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 001920994
007 cr mnu|||uuuuu
008 080115s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001871
Ancheta, Irma B.
The relationship between B-type natriuretic peptide levels and hospital length of stay and quality of life in congestive heart failure patients
h [electronic resource] /
by Irma B. Ancheta.
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: Previous research on quality of life (QOL) and its relation to bnp levels in heart failure (HF) has been widely studied. However, the impact of physicians' knowledge of BNP levels at time of clinic visit on QOL and hospital length of stay (LOS) has yet to be fully investigated. The purpose of this study were to determine if physicians' knowledge of BNP levels affected a change in QOL scores at 90 days and reduce hospital length of stay among heart failure patients. QOL data from HF clinic patients (N = 108, 67.5 Â¨ 12.3, 56% male, ejection fraction 26.5 Â¨ 8.2) were analyzed. QOL was measured at time of clinic visit (T1) and at 90 days (T2) using the Minnesota Living with Heart Failure Questionnaire (MLHFQ). An independent t-test was utilized to compare the two groups. Findings: Both groups were comparable regarding demographic and baseline characteristics.^ There was no significant association observed between the experimental and control group at 90 days, although the data indicated a decrease in the mean QOL scores at 90 days (37.46 Â¨ 28.67) as compared to the mean QOL scores at baseline (46.87 Â¨ 29.63) for both groups. Because the QOL scale is reversed, this indicated that there was a positive change in QOL scores during the 90 day time interval. Hospital LOS was similar for both groups (mean=3 days). BNP levels were significantly correlated with both baseline QOL scores (r=.25, p=.01) and physical subscale scores (r=.24, p=.01). Mortality was higher in the control when compared to the experimental group (t=1.99, df=90, p=.04). Conclusion: While physicians' awareness of BNP levels had not shown a significant change in QOL at 90 days, patients' QOL might already have been quite positive. Chronic HF patients may have adapted to their disease and have adjusted their perception of their QOL.^ Therefore, QOL may be a stable construct at this time. Findings may have been different on newly diagnosed HF patients since they may not have adapted to their health condition.
Dissertation (Ph.D.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 142 pages.
Adviser: Theresa Beckie, Ph.D.
Quality of Life.
Congestive heart failure.
Minnesota Living with Heart Failure Questionnaire.
t USF Electronic Theses and Dissertations.
The Relationship Between B-Type Natriuretic Peptide Levels and Hospital Length of Stay and Quality of Life in Congestive Heart Failure Patients by Irma B. Ancheta A dissertation submitted in partial fulfillment of the requirement s for the degree of Doctor in Philosophy College of Nursing University of South Florida Major Professor: Theresa Beckie, Ph.D, RN Mary Evans, Ph.D, RN Heather Stockwell, Sc.D Yougui Wu, Ph.D Date of Approval: December 12, 2006 Keywords: Heart Failure, BNP, Quality of Life, Copyright 2007, Irma B. Ancheta
i Table of Contents List of Tables iv List of Figures v Abstract viii Chapter One Introduction 1 Definition of congestive heart failure 1 Etiology 2 Pathophysiology 3 Signs and Symptoms 4 Stages of heart failure 5 Classification of heart failure 6 Epidemiology of heart failure 7 Prevalence 7 Incidence 7 Risk Factors 8 Morality Rates 9 Economic Impact 9 Effect of congestive heart fa ilure on hospitalizations 10 Effect of congestive heart fa ilure on hospitalizations 10 Role of B-type natriuretic peptide in heart failure 13 Statement of t he Problem 18 Purpose of Study 19 Research Hypotheses 19 Definition of Terms 20 Underlying Assumptions 21 Delimitations 21 Limitations 22 Significance of Study 22 Chapter Two Review of Literature Introduction 24 Quality of life as experienced by heart failure patients 25 Perception of Life Situation Subheading 25 Gender differences regarding perception of quality of life in heart failure 25
ii Quality of life as perce ived among male heart failure patients 27 Quality of life as per ceived among women with heart failure 28 Quality of life as perceiv ed by both heart failure patients and their partners 29 Subjective perception of quality of life and objective evaluation of the severity of heart failure 30 Psychological adjustment to illness 31 Depression as experienc ed among heart failure patients 31 Functional Limitations 34 Symptom severity and symptom burden 34 Self-care management 37 Patient Education 41 Effect of congestive heart fa ilure on hospitalization 44 Role of BNP in congestive heart failure hospitalization 44 Knowledge of BNP levels predicting treatment 50 Summary 54 Chapter Three Methods Introduction 56 Design 56 Setting 57 Sample 59 Instruments 59 Independent Variables 60 BNP Rapid Assay 60 Demographic Data Form 68 Dependent Variables 69 Minnesota Living with Heart Failure Questionnaire 69 Hospital length of stay 73 Medical Information Form 73 Procedure 74 Approvals 74 Recruitment 75 Random Assignment 76 Baseline Data Collection 77 Follow-up 80 Data Management 81 Data Analysis 81
iii Chapter Four Results Procedure 83 Sample 84 Research Hypothesis Number One 94 Research Hypothesis Number Two 95 Chapter Five Discussion, C onclusions and Recommendations Introduction 98 Discussion 98 Limitations 103 Implications for Nursing 105 Recommendations for Further Study 106 References 108 Appendices Appendix A: Demographic Information Form 128 Appendix B: Minnesota Living with Heart Failure Questionnaire 130 Appendix C: Medical Information Form 131 Appendix D: Letter of Support from Medical Director 132 Appendix E: Informed Consent Form 133 Appendix F: Frequently Asks Questions 141 About the Author End Page
iv List of Tables Table 1 Stages of heart failure 6 Table 2 Coefficient of variation meas ures for BNP test (Total CV%) 65 Table 3 Summary of studies utilizing the BNP Triage Immunoassay Kit 68 Table 4 Summary of studies sho wing reliability of MLWHFQ 72 Table 5 Summary of Reliabilities of Instruments used in the study 74 Table 6 Baseline demographics and clinical characteristics of patients by group 89 Table 7 Baseline descriptives and relia bilities of QOL scores by group 93 Table 8 Bivariate correlational analysis of baseline variables 94 Table 9 Comparison of QOL scor e means by group at 90 days 95 Table 10 Hospital length of stay by group assignment 96 Table 11 Total number and percent of deaths per group assignment 97
v List of Figures Figure 1 Conceptual Model illustrati ng the effect of heart failure on QOL 12 Figure 2 Logic Model: Relationship between disclosed BNP levels to Hospital LOS and Quality of Life in patients treated for heart failure 57 Figure 3 The flow of participants through the trial 86
Preface First and foremost, I would like to thank my soul mate, my second heaven, and my rock, Roy, for providing me the support and encouragement to complete the academic journey I started. His unconditional sacrifice has helped me through the greatest challenges in life. I would also like to recognize the steadfast support and guidance of my major professor and committee members. Special thanks to Dr. Yougui Wu for his expert statistical thoughtprovoking direction. Many thanks to Dr. Heather Stockwell for her stimulating guidance. My most sincere and deepest gratitude to Dr. Mary Evans for always being there for me especially in my trying moments. To Dr. Patricia Burns, for believing in me throughout the pursuit of my academic dreams, I thank you most sincerely. And to Dr. Theresa Beckie, my mentor and teacher, for sharing your knowledge and expertise and allowing me to grow as a person and as an academic colleague. No words are enough to adequately express my appreciation. I thank you from the bottom of my heart. You are an adviser and mentor beyond compare. In addition, my academic success was made possible because of the continued encouragement and support of many other people in my life. I would like to make a special acknowledgement to my children, Jean Murillo, Sofia Banda, Christine Ancheta and Ernesto Ancheta Jr., for giving their mom an opportunity to better herself. My sincere thank you to my mother, Sofia Blanco, who has given me the passion to achieve. To my sister, my confidant, Ramona vi
Wilson, who has given me her unwavering faith to continue on with my dream. To Sarah Cobb, thank you for providing the right words to say in order to convey the meaning across to my readers. To Dr. Jun Chiong, and Dr Alan B. Miller for the unrelentless support throughout my research studies. To Dr. Imelda Nwoga, Dr. June Chandler and Ding Sabio, I would like to say thank you. To my heart failure patients, to those who are living and to those who are deceased, I dedicate this work for you. Also, I am truly grateful for those people who have invested in me by way of scholarships. I hope to continue to pay it forward. To all, I would like you to know how much you all have affected my life in ways beyond words. Finally, and most importantly, I dedicate the glory of this work to JESUS, my GOD. He has provided me the strength, passion and dedication to persevere so I may affect the lives of others in his name. vii
viii The Relationship Between B-type Natriuretic Peptide (BNP) Levels and Hospital Length of Stay and Quality of Life in Congestive Heart Failure Patients Irma B. Ancheta ABSTRACT Previous research on quality of life (QOL) and its relation to BNP levels in heart failure (HF) has been widely studied. Howeve r, the impact of physicians knowledge of BNP levels at time of clinic visit on QOL and hospital length of stay (LOS) has yet to be fully investigated. The purpose of this study were to determine if physicians knowledge of BNP levels affected a c hange in QOL scores at 90 days and reduce hospital length of stay among heart failure patient s. QOL data from HF clinic patients (N= 108, 67.5 .3, 56% male, ejection fraction 26.5 8.2) were analyzed. QOL was measured at time of clinic visit (T 1) and at 90 days (T2) using the Minnesota Living with Heart Failure Questionnaire (ML HFQ). An independent t-test was utilized to compare the two groups. Findings: Both groups were comparable regarding demographic and baseline characteristics. There was no significant association observed between the experimental and cont rol group at 90 days, although the data indicated a decrease in the mean QOL scores at 90 days (37.46 28.67) as compared to the mean QOL scores at base line (46.87 29.63) for both groups.
