26.9 5.3 26.2 5.6 36.0 6.8 36.0 6.0 30 39 51 38 (74.5) 38 (74.5) 25.5 3.2 25.7 3.6 27.5 3.2 27.0 4.3 36.5 4.9 36.7 4.9 40 49 33 25 (75.8) 28 (84.9) 25.0 3.6 24.0 2.2 26.2 5.3 25.5 4.0 35.5 6.3 34.7 5.2 50 or more 29 23 (79.3) 21 (72.4) 24.2 2.8 24.4 1.8 25.8 5.4 25.4 3.2 35.0 6.0 34.7 5.1 Missing 37 19 (51.4) 25 (67.6) 24.9 2.7 24.7 3.1 28.6 6.3 27.3 5.6 37.3 9.1 35.8 6.4 (Continued on the next page) 68

PAGE 80

Table 15 (Continued). Stage 2 or lower *4 Foot Ankle *4 *5 Leg*3 N Left Right Left Right Left Right Left Right Daily income (gde) 25 or less 33 25 (75.7) 23 (69.7) 24.6 2.4 24.8 4.1 26.5 5.1 25.9 4.4 34.7 5.9 34.2 4.8 26 50 28 18 (64.3) 21 (75.0) 25.1 2.5 25.6 3.9 27.7 4.4 27.4 5.3 36.9 5.9 36.6 6.2 51 100 29 / 28 22 (75.9) 21 (75.0) 25.1 3.8 24.6 2.5 27.5 5.4 27.1 5.5 36.1 6.4 36.2 5.0 101 200 25 13 (52.0) 18 (72.0) 26.2 2.9 25.3 3.0 27.5 3.3 27.7 6.1 36.3 4.9 36.8 5.4 201 400 29 18 (62.1) 19 (65.5) 25.4 3.3 25.0 2.2 28.9 9.0 26.7 4.6 37.6 7.6 36.7 6.0 401 or more 28 19 (67.9) 20 (71.4) 25.3 2.6 24.8 2.5 27.2 3.7 26.1 4.0 36.7 4.8 36.8 5.0 Missing/refused *1 142/143 107 (75.4) 111 (77.6) 24.3 2.8 24.2 2.8 26.3 5.2 25.6 4.7 35.3 6.9 34.7 5.7 Occupation Farmer 44 34 (77.3) 38 (86.4) 25.3 2.4 25.3 3.7 26.7 5.2 26.8 6.0 34.3 5.8 34.5 6.3 Seller at home *1 66 / 65 40 (60.6) 45 (69.2) 24.8 2.7 24.7 2.7 26.8 3.5 26.4 4.0 36.0 5.2 36.4 5.2 Seller at market 52 35 (67.3) 37 (71.2) 25.6 3.4 25.2 3.5 28.3 6.7 27.0 4.3 37.7 7.1 37.1 5.4 Tailor / seamstress 17 12 (70.6) 11 (64.7) 25.1 3.2 24.1 2.0 27.9 5.3 25.9 3.7 38.9 7.0 35.8 4.5 Other 64 51 (79.7) 54 (84.4) 24.5 3.0 24.2 2.6 26.6 5.6 25.7 5.4 35.2 5.7 35.0 5.6 Unemployed *1 88 / 89 63 (71.6) 62 (69.7) 24.3 2.7 24.2 2.6 26.4 5.2 25.8 5.3 35.2 6.9 34.7 5.9 Education (yrs) 2 or less (preparatory) 155 102 (65.8) 111 (71.6) 25.2 3.1 24.8 3.0 27.4 5.2 26.7 5.3 35.8 7.0 35.4 6.0 3 4 (elementary) 33 23 (69.7) 23 (69.7) 24.5 3.4 25.1 3.7 27.2 6.1 27.7 6.3 36.5 7.1 37.3 7.0 5 6 (intermediate) 40 25 (62.5) 33 (82.5) 24.7 2.8 23.9 1.7 27.7 7.0 24.8 2.9 36.0 7.5 35.1 5.8 7 10 (secondary) 40 31 (77.5) 30 (75.0) 24.5 2.4 24.5 2.7 25.7 5.1 25.6 4.2 35.2 4.1 35.1 3.8 11 or more (higher) 37 33 (89.2) 31 (83.8) 24.3 2.8 24.3 3.6 25.9 3.8 24.7 3.0 36.1 4.8 34.9 3.7 Missing/refused 9 8 (88.9) 5 (55.6) 24.2 2.2 24.4 2.8 26.6 3.3 27.6 4.0 34.9 3.1 36.0 3.6 (Continued on the next page) 69

PAGE 81

Table 15 (Continued). Stage 2 or lower Foot Ankle Leg N Left Right Left Right Left Right Left Right Knowledge Insect bite / worm 31 25 (80.7) 23 (74.2) 24.5 2.9 24.7 2.4 26.0 4.5 26.4 4.6 37.3 6.9 37.3 5.9 Magic 33 23 (69.7) 24 (72.7) 25.2 3.4 24.8 4.0 28.0 7.2 26.0 4.3 35.0 6.4 35.1 5.0 Sprain foot / injury *1 30 / 29 23 (76.7) 22 (75.9) 24.9 2.6 24.3 2.8 27.9 5.4 25.9 4.0 36.0 5.0 35.7 4.4 Chill *1 41 / 42 29 (70.7) 31 (73.8) 25.0 2.8 25.3 3.4 27.9 5.3 27.7 5.6 35.8 5.1 35.6 5.8 Other 41 29 (70.7) 29 (70.7) 24.6 2.4 25.3 4.1 27.5 3.8 28.0 5.6 35.9 5.0 36.9 6.2 Dont know 177 123 (69.5) 133 (75.1) 24.7 2.9 24.4 3.4 26.5 5.1 25.6 4.4 35.5 6.9 34.8 5.2 Number of acute attacks 0 16 13 (81.3) 11 (68.8) 23.8 2.7 24.2 2.0 25.3 4.6 26.0 3.2 35.1 6.5 35.8 5.3 1 177 123 (69.5) 137 (77.4) 24.8 3.0 24.4 2.6 27.2 6.0 25.9 5.0 36.2 7.1 35.3 5.5 2 *1 79 / 80 57 (72.2) 56 (70.0) 25.0 2.8 25.0 3.7 26.5 4.0 26.5 4.8 35.6 5.2 35.6 6.0 3 39 26 (66.7) 27 (69.2) 24.5 2.8 24.8 3.4 27.3 4.6 27.3 5.1 34.9 5.5 35.7 5.5 70

PAGE 82

Figure 4. Stage of Illness by Socio-demographic Variables. stage 2 or lower (age group)02040608010019 orless20 2930 3940 4950 5960 6970 ormoreage% left right stage 2 or lower (age of onset)0204060 80 1009 orless10 1920 2930 3940 4950 ormoreage% left right stage 2 or lower (daily income)02040608025 orless26 5051 100101 200201 400401 ormoreGde% left right (Continued on the next page) 71

PAGE 83

Figure 4 (Continued). stage 2 or lower (occupation)020406080100Farmer Seller at home Seller atmarket Tailor orseamstressOther Un -employed% respondents left right stage 2 or lower (education)0204060801002 or less3 4 5 67 1011 ormoreyears% respondents left right stage 2 or lower (knowledge)6065707580 85s Insectbite /wormMagicSprainfoot /injuryChillOtherDontknow% respondent left right (Continued on the next page) 72

PAGE 84

Figure 4 (Continued). stage 2 or lower (acute attack)0204060801000123number of acute attacks% respondents left right 73

PAGE 85

74 Self-care Practice and Self-efficacy The data regarding self-care practices and self-efficacy were analyzed to describe respondents health-related behaviors. Chisquare test was conducted to observe the differences of their behavior s by the variables of interest at a significant level of = 0.05. Table 16 shows the number of people who curre ntly practice self-care for affected legs. Hygiene wearing shoes, and herbal remedy we re among the most common self-care practices reported for legs. Of those, mo st people who answered hygiene and wearing shoes performed these practices as a daily care. On the other hand, the frequency of performing it was less often. Only about one-third of people who answered herbal remedy did it as a daily care. No gender diffe rences were found, but regional differences were significant ( 2=58.77, df=22, p<0.01). People in La Plaine reported taking care of their legs more often than those in other regions. Table 17 shows the self-care practices a ssociated with other socio-demographic variables. The table includes the five most common practices (hygiene, wearing shoes, elevation, herbal remedy, and pomade) due to the small number of re sponses in the other practices. The remaining practices were categor ized as other. However, the choice of self-care practices was not significantly associated with the variables of interest (age group, age of onset, education, income, o ccupation, and knowledge of the illness).

PAGE 86

75 Table 16. Gender, Town vs. Self-care Practi ce for Legs (n (% N), *: 1 missing, **: different among towns (p<0.05)). Gender Town* ** Total (N = 316) Male (N = 61) Female (N = 255) Arcahaie (N=120) Cabaret (N = 72) La Plaine (N=123) Hygiene 202 (63.9) 36 (59.0) 169 (66.3) 68 (56.7) 44 (61.1) 92 (74.8) Wearing shoes 238 (75.3) 41 (67.2) 197 (77.3) 77 (64.2) 54 (75.0) 106 (86.2) Permangan ate 41 (13.0) 8 (13.1) 33 (12.9) 10 (8.3) 5 (6.9) 26 (21.1) Creme 49 (15.6) 4 (6.6) 45 (17.6) 16 (13.3) 5 (6.9) 27 (22.0) Elevation 72 (22.8) 9 (14.8) 63 (24.7) 21 (17.5) 12 (16.7) 39 (31.7) Massage 29 (9.2) 6 (9.8) 23 ( 9.0) 5 (4.2) 3 (4.2) 21 (17.1) Exercise 21 (6.7) 6 (9.8) 15 (5.9) 8 (6.7) 5 (6.9) 8 (6.5) Bandage 54 (17.1) 9 (14.8) 45 (17.6) 12 (10.0) 8 (11.1) 34 (27.6) Medicine 59 (18.7) 16 (26.2) 43 (16.9) 26 (21.7) 11 (15.3) 22 (17.9) Herbal remedy 139 (44.0) 23 (37.7) 116 (45.5) 57 (47.5) 35 (48.6) 47 (38.2) Pomade 90 (28.5) 23 (37.7) 67 (26.3) 36 (30.0) 32 (44.4) 22 (17.9) Other 32 (10.1) 6 (9.8) 26 (10.2) 8 (6.7) 10 (13.9) 14 (11.4)

PAGE 87

Table 17. Other Socio-demographic Variables vs. Major Self-care Practices for Legs (n (% N), other indicates the sum of the rest of practices so that the number doesnt reflect the proportion of N). N Hygiene Wearing shoes Elevation Herbal remedy Pomade Other Age group (yrs) 19 or less 33 20 (73.4) 23 (69.7) 8 (24.2) 17 (51.5) 11 (33.3) 34 20 29 42 25 (59.5) 30 (71.4) 7 (16.7) 21 (50.0) 10 (23.8) 46 30 39 49 35 (71.4) 40 (81.6) 13 (26.5) 19 (38.8) 14 (28.6) 43 40 49 64 44 (68.8) 50 (78.1) 16 (25.0) 23 (35.9) 16 (25.0) 66 50 59 52 29 (55.8) 35 (67.3) 9 (17.3) 27 (51.9) 17 (32.7) 44 60 69 39 25 (64.1) 27 (69.2) 10 (25.6) 20 (51.3) 11 (28.2) 26 70 or more 34 25 (73.5) 30 (88.2) 9 (26.5) 12 (35.3) 9 (26.5) 25 Missing 3 2 (66.7) 3 (100.0) 0 (0) 0 (0) 2 (66.7) 1 Age of onset (yrs) 9 or less 15 8 (53.3) 12 (80.0) 1 (6.7) 8 (53.3) 5 (33.3) 17 10 19 92 56 (60.9) 64 (69.6) 18 (19.6) 42 (45.7) 30 (32.6) 94 20 29 58 45 (77.6) 49 (84.5) 16 (27.6) 28 (48.3) 15 (25.9) 49 30 39 51 32 (62.7) 37 (72.5) 13 (25.5) 22 (43.1) 8 (15.7) 59 40 49 34 20 (58.8) 23 (67.6) 10 (29.4) 15 (44.1) 12 (35.3) 23 50 or more 29 17 (58.6) 20 (69.0) 5 (17.2) 11 (37.9) 9 (31.0) 19 Missing 37 27 (73.0) 33 (89.2) 9 (24.3) 13 (35.1) 11 (29.7) 30 Education (yrs) 2 or less (preparatory) 157 101 (64.3) 118 (75.2) 32 (20.4) 65 (41.4) 41 (26.1) 110 3 4 (elementary) 33 19 (57.6) 19 (57.6) 5 (15.2) 18 (54.5) 9 (27.3) 27 5 6 (intermediate) 40 29 (72.5) 34 (85.0) 15 (37.5) 18 (45.0) 14 (35.0) 50 7 10 (secondary) 40 23 (57.5) 29 (72.5) 11 (27.5) 20 (50.0) 18 (45.0) 46 11 or more (higher) 37 28 (75.7) 31 (83.8) 9 (24.3) 15 (40.5) 5 (13.5) 41 Missing/refused 9 5 (55.6) 7 (77.8) 0 (0) 3 (33.3) 3 (33.3) 11 (Continued on the next page) 76

PAGE 88

Table 17 (Continued). N Hygiene Wearingshoes Elevation Herbal remedy Pomade Other Daily income (gde) 25 or less 33 27 (81.8) 28 (84.8) 12 (36.4) 14 (42.4) 9 (27.3) 30 26 50 28 18 (64.3) 22 (78.6) 6 (21.4) 10 (35.7) 10 (35.7) 21 51 100 29 20 (69.0) 22 (75.9) 7 (24.1) 10 (34.5) 5 (17.2) 23 101 200 25 11 (44.0) 19 (76.0) 2 (8.0) 12 (48.0) 7 (28.0) 12 201 400 29 13 (44.8) 16 (55.2) 6 (20.7) 15 (51.7) 9 (31.0) 34 401 or more 28 14 (50.0) 18 (64.3) 6 (21.4) 10 (35.7) 9 (32.1) 27 Missing/refused 144 102 (70.8) 113 (78.5) 33 (22.9) 68 (47.2) 41 (28.5) 138 Occupation Farmer 44 31 (70.5) 34 (77.3) 7 (15.9) 18 (40.9) 18 (40.9) 34 Seller at home 66 44 (66.7) 52 (78.8) 20 (30.3) 31 (47.0) 14 (21.2) 55 Seller at market 52 31 (59.6) 38 (73.1) 9 (17.3) 24 (46.2) 19 (36.5) 37 Tailor / seamstress 17 12 (70.6) 15 (82.4) 6 (35.3) 6 (35.3) 3 (17.6) 22 Other 64 39 (60.9) 48 (75.0) 14 (21.9) 24 (37.5) 17 (26.6) 74 Unemployed 90 64 (71.1) 69 (76.7) 25 (27.8) 41 (45.6) 21 (23.3) 90 Illness knowledge Insect bite / worm 31 21 (67.7) 26 (83.4) 9 (29.0) 15 (48.4) 12 (38.7) 49 Magic 33 18 (54.5) 20 (60.6) 4 (12.1) 20 (60.6) 5 (15.2) 21 Sprain/injure foot 30 19 (63.3) 23 (76.7) 7 (23.3) 14 (46.7) 6 (20.0) 21 Chill 43 24 (55.8) 31 (72.1) 5 (11.6) 26 (60.5) 16 (37.2) 34 Other 41 16 (39.0) 21 (51.2) 9 (22.0) 20 (48.8) 12 (29.3) 32 Dont know 177 125 (70.6) 139 (78.5) 46 (26.0) 71 (40.1) 47 (26.6) 252 77

PAGE 89

78 Table 18 reports other self-care practi ces which respondents do not currently do but can be done to help ones leg. Although th e number of responses was lower than that of self-care in general, its preference and frequency tended to be similar. Only slight gender differences were found in self-efficacy, but regional differences were significant ( 2=120.25, df=22, p<0.01). Because of the number of responses, people in La Plaine were much more willing to do their leg treatment. This tendency is similar to self-care practices currently being done for legs. Peopl e who answered other practices recorded 120, but 53 (16.8 %) and 18 (5.7 %) people res ponded nothing and dont know in that order. Other socio-demographic variables were also included in the anal ysis of potential self-care practices (Table 19). The select ion of the practices was dependent on the frequency in Table 18 (hygiene, wearing shoe s, herbal remedy, pomade, and other). The rest were categorized as other2. Unlike current self-care prac tices for legs, other practices which can be done in the fu ture varied significantly by education ( 2=44.06, df=25, p=0.01) and knowledge of the illness ( 2=73.21, df=25, p<0.01).