ix Because the QOL scale is reversed, this indi cated that there was a positive change in QOL scores during the 90 day time interval. Hospital LOS was similar for both groups (mean=3 days). BNP levels were significantly correlated with both baseline QOL scores ( r =.25, p =.01) and physical subscale scores ( r =.24, p =.01). Mortality was higher in the control when compared to the experimental group ( t =1.99, df =90, p =.04). Conclusion: While physicians awar eness of BNP levels had not shown a significant change in QOL at 90 days, pati ents QOL might already have been quite positive. Chronic HF patients may have adapted to their disease and have adjusted their perception of their QOL. Therefore, QOL may be a stable construct at this time. Findings may have been different on newly diagnosed HF patients since they may not have adapted to their health condition.
1 Chapter 1 Introduction This chapter begins by defining congesti ve heart failure (CHF), discussing the etiology and pathophysiology of heart failure (HF), including the signs and symptoms, stages, and classifications to facilitate understanding of the disease process. This is followed by the presentation of the epidemiology of HF including the prevalence, incidence, risk factors, co-morbid conditions, mo rtality rates, and economic impact on the healthcare delivery system in the United States. The effect of CHF on hospitalization, effect of CHF on quality of life (QOL) and the role of b-type natriuretic peptide (BNP) in HF will be explained. The statement of the problem, purpose of the study, research hypotheses, and definition of terms will be described. The definition of BNP and BNP-related studies will be discussed. Finally, the underlying assumptions, delimitations, limitations, and significance of the st udy will be presented. Definitions of Congestive Heart Failure The Heart Failure Society of America (HFSA) 2006 Comprehensive Heart Failure Practice Guidelines (Adams et al., 2006) defines congestive heart failure as: A syndrome caused by cardiac dysf unction, generally resulting from myocardial muscle dysfunction or loss and characterized by left ventricular dilation or hypertrophy. Whether the dysfunction is primarily systolic or diastolic or mixed, it leads to neurohormonal and circulatory abnormalities, usually resulting in characteristic symptoms such as fluid retention,
2 shortness of breath, and fatigue, espec ially on exertion. In the absence of appropriate therapeutic intervention, HF is usually progressive at the levels of cardiac function and clinical symptoms. The severity of clinical symptoms may vary substantially duri ng the course of the disease process and may not correlate with changes in underlying cardiac function. Although HF is progressive and often fatal, patients can be stabilized, and myocardial dysfunction and remodeling may improve, either spontaneously or as a consequence of therapy. In physiologic terms, HF is a syndrome characterized by elev ated cardiac filling pressure and/or inadequate peripheral oxygen delivery, at rest or during stress, caused by cardiac dysfunction. (p. 14) The American College of Cardiology/American Heart Association (ACC/AHA) (2005) provides a scientific statement defining heart fa ilure as A complex clinical syndr ome that can result fr om any structural or functional cardiac disorder that impairs the ability of the v entricle to fill with or eject blood. The cardinal manife stations of HF are dyspnea and fatigue, which may limit exercise tolerance, and fluid retention, which may lead to pulmonary congestion and peripheral edema. (p. e160) Etiology According to the ACC/AHA guidelines (2005), the underlying causes of HF in a substantial portion of patients are coronary artery disease, hypertension, and dilated cardiomyopathy. One-third of HF patients have nonischemic
3 cardiomyopathy, which may be caused by hy pertension, thyroid disease, valvular disease, alcohol use, or myocarditis (ACC/AHA, 2005). Pathophysiology of Heart Failure In HF, there is decreased cardiac output and increased pulmonary pressures consequently leading to pul monary congestion. Because of this process, the body activates both neurohormonal pathways in order to compensate for low cardiac output. There is increased heart rate and contractility as initiated by the sympathetic nerv ous system. Catecholamines such as epinephrine and noepinephrine ar e circulated to cause artrial vasoconstriction and stimulate secretion of rennin from the kidney (ACC/AHA, 2005). These circulating catecholamines may exacer bate ischemia, cause arrhythmias or promote cardiac remodeling. Increased production or stimulation of the reninangiotensin system (RAAS) result s in arterial vasoconstriction, sodium and water retention, and release of aldosterone, leading to sodium and water retention (ACC/AHA, 2005). On the other hand, nitric oxide an d natriuretic peptides are hormones released by secretory granules in ca rdiac myocytes. Both are produced to counteract the effects of the vasoconstriction caused by catecholamine and rennin-angiotensin producti on. Nitric oxide and natriur etic peptides, to include btype natriuretic peptide (BNP), promote systemic and pulmonary vasodilation and increase sodium and water excretion. Continuous neurohormonal stimulation causes the left ventricle to undergo remodeling consisting of left ventricular dilatation, myocyte hypertrophy, and elongation. Enhan ced neurohormonal
4 stimulation can lead to apoptosis, aggravat ion of ventricular contractility, and death (ACC/AHA, 2005). Chronic HF re fers to the clinical syndrome characterized by signs and symptoms of increased tissue/organ water and decreased tissue/organ perfusion (Zile, 2005). HF is a common outcome for many cardiovascular diseases that result s in symptomatic or asymptomatic left ventricular dysfunction (LVD). HF is a vi cious cycle if left unt reated. Dysfunction begets additional dysfunction t hat culminates in the demise of the patient (Ramakrishnan, et al., 2005). Signs and symptoms of heart failure The ACC/AHA guidelines refer to t he cardinal manifestations of heart failure that include dyspnea, fatigue, whic h may limit exercise tolerance, fluid retention, which may lead to pulmonar y congestion and peripheral edema (Hunt et al., 2005). Signs and symptoms of congestive heart failure depend upon the side of the heart affected. Fatigue, ort hopnea, wheezing or hacking cough, and shortness of breath during mild exertion are symptoms of left-sided failure. Pulmonary edema occurs when too much fluid accumulates in the lungs. Since right-sided congestive heart failure reduces the amount of blood returning to the heart, the main sym ptoms are swelling in the feet, ankles, legs, and abdomen because the tissues throughout t he body fill up with excess fluid. Patients with systolic heart failure experi ence feeling tired more often and have decreased appetite and increased wei ght gain (HFSA, 2006).
5 Stages of Heart Failure The staging of HF was devised to establish the evol ution and progression of this disease. This HF classifica tion was intended to complement but not replace the NYHA functional classi fication (ACC/AHA, 2005; HFSA, 2002). These levels or stages can only advance; the patient always goes forward not backward. This classification focuses on patients with HF as well as those who are at risk of developing HF (ACC/AHA, 2005). The following are the stages of HF as presented by the ACC/AHA guideli nes: Stage A refers to the patient who is at high risk for developing HF but has no structural disorder of the heart (such as, hypertension or coronary artery dis ease); Stage B refers to an asymptomatic patient with a structural disorder of t he heart (such as, LVD either dilation or hypertrophy); Stage C refers to the patient having underlying structural heart disease with past or current symptoms of HF; and Stage D refers to the patient with end-stage disease who requ ires specialized treatment strategies (Hunt, et al., 2001; Ramakrishnan, et al., 2005). Table 1 illu strates the stages of HF.
Table 1: Stages of heart failure American College of Cardiology/American Heart Association Classification of Chronic Heart Failure Stage Description A High risk for developing heart failure Hypertension, diabetes mellitus, CAD, family history of cardiomyopathy B Asymptomatic heart failure Previous MI, LV dysfunction, valvular heart disease C Symptomatic heart failure Structural heart disease, dyspnea and fatigue, impaired exercise tolerance D Refractory endstage heart failure Marked symptoms at rest despite maximal medical therapy Classification of Heart Failure The New York Heart Association Classification (NYHA) criteria are utilized to assess the functional capacity of HF patients. The staging of heart failure has not replaced but complemented the functional classification of HF as categorized by the NYHA (ACC/AHA, 2005). Listed below is the NYHA classification according to the HFSA (2006). Class 1 refers to no limitation of physical activity. Ordinary physical activity does not cause undue fatigue, palpitation, and dyspnea. Class 2 refers to the presence of slight limitation of physical activity. Comfortable at rest, but ordinary physical activity results in fatigue, palpitation, and dyspnea. Class 3 refers to the marked limitation of physical activity: 6
7 comfortable at rest but less than ordinary physical ac tivity causes fatigue, palpitations, and dyspnea. Class 4 refers to the inability to carry out physical activity without discomfort, and symptoms of cardiac insufficien cy at rest. If any physical activity is undertaken, disco mfort is increased (HFSA, 2006). Epidemiology Prevalence As an epidemic disease and a major cause of chronic disability, CHF adversely affects the health of millions. Nearly 5 million Americans are living with HF (ACC/AHA, 2005). CHF is the primary reason for 12 to 15 million office visits and 6.5 million hospital days each year. Prevalence of HF in 2002 indicated that there were 4,900,000 of the total population with HF. The total number of males with HF was 2,400,000 (2.6%) and tota l number of females was 2,500,000 (2.1%) (AHA, 2005). CHF increases with age: 7% of people 65-74 years of age and 10% of people over 75 are affected by HF (CDC, 2005). Incidence The incidence of HF showed 550,000 new cases annually or 10 per 1000 after age 65. An estimated 80% of patient s hospitalized with HF are more than 65 years old (AHA, 2005). HF is a co mmon Medicare diagnosis-related group with more Medicare dollars spent for the diagnosis, treatment, and management when compared to other diagnoses (ACC/ AHA, 2005). Seventy-five percent (75%) of HF cases had a preexisting conditi on for hypertension. It is estimated that 22% of male and 46% of female patients who experience a myocardial infarction (MI) will most likely be diagnosed with HF within 6 years. According to
8 the National Heart Failure Data from the Acute Decompensated Heart Failure National Registry (ADHERE), the mean age of hospitalized HF patients was 71.2 years old, and 52% of them were female (AHA, 2005). Persons diagnosed with high blood pressure greater than 160/90 mm Hg have a twofold risk of having HF as compared to those with blood pressure less than 140/90 mm Hg (AHA, 2005). The annual rates of new HF cases per 1000 of non-black men ages 65-74 are 21.5; ages 75-84 are 43.3; ages 85 and abov e are 73.1. The annual cases for non-black women ages 65-74 are 11.2; ages 75-84 are 26.3; and ages over 85 are 6.9. The annual rates of new HF cases for black men ages 65-74 are 21.1; ages 74-84 are 52; and over age 85 are 66. 7. The prevalence for HF in black women is at 18.9, 33.5 and 48.4, respectively (AHA, 2005). Risk Factors The presence of multiple risk factors makes HF a more complicated diagnosis to manage and predisposes patie nts to frequent hospital admissions. Risk factors include smoking, diabetes, high blood pressure, high cholesterol levels, low high density lipoprotein (HDL) levels, obesity, and sedentary lifestyle (HFSA, 2006). Coronary artery disease, my ocardial infarction (MI), hypertension, abnormal or damaged heart valves, damage to the heart muscle caused by alcohol or drug use, overexposure to radiation or viruses, heart defects from birth, severe lung disease, diabetes, seve re anemia, overactive thyroid gland or cardiac dysrthymias are all considered causes of heart failure (AHA, 2005). Some reasons for HF are unknown. The most common risk factor for HF is hypertension. Uncontrolled hypertension increases the risk of heart failure by
9 200%, and people diagnosed with diabetes have a twoto eight-fold greater risk for developing HF (AHA, 2005). Mortality An estimated 53,0000 patients have died of HF as a primary cause, and the mortality of HF has increased despite advances in treatment. in part due to better treatment and saving patients with MI earlier in life (ACC/AHA, 2005). It is estimated that 1 in 5 HF patients will die within one year of diagnosis (AHA, 2005). Worst of all, HF patients suffer sudden cardiac death six to nine times more frequently than the general population (AHA, 200 5). According to the AHRERE data, in-hospital mortality was 3. 8% (AHA, 2005). From 1992 to 2002, deaths from HF in creased by 35.3%, and death rate increased to 7.7%. In 2000, there were 55,704 deaths occurring from HF; this mortality rate was a 148% increase fr om 1979 (National Center for Health Statistics, 2004). The age-adjusted death rate wa s 20.2 (deaths per 100,000 population) in 2000. Death rates were: for white males (21.0), black males (23.2), white females (19.4) and black females (21.3) (National Center for Health Statistics, 2004). The overall death rate for HF in 2001 was at 18.7 %, with 19.6% for white males, 21. 7% for black males, 18.1 % for white females and 18.8% for black females (AHA, 2005). Economic Impact In 1999, $3.6 billion was paid to Medicare beneficiaries diagnosed with HF (CMS, 2003). For 2005, an estimated direct and indirect cost of HF was $27.9 million (AHA, 2005). The management of HF costs $56 billion a year, with 70%
10 due to hospitalization (Bhalla et al., 2004). Thus, HF has an enormous, escalating financial impact on healthcareexpenditures (AHA, 2005; CDC, 2005). Effect of Congestive Heart Failure on Hospitalization CHF is the most frequent cause for hospital readmissions, higher within the first 30 days after discharge than withi n 60 to 90 days (AHA, 2005). Hospital discharges increased by 152 percent from 377,000 in 1979 to 970,000 in 2002. According to the ADHERE registry, t he mean hospital length of stay for HF patients was 5.8 days (medi an 4.3 days) (AHA, 2005). Several studies indicated that CHF is the primary diagnosis in most hospital admissions. In the Medicare study, chronic obstructive pulmonary disease, renal failure, and diabetes were medical conditions listed as other noncardiac comorbidities prevalent in older HF patients. Results of the Medicare study revealed that 40% of patients wit h HF had five or more noncardiac comorbidities; they accounted for 81% of the total inpatient hospital days. The presence of increased number of comorb idities in this patient population increased the risk of hospitalization (Br aunstein et al., 2003). National databases on HF show that several studies have been conducted to reduce CHF hospitalizations. These studies include clinical trials on HF pharmacological management, HF device treatment, and non-pharmacological interventions to improve patient compliance and self-care behaviours (AHA, 2005). Effect of Congestive Heart Failure on Quality of Life The treatment of HF is complicated and frequently focused on improvement of quality of life (QOL) rat her than on the recovery of the patient.
11 Patients with CHF often experience severe symptoms and deterioration of QOL (Moser, 2002; Rector, 2005; Luttik et al., 2005; Lee et al., 2005). To date, there remains confusion about the definition and measurement of QOL. For purposes of this study, QOL is defined as a multidimensional subjective description of the psychological, physical, and social domains of health as measured by the Living with Heart Failure Questionnaire (MLHFQ) (Rector, 2005). The conceptual model by Rector et al. (2006) is used to guide this study. The model illustrates that the effects of HF on QOL are attributed to symptoms caused by HF. In the Valsartan Heart Fa ilure Trial (VAL-HEFT) study, patients were asked to repeatedly assess seve ral symptoms of HF and complete the MLHWF questionnaire. The purpose of the conceptual model was to assess patients perceptions of how heart failure affected their QOL. Other pathophysiologic measures were also used to assess the severity and secondary effects of HF. Results from both of these measurements were statistically analyzed to determine whether relationshi ps exist among symptom assessments, pathologic measures, and MLHF scores. Using this secondary analysis of data from the VAL-HEFT study, the authors concluded that a si gnificant proportion of the effects of HF on QOL is explained by the presence of symptoms of HF as measured by the MLWHF score and that the effects of QOL varies with age regardless of symptoms. Results of the study indicated t hat effects of HF on QOL with HF symptoms depend on HF pathol ogy and that sym ptoms are the mediating factor on the effect s on QOL (Rector et al., 2006).