PAGE 90

79 Table 18. Gender, Town vs. Possible Leg Care in the Future (n (% N), *: 1 missing, **: different among towns (p<0.05)). Gender Town* ** Total (N = 316) Male (N = 61) Female (N = 255) Arcahaie (N=120) Cabaret (N = 72) La Plaine (N=123) Hygiene 108 (34.2) 18 (29.5) 90 (35.3) 17 (14.2) 17 (23.6) 74 (60.2) Wearing shoes 119 (37.7) 22 (36.1) 97 (38.0) 27 (22.5) 19 (26.4) 73 (59.3) Permangan ate 15 (4.7) 1 (1.6) 14 (5.5) 1 (0.8) 2 (2.8) 12 (9.8) Creme 14 (4.4) 0 (0) 14 (5.5) 1 (0.8) 0 (0) 13 (10.6) Elevation 29 (9.2) 3 (4.9) 26 ( 10.2) 6 (5.0) 6 (8.3) 22 (17.9) Massage 19 (6.0) 4 (6.6) 15 ( 5.9) 2 (1.7) 1 (1.4) 16 (13.0) Exercise 3 (0.9) 0 (0) 3 (1.1) 2 (1.7) 0 (0) 1 (0.8) Bandage 43 (13.6) 4 (6.6) 39 (15.3) 12 (10.0) 5 (6.9) 26 (21.1) Medicine 50 (15.8) 13 (21.3) 37 (14.5) 21 (17.5) 16 (22.2) 13 (10.6) Herbal remedy 114 (36.1) 18 (29.5) 96 (37.6) 49 (40.8) 21 (29.2) 44 (35.8) Pomade 79 (25.0) 16 (37.7) 63 (26.3) 34 (30.0) 26 (44.4) 19 (17.9) Other or dont know 120 (38.0) 20 (32.8) 100 (39.2) 47 (39.2) 35 (48.6) 37 (30.0)

PAGE 91

Table 19. Other Socio-demographic Variables vs. Possible Leg Care in the Future (n (% N), other2 indicates the sum of the rest of practices so that the number doesnt reflect the proportion of N, *: p < 0.05). N Hygiene Wearing shoes Herbal remedy Pomade Other Other2 Age group (yrs) 19 or less 33 13 (39.4) 16 (48.5) 14 (42.4) 13 (39.4) 10 (30.3) 21 20 29 42 11 (26.2) 15 (35.7) 15 (35.7) 11 (26.2) 15 (35.7) 29 30 39 49 20 (40.8) 20 (40.8) 16 (32.7) 8 (16.3) 21 (42.9) 28 40 49 64 15 (23.4) 18 (28.1) 18 (28.1) 11 (17.2) 27 (42.2) 36 50 59 52 20 (38.5) 21 (40.4) 23 (44.2) 15 (28.9) 21 (40.4) 29 60 69 39 14 (35.9) 14 (35.9) 19 (48.7) 11 (28.2) 11 (28.2) 16 70 or more 34 13 (38.2) 13 (38.2) 9 (26.5) 8 (23.5) 15 (44.1) 12 Missing 3 2 (66.7) 2 (66.7) 0 (0) 2 (66.7) 0 (0) 2 Age of onset (yrs) 9 or less 15 7 (46.7) 6 (40.0) 7 (46.7) 4 (26.5) 5 (33.3) 10 10 19 92 26 (28.3) 35 (38.0) 39 (42.4) 28 (30.4) 32 (34.8) 51 20 29 58 21 (36.2) 22 (37.9) 18 (31.0) 11 (19.0) 23 (39.7) 33 30 39 51 15 (29.4) 14 (27.5) 18 (35.3) 9 (17.6) 23 (45.1) 32 40 49 34 11 (32.4) 14 (41.2) 13 (38.2) 7 (20.6) 12 (35.3) 19 50 or more 29 4 (13.8) 14 (48.3) 11 (37.9) 9 (31.0) 9 (31.0) 14 Missing 37 14 (37.8) 14 (37.8) 8 (21.6) 11 (29.7) 16 (43.2) 14 Education (yrs) 2 or less (preparatory) 157 44 (28.0) 45 (28.7) 57 (36.3) 39 (24.8) 66 (42.0) 56 3 4 (elementary) 33 18 (54.5) 20 (60.6) 13 (39.4) 11 (33.3) 13 (34.4) 19 5 6 (intermediate) 40 13 (32.5) 15 (37.5) 16 (40.0) 8 (20.0) 16 (40.0) 26 7 10 (secondary) 40 9 (22.5) 23 (57.5) 16 (40.0) 14 (35.0) 6 (15.0) 43 11 or more (higher) 37 10 (27.0) 12 (32.4) 11 (29.7) 5 (13.5) 15 (40.5) 22 Missing/refused 9 4 (44.4) 4 (44.4) 1 (11.1) 2 (22.2) 4 (44.4) 7 (Continued on the next page) 80

PAGE 92

Table 19 (Continued). N Hygiene Wearing shoes Herbal remedy Pomade Other Other2 Daily income (gde) 25 or less 33 4 (12.1) 5 (15.2) 11 (33.3) 2 (6.1) 16 (48.5) 13 26 50 28 7 (25.0) 6 (21.4) 7 (25.0) 5 (17.9) 10 (35.7) 16 51 100 29 8 (27.6) 8 (27.6) 10 (34.5) 3 (10.3) 17 (38.6) 10 101 200 25 16 (64.0) 17 (68.0) 12 (48.0) 8 (32.0) 6 (24.0) 12 201 400 29 10 (34.5) 13 (44.8) 12 (41.4) 8 (27.6) 13 (44.0) 19 401 or more 28 15 (53.6) 17 (60.7) 7 (25.0) 9 (32.1) 14 (50.0) 21 Missing/refused 144 48 (33.3) 53 (36.8) 55 (38.2) 44 (30.6) 44 (30.6) 78 Occupation Farmer 44 7 (15.9) 8 (18.2) 17 (38.6) 10 (22.7) 18 (40.9) 17 Seller at home 66 24 (36.4) 25 (37.9) 21 (31.8) 12 (18.2) 29 (43.9) 37 Seller at market 52 18 (34.6) 22 (42.3) 20 (38.5) 17 (32.7) 19 (36.5) 27 Tailor / seamstress 17 7 (41.2) 7 (41.2) 5 (29.4) 1 (5.9) 10 (58.4) 18 Other 64 27 (42.2) 33 (51.6) 24 (37.5) 14 (21.9) 16 (25.0) 48 Unemployed 90 29 (32.2) 28 (31.1) 31 (34.4) 25 (27.8) 36 (40.0) 39 Illness knowledge Insect bite / worm 31 12 (38.7) 15 (48.4) 13 (41.9) 11 (35.5) 8 (25.8) 34 Magic 33 2 (6.1) 4 (12.1) 12 (36.4) 12 (36.4) 15 (45.5) 7 Sprain/injure foot 30 2 (6.7) 3 (10.0) 15 (50.0) 6 (20.0) 11 (36.7) 7 Chill 43 13 (30.2) 14 (32.6) 17 (39.5) 14 (32.6) 15 (34.9) 21 Other 41 7 (17.1) 8 (19.5) 18 (43.9) 10 (24.4) 19 (46.3) 12 Dont know 177 78 (44.1) 82 (46.3) 55 (31.1) 36 (20.3) 69 (39.0) 106 81

PAGE 93

82 Table 20 shows the degree of confidence for leg care. Nearly 90 % of people reported that they were at least somewhat conf ident in their ability to care for their legs. Like self-care practices for legs, there we re no significant gender differences, but people in La Plaine reported higher confidence for their leg care than those in other towns ( 2=21.79, df=2, p<0.01). Table 20. Degree of Confidence for Leg Care (n (% N), *: 1 missi ng, **: different among towns (p<0.05)). Gender Town* ** Total (N = 316) Male (N = 61) Female (N = 255) Arcahaie (N = 120) Cabaret (N = 72) La Plaine (N = 123) Very confident 247 (78.2) 50 (82.0) 197 (77.3) 85 (70.8) 49 (68.1) 113 (91.9) Somewhat confident or below 69 (21.8) 11 (18.0) 58 (22.7) 35 (29.2) 23 (31.9) 10 (8.1)

PAGE 94

83 Quality of Life One of the most interesting research ques tions is to what extent the lymphedema condition affects quality of lif e, measured by the standardized QOL and subjective wellbeing instruments. To describe it, the scores of EuroQol, CES-D, and CDC Healthy Days were analyzed. Also, in order to observe the association between the outcomes of the scales and the variables of interest, the T-te st and F-test were used to compare mean differences. At the same time, the chi-squa re test was performe d for the categorical outcome. For the criterion of significance, = 0.05 was used. EuroQol Table 21 presents the results of the Eu roQol questionnaire. More than 70 % of respondents reported no problems in four of fi ve health categories. Above all, none of them had extreme problems in mobility or se lf-care. Only six people were unable to perform their usual activities. Also, anxiety/ depression was less likely to be severe among them. A gender difference was found only for mobility ( 2=5.67, df=1, p=0.02). Females were 2.02 times more likely than males to have problems in walking (RR=2.02, 95% CI: 1.07
PAGE 95

84 but people in La Plaine repor ted less pain. A similar patt ern was also found for overall health status (F=12.19, df=2, 312, p<0.01). Figure 5 shows the clear regional differences visually. Though not significant, the degree of anxiety/depression wa s different across towns ( 2=5.34, df=2, p=0.07). Thus, the results in dicated that people in La Plaine experienced milder symptoms of LF than those in other places. The results of the EuroQol scale we re also compared with other sociodemographic variables, including stage of th e illness. Table 22 shows the number and percent of people not reporting any health problems in the five categories and the average health score, and Figure 6 depicts the difference of overall he alth status by the variables of interest. Of those variables, age group and educational level had a strong association with QOL measures. Mobility, us ual activities, pain/discomfort, and overall health status varied significantly within them (all p-valu es < 0.05). Though not si gnificant, age group was also associated with self-care (p-values < 0.10). In general, older people were more likely to report poorer health status, and people who comple ted more formal education tended to maintain better health status. Also, QOL varied by stage of the illness. There were significant differences in mobility and usual activities by stage of both legs (all pvalues < 0.05); however, the othe r health categories in cluding overall hea lth status varied significantly by stage of one leg only. The ot her variables had a partial impact on the QOL measures. Occupation seems to substantia lly impact mobility, usual activities, and overall health condition (all p-values < 0.05), but the small number of respondents in tailor/seamstress category and th e large number in other occupation may negatively affect the outcomes so that it would be difficult to ob serve the impact of the variable correctly. The QOL scores were also significantly different in usual activities for age of onset, self-

PAGE 96

85 care for knowledge of the illness, and pain/discomfort for both categories (all p-values < 0.05); however, it is difficult to observe the pa tterns of the scores. No differences were observed by income.

PAGE 97

Table 21. Gender, Town vs. EuroQol (#: 1 missing, *: different between gender (p<0.05), **: different among towns (p<0.05)). Gender Town# Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) No problem 231 (73.1) 52 (85.3) 179 (70.2) 83 (69.2) 58 (80.6) 89 (72.4) Mobility (n (% N)) Some problem 85 (26.9) 9 (14.8) 76 (29.8) 37 (30.8) 14 (19.4) 34 (27.6) No problem 287 (90.8) 56 (91.8) 231 (90.6) 107 (89.1) 65 (90.3) 114 (92.7) Self-care (n (% N)) Some problem 29 (9.2) 5 (8.2) 24 (9.4) 13 (10.8) 7 (9.7) 9 (7.3) No problem 233 (73.7) 45 (73.4) 188 (73.7) 84 (70.0) 57 (79.2) 91 (74.0) Some problem 75 (23.7) 12 (19.7) 63 (24.7) 32 (26.7) 13 (18.1) 30 (24.4) Unable to perform 6 (1.9) 3 (4.9) 3 (1.2) 3 (2.5) 2 (2.8) 1 (0.8) Ambiguous response 1 (0.3) 1 (1.6) 0 (0) 0 (0) 0 (0) 0 (0) Usual activities (n (% N)) Missing 1 (0.3) 0 (0) 1 (0.4) 1 (0.8) 0 (0) 0 (0) No problem 138 (43.7) 27 (44.3) 111 (43.5) 37 (30.8) 32 (44.4) 69 (56.1) Some problem 158 (50.0) 30 (49.2) 128 (50.2) 70 (58.3) 35 (48.6) 52 (42.3) Extreme 19 (6.0) 3 (4.9) 16 (6.3) 12 (10.0) 5 (6.9) 2 (1.6) Pain/discomfort (n (% N)) ** Ambiguous response 1 (0.3) 1 (1.6) 0 (0) 1 (0.8) 0 (0) 0 (0) No 237 (75.0) 45 (73.4) 192 (75.3) 83 (69.2) 51 (70.8) 103 (83.7) Moderate 63 (19.9) 14 (23.0) 49 (19.2) 25 (20.8) 18 (25.0) 19 (15.5) Extreme 9 (2.9) 1 (1.6) 8 (3.1) 7 (5.8) 1 (1.4) 1 (0.8) Anxiety /depression (n (% N)) Missing 7 (2.2) 1 (1.6) 6 (2.4) 5 (4.2) 2 (2.8) 0 (0) Overall health status (mean SD) ** 57.1 17.3 59.7 22.8 56.5 18.5 53.8 15.5 53.2 14.0 63.0 18.9 86

PAGE 98

Figure 5. Gender, Town vs. Overall Health Status by EuroQol. Overall health status by EuroQol57.159.756.553.853.263.03040506070totalmalefemaleArcahaieCabaretLa Plainescore (mean, 0 100) 87

PAGE 99

Table 22. Filariasis Related Variables vs. People Answered No Problem in EuroQol Questionnaires (n (% N)) and EuroQol Overall Health Status (mean SD) (*1: different among age group, *2: different among age of onset, *3: different among education level, *4: different among occupation, *5: different among knowledge of the illness, *6: different among right leg, *7: different among left leg (all p<0.05)). N Mobility*1 Self-care Usual activities*1 *2 Pain / discomfort *1 *2 Anxiety / depression Overall health status *1 Age group (yrs) 19 or less 33 28 (84.8) 30 (90.9) 30 (90.9) 20 (60.6) 29 (87.9) 63.6 17.8 20 29 42 33 (78.6) 41 (97.6) 34 (81.0) 22 (52.4) 29 (69.0) 58.1 15.5 30 39 49 39 (79.6) 47 (95.9) 42 (85.7) 21 (42.9) 40 (81.6) 61.2 15.6 40 49 64 51 (79.7) 61 (95.3) 54 (84.4) 33 (51.6) 46 (71.9) 55.0 16.0 50 59 52 33 (63.5) 44 (84.6) 34 (65.4) 16 (30.8) 39 (75.0) 54.6 17.0 60 69 39 26 (66.7) 33 (84.6) 22 (56.4) 12 (30.8) 28 (71.8) 56.4 20.4 70 or more 34 19 (55.9) 28 (82.4) 12 (35.3) 11 (32.4) 23 (67.6) 49.7 14.9 Missing/refused 3 2 (66.7) 3 (100.0) 3 (100.0) 3 (100.0) 3 (100.0) 86.7 23.1 Age of onset (yrs) 9 or less 15 12 (80.0) 14 (93.3) 12 (80.0) 4 (26.7) 10 (66.7) 64.0 10.6 10 19 92 71 (77.2) 84 (92.1) 77 (83.7) 44 (47.8) 70 (76.1) 57.6 17.1 20 29 58 41 (70.7) 52 (89.7) 39 (67.2) 31 (53.4) 43 (74.1) 57.9 18.2 30 39 51 41 (80.4) 45 (88.2) 38 (74.5) 15 (29.4) 36 (70.6) 57.3 15.8 40 49 34 25 (73.5) 31 (91.2) 28 (82.4) 19 (55.9) 28 (82.4) 53.5 19.1 50 or more 29 20 (69.0) 28 (96.6) 17 (58.6) 5 (17.2) 23 (79.3) 56.9 18.1 Missing/refused 37 21 (56.8) 33 (89.2) 22 (59.5) 20 (54.1) 27 (73.0) 55.1 18.5 (Continued on the next page) 88

PAGE 100

Table 22 (Continued). N Mobility Self-care*5 Usual activities*3 *4 Pain / discomfort *3 *5 Anxiety / depression Overall health status *3 *4 Education (yrs) 2 or less 157 106 (67.5) 136 (86.7) 100 (63.7) 54 (34.4) 108 (68.8) 53.1 16.8 3 4 33 23 (69.7) 31 (93.9) 26 (78.8) 15 (45.5) 26 (78.8) 59.4 18.2 5 6 40 32 (80.0) 38 (95.0) 33 (82.5) 25 (62.5) 35 (75.0) 60.5 15.7 7 10 40 33 (82.5) 39 (97.5) 36 (90.0) 20 (50.0) 32 (80.0) 62.0 14.2 11 or more 37 31 (83.8) 35 (94.6) 30 (81.1) 19 (51.4) 28 (75.7) 60.5 18.8 Missing/refused 9 6 (66.7) 8 (88.9) 8 (88.9) 5 (55.6) 8 (88.9) 68.9 21.5 Occupation Farmer 44 34 (77.3) 39 (88.6) 34 (77.3) 19 (43.2) 33 (75.0) 55.2 17.0 Seller at home 66 45 (68.2) 59 (89.4) 42 (63.6) 25 (37.9) 49 (74.2) 55.0 15.5 Seller at market 52 36 (69.2) 48 (92.3) 40 (76.9) 19 (36.5) 37 (71.2) 59.2 16.4 Tailor/seamstress 17 15 (88.2) 15 (88.2) 14 (82.4) 12 (70.6) 14 (82.4) 58.2 18.1 Unemployed 90 57 (63.3) 80 (88.9) 59 (65.6) 40 (44.4) 68 (75.6) 52.6 18.1 Other 64 54 (84.4) 61 (95.3) 56 (87.5) 34 (53.1) 50 (78.1) 64.2 17.4 Illness knowledge Insect bite 31 21 (67.7) 28 (90.3) 22 (71.0) 8 (25.8) 22 (71.0) 58.4 13.9 Magic 33 21 (63.6) 26 (78.8) 19 (57.6) 7 (21.2) 18 (54.5) 53.9 17.1 Sprain 30 23 (76.7) 25 (83.3) 22 (73.3) 10 (33.3) 20 (66.7) 58.7 14.8 Chill 43 31 (72.1) 39 (90.7) 30 (69.8) 13 (30.2) 32 (74.4) 51.9 13.7 Other 41 29 (70.7) 34 (82.9) 27 (65.9) 15 (36.6) 29 (70.7) 52.7 16.7 Dont know 177 133 (75.1) 169 (95.5) 138 (78.0) 93 (52.5) 141 (79.7) 58.0 18.8 (Continued on the next page) 89