Figure 1 illustrates the model as described by Rector (2006). Figure 1: Conceptual model illustrating the effect of heart failure on QOL This conceptual model shows the effects of HF pathology on QOL, with HF symptoms as the mediating factor (Rector et al., 2006). A domino effect is initiated from experiencing the physical symptoms leading to functional limitations suffered by HF patients. As HF advances and progresses, the disease prevents patients from living as they would have wanted. (Rector et al., 2006). Ordinary daily activities and recreational hobbies become difficult to perform without getting short of breath, lacking energy, or becoming easily fatigued. Patients with HF give up their independence since they must depend on others for menial tasks. This situation causes a variety of feelings such as frustration, hopelessness, depression, being a burden to the family, loss of self-control, and lowered self-esteem (Rector et al., 2006; Rector, 2005; Brostrom et al., 2004; Lee et al., 2005; Moser, 2002). Literature reviewed on QOL indicates infrequency of treatment for psychological factors despite their role in the outcomes of heart failure, its association to QOL and to hospital admissions. It has been suggested that perception of life situation, psychosocial adjustment to illness, and functional 12
13 limitations are mult i-dimensional areas that affect the QOL among heart failure patients. The perception of a life situation included ho w HF patients feel about their disease condition (Costelo & Bobl in, 2004). Psychological adjustment to illness describes how depression, anxiety, so cial support, meaning, coping style, and spiritual beliefs affect QOL. The ph ysical dimension includes functional limitations and HF symptomatology, self-care management strategies, and patient education in coping with a lifelong chronic condition. Amelioration of physical symptoms may impr ove functional status, wh ich may improve QOL. An increased awareness of HF patients life situations may lead to patients adapting to their CHF. Increased knowledge and ability to perform self-care may keep patients out of a vicious cycle of limitation and resignation. QOL, therefore, should be targeted as a relevant out come measure when dealing with HF patients (Brostrom et al., 2004; Lee et al., 2005; Moser, 2002). Role of B-Type Natriuretic Pe ptide (BNP) in Heart Failure CHF is a complex progression of dom ino effects involving cardiac and neurohormonal systems. BNP is a 32-amino acid hormone that was first found in the porcine brain. Subsequent studies found that b-type natriuretic peptide (BNP) level is a neurohormone produced by the left ventricle in response to fluid overload and released to the bodys systemic circulation (Masson et al. 2006; Brenden et al. 2006; HFSA,2006). BNP is a biomarker of HF, and early detection of increased BNP levels may lead to early diagnosis and treat ment of CHF, thereby decreasing readmissions and improving pat ients quality of life (QOL) (Heidenrich et al.,
14 2004; Maisel et al., 2001; MacMahon et al ., 2002). BNP is released in a pulsatile manner, approximately every 30 to 90 minut es in both healthy individuals and those with HF. However, in the presence of volume overload, there is a rapid elevation of BNP levels as evidenced by an increase in its pulsatile release (White, 2005). BNP inhibits sodium reabsorption in the distal tubules, increases globular filtration, and is involved in t he regulation of diuresis. BNP antagonizes the vasoconstricting effects of the renin angiotensin-aldosterone system (RAAS), thereby regulating blood pressure and fl uid balance (Chiong & Miller, 2002). BNP levels are elevated with cardiac overload and increased ventricular volume; they are sensitive to increased ventricular stre tch (Mark & Felker, 2004). Ventricular volume expansion and fluid overload are evident in the early phases of CHF, thus becoming a marker for heart failure (J iang et al., 2001; Maisel et al., 2002; Cheng et al., 2001; Ishii et al., 2003; Tabbizar et al., 2002; Anand et al., 2002) Activation of the neurohormonal system leads to progressive myocardial dysfunction and heart failure (Eichhorn & Bristow, 2001; Venugopal, 2001; Chiong & Miller, 2002). In the presenc e of volume overload, the cardiac ventricles stimulate BNP producti on and electrophysiologic arrthymias, suggesting the relationship between BN P and sudden death. Increased BNP levels are therefore a st rong predictor of sudden deat h (Berger et al, 2002). Currently, BNP levels are not routinely tested during clin ic visits (CMS, 2003), nor are they part of current CHF guidelines (HFSA, 2006; Howie, Caldwell & Dracup, 2003).
15 Studies of BNP in Heart Failure The following research studies have shown how BNP levels are an indicator and a useful diagnostic tool for ear ly stages of HF (Morrison et al. 2002, Ninuma et al., 1998; Wieczorek et al ., 2002; Lubarsky & Mandell, 2004; Hirata et al., 2001; Teboul et al., 2004; Sagnella, 1998; Vanderheyden et al., 2004; Heidenreich et al., 2004; Mair et al., 1999). Taniguchi et al. (2006) conducted a study investigating the relationships between BNP and (QRS) duration to determine the prognostic value in HF pat ients. QRS duration and BNP levels were measured after patients (n = 93) were treated in the emergency department. Results showed that sudden deat h (6 patients, 348 128 pg per ml) for progressive heart failure (9 patients, 390 97 pg per ml), and readmission for worsening heart failure in (20 patients, 354 79 pg per ml) occurred in 35 patients. The authors concluded that high levels of BNP and prolonged QRS duration was associated with poor prognosis regardless of any type of cardiac events. The study also suggested that a combination of both BNP levels and QRS duration may be useful in predict ing the prognosis of HF patients. Masson et al. (2006) investigated the prognostic value of BNP and amino terminal probrain natriuretic peptide (NT-proBNP) levels in stable, chronic HF patients. Baseline BNP and NT-proBNP le vels were drawn from 3,916 patients enrolled in the Val-HEFT study. Findings reported t hat receiver-operator characteristic curves for all-cause mort ality (area under the curve (SD) was BNP 0.665 (0.011) vs. NT-proBNP 0.679 (0.011), p = 0.0734). Sensitivity and specificity ranged from 0.590 to 0.696. The authors concluded that both BNP and
16 NT-proBNP showed slight differences in t heir relation to clinical characteristics and prognostic performance as a diagnostic tool in a large HF population and were the most powerful independent markers of outcome in HF. Cardarelli and Lumicao (2003) conduct ed an extensive literature review on the prognostic and therapeutic monitoring value of BNP levels. They concluded that symptomatic patients without a history of CHF had BNP levels proportional to the severity and survival in CHF pati ents (BNP 80 pg per ml, sensitivities 93%98%, PV92%-98%). Those with BNP levels over 256.9 pg per ml deteriorated within the ensuing 12 months as compar ed to those with a BN P level of 42.4 8.6 pg per ml who remained improved in t heir functional class. Another study by Berger et al. (2002) indicated that wit h cut-off point of BNP 130 pg per ml Kaplan-Meier survival rates were significantly higher in those patients with lower BNP levels than in those with values higher than the cut-off score (n = 452, p = .0001). Wieczorek et al. (2002) investigat ed the performance of the BNP rapid assay as a diagnostic tool in CHF, ev aluating it in inpatient, outpatient, and healthy control subjects (n = 1050). Participants were classified into categories of those without CHF (n = 473), those with hypertension but no cardiovascular disease (n = 168), NYHA Class I (n = 73); Class II (n = 135); Cla ss III (n = 141); and Class IV (n = 60). Using receiver oper ator characteristic (ROC) curves, results indicated that with a cut-off of 100 pg per ml, the assay showed a sensitivity of 82% and a specificity of 99% validating the usefulness of BNP in the diagnosis of CHF and staging the severity of HF.