PAGE 101

Table 22 (Continued). N Mobility*6 *7 Self-care Usual activities*6 *7 Pain / discomfort *7 Anxiety / depression Overall health status *6 Daily income (gde) 25 or less 33 25 (75.8) 27 (81.8) 20 (60.6) 14 (42.4) 24 (72.7) 54.5 17.0 26 50 28 22 (78.6) 26 (92.9) 22 (78.6) 12 (42.9) 18 (64.3) 53.2 19.3 51 100 29 21 (72.4) 26 (89.7) 21 (72.4) 11 (37.9) 21 (72.4) 52.8 17.1 101 200 25 20 (80.0) 23 (92.0) 20 (80.0) 9 (36.0) 19 (76.0) 58.4 14.9 201 400 29 23 (79.3) 29 (100.0) 24 (82.8) 16 (55.2) 23 (79.3) 59.7 15.9 401 or more 28 20 (71.4) 27 (96.4) 23 (82.1) 9 (32.1) 21 (75.0) 58.9 15.9 Missing/refused 144 100 (69.4) 129 (89.6) 103 (71.5) 67 (46.5) 111 (77.1) 58.3 17.9 Stage of right leg Normal 43 40 (93.0) 41 (95.3) 39 (90.7) 25 (64.1) 34 (79.1) 67.4 16.0 Stage 1 106 80 (75.5) 96 (90.6) 81 (76.4) 50 (47.2) 79 (74.5) 56.1 15.5 Stage 2 84 60 (71.4) 80 (95.2) 64 (76.2) 35 (41.7) 65 (77.4) 57.3 14.1 Stage 3 52 34 (65.4) 44 (84.6) 29 (55.8) 15 (28.8) 38 (73.1) 51.9 17.5 Stage 4 or more 29 15 (51.7) 24 (82.8) 19 (65.5) 12 (41.4) 19 (65.5) 55.9 26.1 Missing/NA 2 2 (100.0) 2 (100.0) 1 (50.0) 1 (50.0) 2 (100.0) 35.0 7.1 Stage of left leg Normal 26 21 (80.8) 25 (96.2) 24 (92.3) 18 (69.2) 20 (76.9) 61.2 21.4 Stage 1 71 61 (85.9) 70 (98.6) 60 (84.5) 37 (52.1) 60 (84.5) 56.8 14.4 Stage 2 125 88 (70.4) 109 (87.2) 91 (72.8) 51 (40.8) 84 (67.2) 57.8 17.3 Stage 3 58 38 (65.5) 51 (87.9) 39 (67.2) 17 (29.3) 46 (79.3) 53.6 16.3 Stage 4 or more 34 21 (61.8) 30 (88.2) 18 (52.9) 14 (41.1) 25 (73.5) 59.7 20.1 Missing/NA 2 2 (100.0) 2 (100.0) 1 (50.0) 1 (50.0) 2 (100.0) 30.0 0.0 90

PAGE 102

Figure 6. Other Socio-demographic Variables vs. Overall Health Status by EuroQol. EQ-VAS (age group)63.658.161.255.054.656.449.7304050607019 orless20 2930 3940 4950 5960 6970 ormoreage groupscore (mean, 0 100) EQ-VAS (age of onset)64.0 57.6 100) 57.957.353.556.930.040.050.060.070.09 or less10 1920 2930 3940 4950 ormoreagescore (mean, 0 EQ-VAS (daily income)54.553.252.858.459.758.9304050607025 orless26 5051 100101 -200201 -400401 ormoregdescore (mean, 0 100) (Continued on the next page) 91

PAGE 103

Figure 6 (Continued). EQ-VAS (education)53.159.460.562.060.530405060702 or less3 45 67 1011 ormoreyearscore (mean, 0 100) EQ-VAS (occupation)55.255.059.258.252.664.23040506070farmerseller at home seller at markettailor/seamtressunemployedotherscore (mean, 0 100) EQ-VAS (knowledge) 58.4 100) 53.958.751.952.758.03040506070insectbitemagicsprainchillotherdon'tknowscore (mean, 0 (Continued on the next page) 92

PAGE 104

Figure 6 (Continued). EQ-VAS (stage of illness)67.456.157.351.955.961.256.857.853.659.73040506070normalstage 1stage 2stage 3stage 4 ormorestagescore (mean, 0 100) left right 93

PAGE 105

94 CES-D Table 23 shows the total scores on the CES-D scale. The mean score and its standard deviation were 13.2 and 9.5, respectively. For gender comparison, females were more likely to have higher score than males (T=1.95, df=314, p=0.05), but regional differences were more marked (F=58.03, df =2, 312, p<0.01). Especially, people living in La Plaine had much higher depression scores than those from the other areas. Since people with a total scor e of 16 or above are generally c onsidered as a depressive case (Eaton et al., 2003), Table 20 also shows the proportion of respondents scoring above 16. More than one-third of the total sample had a score of 16 or more. The score of people in La Plaine was prominently higher than the comparison groups. Total CES-D scores and the proportion of depressive cases in relation to other socio-demographic variables are shown in Ta ble 24. There were no significant statistical differences in both scores and depressive cases for current age, age of onset, income, educational level, and knowledge of the illne ss. Only the stage of the illness was strongly associated with depression scores (p-values < 0.05); however, no linear or systematic variation was observed.

PAGE 106

95 Table 23. Gender, Town vs. Total CES-D Score (mean SD, (Range)) and the Proportion of the Score Indicating Depres sive Cases (n (% N)) (*: 1 missing, **: different among towns (p<0.05)). Gender Town Total (N=315) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) CES-D score ** 13.2 9.5 (0 41) 11.1 8.4 (0 27) 13.7 9.7 (0 41) 9.3 7.0 (0 33) 9.1 6.7 (0 24) 19.4.8 (0 41) below 16 198 (62.6) 42 (68.9) 156 (61.2) 102 (85.0) 58 (80.6) 37 (30.1) 16 or more 118 (37.3) 19 (31.2) 99 (38.9) 18 (15.0) 14 (19.4) 86 (69.9)

PAGE 107

96 Table 24. Other Socio-demographic Variable s vs. Total CES-D Score (mean SD) and the Proportion of the Score Indicating De pressive Cases (n (% N)) (*: p<0.05 for both score and depression level). N Score Depress ion level N Score Depress ion level Age group (yrs) Occupation 19 or less 33 10.9 9.1 12(36.4) Farmer 44 10.8 7.5 9 (20.5) 20 29 42 12.3 8.7 14(33.3) Sell home 66 15.7 10.6 31(47.0) 30 39 49 14.4 9.9 19(38.8) Sell market 52 12.3 10.4 16(30.8) 40 49 64 12.6 8.7 21(32.8) Tailor 17 14.3 8.5 6 (35.3) 50 59 52 11.9 8.6 15(28.8) Other 64 12.9 9.4 29(45.3) 60 69 39 13.6 9.9 18(50.0) Unemployed 90 13.7 9.1 35(38.9) 70 or more 34 16.3 .2 17(50.0) Missing 3 24.3 .5 2 (66.7) Illness knowledge Insect bite 31 12.2 9.9 11 (35.5) Age of onset (yrs) Magic 33 11.6 7.4 10 (30.3) 9 or less 15 13.7 9.0 6 (40.0) Sprain 30 11.9 7.4 10 (33.3) 10 19 92 11.2 8.4 26(28.3) Chill 43 11.6 9.0 14 (32.6) 20 29 58 15.3 9.6 27(46.6) Other 41 10.5 7.7 10 (24.4) 30 39 51 13.4 8.9 19(37.3) Dont know 177 14.6 10.3 76 (42.9) 40 49 34 11.8 .1 11(32.4) 50 or more 29 14.3 .0 12(41.4) Stage of right leg Missing 37 14.9 .8 17(45.9) Normal 43 15.2 10.3 20 (46.5) Stage 1 106 12.4 8.6 34 (32.1) Daily income (gde) Stage 2 84 10.3 9.2 22 (26.2) 25 or less 33 15.4 7.3 16(48.5) Stage 3 52 14.7 9.7 23 (44.2) 26 50 28 12.9 8.5 8(28.6) Stage 4 plus 29 18.8 9.3 18 (62.1) 51 100 29 12.9 9.4 10(34.5) Missing 2 16.5 3.5 1 (50.0) 101 200 25 11.4 .8 9(36.0) 201 400 29 13.0 .6 9(31.0) Stage of left leg 401 or more 28 14.1 .3 11(39.3) Normal 26 14.7 11.2 14 (53.9) Missing 144 13.0 8.9 55(38.2) Stage 1 71 12.3 7.6 20 (28.2) Stage 2 125 11.1 8.7 38 (30.4) Education (yrs) Stage 3 58 13.7 10.3 20 (17.0) 2 or less 157 13.0 8.9 58(36.9) Stage 4 plus 34 21.1 9.7 26 (76.5) 3 4 33 11.9 .7 11(33.3) Missing 2 11.5 3.5 0 (0.0) 5 6 40 14.1 9.2 15(37.5) 7 10 40 12.6 .1 17(42.5) 11 or more 37 13.6 8.3 12(32.4) Missing 9 18.3 .3 5 (55.6)

PAGE 108

97 CDC Healthy Days Table 25 and Figure 7 present the self-rate d general health status as measured by the CDC Healthy Days questionnaire. Slightly more than half (52.8 %) of the respondents felt they maintained at least a good health status, and 88 % thought their condition was fair or above. There were no significant gende r differences, but more people in La Plaine fell into excellent or fair health conditions than others ( 2=35.21, df=8, p<0.01). Table 26 shows the number of unhealthy da ys reported for each health condition. The first four health categories derived fr om the core questions which asked general unhealthy days in the past 30 days (physically unhealthy days, mentally unhealthy days, activity limitation days, overa ll unhealthy days in past 30 days ). The average physically and mentally unhealthy days recorded 5 to 6 days; in contrast, fewer activity limitation days were reported. Overall unhealthy days in past 30 days are the combination of both physically and mentally unhealthy days in th e 30-day period. The average unhealthy days was 9.9 days. There were no significant ge nder differences or regional differences; however, females tended to experience more overall unhealthy days in general but pvalue of the statistical te st slightly exceeded at = 0.05 (T=1.81, df=299, p=0.07). Likewise, residents of La Plaine were lik ely to report more mentally unhealthy days (F=2.39, df=2, 306, p=0.09). The result of mentally unhealthy days was similar to the outcome of CES-D score. Additional five health categories focus on more specific health conditions in the same manner ( pain days, depression days, anxiety days, sleepless days, vitality days ).

PAGE 109

98 None of the unhealthy days among all res pondents exceeded more than a week. There were no significant statistical differences by gender, but females were more likely to report more frequent days in pain (T= 1.95, df=310, p=0.05). Also, no significant regional differences were observed, but the differences between Cabare t and La Plaine were more likely to be obvious in depression days and anxiety day (p-values 0.11). There were no significant differences in pain day by locality. Table 27 illustrates the details of repo rted activity limitati on. About half of respondents thought their activities were limited due to their health impairments. There were no significant differences for gender or region. The rest of the questions were answered only by them. Due to an error in data collection, the major health problem had multiple answers per respondent instead of a single response, which is required by the CDCs guidelines. Therefore, the rest of que stionnaire can be regarded as a general activity limitation instead of a single health problem. Sligh tly over three-fourths (75.5 %) of respondents reported a problem with arth ritis, and over one-ha lf (52.8 %) reported having lymphedema. The other problems mainly included headache and stomachache, but the arthritis and lymphedema were more common health problems. There were significant impairment differences across gender ( 2=9.12, df=3, p=0.03) and towns ( 2=13.90, df=6, p=0.03). Particularly, males voiced more complaints about lymphedema but less about other problems such as walking or high blood pressure. Additional activity limitation indicators are shown in Table 27. The average length of activity limitation due to a major cau se of impairment or health problem was 5.8 years. The length ranged from a minimum of one week up to 48 years. About 45 % of those who had activity limitati ons had difficulty in their r outine care, but the personal

PAGE 110

care limitation was less serious. In any case, neither gender nor regional differences were found. Table 25. Gender, Town vs. Self-rated Health Status by CDC Healthy Days (n (% N), *: 1 missing, **: difference among towns (p<0.05)). Gender Town ** Total (N = 316) Male (N = 61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Excellent 9 (2.9) 1 (1.6) 8 (3.2) 0 (0) 0 (0) 9 (7.3) Very good 38 (12.1) 10 (16.4) 28 (11.0) 14 (11.8) 4 (5.6) 20 (16.3) Good 119 (37.8) 27 (44.3) 92 (36.2) 55 (46.2) 35 (48.6) 29 (23.6) Fair 111 (35.2) 16 (26.2) 95 (37.4) 37 (31.1) 21 (29.2) 53 (43.1) Poor 36 (11.4) 7 (11.5) 29 (11.4) 11 (9.2) 12 (16.7) 12 (9.8) Do not know 2 (0.6) 0 (0) 2 (0.8) 2 (1.7) 0 (0) 0 (0) Figure 7. Gender, Town vs. Self-rated Health Status by CDC Healthy Days. Self-rated health status by CDC Healthy Days0102030405060PoorFairGoodVery goodExcellent% respondents total male female Arcahaie Cabaret La Plaine 99

PAGE 111

Table 26. Gender, Town vs. Healthy and Unhealthy Days (mean SD, all ranges are 0 30). Gender Town Total Male Female Arcahaie Cabaret La Plaine Physically unhealthy days 5.5 7.1 4.7 7.1 5.7 7.1 5.4 8.0 6.5 7.3 5.0 5.9 Mentally unhealthy days 5.3 7.4 4.1 7.9 5.6 7.3 4.4 6.9 4.8 6.9 6.4 7.8 Activity limitation days 3.8 5.6 3.0 5.5 4.0 5.6 3.4 5.8 4.5 6.7 3.8 4.8 Unhealthy days in past 30 days 9.9 9.9 7.8 10.1 10.4 9.8 8.5 10.0 10.7 10.2 10.5 9.5 Pain days 5.0 6.5 3.6 6.1 5.4 6.6 4.8 6.7 5.9 7.5 4.8 5.8 Depression days 5.0 7.0 3.8 6.5 5.3 7.1 5.1 7.1 3.7 5.9 5.9 7.4 Anxiety days 4.8 6.8 3.5 5.7 5.1 7.0 4.4 6.2 3.8 5.5 5.8 8.9 Sleepless days 4.7 6.7 4.5 7.2 4.8 6.6 4.4 6.6 3.9 6.5 5.5 6.9 Vitality days 6.3 8.5 5.4 8.2 6.5 8.5 5.1 7.7 7.0 9.8 6.9 8.0 100

PAGE 112

Table 27. Gender, Town vs. Major Cause of Impairment or Health Problem (#: 1 missing, *: different between gender (p<0.05), **: different among towns (p<0.05)). Gender Town# Total (N=316) Male (N=61) Female (N=255) Arcahaie (N=120) Cabaret (N=72) La Plaine (N=123) Activity limitation (n (% N)) ** 163 (51.6) 28 (45.9) 135 (52.9) 56 (46.7) 40 (55.6) 66 (53.7) Arthritis 122 (74.8) 19 (67.9) 103 (76.3) 38 (67.9) 31 (77.5) 53 (80.3) Back or neck problem 23 (14.1) 3 (10.7) 20 (14.8) 12 (21.4) 2 (5.0) 8 (12.1) Fractures, bone/joint injury 21 (12.9) 2 (7.1) 19 (14.1) 10 (17.9) 2 (5.0) 8 (12.1) Problem walking 25 (15.3) 1 (3.6) 24 (17.8) 10 (17.9) 2 (5.0) 13 (19.7) High blood pressure 23 (14.1) 1 (3.6) 22 (16.3) 6 (10.7) 3 (7.5) 14 (21.2) Depression/anxiety/emotional problem 20 (12.3) 3 (10.7) 17 (12.6) 6 (10.7) 6 (15.0) 8 (12.1) Lymphedema 84 (51.5) 20 (71.4) 60 (44.4) 37 (66.1) 25 (62.5) 21 (31.8) Health problem (n (% per activity limitation)) ** Other problem 121 (74.2) 21 (75.0) 100 (74.1) 45 (80.4) 26 (65.0) 49 (74.2) Length of activity limitation due to health problems (mean year SD (Range: days years)) 5.8 9.0 (7 48) 7.4 11.2 (10 45) 5.5 8.5 (7 48) 5.8 8.6 (7 35) 7.5 12.9 (14 48) 4.7 5.9 (7 27) Personal care for activity limitation (n (% per activity limitation)) 39 (23.9) 9 (33.3) 30 (22.4) 13 (24.1) 12 (30.0) 14 (21.2) Routine care for activity limitation (n (% per activity limitation)) 73 (44.8) 12 (42.9) 61 (45.5) 22 (39.3) 16 (40.0) 35 (53.9) 101

PAGE 113

102 The next three tables (Table 28 Table 30) describe the association of the CDC Healthy Days measures with other socio-dem ographic variables. Table 28 reports selfrated health status and core unhealthy days. The self-rated health status indicates the number of people who described their health condition as good or above. The self-rated health condition was strongly associated with age ( 2=31.88, df=6, p<0.01), educational level ( 2=22.87, df=4, p<0.01), and income level ( 2=11.85, df=5, p=0.04). Peoples health condition was more likely to be poor as they got older. Also, those completing higher levels of education seemed to be healthier, as did peopl e with higher income levels. Stage of leg was also associated w ith general health status (p-values < 0.01). Though no systematic pattern was observed by age of onset, those with higher stages of illness tended to report lower general heal th status. Age of onset, occupation, and knowledge of the illness were not significantly associated with self-rated health. The details of the self-rated health status ar e summarized graphically in Figure 8. The relationships between unhealthy days and the variables of interest were dependent on the variables. Physically unhealthy days were significantly associated with age of onset solely (F=3.13, df=5, 263, p=0.01). Activity limitation days were associated with education (F=2.56, df=4, 298, p=0.04). On th e other hand, mentally unhealthy days varied by age group (F=2.61, df=6, 300, p=0.02) and stage of left leg (F=2.41, df=4, 303, p=0.05), and overall unhealthy days were significantly di fferent within age groups (F=2.99, df=6, 292, p=0.01), age of onset (F=2.52, df=5, 259, p=0.03), and educational level only (F=2.72, df=4, 287, p=0.03). Table 29 reports the results from five specific statements ( pain, depression, anxiety, sleeplessness, vitality ) and demographic variables. Compared with four

PAGE 114

103 unhealthy day indicators, the socio-demogra phic variables seem to be less important. Significant differences were f ound only in anxiety days for th e stage of left leg (F=4.43, df=4, 154, p<0.01). Table 30 corresponds to activity limitati on indicators in Tabl e 24. Because of the small number of responses for health probl ems in Table 24, only arthritis, lymphedema, and other problem were left in the categor y. The rest of the health problems were summarized as other2. Generally, the hea lth problems were less likely to be an important influence on activity limitation. Only the number of respondents who indicated any activity limitation was significantly differe nt for age group, age of onset, educational level, occupation, and stage of right le g (all p-values < 0.05). Length of activity limitation, personal care, and routine care we re not significantly different among the variables, except for personal care vs. educati on level and stage of left leg (p-values < 0.05). Similarly, the type of health problem was not significantly different within the socio-demographic variables. In general, pe rhaps due to the purpose of this research, arthritis and lymphedema were amongst the major problems reported among patients.

PAGE 115

104 Table 28. Other Socio-demographic Variables vs. Self-rated Health Condition (n (% N)) and Unhealthy Days (mean SD) (*1: different among agegroup, *2: different among age of onset, *3: different among daily income, *4: different among educational level, *5: different among right leg, *6: different among left leg (all p<0.05)). Unhealthy days N Health condition *1 *2 *3 *4 Physical *2 Mental *1 Activities *4 Overall *1 *2 *4 Age group (yrs) 19 or less 32 28 (87.5) 3.5 6.3 1.9 5.1 1.3 3.2 4.8 8.3 20 29 42 27 (64.3) 6.0 8.2 4.5 7.3 4.0 6.0 8.6 9.8 30 39 49 27 (55.1) 5.9 6.7 5.7 7.3 4.0 5.1 10.7 9.9 40 49 64 28 (43.8) 5.7 7.4 5.8 7.0 4.3 6.3 10.7 10.0 50 59 52 27 (54.0) 4.1 4.7 5.1 7.0 3.3 5.4 9.3 9.4 60 69 39 19 (48.7) 4.9 6.3 5.0 7.2 3.9 4.8 9.4 9.6 70 or more 34 8 (23.5) 8.6 9.1 9.0 10.1 5.7 7.0 14.7 11.0 Missing/refused 3 1 (33.3) 5.3 3.9 4.3 4.0 2.7 2.5 11.8 1.8 Age of onset (yrs) 9 or less 15 11 (73.3) 1.8 2.7 3.3 8.1 1.1 2.1 3.1 5.3 10 19 91 53 (58.2) 4.9 6.6 4.4 6.8 3.3 5.4 8.4 9.7 20 29 58 30 (51.7) 6.5 8.0 5.6 7.8 4.6 5.9 10.6 10.0 30 39 51 30 (61.2) 3.8 4.7 6.1 8.9 3.0 4.8 9.3 9.6 40 49 34 18 (52.9) 5.6 7.3 4.6 4.6 5.0 7.1 9.7 9.8 50 or more 29 14 (48.3) 9.0 9.3 6.9 8.0 4.4 4.6 13.7 10.7 Missing/refused 37 10 (27.0) 6.2 6.9 6.6 7.7 4.5 6.5 12.7 10.2 Daily income (gde) 25 or less 33 13 (39.4) 6.6 7.4 6.7 7.5 5.4 5.9 11.3 9.8 26 50 28 8 (28.6) 4.2 4.7 8.1 11.3 2.9 4.2 11.2 11.9 51 100 29 15 (55.6) 4.4 6.4 3.8 3.7 3.5 5.9 8.4 9.3 101 200 25 15 (60.0) 5.8 5.3 5.0 6.8 3.1 3.9 10.5 9.8 201 400 29 18 (62.1) 6.4 8.5 5.8 8.3 5.1 6.4 10.2 10.5 401 or more 28 18 (64.3) 6.5 8.5 5.9 8.2 4.8 6.8 11.1 10.6 Missing/refused 143 79 (55.3) 5.3 7.2 4.6 6.7 3.4 5.5 9.1 9.5 Education (yrs) 2 or less 157 63 (40.7) 6.4 7.5 6.2 7.9 4.6 6.3 11.5 10.1 3 4 33 19 (57.6) 5.5 6.6 5.8 6.6 3.8 4.1 10.7 10.0 5 6 40 25 (62.5) 4.1 6.2 4.7 7.9 2.7 4.7 7.2 9.4 7 10 40 28 (70.0) 4.1 5.7 3.6 5.7 2.2 4.2 7.1 9.3 11 or more 36 27 (75.0) 5.7 7.9 4.0 6.4 2.6 3.6 8.5 9.4 Missing/refused 9 4 (44.4) 2.3 2.8 5.2 9.8 7.2 9.8 6.7 10.1 (Continued on the next page)

PAGE 116

105 Table 28 (Continued). Unhealthy days N Health condition *5 *6 Physical Mental *5 *6 Activities Overall *6 Occupation Farmer 43 22 (51.2) 7.1 8.6 5.8 8.9 4.3 6.1 11.3 11.4 Seller at home 65 29 (44.6) 5.7 6.4 6.3 7.9 3.9 5.1 11.2 10.1 Seller at market 52 28 (53.9) 4.9 5.6 4.3 5.1 4.8 6.3 9.0 8.7 tailor/seamstress 17 9 (52.9) 3.8 4.8 5.9 8.1 2.8 4.2 9.1 9.8 Unemployed 90 43 (47.8) 6.4 8.0 5.8 7.4 4.2 6.3 11.1 10.1 Other 63 43 (68.3) 4.1 7.0 3.9 6.9 2.2 3.9 6.8 9.1 Illness knowledge Insect bite 31 21 (67.7) 8.6 8.3 9.6 10.8 4.5 6.3 15.4 12.4 Magic 31 16 (51.6) 9.2 5.8 6.4 8.4 6.8 6.8 14.5 10.0 Sprain 29 15 (51.7) 2.8 3.2 4.7 8.1 1.8 2.3 7.2 8.8 Chill 43 21 (48.9) 8.4 7.3 7.3 8.8 4.7 3.5 14.7 9.2 Other 41 25 (61.0) 9.8 8.7 7.4 7.8 7.5 6.8 15.7 11.2 Dont know 177 90 (50.9) 7.7 8.5 7.1 8.1 5.9 7.2 13.3 10.4 Stage of right leg Normal 43 28 (65.1) 3.5 3.3 4.6 7.4 1.8 2.3 7.8 8.5 Stage 1 106 61 (57.5) 6.1 8.5 5.1 7.4 4.2 6.1 9.9 10.5 Stage 2 83 53 (63.9) 5.2 6.5 4.1 5.2 4.2 6.2 9.2 9.5 Stage 3 52 16 (30.8) 7.0 7.4 7.7 9.0 3.2 4.5 13.0 10.6 Stage 4 or more 29 8 (27.6) 4.4 6.1 5.6 8.0 5.5 6.7 8.0 8.1 Missing/NA 2 0 (0) 4.0 5.7 19.0 .6 1.0 1.4 23.0 9.9 Stage of left leg Normal 26 15 (57.7) 9.0 .4 8.0 8.0 5.9 6.1 14.5 11.4 Stage 1 71 40 (56.4) 5.4 6.8 5.0 7.0 3.8 5.2 9.7 9.9 Stage 2 124 77 (62.1) 5.7 7.7 4.3 6.9 4.0 6.6 9.0 10.2 Stage 3 58 25 (43.1) 4.4 4.5 5.1 7.2 2.9 3.7 9.0 8.3 Stage 4 or more 34 9 (27.0) 4.6 5.0 7.7 8.6 3.4 4.4 9.0 8.3 Missing/NA 2 0 (0) 4.0 5.7 17.0 .4 1.5 2.1 21.0 12.7

PAGE 117

Figure 8. Other Socio-demographic Variables vs. Self-rated Health Status by CDC Healthy Days (good or above). Self-rated health condition (age group)87.564.355.143.854.048.723.502040608010019 orless20 2930 3940 -4950 -5960 6970 ormoreage group% Self-rated health condition (age of onset) 73.380 58.251.761.252.948.302040601009 or less10 1920 2930 3940 4950 ormoreage% Self-rated health condition (daily income)39.428.655.660.062.164.302040608010025 orless26 5051 100101 200201 400401 ormoregde% (Continued on the next page) 106

PAGE 118

Figure 8 (Continued). Self-rated health condition (education)40.757.662.570.075.00204060801002 or less3 45 67 1011 ormoreyear% Self-rated health condition (occupation)51.244.653.952.947.868.3020406080100FarmerSeller at homeSeller at markettailor/seamstressUnemployedOther% Self-rated health condition (knowledge) 67.780 51.651.748.961.050.90204060100InsectbiteMagicSprainChillOtherDontknow% (Continued on the next page) 107

PAGE 119

Figure 8 (Continued). Self-rated health condition (stage of illness)57.756.462.143.127.065.157.563.930.827.6020406080100NormalStage 1Stage 2Stage 3Stage 4or morestage% left right 108

PAGE 120

109 Table 29. Other Socio-demographic Variables vs. Unhealthy/healthy Days (mean SD) (*: different among left leg (p<0.05). Pain days Depressio n days Anxiety days Sleepless days Vitality days Age group (yrs) 19 or less 2.4 4.9 2.3 5.9 1.8 3.3 2.6 5.0 3.1 6.7 20 29 5.7 7.1 5.9 8.1 5.7 8.1 5.3 7.3 7.1 9.7 30 39 5.7 6.7 5.7 8.5 5.2 7.7 5.2 7.4 6.2 7.9 40 49 5.5 7.1 5.0 6.5 4.9 7.0 4.2 6.6 6.7 8.6 50 59 4.5 5.8 4.7 5.6 5.1 6.4 4.7 6.8 4.4 6.2 60 69 4.2 5.0 3.8 4.8 4.0 4.9 5.6 6.8 7.9 8.9 70 or more 6.9 7.8 8.0 8.5 6.3 7.6 5.8 6.4 9.7 10.3 Missing/refused 3.2 3.0 0.8 1.4 5.0 8.7 3.7 3.5 0.3 0.6 Age of onset (yrs) 9 or less 2.4 3.4 5.1 8.6 5.6 8.7 5.9 8.6 4.2 6.0 10 19 4.5 6.1 4.2 6.4 4.0 6.5 4.5 6.6 6.4 9.1 20 29 4.6 6.4 4.8 7.6 5.1 6.9 4.9 7.5 6.5 8.9 30 39 4.6 6.0 4.8 6.7 5.1 7.7 4.0 5.6 4.3 5.8 40 49 6.4 7.9 5.5 8.0 4.6 6.4 4.9 7.9 6.6 9.2 50 or more 5.7 6.5 6.9 6.9 5.9 6.3 6.6 7.1 9.0 9.3 Missing/refused 7.0 7.8 5.9 6.6 5.0 6.2 4.0 4.6 7.2 8.4 Daily income (gde) 25 or less 5.2 5.4 5.8 8.0 5.3 6.7 4.0 6.3 9.1 11.1 26 50 5.9 7.2 3.8 4.7 5.4 8.1 4.8 6.4 5.6 8.1 51 100 4.3 6.3 3.9 4.7 3.8 4.7 4.9 7.0 5.6 7.3 101 200 5.9 5.4 5.1 5.1 4.4 4.7 6.3 6.2 5.9 5.2 201 400 6.7 8.3 5.5 8.1 6.3 8.2 7.9 10.1 6.0 7.6 401 or more 7.1 8.6 4.7 9.6 6.2 8.5 5.4 7.1 6.6 8.5 Missing/refused 4.1 5.9 4.9 7.1 4.3 6.6 3.9 5.8 6.0 8.7 Education (yrs) 2 or less 5.6 6.6 5.8 7.1 5.0 6.2 4.7 6.6 7.5 9.1 3 4 4.7 6.4 4.9 7.5 5.7 7.2 6.9 9.0 6.3 8.9 5 6 4.6 4.7 4.5 6.3 4.7 7.2 4.6 6.6 4.7 5.7 7 10 3.4 5.3 3.5 6.1 4.7 8.2 4.3 7.2 4.8 6.9 11 or more 4.3 6.2 4.6 8.1 3.3 7.1 3.2 4.8 4.9 9.5 Missing/refused 8.2 10.1 4.0 5.8 4.2 5.2 5.7 3.2 5.8 6.1 Occupation Farmer 4.9 7.0 4.6 8.0 3.6 5.8 4.1 7.1 4.8 8.1 Seller at home 5.9 5.9 5.4 6.4 5.4 6.7 5.3 7.0 7.1 7.8 Seller at market 6.2 6.9 4.9 5.2 5.1 5.4 5.5 7.5 5.7 7.5 Tailor/seamstress 2.9 3.5 3.8 6.1 5.5 8.3 5.4 7.6 6.0 7.5 Unemployed 4.9 6.5 6.4 8.0 5.4 7.6 4.7 6.5 8.3 10.1 Other 3.7 6.7 3.9 6.9 3.6 7.0 4.1 5.9 4.5 7.7 (Continued on the next page)

PAGE 121

110 Table 29 (Continued). Pain days Depression days Anxiety days Sleepless days Vitality days Illness knowledge Insect bite 6.8 7.3 9.1 9.9 9.4 9.8 7.4 9.3 7.4 8.5 Magic 6.7 6.7 8.8 8.8 5.5 6.1 6.8 9.3 12.7 11.5 Sprain 3.4 3.1 2.4 2.6 4.6 7.9 1.8 2.7 4.1 4.4 Chill 6.9 6.5 8.2 9.1 5.3 6.7 5.9 6.7 10.9 9.5 Other 9.7 8.9 7.2 8.3 7.1 8.3 7.7 8.8 10.2 10.1 Dont know 6.7 7.1 6.6 8.1 5.9 7.7 5.9 7.5 9.0 10.0 Stage of right leg Normal 3.5 4.1 4.3 6.3 3.4 4.0 4.0 4.5 5.2 7.4 Stage 1 5.0 6.6 4.4 7.1 3.8 6.2 4.1 7.2 6.0 8.9 Stage 2 5.4 7.0 5.1 6.2 4.5 5.9 5.0 6.2 5.4 7.0 Stage 3 6.1 7.9 5.8 7.6 7.1 9.4 6.3 7.9 8.7 9.9 Stage 4 or more 4.6 4.9 6.2 7.8 6.8 9.4 5.2 6.9 6.3 8.1 Missing/NA 4.5 4.9 16.0 19.8 12.5 13.4 1.0 1.4 22.5 10.6 Stage of left leg Normal 6.1 8.3 7.5 8.0 8.8 9.1 7.0 7.9 8.5 9.8 Stage 1 4.6 5.0 5.0 7.5 3.1 3.7 2.9 5.5 6.8 10.3 Stage 2 5.2 7.2 4.3 6.4 4.3 6.7 4.8 6.6 5.6 7.5 Stage 3 4.3 4.4 4.6 5.6 4.6 6.2 5.9 7.3 5.6 7.4 Stage 4 or more 6.0 4.4 6.2 8.3 6.7 9.1 5.0 7.0 6.5 7.4 Missing/NA 3.0 2.8 16.0 19.8 12.0 14.1 1.0 1.4 19.0 15.6

PAGE 122

Table 30. Other Socio-demographic Variables vs. Major Cause of Impairment or Health Problem (Other problem and other2 indicate the sum of the health problems so that the proportion of AL may exceed 1, *: different in activity limitation (p<0.05), **: different in personal care (p<0.05)). Health problem (n (% per Activity limitation (AL))) Activity limitation (n (%total)) Arthritis Lymphedema Other problem Other2 Year of activity limitation (mean SD) Personal care (n (% AL)) Routine care (n (% AL)) Age group (yrs) 19 or less 8 (24.2) 6 (75.0) 2 (25.0) 6 (75.0) 7 (87.5) 1.5 1.0 1 (12.5)3 (37.5) 20 29 20 (47.6) 14 (70.0) 9 (45.0) 13 (65.0) 9 (45.0) 2.9 3.9 2 (10.0)5 (25.0) 30 39 22 (44.9) 19 (86.4) 6 (27.3) 11 (50.0) 8 (36.4) 3.7 5.1 4 (13.6)8 (36.4) 40 49 32 (50.0) 18 (56.3) 17 (53.1) 30 (93.8) 17 (53.1) 8.2 11.0 10 (31.3)13 (40.6) 50 59 28 (53.8) 22 (78.6) 19 (67.9) 16 (57.1) 27 (96.4) 7.8 9.5 8 (28.6)12 (42.9) 60 69 24 (61.5) 20 (83.3) 14 (58.3) 14 (58.3) 17 (70.8) 8.3 13.4 7 (29.2)12 (50.0) 70 or more 28 (82.4) 23 (82.1) 16 (57.1) 31 (110.7) 27 (96.4) 4.2 5.9 8 (28.6)20 (71.4) Missing/refused 1 (33.3) 0 (0) 1 (100.0) 0 (0) 0 (0) 1.0 0.0 0 (0) 0 (0) Age of onset (yrs) 9 or less 3 (20.0) 3 (100.0) 1 (33.3) 2 (66.7) 1 (33.3) 6.7 7.0 0 (0) 0 (0) 10 19 39 (42.4) 25 (64.1) 23 (59.0) 29 (74.4) 23 (59.0) 8.6 12.9 9 (23.1)16 (41.0) 20 29 32 (55.2) 25 (78.1) 18 (56.3) 23 (71.9) 28 (87.5) 7.0 10.1 8 (25.0)16 (50.0) 30 39 20 (39.2) 14 (70.0) 10 (50.0) 15 (75.0) 9 (45.0) 6.6 7.6 5 (25.0)8 (40.0) 40 49 22 (64.7) 16 (72.7) 6 (27.3) 16 (72.7) 18 (81.8) 2.6 3.1 5 (22.7)10 (45.5) 50 or more 21 (72.4) 19 (90.5) 10 (47.6) 18 (85.7) 15 (71.4) 4.1 5.4 3 (14.3)9 (42.9) Missing/refused 26 (70.3) 20 (76.9) 16 (61.5) 18 (69.2) 18 (69.2) 3.4 5.7 9 (34.6)14 (53.9) (Continued on the next page) 111