17 The seven-site, international Breathing Not Properly Mu ltinational Study (Maisel, 2002) examined if BNP was useful in predicting CHF patients with acute dyspnea in 1586 emergency dep artment (ED) patients. Two independent cardiologists were blinded to patients BNP measurements. A receiver operator characteristic was used to illustrate vari ous BNP levels. With a cut-off value of 100 pg per ml, diagnostic accuracy of 83.4% predictive negative value (PV-) of 96% at 50 pg per ml, findings were: BN P alone was more accurate than any historical or laboratory values in predict ing CHF as a cause of dyspnea. At least a non-systolic CHF patients showed signif icantly lower BNP levels than those with systolic heart failure (413 pg per ml vs. 821 pg per ml, p <0.001 for each pairwise comparison). The study conclud ed that, used in conjunction with other clinical assessment, BNP is useful in es tablishing or excluding the diagnosis of CHF. Morrison et al. (2002) enrolle d 321 ED patients with dyspnea and determined if BNP levels could differentia te cardiac from pulmonary causes of dyspnea. Two physicians blinded to the BN P levels were asked to give their opinions on the probability of the patient having HF and their final diagnosis. The area under the ROC, which plots sensitivity and specificity of BNP levels differentiating cardiac from pulmonary, was .96 (p< 0.001). Resu lts revealed that CHF patients (n = 137) had high BNP levels (758 to 798 pg per ml) as compared to those with pulmonary diseases (n = 85, 61 to 10 pg per ml). Hirata et al. (2001) exami ned the utility of BNP levels for early diagnosis of CHF and severity of the disease process in daily clinical practice. For 415 heart-
18 disease patients and 65 control-group subj ects, comparison of BNP and atrial natriuretic peptide (ANP) levels usi ng nonparametric Tukey type multiple comparison showed that BNP was higher than ANP (.864 vs. .787, p = 0.06). The area under the ROC curve (AUC) was used to evaluate the usefulness of both levels. With cut-off values of 15 pg per ml, BNP sensitivity was 74% and specificity of 83% in patients with cardiac disease. BNP levels correlate well with LVD, pulmonary artery wedge pressures, left ventricular hypertrophy, and systolic/diastolic dysfunction; levels higher than 100 pg per ml are highly suggestive of heart failure (ACC/AHA, 2001). BNP levels not only accurately confirm the diagnosis of CHF but differentiate HF from other diseases (McCullough et al., 2003; Maisel et al., 2002). A study conducted by Ninuma et al. (1998) examined if at rial natriuretic peptide (ANP) and BNP are effective me thods of predicting heart disease irrespective of LVD. Examining 481 pat ients, the study conc luded that BNP was effective for screening asymptomatic pat ients with left ventricular dysfunction (BNP<13 pg per ml gave a predictive negative value of 100%). The area under the receiver operating curve (ROC) fo r BNP was signific antly greater when compared to ANP (0.94 vs. 081; p = .001). Knowing the role the neurohormones play in the pathophysiology of HF makes for a better understanding and appreciation of BNP levels as a valuable test in the diagnosis of heart failu re. Statement of the Problem Few studies have examined the relationship between clinicians knowledge of BNP levels and hospital length of stay (LOS) and quality of life in
19 heart failure (HF) patients. Troughton et al. (2000) concluded that current treatment strategies in the clinic ignore plasma neurohormone concentrations (BNP), even though they are independent markers of cardiac status and prognosis of heart disease in cluding heart failure. Recent CHF guidelines do not target any hemodynamic criter ia such as BNP levels prior to hospital discharge since most efforts have been focused on t he use of pharmacological therapy and CHF management clinics (Tr oughton et al., 2000). Therefore, the goal of this study is to compare two CHF clinic groups: one with and one wit hout clinicians knowledge of BNP levels and examine the relationship of physi cian knowledge to hospital LOS and QOL. Purpose of the Study The purpose of this study is  to determine if clinic ians knowledge of BNP levels would make any difference in the QOL scores between the experimental and control groups at 90 da ys and  to determine if physicians knowledge or lack of knowledge of BNP levels at time of CHF clinic visit affect hospital LOS on all hospital admissions regardless of how many hospital admissions occur in 90 days. Research Hypotheses The effects of the study will be as sessed by testing the following hypotheses: Hypothesis 1: It is hypothesized that c linicians knowledge or lack of knowledge of BNP levels at time of clinic visit ma kes a difference in the quality of life scores
20 between the experimental group and the control group at 90 days. An independent t-test between experiment al and control groups was used to compare their mean QOL scores at 90 days. Hypothesis 2: It was hypothesized that clinicians knowledge or lack of knowledge of BNP levels at time of CHF clinic visit would affect hospital LOS on all hospital admissions of CHF patients within 90 days. A comparison of means for both experimental and control groups was used to examine the relationship between BNP levels and hospital LOS withi n 90 days. Definition of Terms For the purpose of the study, the fo llowing terms are identified:  Congestive heart failure: a complex clinic al syndrome that can result from any structural or functional cardiac disorder t hat impairs the ability of the ventricle to fill with or eject blood (ACC/AHA, 2005). The cardinal manifestations of HF are dyspnea and fatigue, which may limit exer cise tolerance, and fluid retention, which may lead to pulmonary congesti on and peripheral edema (ACC/ AHA, 2005).  Systolic dysfunction: a defect in the ability of the cardiac muscles to shorten against a volume load. The ventricle loses its ability to eject blood into the aorta, and the LV systolic properties become abnormal (Zile et al., 2005).  Diastolic dysfunction: the inability of the cardiac muscle to rapidly or completely return to a resting state. At this point, the ventricle cannot accept blood at low pressures, and ventricular filling is slow or incomplete unless atrial pressure increases (Zile et al., 2005).
21  B-type natriuretic peptide (BNP): a cardiac neurohormone secreted from the cardiac ventricles as a response to v entricular volume expansion and fluid overload, evident in the early phases of congestive heart failure (Maisel, 2001; Mark & Felker, 2004).  Hospital length of stay: number of da ys a patient stays in the hospital from time and date of admission to time and date of discharge. Portions of a day were considered as one day of hospital stay.  Quality of life: a mult idimensional subjective descr iption of the psychological, physical, and social domains of health as measured by the Living with Heart Failure Questionnaire (MLHFQ) (Rector, 2005). Underlying Assumptions The proposed study has some assumptions that are specific to the population of interest. The first assumption is that there is a correlation between BNP and HF. The second assumption is that early detection and knowledge of BNP levels may enable physicians to pr ovide more aggressive treatment and titration of medications. T he third assumption is that increased levels of BNP are directly related to poor QOL scores am ong heart failure patients. The fourth assumption is that timely detection and knowledge of BNP levels decreases hospital LOS. Delimitations The sample included patients diagnosed with congestive heart failure. They had a wide range of ethnicity; were ov er 21 years of age; ab le to read, write and speak English; and hadserum creatinine levels not greater than 2.5.
22 Limitations The limitations of the study inclu ded those that are generally experienced with quantitative study. First of all, the study population is focused on one clinic center, thereby limiting any generalizat ion of findings to other geographical areas. Secondly, the partici pants of the study may have already been exposed to the same questionnaire in the past. Thir dly, there is a po ssibility of having measurement error in the study questionna ire. Lastly, there is a possibility of having random error in the BNP rapid assay machine. Significance of the Study Congestive heart failure is becoming one of the most chronic, debilitating, and progressive diseases in the Unit ed States. Numerous factors have been implicated in the disease, including cor onary artery disease, hypertension, high cholesterol levels, diabetes, smoking, and diet and lifestyle behaviors. Despite a high level of public awareness of this dis ease, the majority of the population are unaware of their risks of having HF. By contributing to the body of knowledge concerning HF management, this study is intended to increase the understanding of the various factors that affect the success of managing HF patients. By measuring BNP levels, researchers will k now how important the link is between BNP levels and QOL. By examining the rela tionship of B-type natriuretic peptide levels on CHF hospital LOS and QOL, this study may provide suggestions for future interventions that would allow patients to properly and timely treat the syndrome of HF.
23 The research study may result in formulating and developing protocols that could provide heart failure pa tients with self-management strategies, allowing them more autonomy despite t heir health conditions, decreased hospital LOS, and enhanced QOL. The study furthers the science of nursing in that it seeks to investigate how clinicians knowle dge of BNP levels affect hospital LOS and QOL among patients with HF.
24 Chapter 2 Review of Literature Chapter 2 describes an overview of congestive heart failure (CHF) as it relates to the study. The first area of t he literature review de scribes quality of life (QOL) as experienced by hear t failure patients. The sec ond area of the literature review describes the effect of CHF on hospitalization. The third area discusses the role of b-type natriuretic peptide (BNP) in CHF hospitalization. Finally, physicians knowledge of BNP levels pr edicting heart failure (HF) treatment will be discussed. Introduction Congestive heart failure (CHF) is the end stage of heart disease (AHA, 2005). It is a common diagnosis among elderly people who have multiple risk factors and co-existing illnesses and are on multiple medications. It is a chronic, disabling disease that c auses patients to become dependent on others for their daily needs. Heart failure patients ex perience symptoms characterized by shortness of breath, edema, easy fati gability, and decreased physical endurance; these symptoms deter pat ients from activities that they enjoy. The prognosis is poor; patients experience det erioration of quality of life and frequent and regular hospital readmissions. The following is literature reviewed on QOL as experienced by patients living with heart failure, the effect of CHF on hospitalization, the role of BNP in CHF hospitalization, and physicians knowledge of BNP levels predicting treatment.