PAGE 123

112 Table 30 (Continued). Health problem (n (% per Activity limitation (AL))) Activity limitation (n (% total)) Arthritis Lymphedema Other problem Other2 Year of activity limitation (mean SD) Personal care (n (% AL)) Routine care (n (% AL)) Education (yrs) ** 2 or less 99 (63.1) 72 (72.7) 57 (57.6) 79 (79.8) 73 (73.7) 6.5 10.1 33 (33.3)50 (50.5) 3 4 17 (51.5) 14 (82.4) 10 (58.9) 8 (47.1) 14 (82.4) 7.1 10.2 3 (17.7)9 (52.9) 5 6 18 (45.0) 11 (61.1) 8 (44.4) 18 (100.0) 10 (55.6) 5.2 7.6 1 (5.6) 3 (16.7) 7 10 12 (30.0) 11 (91.7) 3 (25.0) 6 (50.0) 6 (50.0) 1.8 2.0 2 (16.7)5 (45.5) 11 or more 13 (35.1) 11 (84.6) 6 (46.2) 10 (76.9) 10 (76.9) 3.9 3.0 0 (0) 4 (30.8) Missing/refused 4 (44.4) 4 (100.0) 0 (0) 0 (0) 1 (25.0) 6.5 5.9 0 (0) 2 (50.0) Occupation Farmer 20 (45.5) 13 (65.0) 12 (60.0) 15 (75.0) 11 (55.0) 7.0 11.3 6 (30.0)7 (35.0) Seller at home 42 (63.6) 34 (81.0) 23 (54.8) 30 (71.4) 32 (76.2) 6.1 8.6 10 (24.4)18 (42.9) Seller at market 26 (50.0) 19 (73.1) 15 (57.7) 12 (46.2) 16 (61.5) 6.2 8.0 4 (15.4)12 (46.2) Tailor/seamstress 6 (35.3) 6 (100.0) 3 (50.0) 5 (83.3) 4 (66.7) 5.0 5.3 1 (16.7)1 (16.7) Unemployed 62 (68.9) 43 (69.4) 30 (48.4) 59 (95.2) 55 (88.7) 5.6 9.0 14 (22.6)32 (51.6) Other 18 (28.1) 16 (88.9) 7 (38.9) 8 (44.4) 5 (27.8) 4.1 8.3 5 (27.8)8 (44.4) Illness knowledge Insect bite 17 (54.8) 12 (70.6) 8 (47.1) 15 (88.2) 15 (88.2) 5.6 6.9 5 (29.4)10 (58.8) Magic 13 (39.4) 9 (69.2) 9 (69.2) 12 (92.3) 15 (115.4) 4.5 5.6 3 (23.1)6 (46.2) Sprain 13 (48.2) 9 (69.2) 8 (61.5) 18 (138.5) 11 (84.6) 10.1 9.7 3 (23.1)3 (23.1) Chill 27 (62.8) 21 (77.8) 19 (70.4) 17 (63.0) 11 (40.7) 4.9 10.9 7 (25.9)14 (51.9) Other 21 (51.2) 19 (90.5) 14 (66.7) 16 (76.2) 17 (81.0) 4.3 9.9 8 (38.1)10 (47.6) Dont know 94 (53.1) 69 (73.4) 40 (42.6) 69 (73.4) 68 (72.3) 5.5 8.4 18 (19.1)39 (41.5) (Continued on the next page)

PAGE 124

Table 30 (Continued). Health problem (n (% per Activity limitation (AL))) Activity limitation (n (% total)) Arthritis Lymphedema Other problem Other2 Year of activity limitation (mean SD) Personal care (n (% AL)) Routine care (n (% AL)) Daily income (gde) 25 or less 17 (51.5) 9 (52.9) 8 (47.1) 17 (100.0) 10 (58.9) 5.2 7.9 1 (5.9) 4 (23.5) 26 50 15 (53.6) 12 (80.0) 12 (80.0) 9 (60.0) 11 (73.3) 10.8 13.0 5 (33.3) 5 (33.3) 51 100 11 (37.9) 11 (100.0) 8 (72.7) 11 (100.0) 16 (145.5)4.4 4.9 2 (18.2) 5 (45.5) 101 200 15 (60.0) 13 (86.7) 6 (40.0) 1 (6.7) 4 (26.7) 5.5 11.4 5 (33.3) 8 (53.3) 201 400 15 (51.7) 12 (80.0) 6 (40.0) 12 (80.0) 2 (13.3) 2.4 5.4 3 (20.0) 7 (46.7) 401 or more 11 (39.3) 10 (90.1) 6 (54.5) 9 (81.8) 6 (54.5) 2.8 5.4 3 (27.3) 6 (54.5) Missing/refused 79 (54.9) 55 (69.6) 38 (48.1) 62 (78.5) 63 (79.7) 6.3 8.9 20 (25.3) 38 (48.1) Stage of right leg Normal 12 (27.9) 9 (75.0) 4 (33.3) 6 (50.0) 5 (41.7) 2.9 4.2 2 (16.7) 6 (50.0) Stage 1 56 (52.8) 39 (69.6) 27 (48.2) 46 (82.1) 43 (76.8) 4.5 7.8 11 (20.4) 17 (30.4) Stage 2 46 (54.8) 34 (73.9) 26 (56.5) 28 (60.9) 24 (52.2) 6.4 9.9 14 (30.4) 24 (52.2) Stage 3 30 (57.7) 23 (76.7) 18 (60.0) 27 (90.0) 23 (76.7) 5.9 7.2 5 (16.7) 16 (53.3) Stage 4 or more 17 (58.6) 16 (94.1) 9 (52.9) 8 (47.1) 12 (70.6) 10.6 14.2 6 (37.5) 9 (56.3) Missing/NA 2 (100.0) 1 (50.0) 0 (0) 6 (300.0) 5 (250.0) 6.5 4.9 1 (50.0) 1 (50.0) Stage of left leg ** Normal 12 (46.2) 8 (66.7) 3 (25.0) 13 (108.3) 11 (91.7) 4.5 5.8 0 (0) 5 (41.7) Stage 1 35 (49.3) 26 (74.3) 13 (37.1) 31 (88.6) 17 (48.6) 5.8 9.9 4 (11.8) 12 (34.3) Stage 2 63 (50.4) 48 (76.2) 38 (60.3) 46 (73.0) 45 (71.4) 5.7 9.5 21 (34.4) 27 (42.9) Stage 3 32 (55.2) 25 (78.1) 21 (65.6) 16 (50.0) 21 (65.6) 6.8 9.7 6 (18.8) 19 (59.4) Stage 4 or more 19 (55.9) 15 (78.9) 8 (42.1) 11 (57.9) 17 (89.5) 5.7 6.7 6 (31.6) 8 (44.5) Missing/NA 2 (100.0) 0 (0) 1 (50.0) 4 (200.0) 1 (50.0) 3.5 0.7 2 (100.0) 2 (100.0) 113

PAGE 125

114 Reliability and Validity The last research question, reliability and validity of the standardized instruments, is an important issue for lymphedema mana gement in the Haitian context. Table 31 shows the results of internal consistency re liability for the QOL instruments (EuroQol, CDC Healthy Days) and a subjective well-being assessment tool (CES-D). Prior to the analysis, the scores of EQ-VAS were revers ed because EQ-VAS had a negative direction in relation to the items in EQ-5D. Some que stions in CDC Healthy Days were excluded due to the non-directional measurements (e .g., major cause of impairment or health problem). Also, all missing values and am biguous responses were ignored prior to calculation. Overall, all measurements have at least acceptable inter-rater reliability (alpha > 0.70). Table 31. Internal Consistency Reliab ility in QOL Scales and CES-D. N Cronbach coefficient alpha EuroQol 306 0.72 CES-D 280 0.85 CDC Healthy Days 281 0.87 Core questions only298 0.71 In order to examine the criterion-related validity, the results of three scales were compared by correlational analysis. Correlationa l analysis was also used for the analysis of convergent and discriminant validity. Table 32 shows the results of validity analyses. The rows consist of the appropriate CDC Hea lthy Days health topics, while the columns correspond to the EuroQol five-dimensioned health categories. The CES-D score was added in both rows and columns for convenien ce. For criterion-rela ted validity, EuroQol

PAGE 126

115 questionnaires were fairly associated with CDC Healthy Days. Particularly, five health dimensions ( mobility, self-care, usua l activities, pain/discomfort, anxiety/depression ) in EuroQol were strongly associated with CDC Healthy Days self-rate d health condition (all p<0.01), and three core h ealth questionnaires ( physically unhealthy days, mentally unhealthy days, activity limitation days ) in CDC Healthy Days were also significantly associated with EuroQol overall healthy st atus (all p<0.01). Likewise, EuroQol agreed with CDC Healthy Days in terms of physical health, activity limitation, and overall/general health status (all p<0.05), and disagreed with each other in the other discordant domains. These outcomes indicate satisfactory converge nt and discriminant validity. However, the mental health indica tors in EuroQol had non-significant, weak relationships with those of both CES-D a nd CDC Healthy Days, even though the other two forms had slightly less strong but statistically signifi cant relationships.

PAGE 127

116 Table 32. Correlations between Representativ e Questions in QOL Instruments and CESD (*: p<0.05, **: p<0.01). Mobility Selfcare Usual activitie s Pain / discomf ort Anxiety / depressi on Overall health CES-D CES_D 0.12 0.12 0.15 ** -0.08 0.01 0.15 ** Self-rated health 0.26 ** 0.25 ** 0.32 ** 0.28 ** 0.19 ** 0.57 ** 0.20 ** Physically unhealthy days 0.10 -0.00 0.11 0.14 0.13 0.28 ** 0.08 Mentally unhealthy days 0.05 -0.01 0.13 0.05 0.07 0.21 ** 0.24 ** Activity limitation (AL) days 0.10 0.07 0.14 0.13 0.18 ** 0.21 ** 0.10 Any AL -0.30 ** -0.04 -0.35 ** -0.27 ** -0.16 ** 0.41 ** -0.12 Length of AL 0.02 0.24 ** 0.15 -0.10 0.07 0.04 0.45 ** Personal care for AL -0.26 ** -0.31 ** -0.36 ** -0.26 ** -0.17 0.02 0.09 Routine care for AL -0.31 ** -0.16 -0.29 ** -0.25 ** -0.10 0.05 0.01 Pain days 0.06 0.03 0.03 0.12 0.14 0.26 ** 0.01 Depression days 0.15 ** 0.00 0.12 0.07 0.07 0.18 ** 0.21 ** Anxiety days 0.17 ** 0.03 0.09 0.11 0.06 0.10 0.19 **

PAGE 128

117 Chapter Five Discussion and Conclusion Morbidity control of lymphatic filariasis is one of the most important public health issues in Haiti. In order to increase QOL among Haitian lymphedema patients due to LF as well as contribute to eradicate LF worldwide, it is critical to implement appropriate control strategies in endemic areas. This thesis aimed to observe the association among filariasis-related vari ables among lymphedema patients in three rural Haitian towns and assess QOL among affected persons by using es tablished QOL instruments. Particularly, attention was given to gender differences in the impact of the disease on peoples daily lives. There were numerous significant findi ngs in the results of data analysis. Regional Differences In general characteristics of lymphedema patients, there are si gnificant differences among the three communities, Arcahaie, Cabaret, and La Plaine. Though lymphedema conditions in Arcahaie are very similar to those in Cabaret, people in La Plaine experienced more severe symptoms than in other areas. Although sample characteristics are slightly different among regions, the results confirm th e finding by Dreyer et al. (1998) that the characteristics of filariasis vary by geographic region. Regional differences are more obvious for illness history. No consistent

PAGE 129

118 symptomatic conditions were observed among towns. Although the interview relied on patients recall, there were significant differ ences among the three commu nities in the first impression of the illness and the first symptom noticed. This different illness history might influence perception and knowledge of the di sease and future prevention, treatment, and control. In particular, since lymphedema treatme nt requires sustained intensive efforts, it is critical to understand the background and percep tion of the disease prior to introducing the prevention and treatment regimen at the local le vel. For example, people in La Plaine were more likely to utilize health services and r outine health care practices. This is likely explained by the greater accessibility to the cap ital. Since there are many medical services available in Port-au-Prince and its suburbs in comparison with other cities, residents in La Plaine tend to talk to health professionals more often. Overa ll, they were more concerned about leg treatment and took care of their legs more frequently with confidence. Particularly, they reported washing legs, use of massage, and avoiding walking on bare feet more often than people in the other zones. Main taining hygiene is one of the most critical practices for LF morbidity control. Therefore, considering these fact s, the introduction of a morbidity control program in La Plaine woul d likely be received well. On the other hand, people in Arcahaie seem more conservative re garding leg treatments. More people prefer to visit a traditional healer and use herbal re medies. Their knowledge of the illness is more related to traditional cultural and spiritual dimensions. However, though knowledge of the illness is a fundamental key to the future prevention of the disease, the results reflect poor understanding of the disease nationwide (Coreil et al., 1998; Eberhard et al., 1996). Therefore, it would be effective to plan and implement morbidity control regimens on a regional basis.

PAGE 130

119 Gender Perspective The sample design was based on non-pr obability sampling, but the gender proportion was consistent with previous findings. In Haiti, women suffer from lymphedema 5 10 times more often than men (Lammie et al., 1993), and there were 4.2 times more females than males in the sample so that the gender comparison in this thesis can be representative of the general lymphedema conditions in Haiti. However, overall gender differences were less pronounced than regional variation. Significant gender differences were found in aspects of acute atta cks, but there was no statistically significant discordance in illness histor y or knowledge of the illne ss. Therefore, in terms of lymphedema, gender differences can be accorded comparatively low priority in these Haitian towns. Lymphedema Condition and Its Related Variables The lymphedema conditions among people in three Haitian communities are consistent with the previous findings in other countries. Dreyer et al. (2002) mention that many lymphedema cases fall within stage 3 or less, which indicates shallow skin folds or swollen legs. In three towns, nearly two-thirds of people were diagnosed as falling within the first two stages, and about 90 % were in th e first three stages. More severe conditions were found among people in La Plaine. Because th e area of La Plaine experiences frequent flooding, residents are more like ly to be exposes to mosquito-favorable environments.

PAGE 131

120 However, the differences across towns were not significant. These results are also similar to findings in another Haitian town, Leogane (Dahl, 2001). Therefore, the conditions of lymphedema in these towns appear to reflect a general lymphedema profile. Socio-demographic variables had a pa rtial association with lymphedema conditions. In this analysis, age group wa s strongly associated with lymphedema characteristics. In both foot size and stage of the illness, lymphedema conditions worsened as people became older, with a peak around ag e 50. This supports th e relationship between age and lymphedema conditions in numerous studies from other parts of the world (Gasarasi et al., 2000; Gyapong et al., 1996b; Hyma et al., 1989; King & Freedman, 2001; Shriram et al., 2002; Weerasooriya et al., 2001). In contrast, other variables of interest such as gender, SES (occupation, income, and education), knowledge of the illness, and history of acute attacks were poorly associated with lymphedema conditions. This is probably due to the type of study design, a cross-sectional study, which make s it difficult to establish a causal relationship due to the collection of data at a single point in time. In the LF literature, these variables may be the potential indicators of the lymphedema conditions more or less. Therefore, in order to find out the clear rela tionship, it would be necessary to conduct more thorough and prospective re search in the future. Health-related Behavior Health-related behaviors are critical in establishing an effective preventive and control strategy for LF. In this thesis, health-related behaviors were assessed as seeking care for leg treatment, taking precautions with on es legs, self-care practices for legs which

PAGE 132

121 are currently done or can be done in the future, and confid ence in leg care (self-efficacy). The proportion of people who indicated choos ing herbal remedies for treatment and engaging in self-care practices is substant ially high. Although no sc ientific benefit of herbal remedies is known for lymphedema, more than 40 % of people usually use them for their treatment and self-c are practices. This suggests that tr aditional therapy for LF is still common and widespread in the study setting. Ho wever, Western medicine is reported as an alternative choice. Seeing health professi onals and purchasing pharmaceutical medicine are practiced in the communities, although availability is still limited due to the small rural community far from the capital and the prohib itive cost of medicine. In addition, regional differences are evident. Traditional medicine is more widely practiced in Arcahaie, but Western medicine is more common in La Plaine probably due to the hi gher accessibility to the metropolitan area. Also, the confidence of se lf-care practice for le gs is significantly different among towns. More people in La Plaine are very confiden t in their leg care. Therefore, since the treatment and prevention strategy of lymphedema is generally based on the Western regimen, the regional characteristics of health-related behaviors would influence acceptability of future interventions. Socio-demographic variables rather than th e regional perspective appear to be less important determinants of health-related behaviors. Gender, age group, age of onset, income, and occupation were not significantly as sociated with current or future self-care for legs or self-efficacy. Only educational le vel and knowledge of the illness were related to future possible self-care practice. Howeve r, knowledge of the illn ess was quite low so that it would be possible to promote self -care practices by impr oving illness knowledge. Reports from other countries indicate that poor LF knowledge significantly contributes to

PAGE 133

122 high risk behaviors as well as exacerbates the di sease (Ahorlu et al., 199 9; Eberhard et al., 1996; Gyapong et al., 1996a; Rauyajin et al., 1995). Therefore, as sessment of the effect of health education about LF would be benefici al for future morbidity control strategies. Quality of Life and Subjective Well-being Scales Reliability and Validity The other noteworthy finding of the resear ch is the successful introduction of the QOL instruments for evaluation of the impact of lymphedema on peoples lives. Since there has been little focused research to assess QOL in LF populati ons, two different QOL measurements (EuroQol and CDC Healthy Days) and the CES-D as a mental health indicator were used in Haitian lymphedema cases. Reliability of the instruments was established. Since the instruments are considered reliable when the internal consistency coefficient or Cronbach alpha falls at 0.70 or a bove, and all the instruments reached at least 0.71 or more. This indicates that the scales are ab le to measure a consistent aspect of LF in Haiti. In contrast, it is usually more difficult to demonstrate the crosscultural validity of instruments. In this thesis, criterion-related validity and construct validity were assessed. Criterion-related validity was acceptable but not really strong. The health domain in the EuroQol was correlated with similar outco mes in the CDC Healthy Days, except for mental health indicators. In particular, overa ll health status in the EuroQol was strongly associated with self-rated health status in the CDC Healthy Days scale. The other health domains also showed similar results in bot h questionnaires. However, mental health