25 Quality of Life as Experienced by Heart Failure Patients Quality of life (QOL) among HF patients is a subjective, multi-dimensional, concept that includes the ph ysical, psychosocial, emotional, and spiritual aspects of life that change with time. It includes [a] the perception of life situation; [b] psychosocial adjustment to illness; [c] functi onal limitations of HF ; [d] HF self-care management programs; and [e] patient education in HF (Brostrom et al., 2004; Sneed et al., 2001; Lee et al., 2005; Boswor th et al., 2004). Research correlating CHF and QOL indicate that the psychological factors are as important as the physiological factors (Moser, 2002; Riegel, 2006; Rector et al ., 1993; Bennett et al., 1997; Clark et al., 2003; Konstam et al., 1996). [A] Perception of Life Situation The empirical literatur e has found that several fa ctors affect how QOL may be influenced by ones perception of lif e situation: gender differences, HF patients and spouses percept ion of QOL, HF patients subjective perception of QOL, and objective evaluation of the seve rity of HF (Martensson et al., 2005; Costelo & Boblin, 2004; Luttik et al., 2005). Gender differences regarding perception of QOL in heart failure Several studies show conflicting re sults regarding gender differences in the perception of QOL in HF. In a qual itative study conducted by Costelo and Boblin (2004), open, semi-structured interv iews were used for clinic CHF patients (n = 6, 3 men and 3 women) with New Yo rk Heart Association (NYHA) Classes IIIIV, with ages ranging from 37 to 87 years old. The objective of the study was to identify whether gender differences influence QOL, treatment, and survival.
26 Data collection focused on the experi ence of the women and men with CHF; the two sources were those individuals with CHF and a family member for each participant. The study utilized a semi-str uctured one-on-one interview that lasted for one hour and occurred at the patients homes or in the CHF clinic. Openended research questions were used to explore participants responses. Results revealed a total of 13 themes: burden to ot hers, frustration, loss, acceptance, hope of the future, fatigue, maintaining independence, fear, physical symptoms, confusion due to lack of knowledge, isolation, depression, and shock and disbelief. Three themes were identified from the subjects responses. First, the psychosocial impact of CHF is greater than the physical impact. The author recommended that in addressing this iss ue, emphasis should be geared toward time for patients to verbalize their feeli ngs and more time provided for the nurse to promote holistic assessment and to develop a therapeutic rapport with the patient. Second, men experi enced more social isolation and loss, while women experienced fear. They recommended that healthcare providers should be aware of gender differences since men and women re spond differently in coping with or accepting their illness. Th ird, the depression experienced by patients with HF is influenced by age. The younger patients experience more physical limitations and depression when compared to older subjects. Nursing implications from this study included using depression scales as part of nursing assessment and thoroughly performing a comprehensive and holisti c evaluation of the patients.
27 On the other hand, Riegel et al. (2003) using secondary analysis from a previous study, examined 320 CHF m en and women (N = 640) with matched functional status, age, ejection fracti on, and marital status. Results showed minimal gender differences in QOL in pat ients with HF. Data from a convenience sample of nine experimental or quasiexperimental studies conducted in eight sites were used for the study. The ML WHF questionnaire was used to assess QOL between the treatment and control groups. The survey was administered at baseline and at 3 months. Results indicat ed that QOL was statistically significant and that QOL was minimally worse among women when compared to men (1-3 points) at baseline and at 3 months. Emotional dimensio ns of QOL were lower in women than in men at baseline ( p <0.03) but were small and statistically nonsignificant in 3 months. Therefor e, the study conc luded that gender differences in the perception of QO L are minimal in patients with HF. QOL as perceived among male heart failure patients In 1997, Martensson et al. conduct ed a qualitative study describing how male patients with heart failure perceive their life situations Once again, the phenomenographic approach was the method ut ilized for the study. Open semistructured questions were used for the interview. Twelve men diagnosed with CHF, with ages ranging from 48 to 80 year s old, were enrolled in the study. Interview questions targeted the biophysi cal, socio-cultural, emotional, intellectual, and spiritual-existential dimensions. The data analysis compared different statements with similarities and differences. Six themes were identified:  a belief in the future that gave CHF patients a feeling of expectancy or being
28 self-influential;  gaining awareness was conceived of as being able to adapt to the symptoms and make the best of the situation;  feeling support from the environment;  feeling limit ations was perceived as either social or physical limitation;  feeling a lack of energy was described as mental and physical inabilities of setting about doing things that needed to be done; and  feeling resignation was conceived of as indifference, in which death was the only thing expected, or as powerlessness to influenc e their life situation. Recommendations from the study included patient educ ation regarding CHF and its symptoms and focusing on self-care a nd other possibilities. QOL as perceived among wome n with heart failure Martensson et al. (1998) conduct ed a study that showed how women with HF have a different percept ion of their life situat ion. The phenomenographic approach was utilized as the research design, describing something from a second persons perspective (the patient s experience of something or how something appears to someone). Methods used were open, semi-structured interviews based on the five dimensions of the holistic theory of Savrimaki and Stenbock-Hult (1993). The five dimensions were: biophysical, socio-cultural, emotional, intellectual, and sp iritual-existential. Subjects were 12 CHF patients between 65 and 85 years old with various etiologies of heart failure. The study found that risk factors were diffe rent in women than in men. Five categories or themes were identified:  feeling content with ones past and present life;  a sense of support was conceived of as feeling abandoned or having a sense of devotion;  a sense of limitation was conceived
29 of as physical or social limitation;  feeling anxiety ranged from insecurity in relation to ones self or in relation to ones surrounding; and  powerlessness was perceived to be a feeling of worthle ssness or being a burden. In this study, women were likely to experience gu ilt, anxiety, and decreased self-worth. Recommendations for future studies in clude nursing interventions that would focus on self-care abilities, setting realis tic goals and expectations, providing a hopeful perspective, and empowering patient s to have self-control and improved self-esteem. QOL as perceived by both heart fail ure patients and their partners Ekman et al. (2002) described heal th related quality of life (HRQOL) and sense of coherence (SOC) in a group of elderly HF patients with moderate to severe heart failure in comparison to a healthy control group (n = 94). Methods used included matching HF patients to health y control subjects. The SF-36 health survey was used to assess QOL health status, and the Antonovsky Sense of Coherence Scale (SOC) measured overall orientation toward demanding life situations. Significant differences we re found between men and women. Male subjects diagnosed with HF scored higher in the physical dimension of the SF-36 (33 vs. 45, p = 0.0005) vs. healthy controls (57 vs. 77, p = 0.036). Findings revealed that old-age and severe heart failure were associated with lower levels of HRQOL scores as compared to healthy controls. The study claimed that a state of tension occurs when a stressor in life is present and that successful coping or management of stress can lead to better health. This means that the presence of an individuals social cultural, and historical contexts is
30 helpful in making stressors in life more manageable, comprehensible, and meaningful. Luttik et al. (2005) conducted an explorative study of QOL as perceived by HF patients and QOL as perceived by thei r partners. The study enrolled 38 heart failure couples, 31 male and 7 female pat ients. The Cantr il Ladder of Life instrument was used to assess the QOL. Results indicated that the mean QOL scores for QOL (present) for HF patients was 6.8 as compared to partners at 5.0 ( p <0.025). Mean QOL (past) scores for HF patients was 4.9 as compared to partners at 6.1 ( p <0.03). In this study, HF pati ents experienced a poor QOL both in the past and in the present as compared to their par tners. However, the QOL expectation scores 3 years in the future did not differ si gnificantly (6.7 vs. 6.4, p = 0.60). Subjective perception of QOL and obj ective evaluation of the severity of heart failure Grigoni et al. (2003) studied the distance between patients subjective perception of QOL and objecti ve evaluation of the seve rity of HF. The study investigated the relationship between QOL (what patients are most interested in) and objective parameters of CHF severi ty (largely physician care). QOL was evaluated using the MLWHF questi onnaire; objective clinical indicators used were the electrocardiographic, echocar diographic, hemodynamic and functional capacity. Results revealed that sinus rhythm ( p = 0.007), NYHA class ( p <0.001), and the distance covered with the 6-minute walk test ( p <0.001) were correlated with QOL. Therefore, t he study recommended the possibility of cost-effective
31 non-pharmaceutical therapeutic approac h in improving QOL heart-failure management as a much needed approach in the management of heart failure. [B] Psychosocial Adjustment to Illness The empirical review found that psychological factors such as depression, anxiety, and coping styles affect QOL in HF patients (Zambroski et al., 2005; Artinian et al., 2004; Carels, 2004; Martenss on et al., 2003; Dracup et al., 1992). Depression in CHF has been well documented in several studies of heart failure (Johansson et al., 2006; Turvey et al., 2006; Konstam et al., 2005; Costelo & Boblin, 2004). An estimated $5 billion of the total $20 billion cost associated with heart failure may be associated with depres sion. The prevalence of depression among hospitalized HF patients ranged fr om 15% to as high as 77.5%; outpatients with HF and depression ranged fr om 13% to 42% (Gottlieb et al., 2004). The summary of the cross-sectional studies conducted by Havranek et al. (1999), Majani et al. (1999) and Koenig et al. (1998) indicated the presence of higher levels of depression in the heart failure population. The longitudinal study conducted by Murberg et al. (1999) reveal ed that depressed mood is a significant indicator of mortality at 2-year follow-up of CHF patients (MacMahon & Lip, 2002). Depression as experienced among heart failure patients Evangelista et al. (2006) studied the relationship of depression and obesity on health related quality of life (HRQ OL) in patients from a tertiary HF clinic (n = 358). The Beck Depression Inventory (BDI), MLWHF questionnaire, and body mass index were utilized in measur ing the variables of the study. The authors reported BMI results in relation to overall MLWHF, physical subscale, and
32 emotional subscale scores were significant ( p <0.001). The stud y concluded that obese HF patients have significantly poorer HRQOL, physical health, and emotional well-being; they also have more depressed symptoms. Lesman-Leegte et al. (2006) ev aluated depressive symptoms among elderly hospitalized HF patients (n = 572). Depression was measured using the Center for Epidemiologica l Studies Depression Scale (CES-D). Findings reported that 41% had symptoms of depression, women more than men (48% vs. 36%, X 2 = 8.1, p <0.005). Multivariable logistic regr ession showed that women had more depressive symptoms (OR.68, 95% CI 1.14-2.48), COPD (OR 2.11, 95% CI 1.353.30), sleep disturbance (OR 3.45, 95% CI 2.03-5.85) and lo ss of appetite (odds 2.61, 95% CI 1.58-4.33). The authors concluded that depression was more prevalent in elderly women than in elderly men hospitalized with HF. Martensson et al. (2003) conducted a study on CHF patients and spouses regarding different levels of depression and health related quality of life (HRQOL). The study used a two-group comparative de sign and enrolled 48 couples with all men diagnosed with heart failure. Depr ession was assessed using the Beck Depression Inventory (BDI), and HRQOL was assessed using the SF-12 health survey. Results revealed gender differences regarding the presence of depression. Patients with HF had signific ant differences in depressive symptoms, with men having a mean of 10. 5 7.3, whereas their s pouses had a mean of 7.0 5.6 ( p <0.006). The authors recommended stra tegies and interventions that would include enhancing education and co mmunication between the couples (patients and spouses).
33 Sullivan et al. (2002) conduct ed a study evaluating 1,098 health maintenance organization patients cat egorized into three groups. Group 1 includes those with no depression (n = 672; cost of $7,474 per patient per year), group 2 includes those with antidepressant prescription only (n = 312; cost, $11 012 per patient per year), and group 3 includes those with antidepressant prescription and depression diagnosis reco rded (n = 114; cost, $9550 per patient per year). Healthcare costs were 26% higher in group 2 (antidepressant prescription only), 29% higher in gr oup 3 (antidepressant prescription and depression diagnosis) as compared to the no-depression group 1 ( p <.001), suggesting that healthcare costs are si gnificantly higher for patients with depression. Moser et al. (2005) studied the pr evalence of psychological, social, and behavioral risk factors in patients recent ly hospitalized with HF. The randomized study recruited participants from thr ee community hospitals (n = 202). The modifiable risk factors measured were depression, QOL, assessment of symptoms, health, and medical complianc e. The Multiple Adjective Affect Checklist was utilized to measure depression and anxiety. The MLWHF questionnaire was used to measure HRQOL The dyspnea and fatigue index was used to measure symptom status. Health compliance was measured by analyzing the documentation of the inte rvention nurse. Using the Mu ltiple Adjective Affect Checklist, a score > 7 represented pres ence of anxiety and a score of > 11 indicated presence of depre ssion. Results showed that the mean anxiety level was 7.8 4.6 (median 7.0, range 0-21) and mean depressi on level was 15.6 8.4
34 (median 16, range 1-37). Anxiety was presen t in half of the participants (50%), and 5% had double the anxiety levels from the cut-off point. The authors concluded that the presence of psychol ogical, social, and behavioral risk factors are prevalent among discharged HF patients. [C] Functional Limitations The lack of mental and physical ener gy may cause CHF patients to have functional limitations. These limitations may cause patients with heart failure to lose hope and feel a sense of resignation t hat neither they nor their environment can influence their medical predicam ent (Martensson, 1997). Symptoms of congestive heart failure include shortness of breath, fluid retention, which may lead to pulmonary congestion and peripheral edema, fatigue that limit exercise tolerance, and general body malaise that lim it daily activities (Hunt et al., 2005). Symptom severity and symptom burden Insomnia and sleep disordered breathing are the most severe and burdensome symptoms of HF associated wit h poor QOL (Zambroski et al., 2005; Konstam et al., 2005; Brostrom et al., 2004). Sleep disordered breathing (SDB) is a cardinal symptom of HF associated with poor QOL (Ferri er et al., 2005; Trupp et al., 2004; Brostrom et al., 2004). SDB is inclusive of both central sleep apnea (CSA) and obstructive sleep apnea (OSA). Zambroski et al. (2005) conducted a study regarding symptom severity, prevalence, and burden on QOL among patients with heart fa ilure (n = 58). The method used for the study was a cross-sect ional descriptive design. Physical and emotional symptoms were assessed us ing the Memorial Symptom Assessment
35 Scale-Heart Failure. Functional status was assessed using the Dyspnea Fatigue Scale and NYHA classification. HRQOL was assessed by the MLWHF questionnaire. Total mean scores reported for th e MLWHF questionnaire were 60.1 21.5. Results indicated that there was a high symptom prevalence of shortness of breath (2.7 1.0), lack of energy (2.9 0. 1), dry mouth (2.5 1.1), feeling drowsy (2.4 1.0), and difficulty sleeping (3.0 1.0). The prevalence of psychological symptoms included difficulty of concentra ting, worry, sadness, nervousness, and irritability. Difficulty sleeping was rated as the most frequent and severe symptom at 94% and 98%, respectively. When compar ed to chest pain, difficulty sleeping was the most burdensome symptom (2.1 .1 vs. 2.8 0.8, p <0.001). Results included both the total prevalence score and the total burden score. The four variables explained 67% of the variance for the HRQOL model; age ( b = .30, p < .01), NYHA functional class ( b = .22, p = .02), total burden ( b = .32, p < .01), and finally, total prevalence ( b = .32, p = .01). This meant that younger patients and those with higher NYHA classification had greater symptom burden and greater symptom prevalence an d therefore predicted a worse QOL. The study concluded that patients with HF experience high le vels of symptoms and symptom burden that af fect their QOL. The aut hors recommended thorough assessment, identification, and treatment of sleep disorder symptoms as crucial steps in providing a better QOL. Brostrom et al. (2004) conducted a study describing the relationship of self-assessed sleep difficulty, daytime sleepiness to HRQOL in HF, and
36 compared to the healthy population. A cros s-sectional design was used for the study. Outcome measures were assess ed using the Uppsala Sleep InventoryChronic Heart Failure, Epworth Sleepi ness Scale, SF-36 and the Minnesota Living With Heart Failure Questionnai re (MLWHFQ). For the study, 223 CHF patients with NYHA Classes II-IV were recr uited. Results revealed that sleep was shorter for women ( p <0.05) while men had an in creased number of sleep awakenings ( p <0.001). Patients who had a difficu lt time maintaining sleep, initiating sleep, and had early morning awakenings reported lower HRQOL as compared to the normal population ( p <.05p <.001). The most significant difference were in the areas of general health, vitality, and social functioning ( p <0.001). The MLWHF questionnaire total was 37.4 21.9 for men vs. 39.9 25.6 for women. The score on the physical subscale was 16.3 10.3 for men as compared to 17.9 9.8 for women, and on the emotional subscale 6.8 6.3 for men as compared to 6.0 5.4 for wom en. The number of frequent awakenings per night were significantly more in men ( p <.001), and the ratio of habitual sleep to the amount of estimated need for sl eep was significantly shorter in women ( p <.05). Sleep apnea can stim ulate neurohormonal and hemodynamic changes in HF. Fifty percent (50%) of HF patient s have sleep apnea and significantly worse outcomes when compared to those HF patients who do not have SDB. Daytime tiredness is the most prominent symptom of sleep apnea (Brost rom et al., 2004). HF patients suffering from SDB have poorer HRQOL, as seen in significant decrease in seven to eight domains measured by the SF-36 (n = 223, p <.05 to
37 p <.001) as compared to the normal population (Brostrom at al., 2004). Patients suffering from difficulties initiating sl eep (DIS), difficulties maintaining sleep (DMS), early morning awakenings (EMA ), and excessive daytime sleepiness (EDS) showed significantly decreased HR QOL as measured by MLWHF. Total MLWHF score in HF patients with DIS wa s 49.0 24.1 (n = 42); 51.2 22.9 (n = 48) in HF patients with DMS; 54.9 22.4 (n = 33) in HF patients with EDS; and 47 21.7 (n = 47) with EDS. HF pati ents suffering from DIS, DMS, EMA, and EDS had significantly decreased HR QOL, measured by MLWHF, compared to the whole group of patients with HF ( p <.05 to p <.001). Physical symptoms as indicators of heart failure are not obvious when compared to patients with diabetes (K odiath, 2005). For example, diabetic patients can monitor their blood glucose le vels, but when HF patients gain 3 to 5 pounds, they are already ill. Empowering HF patients to become active participants rather than passive observers of their health situation is, therefore, critical and essential. [D] Self-Care Management Disease state management is an ev idenced-based program designed to provide prevention, screening, and monitori ng of patients health and education, thereby increasing complianc e and autonomy regarding pati ents control of their health (ACC, 2001; HFSA 2006). Disease state management follows ACC/AHA guidelines in the use of beta blocke rs, diuretics, digoxin, and angiotensin converting enzyme (ACE) inhibitors for patients with left ventricular systolic dysfunction with or without symptom atic heart failure, and the use of