PAGE 134

123 indicators were poorly associated with QOL measures. Particularly, the EuroQol mental health domain (anxiety/depress ion) was problematic, probably due to the simplicity of the questionnaires. On the other hand, the CDC Healthy Days instrument has a few items regarding mental health, and the CES-D is en tirely focused on mental health assessment. Scores for the two instruments were significan tly correlated. Therefore, both the EuroQol and CDC Healthy Days would be acceptable fo r evaluating lymphedema conditions, but in terms of mental health, CD C Healthy Days would provi de a more valid tool. On the other hand, construct validity esta blished more reasonable evidence for validity of the instruments. Dahl (2001) re ported that lymphedema patients in Haiti experienced limited physical activ ities due to acute attacks. Nearly 95 % of respondents had acute attacks in the previous year, a nd more than 50 % of them reported activity limitations especially due to arthritis a nd lymphedema in the CDC Healthy Days instrument. Bandyopadhyay (1996) and Coreil et al. (1998) indicate that women with abnormal physical features due to lymphedema experience mental a nd psychological stress. Although no gender differences were found in ment al health indicators in any of the QOL instruments, respondents in La Plaine, where there are 6.8 times more females than males in the sample, showed significantly worse mental health scores in two of three instruments. Thus, construct validity is somewhat demonstrated. Related to construct validity, convergent and discriminant validity were also assessed. The results showed that the multiple health indicators in different domains could operate consistently at least at acceptable levels, except for the mental health indicators mentioned in criterion-related validity. Although the correlation coefficients among convergent domains are slightly lower that expe cted, most of the matched pairs indicated a

PAGE 135

124 significant relationship between them. Likewise, most of disc riminant pairs showed no significant relationship. Therefore, the QOL inst ruments are fairly applicable in assessing lymphedema patients in Haiti, but further investigation would be beneficial. Outcome of the Scales vs. Socio-demographic Variables The results of QOL instruments s how significant relationships with socio-demographic variables. Like lymphede ma conditions and health-related behaviors, regional differences are strongl y associated with the result s of QOL instruments. Though the significant relationship between towns and EuroQol mental health indicators is questionable due to poor validity of the instrument discussed above, the relationship in CES-D and CDC Healthy Days is significant. In the CES-D, those in La Plaine experienced significantly more se vere mental health symptoms than the other areas, and a similar tendency was observed in the CDC Healthy Days scores. Particularly, the outcomes in CDC Healthy Days indicate apparent regi onal differences between La Plaine and the other towns for the other health domains. Al so, both EuroQol and CDCs general health condition indicators showed significant differences among these three towns. Other socio-demographic variables were al so associated with QOL outcomes. The most significant relationship was found with age. As in general healthy populations, physical health conditions including activity li mitations are strongly co rrelated with age. As the age of the patients increases, the scor es for physical health conditions and activity limitations in EuroQol and CDC Healthy Days showed a negative relationship. Noteworthy, the tendency was most obvious in usual activities, pain/dis comfort, and overall health

PAGE 136

125 status in EuroQol, and overall health status and activity limitation in CDC Healthy Days. Some other mental health indi cators are also associated w ith age group, but the results are inconsistent. Thus, age would be a potential indicator of health status among lymphedema patients. Additional variables such as age of ons et, educational level, occupation, and knowledge and stage of the illness were also potential indicators of he alth status, but the results were variable. Though gender, income, and knowledge of the illness seem to be less important to evaluate hea lth conditions among lymphedema patients, age of onset, educational level, and occupatio n were significantly associated with indicators of physical health and activity limitations. In age of onset, for example, those who had the first symptom in early life seem to have better health status and less activity limitation. Similarly, as people complete higher educati on, their health status and activity limitation appear to be better. Regarding occupation, une mployed people are more likely to have poor health status and more activity limitation. Thoug h it is difficult to identify which type of occupation influences health status probably due to the imbalanced sample among job categories, tailor/seamstress seems to be healthier than others. Stage of the illness is another variable a ssociated with QOL outcomes. In addition to the relationship with physical health and activity limita tion, stage of the illness seems to be a good mental health indicator among lym phedema patients. In both CES-D and CDC Healthy Days, there were significant differences in scores across the stage levels. However, there is no tendency that people at higher stag es of the illness have worse mental health conditions than those at lower. The results indicate that pe ople with normal leg conditions had very poor mental health status, similar to stage 4 or more. This might imply that their

PAGE 137

126 mental health is influenced by additional fact ors in their lives. Although the condition of lymphedema may be associated with mental health, there is poor control of possible confounders due to the characteri stics of the study. Thus, fu rther investigation would be valuable. Overall, the generic QOL instruments we re useful in evaluating the health conditions among lymphedema patients. Without the analysis of QOL, it was found that only regional differences and age groups were significant determinants. However, by analyzing the results of QOL instruments, other socio-demographic variables appeared to be potential correlates of lymphedema patients health, especially physical health and activity limitation. Though no compar ison with people who have normal health status was made, the results of the QOL instruments suggest it would be beneficial to look at such a comparison group. That additional information would help design effective prevention, treatment, and morbidity control strategies. Limitation There are several limitations on this resear ch. The first one is the sampling method. Since non-probability sampling was chosen, there are possible biases in sample characteristics. The common one is selection bias Since it was difficult to select a member of the study group probabilistically, the possibility of sample misrepresentation was unable to be excluded throughout the study. Partic ularly, in comparison with estimated total population in each community, the area of La Pl aine might be overrepresentative in the sample. However, sample characteristics in this thesis were more likely to be very similar

PAGE 138

127 to previous studies (eg. Eber hard et al., 1996; Lammie et al., 1993), even though several minor disparities such as gender ratio among three towns might have skewed the outcomes and their interpretation. Other limitati ons due to non-probability sampling are susceptibility to confounding and weak exte rnal validity. Due to the inability of randomization, it might be difficult to rule ou t potential explanations and generalize the results to and/or across communities. Theref ore, since population characteristics were unavailable in these regions, further studies will need to support the findings in this study. Also, regarding the data colle ction, interviewer bias might be a critical factor. Since the data were collected by five trained interviewers in three different locations in a short period of time, there were some discrepanc ies in the dataset and deviation from the protocol of data collection. Particularly, the faulty administration of one of CDC Healthy Days questionnaires made it difficult to follow the guidelines of the analysis and made the outcomes less useful. Therefore, careful attention to the data collecting process is essential. The second limitation stems from the purpose of the dataset. The data came from an ongoing project called Evaluation of Support Groups in the Management of Lymphedema Caused by Lymphatic Filariasis. Although this thesis is aimed at the description of lymphedema conditions in new areas of Haiti, the dataset has already been designed as a baseline survey for evaluating a planned suppor t group intervention. This is appropriate for the preparation and introduction of morbidit y control in the study community, but some important variables might be excluded for the purpose of the desc ription of general filariasis conditions. In order to obtain more universal info rmation in the new areas, it would be necessary to conduct the primary data collection at the neutral position. The third limitation is the method of analys is. The purpose of the data analysis was

PAGE 139

128 descriptive rather than analytical so that the socio-demographic variables were analyzed one by one. This thesis concluded that the ge nder difference seemed to be less important, but no gender variation within three different towns was examined. Also, other variables which had no association with lymphedema conditions might be potential confounders. Thus, more complex analyses such as multiple regression analysis would be required for deeper analysis. In this thesis, however, data collection was at a single point of time so it is difficult to determine the causal relationships between variables. Therefore, the results of this thesis can be viewed as a preliminar y assessment of QOL among lymphedema patients in Haiti. Conclusion Lymphedema conditions due to LF in th ree rural towns in Haiti are explained by several interesting variables. Regional variat ion was noteworthy. Generally, people in La Plaine reported more severe symptoms than those in other communities. This was indicated by foot examinati on and scores on the QOL instru ments. Particularly, physical health status and activity limitation were significantly different among towns. Also, health-related behaviors vari ed across communities. Tradit ional medicine was more common in Arcahaie, and use of Western medici ne was higher in La Plaine. These findings should help design a culturally competent morb idity control strategy at the local level. Likewise, age is an importan t aspect of public health in terventions among lymphedema patients, especially in the treatment of legs. Other variables might be potential indicators for lymphedema control, but their effects se em to be limited. Thus, further investigation

PAGE 140

129 would be valuable. Lastly, the standardized QOL instrument s are useful tools to evaluate health conditions among lymphedema patients. They showed the influence of the socio-demographic variables on the health conditions among patients, particularly on physical health condition and activity limitation. Mental health condition seems to be less important, but additional study would be help ful. The reliability and validity of the instruments in this population are acceptable. Though more careful attention to validity of the EuroQol mental health indicators is needed general applications of the instruments for lymphedema patients are appropriate. Thus, th e information obtained in this research will contribute to the development of morb idity control programs in these areas.

PAGE 141

130 References Ahorlu, C.K., Dunyo, S.K., Koram, K.A., Nkrumah, F.K., Aagaard-Hansen, J., & Simonsen, P.E. (1999). Lymphatic filariasis related perceptions and practices on the coast of Ghana: implications for prevention and control. Acta Tropica, 73 251-264. Ahorlu, C.K., Dunyo, S.K., Asamoah, G., & Si monsen, P.E. (2001). Consequences of hydrocele and the benefits of hydrocelec tomy: a qualitative study in lymphatic filariasis endemic communities on the coast of Ghana. Acta Tropica, 80 215-221. Ary, D., Jacobs, L.C., & Razavieh, A. (2002). Validity and Reliability. In Introduction to research in education (6th ed., pp. 241-274). Belmont, CA: Wadsworth / Thomson Learning. Babu, B.V., Nayak, A.N., Dhal, K., Acharya, A. S., Jangid, P.K., & Mallick, G. (2002). The economic loss due to treatment costs and work loss to individuals with chronic lymphatic filariasis in rural communities of Orissa, India. Acta Tropica, 82 31-38. Bandyopadhyay, L. (1996). Lymphatic filariasis and the women of India. Social Science and Medicine, 42(10), 1401-1410. Beau de Rochars, M., Milord, M.D., Saint Jean, Y., Desormeaux, A.M., Dorvil, J., Lafontant, J., et al. (in press). Geographic Distribution of Lymphatic Filariasis in Haiti. American Journal of Tropica l Medicine and Hygiene. Brazier, J., Jones, N., & Kind, P. (1993). Testing th e validity of the Euroqol and comparing it with the SF-36 health survey questionnaire. Quality of Life Research, 2 169-180. Brooks, R. (1996). EuroQol: th e current state of play. Health Policy, 37, 53-72. CDC. (1993). Recommendations of the internat ional task force for disease eradication. MMWR, 42, 1-38. CDC. (2000). Measuring Health Days Population Assessment of Health-Related Quality of Life Atlanta, GA: Centers for Disease Control and Prevention. Coons, S.J., Rao, S., Keininger, D.L., & Hays, R.D. (2000). A comparative review of generic quality-of-life instruments. Parmacoeconomics, 17(1) 13-35.

PAGE 142

131 Coreil, J., Mayard, G., Louis-Charles, J., & Addiss, D. (1998). Filarial elephantiasis among Haitian women: social context and be havioural factors in treatment. Tropical Medicine and Interna tional Health, 3(6), 467-473. Coreil, J., Mayard, G., & Addiss, D. (2003). Support groups for women with lymphatic filariasis in Haiti. TDR Social, Economic, and Behavioral Research Report Series No.2. Geneva: World Health Organization. Cronbach, L.J. (1951). Coefficient alpha a nd the internal structure of tests. Psychometrika, 16, 297-334. Dahl, B.A. (2001). Lymphedema treatment in Leogane, Haiti: An effective, sustainable and replicable model program for lymphatic filariasis morbidity control. (MPH thesis, Emory University, 2001). DeVellis, R.F. (2003). Validity. In Scale Development: Theory and Application (2 nd ed). Thousand Oaks, CA: Sage Publications. Dreyer, G., Figueredo-Silva, J., Neafie, R.C., & Addiss, D.G. (1998). Lymphatic Filariasis. In A.M.Nelson& C.R.Horsburgh Jr (Eds), Pathology of Emerging Infections 2 (p.317-342). Washington, D.C.: ASM press. Dreyer, G., Addiss, D., Dreyer, P., & Noroes, J. (2002). Basic Lymphoedema Management: Treatment and Prevention Problems Asso ciated with Lymphatic Filariasis Hollis Publishing Company. Eaton, W., Muntaner, C, Smith, CB, & Tien, AY. (2003) Center for Epidemiologic Studies Depression Scale: Review a nd Revision (CESD and CESDR). The Use of Psychological Testing for Treatmen t Planning and Outcomes Assessment, (3 rd ed, volume III, chapter 40). Retrieved on November 10, 2003, from http://www.mdlogix.com/cesdrpaper.pdf Eberhard, M.L., Walker, E.M., Addiss, D.G., & Lammie, P.J. (1996). A survey of knowledge, attitude, and percep tions (KAPs) of lymphatic filariasis, elephantiasis, and hydrocele among residents in an endemic area in Haiti. American Journal of Tropical Medicine and Hygiene, 54(3), 299-303. Fransen, M., & Edmonds, J. (1999). Reliability an d validity of the EuroQol in patients with osteoarthritis of the knee. Rheumatology, 38, 807-813. Gasarasi, D.B., Premji, Z.G., Mujinja, P.G.M., & Mpembeni, R. (2000). Acute adenolymphangitis due to bancroftian fila riasis in Rufiji district, south east Tanzania. Acta Tropica, 75, 19-28.

PAGE 143

132 Gudex, C., Dolan, P., Kind, P., & Williams, A. (1996). Health state valuations from the general public using the vi sual analogue scale. Quality of Life Research, 5(6) 521-531. Gyapong, J.O., Gyapong, M., Evans, D.B., Aikins, M.K., & Adjei, S. (1996a). The economic burden of lymphatic fila riasis in northern Ghana. Annals of Tropical Medicine and Parasitology, 90(1), 39-48. Gyapong, J.O., Gyapong, M., & Adjei, S. (1996b). The epidemiology of acute adenolymphangitis due to lymphatic filariasis in northern Ghana. American Journal of Tropical Medicine and Hygiene, 54(6), 591-595. Gyapong, M., Gyapong, J., Weiss, M., & Tanne r, M. (2000). The burden of hydrocele on men in Northern Ghana. Acta Tropica, 77 287-294. Henry, G.T. (1990). Sample selection approaches. In Practical Sampling Thousand Oaks, CA: Sage Publication. Hurst, N.P., Kind, P., Hunter, M., & Stubbing, A. (1997). Measuring he alth-related quality of life in rheumatoid arthri tis: validity, responsiveness and reliability of EuroQol (EQ-5D). British Journal of Rheumatology, 36 551-559. Hyma, B., Ramesh, A., & Gunasekaran, K. (1989). Lymphatic f ilariasis in Madras, India. Social Science and Medicine, 29(8) 983-990. King, C.L., & Freedman, D.O. (2000). Filariasis. In G.T.Stickland, Hunters Tropical Medicine and Emerging Infectious Disease, (8th ed., pp.740 752) W.B. Sanders Company. Konig, H., Ulshofer, A., Gregor, M., von Tirpitz, C., Reinshagen, M., Adler, G, et al. (2002). Validation of the EuroQol questionnaire in patients with inflammatory bowel disease. European Journal of Gastroenterology and Hepatology, 14(11) 1205-1215. Lammie, P.J., Addiss, D.G., Leonard, G., Hightower, W., & Eberhard, M.L. (1993). Heterogeneity in filarial-specific i mmune responsiveness among patients with lymphatic obstruction. Journal of Infectious Diseases, 167(5) 1187-1183. McPherson, T. (2003). Impact on the quality of life of lymphedema patients following introduction of a hygiene and skin care re gimen in a Guyanese community endemic for lymphatic filariasis: A preliminary clinical intervention study. Filaria Journal, 2(1) 1.

PAGE 144

133 Mwobobia, I.K., & Mitsui, Y. (1999). Demogr aphic and socio-economic factors with implications for the control of lymphatic filariasis in Kwale District, Kenya. East African Medical Journal, 76(9), 495-498 (abstract). Myers, C. & Wilks, D. (1999). Comparison of Euroqol EQ-5D and SF-36 in patients with chronic fatigue syndrome. Quality of Life Research, 8 9-16. Ottesen, E.A. (1987). Introduction. In Ciba Foundation Syposium 127, Filariasis Chichester, Swiss: A Wile y-Interscience Publication. Ottesen, E.A. (2000). Towards eliminating lymphatic filariasis. In T.B.Nnutman, Lymphatic Filariasis (pp.201-215). Imperial College Press, London. PAHO. (1998). Haiti. Health in the Americas. (1998 ed, Volume II, pp.316-330). Washington, D.C.: Pan American Health Organization. PAHO. (2001). Country Health Profile: Haiti Washington, D.C.: Pan American Health Organization. Retrieved on May 20, 2003, from http://www.paho.org/English/SHA/prflHAI.htm PAHO. (2003). Epidemiological Bulletin. (Mar 24(1), pp.1-24). Washington, D.C.: Pan American Health Organization. Pereira de Godoy, J.M., Braile, D.M., de Fatima Godoy, M., & Longo, O. Jr. (2003). Quality of life and peripheral lymphedema. Lymphology, Jun, 35(2) 72-75. Radloff, L.S. (1977). The CES-D scale: a self-r eported depression scale for research in the general population Applied Psychological Measurement, 1 385-401. Ramaiah, K.D., Kumar, K.N., Ramu, K., Pa ni, S.P., & Das, P.K. (1997). Functional impairment caused by lymphatic filarias is in rural areas of South India. Tropical Medicine and interna tional Health, 2(9), 832-838. Ramaiah, K.D., Radhamani, M.P., John, K.R., Ev ans, D.B., Guyatt, H., Joseph, A., et al. (2000). The impact of lympha tic filariasis on labour i nputs in southern India: results of a multi-site study. Annals of Tropical Medicine and Parasitology, 94(4), 353-364. Rauyajin, O., Kamthornwachara, B., & Yablo, P. (1995). Socio-cultur al and behavioural aspects of mosquito-borne lymphatic filariasis in Thailand: A qualitative analysis. Social Science and Medicine, 41(12) 1705-1713.

PAGE 145

134 Shriram, A.N., Murhekar, M.V., Ramaiah, K. D., & Sehgal, S.C. (2002). Prevalence of diurnally subperiodic bancroftian filariasis among the Nicobarese in Andaman and Nicobar Island, India: effect of age and gender. Tropical Medicine and International Health, 7(11) 949-954. Sitzia, J., & Sobrido, L. (1997). Measurement of health-related quality of life of patients receiving conservative treatment for limb lymphoedema using the Nottingham Health Profile. Quality of Life Research, 6, 373-384. van Agt, H.M., Essink-Bot, M.L., Krabbe P.F., & Bonsel, G.J. (1994). Test-retest reliability of health state valuations collected with the EuroQol questionnaire. Social Science & Medicine, 39(11), 1537-1544. Ware, J.E., & Sherbourne, C.D. (1992). The MOS 36-Item Short-Form Health Survey (SF-36) I. Conceptual Fr amework and Item Selection. Medical Care, 30(6), 473-483. Weerasooriya, M.V., Weerasooriya, T.R., G unawardena, N.K., & Samarawickrema, W.A. (2001). Epidemiology of bancro ftian filariasis in three suburban areas of Matara, Sri Lanka, Annals of Tropical Medicine and Parasitology, 95(3), 263-273. WHO. (1995). The World Health Report 1995: Bridging the gaps Geneva: World Health Organization. WHO. (2000a). Fact sheets: Lymphatic filariasis (Fact sheet No.102). Geneva: World Health Organization. Retrieved on May 14, 2003, from http://www.who.int/inf-fs/en/fact102.html WHO. (2000b). The Programme to Eliminate Lymphatic Filariasis PELF. Geneva: World Health Organization. Retrieved on May 14, 2003, from http://www.who.int/ctd/filariasis/docs/pelf.zip ( Slide Presentations) WHO. (2002a). Lymphatic filariasis endemic countries and territories. Geneva: World Health Organization. Retrieved on July 7, 2003, from http://www.filariasis .org/index.pl?iid=2695 WHO. (2002b). Country profile: Haiti Geneva: World Health Organization. Retrieved on July 10, 2003, from http://www.who.int/country/hti/en/ World Bank. (2002). Haiti at a glance Washington, D.C.: The Worldbank. Retrieved on July 14, 2003, from http://www.worldbank.org/cgi-bin/se ndoff.cgi?page=%2Fdata%2Fcountrydata%2 Faag%2Fhti_aag.pdf

PAGE 146

135 Bibliography Addiss, D.G., Dimock, K.A., Eberhard, M.L., & Lammie, P.J. (1995). Clinical, parasitologic, and immunologic observa tions of patients with hydrocele and elephantiasis in an area with endemic lymphatic filariasis. Journal of Infectious Diseases, Mar; 171(3), 755-758. Beach, M.J., Streit, T.G., Houston, R, May, W.A., Addiss, D.G., & Lammie, P.J. (2001). Short report: documentation of iodine deficiency in Haitian schoolchildren: implication for lymphatic filariasis elimination in Haiti. American Journal of Tropical Medicine and Hy giene, Jan-Feb, 64(1-2) 56-57. CDC. (2000). Control of communicable diseases manual (17 th ed). American Public Health Organization. Cody, R.P., & Smith, J.K. (1997). Applied Statistics and the SAS Programming Language. (4 th ed). Precentice Hall. Dreyer, G., Nores, J., & Addiss, D. (1997) The silent burden of sexual disability associated with lymphatic filariasis. Acta Tropica, 63 57-60. Freeman, A.R., Lammie, P.J., Houston, R., LaPointe, M.D., Streit, T.G., Jooste, P.L., et al. (2001). A community-based trial for the cont rol of lymphatic filariasis and iodine deficiency using salt fortified w ith diethylcarbamazine and iodine. American Journal of Tropical Medicine and Hygiene, Dec, 65(6), 865-871. Lammie, P.J., Hightower, A.W., & Eberhard, M.L. (1994). Age-specific prevalence of antigenemia in a Wuchereria bancrofti-exposed population. American Journal of Tropical Medicine and Hygiene, Sep, 51(3) 348-355. Melrose, W.D. (2002). Lymphatic filarias is: new insights into an old disease. International Journal for Parasitology, 32 947-960. MPCE. (2002). Carte de pauvret pour Haiti Version interimaire. (in French) Newman, W.L. (2003). Social research methods, qualitative and quantitative approaches (5 th ed). Allyn and Bacon. Pani, S.P., Balakrishnan, N. Srividya, A., Bundy, D.A., & Grenfell, B.T. (1991). Clinical epidemiology of bancroftian filari asis: effect of age and gender. Transactions of the Royal Society of Tropical Me dicine and Hygiene, Mar-Apr, 85(2) 260-264.

PAGE 147

136 Appendices

PAGE 148

137 Appendix A: Lymphedema Stages (Dreyer et al., 2002).

PAGE 149

138 Appendix B: Filariasis Baseline Evaluation Survey Arcahaie. Anket de Baz Pou Filaryoz Akaye (Filariasis Baseline Evaluation Survey Arcahaie) Dat anket la (Date of survey) _____________________________2003 Nom ankete a (Name of interviewer) ____________________________________ Nom enfomate a (Nam e of respondent) __________________________________ Eske-w kann vizite Klinik Gwopye nan lHpital Ste. Croix dj? Wi Non (si Wi pa ranpli kesyon a) Demografik (Demographics) Lokalite (address)_________________________________________________________ Sex (Gender) Fiy (Female) ________ Gason (Male) ______ Ki laj ou? (How old are you?) _________ Eske ou marye/plase/ etc.? (Are you married/co-habiting, etc.?) ___________ [1 marye 2 plase 3 viv avek 4 renmen 5 fyanse 6 selibate 7 separe ou divose 98 Lot, di kisa] [1married 2plase 3-live together 4in relationship 5engaged 6single 7separated or divorced 98Other, explain] Kombyen pitit you gen kap viv kounye a (How many living children do you have?) _________ Ki laj premye la genyen? ____ ans Ki laj denye la genyen? ____ans (How old is the oldest child?) (How old is the youngest child?) Kombyen petit ale lekol? _______ (How many of your children go/went to school?) Nan ki relijon ou mache (Wha t is your religion?) ______________ [1katolik 2protestan 3vodouyizan 4pa gen relijon 98lot, di kisa] [1Catholic 2Protestant 3Voudouiste 4No religion 5Other, explain] Kisa ou ap fe pou viv/ki metye ou genyen? _________________

PAGE 150

139 Appendix B: (Continued) (What do you do for a living/what is your occupation?) [1kiltivate 2komesan nan lakay 3 komesan nan marche 4 Couture 5 Lot travay 6 Pa gen travay 98Lot, di kisa] [1farmer 2seller at home 3 seller at market 4 tailor/ seamstress 5Other work 4Unemployed 98Other, explain] Normalman, kombyen jou ou travay nan seman? _____ jou pa semane Kombyen kb ou kabap f? _____ Gdes pa jou pa semane pa mwa Kombyen ane ou te fe nan lekol? _______________ [How many years of school did you complete?] Eske ou konn li ak ekri? (Can you read and write?) wi (Y) _______ non (N) _______ Eske genyen nan kay la? (Do you have the following in your house?) Radyo (radio) ________ Office/china/gade manje/armoire (Storage chest) ________ Salon (living room) ________ Bicyclette/motocyclette (Bicycle/motor cycle) ________ Istwa de maladi (Illness history) Ki laj ou te genyen le ou te premye santi maladi ya? ________________ (What age were you when you fi rst became aware of the illness?) Kisa ou ta panse ou te genyen? ________________ (What did you think you had?) 1-Fredite 2Ekzema 3-Glan 4Poud maji 5Piki insek 6-Ansent 7Antoch 8Yon maladi 9Gwopye 10Filariose 11Pakonnen 12Lot bagay, di kisa (1-Chill 2Eczema 3Gland 4Magical powder 5Insect bite 6Pregnancy 7Sprain 8An illness 9Big f oot 10Filariasis 11Dont know 12Other, explain) Ki premye sintom ou te remake pou maladi sa a? 1pye a anfle 2doule 3glann 4 fyev 5 tet fe mal 6 pye a cho 7Lot, di kisa

PAGE 151

140 Appendix B: (Continued) _________________________________________ (What was the first symptom you noticed? 1foot swollen 2pain 3swollen gland 4 fever 5 headache 6 foot hot 7Other, explain) Kisa ou te fe pou jwenn soulajman? Eksplike nou sa. 1Kay ougan 2Remed fey 3pommade 4mete med fey 5Ti kwi/Sansi 6Mete glac 7Medicaman famasi 8Lot bagay, di kisa ________________________ (What did you do to treat the illness? 1Traditional healer 2Herbal remedy 3Pommade 4Herbal leaves on leg 5Cupping/leeches 6put ice on leg 7Pharmaceutical medicine 8Other, explain) Kijan de prekosyon ou pran ak pye ou? Di mwen tou sa ou fe pou li. _______________ _______________________________________________________________________ (What precautions do you take with your foot? Tell me everything you do for it.) Eske ou konn ki sa atak la-ye? Eske ou kapab dim, sil vous pl ait. (Pa bezwen ekrit anyen)

PAGE 152

Appendix B: (Continued) Ann ki sot pase, konbyen atak ou te genyen? ___________ (During the past year, how many acute attacks did you have?) 141 Atak #1 Atak #2 Atak #3 Nan ki mwa ou te genyen atak-sa? Mwa ____________ Mwa ____________ Mwa ____________ Konbyen jou atak-sa te dire? ______ Jou ______ Jou ______ Jou Eske gen kote ou te ale pou jwenn tretman? Di ki kote ou te ale. Klinik _________________ Kay Medsen Fey _________ Kay Ougan ____________ Ajan de Sante ___________ Lot kote, di ki kote ___________________ Okenn kote Klinik _________________ Kay Medsen Fey _________ Kay Ougan ____________ Ajan de Sante ___________ Lot kote, di ki kote ___________________ Okenn kote Klinik _________________ Kay Medsen Fey _________ Kay Ougan ____________ Ajan de Sante ___________ Lot kote, di ki kote ___________________ Okenn kote Kombyen jou ou te tan avan ale la? Pou ki sa? ____ Jou Mwen pat tan Pat gen kb ____ Jou Mwen pat tan Pat gen kb ____ Jou Mwen pat tan Pat gen kb

PAGE 153

Appendix B: (Continued) Si ou pat ale okenn kote, pou ki sa? Malade pat kapab soti Pa renmen tretman Pa bezwen ale mwen ka f tretman lakay mwen Lot __________________ Malade pat kapab soti Pa renmen tretman Pa bezwen ale mwen ka f tretman lakay mwen Lot __________________ Malade pat kapab soti Pa renmen tretman Pa bezwen ale mwen ka f tretman lakay mwen Lot __________________ 142

PAGE 154

Appendix B: (Continued) Kombyen fwa ou te ale la pou meme atak? ______ Fwa ______ Fwa ______ Fwa Eske mounn konn ale ak ou? Non Wi Si Wi, ki moun?_____________ Si Wi, ki moun? _____________ Kijan de travay li fe? ______________________ Non Wi Kijan de travay li fe? _________________________ Non Wi Si Wi, ki moun? _____________ Kijan de travay li fe? _________________________ Atak #1 Atak #2 Atak #3 Kombyen tan wap pran pou kabap rive la? ______ Minutes ______ Minutes ______Minutes Eske ou te bezwen peye transportasyon pou ale la? Non Wi Si wi, $ _____ Gdes chak moun Non Wi Si wi, $ _____ Gdes chak moun Non Wi Si wi, $ _____ Gdes chak moun Kijan de tretman ou te jwenn nan lokalite-sa? Di kombyen kb ou te peye pou chak tretman ou te Ki medikaman?_____________ Piki, $ ______ Gdes Medikaman, $ _____ Gdes Ki medikaman? _____________ Piki, $ ______ Gdes Medikaman, $ _____ Gdes Piki, $ ______ Gdes Medikaman, $ _____ Gdes Ki medikaman? _____________ 143

PAGE 155

Appendix B: (Continued) jwenn. Med fey, $ _____ Gdes Tikwi/ Sansi, $ _____ Gdes Pommade, $ _____ Gdes Mesaj, $ ______ Gdes Bandaj, $ ______ Gdes Ts, $ ______ Gdes Ki ts? ________________ Lot, di ki sa ______________ Med fey, $ _____ Gdes Tikwi/ Sansi, $ _____ Gdes Pommade, $ _____ Gdes Mesaj, $ ______ Gdes Bandaj, $ ______ Gdes Ts, $ ______ Gdes Ki ts? ________________ Lot, di ki sa _______________ Med fey, $ _____ Gdes Tikwi/ Sansi, $ _____ Gdes Pommade, $ _____ Gdes Mesaj, $ ______ Gdes Bandaj, $ ______ Gdes Ts, $ ______ Gdes Ki ts? ________________ Lot, di ki sa _______________ Kombyen tan ou te f la pou jwenn tretman? ______ Minutes ______ Minutes ______ Minutes Kombyen kb ou te peye pou konsiltasyon? Eske se pou chak Chak fwa _______ Gdes Total Chak fwa _______ Gdes Total Chak fwa _______ Gdes Total 144

PAGE 156

Appendix B: (Continued) vizit ou byen pou total? Atak #1 Atak #2 Atak #3 Ki sa ou te f lakay ou pou atak sa? Si ou te bezwen achete bagay pou tretman sa lakay ou, di kombyen kb ou te peye. Bwe medikaman Ki medikaman?_____________ Ki medikaman? _____________ Pou kombyen jou? ___ jou Kombyen kb? $ ______ Gdes Mete med fey Mete pomade Kombyen kb? $ ______ Gdes Lave pye ____ fwa chak jou Leve pye Mete konpres fre nan janm Priye Anyen Bwe medikaman Pou kombyen jou? ___ jou Kombyen kb? $ ______ Gdes Mete med fey Mete pomade Kombyen kb? $ ______ Gdes Lave pye ____ fwa chak jou Leve pye Mete konpres fre nan janm Priye Anyen Bwe medikaman Ki medikaman? _____________ Pou kombyen jou? ___ jou Kombyen kb? $ ______ Gdes Mete med fey Mete pomade Kombyen kb? $ ______ Gdes Lave pye ____ fwa chak jou Leve pye Mete konpres fre nan 145

PAGE 157

Appendix B: (Continued) Lot, di ki sa _____________ Kombyen kb? $ ______ Gdes Lot, di ki sa _____________ Kombyen kb? $ ______ Gdes janm Priye Anyen Lot, di ki sa _____________ Kombyen kb? $ ______ Gdes Eske ou pat genyen mounn ki te d ou ak tretman pou atak-sa? Non Wi Ki moun? _______________ Kijan de travay li fe? ______________________ Non Wi Ki moun? _______________ Kijan de travay li fe? ______________________ Non Wi Ki moun? _______________ Kijan de travay li fe? ______________________ Atak #1 Atak #2 Atak #3 Eske ou te kapab travay pendan ou te genyen atak-sa? Non Wi Si Non, pou kombyen jou ou pat ka travay? ______ jou Non Wi Si Non, pou kombyen jou ou pat ka travay? ______ jou Non Wi Si Non, pou kombyen jou ou pat ka travay? ______ jou 146

PAGE 158

Appendix B: (Continued) Eske ann ki sot pase-a, ou te achet materiel paske ou gen gwo pye (pa pendan atak-yo)? Non Wi Si Wi, ki sa ou te achet? Souliye/Sandal Kombyen kb? $ ______ Gdes Pomade Kombyen kb? $ ______ Gdes Bandage Kombyen kb? $ ______ Gdes Ti ban Kombyen kb? $ ______ Gdes Kivet Kombyen kb? $ ______ Gdes Lot, di sa ____________ Kombyen kb? $ ______ Gdes Eske semain ki sot pas ou te: Lav rad? Non Wi Si Wi, _______ fwa Al nan mach? Non Wi Si Wi, _______ fwa Al vann? Non Wi Si Wi, _______ fwa Al leglise? Non Wi Si Wi, _______ fwa Al fonksyone lekol pou pitit-w? Non Wi Si Wi, _______ fwa 147

PAGE 159

Appendix B: (Continued) 148 Examen de pye a (Foot exam) Mesi jam malad la (Leg measurements) Pye dwat ( R) Pye goch (L) 1Do pye (foot) (10 cm from toe) __________ __________ 2Chevill (ankle) (10 cm from floor) __________ __________ 3Jam (leg) (20 cm from floor) __________ __________ Staj maladi a (Stage of illness) __________ __________ Presence des lesions (Lesions present) Oui /Non ________ ________ Si Wi, lokalite lesion-la _________________________________________________ Lesyon fongal (entry lesions) (indicate on the drawing where all entry lesions are, especially those between the toes) Konesyans maladi a (Knowledge about the illness) Eske ou ka di nou kisa ki bay maladi sa a? (Can you tell me what causes this illness?) 1Piki insek 2Maji 3Antoch/frappe pye 4Ve 5Fredi 6Mank de vitamin 7-Lot, di kisa [1-Insect bite 2Magic 3Sprain/injure foot 4Worms 5Chill 6Vitamin deficiency 7Other, explain] What kinds of care can help your gwopye? (Circle all that patient says)

PAGE 160

Appendix B: (Continued) 149 1Hygiene/washing 2Wear shoes 3Permanganate 4Crme 5Elevation 6Massage 7Exercise 8Bandage 9Medicine 10Nothing 11 Other, explain Are there things you can do to help prevent acute attacks? No _____ Yes ______ (If yes:) What can you do? 1Hygiene/washing 2Wear shoes 3Permanganate 4Crme 5Elevation 6Massage 7Exercise 8Bandage 9Medicine 10Nothing 11 Other, explain What can be done to provide relief during an acute attack? (Circle all that patient says) 1Hygiene/washing 2Wear shoes 3Permanganate 4Crme 5Elevation 6Massage 7Exercise 8Bandage 9Medicine 10Nothing 11 Other, explain Pratik ke malad fe li menm pou swenye pye a (Self-care practices for leg) Di mwen tou sa ou fe pou pye a ak fwekans (Tell me everything you do for your leg and how often). Chak jou Yon fwa/semen Yon fwa/mwa Mwens ke sa (Daily) (Once/week) (Once/month) (Less often) 1Ijenn ______ ______ ______ ______ 2Mete sandal ______ ______ ______ ______ 3Perman. ______ ______ ______ ______ 4Krm ______ ______ ______ ______ 5Elevasyon ______ ______ ______ ______ 6Masaj ______ ______ ______ ______ 7Eksersis ______ ______ ______ ______ 8Bandaj ______ ______ ______ ______ 9Medikaman ______ ______ ______ ______ 10Remed fey ______ ______ ______ ______ 11Pomad ______ ______ ______ ______ 12Lot:_____ ______ ______ ______ ______ Self-efficacy Ki lot bagay ou kapab fe pou pye-a ke ou pap fe kounye-a? Fwekans? What other things can you do to help your leg that you do not currently do? Frequency?

PAGE 161

Appendix B: (Continued) 150 Chak jou Yon fwa/semen Yon fwa/mwa Mwens ke sa (Daily) (Once/week) (Once/month) (Less often) 1-Ijenn ______ ______ ______ ______ 2-Mete sandal ______ ______ ______ ______ 3-Perman. ______ ______ ______ ______ 4-Krm ______ ______ ______ ______ 5-Elevasyon ______ ______ ______ ______ 6-Masaj ______ ______ ______ ______ 7-Eksersis ______ ______ ______ ______ 8-Bandaj ______ ______ ______ ______ 9-Medikaman ______ ______ ______ ______ 10-Remed fey ______ ______ ______ ______ 11-Pomad ______ ______ ______ ______ 12-Lot:_____ ______ ______ ______ ______ Eske ou gen konfyans nan kapasite pa-ou pou fe tout bagay posib pou pran swen pye-ou nan meye fason posib? How confident are you in your ability to do all things possible to take care of your leg the best way you can? 1Anpil confyans 2Ti kras confyans 3A penn confyans 4Manke confyans 5Pa ditou konfyans 1Very confident 2Somewhat confident 3A little confident 4Not very confident 5Not at all confident Quality of Life Euro-Qual 5D By placing a check in one box in each group below, pl ease indicate which statements best describe your own health state today. (Tanpri, pou chak gwoup nan paj sa a, ty eke nan ti kare yo pou ou ka f nou konnen ki fraz ki esplike pi byen kijan sante ou ye jodi a Mobility (Mouvman) I have no problems in walking about (Mwen pa gen pwoblm pou m mache) I have some problems in walking about

PAGE 162

Appendix B: (Continued) 151 (Mwen gen kk pwoblm pou m mache) I am confined to bed (Mwen oblije ret nan kabann toutan) Self-Care (Pwpte k w) I have no problems with self-care (Mwen pa gen pwoblm pou m pwpte tt mwen pou kont mwen) I have some problems washing or dressing myself (Mwen gen ti pwoblm pou m f twalt mwen, pou m benyen osnon pou m abiye pou kont mwen) I am unable to wash or dress myself (Mwen pa ka ni f twalt mwen, ni abiye m pou kont mwen) Usual Activities (e.g. work, study, housework, family or leisure activities) {Aktivite Ou Abitye F (pa egzanp: ale travay, etidye, f travay nan kay, f aktivite ak fanmi an osnon amizman)} I have no problems with perf orming my usual activities Mwen pa gen pwoblm pou mwen f aktivite mwen toujou abitye f yo I have some problems with pe rforming my usual activities (Mwen gen ti pwoblm pou mwen f aktivite mwen abitye f yo) I am unable to perform my usual activities (Mwen pa kapab f aktivite mwen te konn abitye f yo ank) Pain/Discomfort (Doul ak malalz) I have no pain or discomfort (Mwen pa gen doul, mwen pa malalz ak janm lan) I have moderate pain or discomfort (Mwen gen ti doul epi mwen yon ti jan malalz ak janm lan)

PAGE 163

Appendix B: (Continued) 152 I have extreme pain or discomfort (Mwen gen anpil doul, mwen ma lalz anpil ak janm lan) Anxiety/Depression (Enk yetid ak Dekourajman) I am not anxious or depressed (Mwen pa gen enkyetid; mwen pa dekouraje non plis) I am moderately anxious or depressed (Mwen gen yon ti enkyetid, mwen yon ti jan dekouraje) I am extremely anxious or depressed (Mwen gen anpil enkyetid ; mwen dekouraje anpil)

PAGE 164

Appendix B: (Continued) 153 To help people say how good or bad a health state is, we have drawn a scale (rather like a thermometer) on which the best state you can imagine is marked 100 and the worst state you can imagine is marked 0. (Pou ede moun jwenn yon fason pou yo ka esplike si sante yo anfm osnon si sante yo pa anfm, nou trase pakt ba sa yo (ki sanble ak yon tmomt). Pi bon sante yon moun ka imajine rive sou liy san (100) epi l sante a pa bon ditou li rive jouska liy 0 (zewo). We would like you to indicate on this scale how good or bad your own health is today, in your opinion. Please do this by drawing a line from the box below to whichever point on the scale indicates how good or bad your health state is today. (Nou ta renmen ou di nou dapre pakt ba sa yo kijan dapre ou sante ou anfm osnon si li pa bon jodi a. Tanpri, trase yon liy apatid kare nwa sa a, jiska pwen sou pakt ba sa yo ki montre kijan sante ou bon osnon kijan li pa bon jodi a.) Best Imaginable Health State (Sante ki pi bon ou ka imajine) 100 9 0 8 0 7 0 6 0 5 0 4 0 3 0 2 0 1 0 Y our own Health State 0 Worst Imaginable Health State (Sante ki pi mal ou ka imajine)

PAGE 165

Appendix B: (Continued) Center for Epidemiologic Studies Depression Scale (CED-D), NIMH Below is a list of the ways you might have felt or behaved. Please tell me how often you have felt this way during the past week. Make bwat ki pi byen montre jan ou te santi-ou oubyen jan ou te aji PANDAN SEMENN PASE-A During the Past Week Rarely or none of the time (less than 1 day) Raman. Ou, pa minm yon fwa (mwens ke yon jou) Some orr a little of the time (1-2 days) Kek fwa. Ou, yon ti kras tan. (1 a 2 jou) Occasionally or a moderate amount of time (3-4 days) Ase souvan oubyen, yon kantite modere (3 a 4 jou) Most or all of the time (5-7 days) Pi fo Oubyen tout tan. (5 a 7 jou) 1. I was bothered by things that usually dont bother me. Kek bagay te deranje-m ki pa abitye deranje-m 2. I did not feel like eating; my appetite was poor. Mwen pat santi-m mwen ta manje Apeti-m te ba 3. I felt that I could not shake off the blues even with help from my family or friends. Mwen pat santi-m te kapab souke tristes mwen malgre 154

PAGE 166

Appendix B: (Continued) fanmi-m ak sanmi-m yo ede-m 4. I felt I was just as good as other people. Mwen te santi-m byen menm jan ak tout lot mounn. 5. I had trouble keeping my mind on what I was doing. Mwen te genyen difikilte konsantre sou sa mwentap fe. 6. I felt depressed Mwen te santi-m demoralize 7. I felt that everything I did was an effort. Mwen santi tout sa mwen fe se ak gwo efo. Rarely or none of the time (less than 1 day) Raman. Ou, pa minm yon fwa (mwens ke yon jou) Some orr a little of the time (1-2 days) Kek fwa. Ou, yon ti kras tan. (1 a 2 jou) Occasionally or a moderate amount of time (3-4 days) Ase souvan oubyen, yon kantite modere (3 a 4 jou) Most or all of the time (5-7 days) Pi fo Oubyen tout tan. (5 a 7 jou) 8. I felt hopeful about the future. 155

PAGE 167

Appendix B: (Continued) 156 Mwen santi mwen genyen espwa avek avni mwen. 9. I though my life had been a failure. Mwen te panes la vi-m te yon gwo echek. 10. I felt fearful. Mwen te santi mwen pe 11. My sleep was restless. Mwen te san somey 12. I was happy. Mwen te kontan. 13. I talked less than usual. Mwen te pale mwens ke mwen Abitye. 14. I felt lonely Mwen te santi mwen poukont Mwen. 15. People were unfriendly Mwen te konn banmwen tretman fret. 16. I enjoyed life. Mwen te jwi lavi mwen. 17. I had crying spells. Mwen te pran kriye yon paket fwa. 18. I felt sad. Mwen te santi mwen tris.

PAGE 168

Appendix B: (Continued) 19. I felt that people dislike me. Mwen te santi ke mounn pat renmen mwen. 20. I could not get going Mwen pat kapab souke kom pou mwen demare. 157

PAGE 169

Appendix B: (Continued) 158 CDC Healthy Days Questions 1. Eske-w ka di an jeneral kijan sante-w ye: (would you say that in general your helath is:) Read responses 1a-e. a. Ekselan b. tre byen c. byen d. pa mal ou e. mal. f Mwen pa konen / Mwen pa si g. Refuse Pa li repons pou kesyon swivan. (Do not read the responses for the following questions) 2. Kounye-a nap panse a sante ko ou, mala di ko ou ak aksidan ladan tou, konbyen jou nan mwa ki sot pase sante ko ou pat byen? (Now thinking about your physical health, which includes physical illness and injury, for how many days during the past 30 days was your physical health not good?) a. Nomb jou ____Number of days ____ b. Ayen____ None____ c. Pa konnen ____Dont know____ d. Refuse____ Refused____ 3. Kounye-a nap panse a maladi nan tet, maladi stres, depresyon, e pwoblem avek emosyon, konbyen jou nan mwa ki sot pase ya ou te malad nan tet? (Now thinking about your mental health, which includes stress, depression, and problems with emotions, for how many da ys during the past 30 days was your mental health not good?) a. Nomb jou ____Number of days ____ b. Ayen____ None____ c. Pa konnen ____Dont know____ d. Refuse____ Refused____ 4. Nan mwa ki sot pase a, konbyen jou maladi ko ou ak maladi tet fe ou pa kab fe aktivite-w abitye fe, tankou oki pe tet ou, travay, ak amizma n? (During the past 30

PAGE 170

Appendix B: (Continued) 159 days, for about how many days did poor phys ical or mental health keep you from doing your usual activities, such as self-care, work, or recreation?) a. Nomb jou ____Number of days ____ b. Ayen____ None____ c. Pa konnen ____Dont know____ d. Refuse____ Refused____ Kesyon kap ven you se sou limitasyon ou ka p genyen chak jou nan lavi-w. (The next questions are about limitations you may have in your daily life) 1. Eske ou limite nan aktivite-w paske ou domaje oubyen gen pwoblem sante? (Are you limited in any way in any activities because of any impairment or health problem?) a. Wi (Yes) b. Non ale nan Kesyon 6 (No go to question 6) c. Pa konnen Ale nan kesyon 6 (Dont know go to question 6) d. Refuse Ale nan kesyon 6 (Refused go to question 6) 2. Ki domaj oubyen pwoblenm sante presipal ou gen ki fe ou pa kap fe aktivite? (What is the MAJOR impairment or health problem that limits your activities?) DO NOT READ. SELECT ONLY ONE CATEGORY. a. arthrit (arthritis) b. pwoblem avek do oubyen kou (back or neck problems) c. fwakti, mal zo (fr actures, bone/joint injury) d. pwoblem mashe (problems walking) e. pwoblem avek respira tion (lung/breathing problem) f. difikilti koute (hearing problem) g. pwoblem avek je (eye / vision problem) h. maladi ke (heart problem) i. maladi atak stroke (stroke problem) j. tansyon (high blood pressure) k diabet/sik (diabetes) l. kanse (cancer) m. depression / anxiety / emotional problems n. gwo pye (lymphedema) o. lot pwoblem (other problem) ___________________________________ p. Pa konnen (Dont know) q. Refuse (Refused) 3. Kombyen tan sa genyen depi ou gen domaj presipal sa? (For how long have your

PAGE 171

Appendix B: (Continued) 160 activities been limited because of your major impairment or health problem?) PA LI REPONS. MAKE SA LI DI. ( DO NOT READ. CODE APPROPRIATE UNIT OF TIME.) a. _____ jou (days) b. _____ semann (weeks) c. _____ mwa (months) d. _____ ane (years) e. Pa konnen (Dont know) f. Refuse (Refused) 4. Poutet domaj oubyen pwoblem sante, eske ou beswen lot moun okipe ou, tankou pou manje, benyen, mete rad sou ou, ou pou mache nan kay la? (Because of any impairment or health problem, do you n eed the help of other persons with your PERSONAL CARE needs, such as eating, bathing, dressing, or getting around the house)? a. Wi (Yes) b. Non (No) c. Pa konnen (Dont know) d. Refuse (Refused) 5. Poutet ou domaje oubyen pwoblem sante, eske ou beswen ed pou fe aktivite chak jou-w, tankou menaj nan kay, komes, fe pwovizyon, oubyen soti pou fe lot bagay? (Because of any impairment or health problems, do you need the help of other persons in handling your ROUTINE needs, such as everyday household chores, doing necessary business, shopping, or getting around for other purposes?) a. Wi (Yes) b. Non (No) c. Pa konnen (Dont know) d. Refuse (Refused) 6. Nan trant jou ki sot pase yo konbyen jou ko w te fe telman mal li te di pou ou fe acktivite nomal tankou okipe te t ou, travay, ou rekreyasyon? (During the past 30 days, for about how many days did PAIN make it hard for you to do your usual activities, such as self-care, work, or recreation?) a. ______ Nomb jou (number of days)

PAGE 172

Appendix B: (Continued) 161 b. Ayen (None) c. Pa konnen (Dont know) d. Refuse (Refused) 7. Nan trant jou ki sot pase-a konbyen fwa ou te sant i kew pa kontan, tris, ou chagren? (During the past 30 days, for a bout how many days have you feldt SAD, BLUE, or DEPRESSED?) a. ______ Nomb jou (number of days) b. Ayen (None) c. Pa konnen (Dont know) d. Refuse (Refused) 8. Nan trant jou ki sot pase-a konbyen fwa ou te santi ke pa kontan, enkye, tandi, ke pa pose? (During the past 30 days, for about how many days have you felt WORRIED, TENSE, or ANXIOUS?) a. ______ Nomb jou (number of days) b. Ayen (None) c. Pa konnen (Dont know) d. Refuse (Refused) 9. Nan trant jou ki sot pase-a konbyen jou ou te santi ou pa pran ase repo oubyen ou pa ase domi? (During the past 30 days, for about how many days have you felt that you did not get ENOUGH REST OR SLEEP?) a. ______ Nomb jou (number of days) b. Ayen (None) c. Pa konnen (Dont know) d. Refuse (Refused) 10. Nan trant jou ki sot pase a konbyen jou ou te santi ou pat an fom oubyen avek anpil eneji? (During the past 30 days, for about how many days have you felt VERY HEALTHY AND FULL OF ENERGY?) a. ______ Nomb jou (number of days) b. Ayen (None) c. Pa konnen (Dont know) d. Refuse (Refused)

PAGE 173

Appendix C: Map of Port-au-Prince Area. 162


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 Ka
controlfield tag 001 001478765
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 040811s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000407
035
(OCoLC)56389920
9
AJS2455
b SE
SFE0000407
040
FHM
c FHM
090
RA425 (ONLINE)
1 100
Kanda, Koji.
4 245
The quality of life among lymphedema patients due to lymphatic filariasis in three rural towns in Haiti
h [electronic resource] /
by Koji Kanda.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (M.S.P.H.)--University of South Florida, 2004.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 173 pages.
520
ABSTRACT: The worldwide eradication of lymphatic filariasis has recently started with two strategies, interruption of transmission and morbidity control. One of the most endemic countries, Haiti has experienced successful interventions through national and international efforts, but the morbidity control is still hindered by a lack of adequate information on quality of life (QOL) issues among those suffering from the chronic manifestations of the disease such as lymphedema. In addition, previous interventions have been focused primarily in a single community where an established lymphedema treatment clinic serves as a national reference center, so it is critical to expand programs to other areas in Haiti. The purpose of the study was to understand the issues of morbidity control and QOL among lymphedema patients due to lymphatic filariasis in three rural Haitian towns. Secondary data (n = 316) collected in an ongoing filariasis support group project was analyzed in terms of socio-demographic characteristics, including gender age, and regional perspectives. Also, two different commercial QOL instruments (EuroQol, CDC Healthy Days) and a subjective well-being assessment tool (CES-D) were introduced to describe their QOL and mental health status, respectively. The reliability and validity of the measurements were established at the same time. Regional differences were evident in patients illness history, knowledge of the illness, self-care and self-efficacy for legs, and major QOL indicators related to physical and mental health. Age of patients also influenced foot size, illness stage, and the QOL scores. However, other socio-demographic factors were poorly associated with filariasis related variables, including gender. The commercial QOL instruments and a standardized mental health tool satisfied a reasonable level of reliability and validity. Though additional discussion is needed regarding the validation of the mental health scales between EuroQol and the other instruments, they nevertheless offer utility for enhancing the quality of morbidity control programs. These findings offer a significant contribution for the development of filariasis prevention programs such as community-based morbidity control and support group activities in Haiti, as well as other areas of the filariasis-endemic world.
590
Adviser: Coreil, Jeannine
653
reliability and validity.
health behavior.
gender.
morbidity control.
global health.
0 690
Dissertations, Academic
z USF
x Public Health
Masters.
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.407