38 spironolactone in patients with severe heart failure. It is a systematic, proactive case management model that utilizes an organized approach in providing early intervention and an active patient self -care participation for optimum health maintenance (ACC/ AHA, 2005; HFSA, 2006). The empirical literature review indicated that non-pharmacologic interventions such as a supportive nursing in tervention may be an effective tool in improving QOL in the HF pat ient population (Riegel, et al., 2006; Karlsson et al., 2005; Harrison et al., 2002; Jaarsma et al., 2000). Home care nursing influences self-care and self-efficacy in the HF popul ation by providing the following: a supportive familiar environment wherein pat ients can perform physical activities (performance mastery); encouragement and support (verbal persuasion); guidance and sharing of vicarious exper iences; and a realistic approach and evaluation of individual abilities (ph ysiologic state) (Bor sody et al., 1999). Riegel et al. (2006) conducted a mi xed method, pretest posttest design evaluating the motivational intervention pr ogram to improve HF self-care. Both quantitative and qualitative data were used to analyze participants responses. Participants (n = 15) received a moti vational intervention designed to help patients increase their intention to change behavior by changing attitude about change. An advanced practice nurse trai ned in motivational interviewing and counseling did an average of 1.5 to 3.0 hom e visits over a 3-month period to HF patients. Self-care was measured using the Self-Care of HF Index (SCHFI), designed specifically for HF patients. The authors reported t hat both qualitative and qualitative analysis showed participants improved their self-care behaviors
39 after receiving intervention by 71.4% (10 of 14). The study concluded that motivational intervention progr ams provided to HF patients improve HF self-care. Karlsson et al. (2005) conducted a nurse -based outpatient clinic study that randomized HF patients to either the in tervention group (followed up at a nursebased outpatient clinic; n = 103) or the c ontrol group (followed up at the primary healthcare clinic; n = 105). Patients cogni tive functioning was assessed using the Mini Mental State Examination (MMSE) at baseline and at 6 months. Results revealed that men knew more about CHF at baseline as compared to women ( p <0.01). However, women in the inte rvention group increased their knowledge regarding self-care between baseline and 6 m onths as compared to the women in the control group ( p <0.05). The study concluded that women gained more than men from a nurse-based management program. Pa tients with cognitive dysfunction who had low scores on the MMSE (<24) presented lower scores on knowledge of CHF as compared to t hose with higher (>24) at baseline ( p <0.001). Differences in scores disappeared after intervention was given. The authors, therefore, recommended that patients with cognitive dysfunction should not be discouraged from participat ing in the program. Harrison et al. (2002) conducted a prospective randomized study of a nurse-led intervention focused on the transition from hospital-to-home and supportive care for self-management 2 weeks after hospital discharge. The purpose of the study was to find out if providing a nurse-led intervention would promote a better QOL. The study enrolled patients (n = 75 per group) randomized to either the transitional care or the usual care. Three months after discharge,
40 31% of the usual care patients were readmitted as compared to 21% of the transitional care groups. Emergency department visits were higher in the usual care group at 46% as compared to t he transitional care group at 29% ( t = 4.86, df = 1, p <0.03). Results showed that at baseline, physical dimension mean scores for the usual care were 25.45 (SD = 9.77) and transitional care mean score was 25.46 (SD = 9.55). At 6 weeks after hospi tal discharge, the total MLHFQ score was better among the transitional care patients (27.2 19 .1) than among the usual care patients (37.5 20.3, p = 0.002). The MLWHF questionnaire total score in 12 weeks for the usual care had a mean score of 38.39 (SD = 18.24); the transitional care had a mean sco re of 25.76 (SD = 19.44) ( p <0.001). The study concluded that significant improvements in health-rela ted quality of life (HRQL) were associated with transitional care and decreased visits to the emergency rooms were noted. Jaarsma et al. (2000) conducted a st udy to test the effectiveness of a supportive nursing intervention regarding se lf-care abilities, self-care behavior, and QOL in CHF patients (n = 179). An experimental design with random assignment to two groups, an interventi on group and a treatment control group. Outcome measures on self-care abilities ut ilized the Appraisal of Self-Care Scale. Three dimensions of QOL were measured. These included functional capabilities, symptoms, and psychosocial adj ustment to illness. Functi onal capabilities were measured using the Heart Failure Functional Status Inventory. Symptoms were measured using a ten-point scale questi onnaire. Psychosocial adjustment to illness was measured using t he Psychosocial Adjustment to Illness Scale (PAIS).
41 Overall well-being was measured utilizing the Cantril's Ladder of Life Survey. Education took place at the hospital and at home. The home visit was scheduled a week after discharge. Data collected included self-care abilities, self-care behavior, hospital readmissions, visits to the emergency department, and use of other healthcare resources. Findings indicated that the interv ention group report ed better compliance as compared to the control group at 3 months (12.2 vs. 10.6; t = 2.9, p = .005) but was not significant different at 9 months (11.2 vs. 10.3; t = 1.6, p = .11). Symptoms decreased significantly in bot h groups, from 3.9 at baseline to an average of 1.9 symptoms in the contro l group versus 2.2 symptoms in the intervention group at 3 months of follow-up ( p <.001).). Total PAIS scores for both groups decreased significantly, indicati ng better psychosocial adjustment to illness (control: t = 2.3, p = .03; intervention: t = 2.3, p = .03). There was no difference demonstrated between the two groups at the three point s in time. Heart failure self-care behavior correlated slight ly with the overall score of well-being only at 9 months after discharge ( r = 0.24, p <.05). The study concluded that education from a nurse provided at t he hospital and at home significantly increased self-care behavior for both groups at one month but was significantly increased in the intervention group (control: 12.2 2. 9 vs. 13.8-3.4, p <0.001). [E] Patient Education Kuztleb and Reiner (2006) conduct ed a prospective quasi-experimental multi-center study on the impact of nurse-directed pat ient education on QOL and functional capacity in people with HF. Patients were grouped to either the nurse-
42 directed care (NC) (n = 13) and the rout ine care (RC) (n = 10). The NC group received comprehensive disease stat e management education and weekly telephone follow-up while the RC receiv ed protocol-driven medical management. Both groups were followed up to 12 months. QOL was measured using the Ferrans and Powers (1992) Psychometric Tool. This QOL instrument consists of two parts. The first part meas ures satisfaction with various aspects of life, and the second part measures the significance of those aspects. Scores were measured for overall QOL and in four domains: health and functioning, psychological and spiritual, social and economic, and family Scores range from 0 to 30, and a significant correlation exists between higher QOL and higher scores. Findings showed that the domains of total QOL we re significantly improved in the NC group ( F = 13.569, p = 0.000), health and function (F = 3.995, p = 0.003), social and economic ( F = 14.109, p = 0.000), psychological and spiritual ( F = 13.212, p = 0.000) and family ( F = 2.384, p = 0.048). The study c oncluded that a nursedirected patient education improved QOL an d improved functional capabilities. Gonzalez et al. (2005) conducted a prospective study to evaluate if a nurse-guided education c hanges self-care behavior in an outpatient HF population. The study utilized a questionnaire given to HF pat ients at time of visit and at one year; it evaluated how the nurse electronic-guided education changed self-care behavior in regard to kno wledge of the disease and treatment, and weight and blood pressure monitoring (n = 2 98). Results indicated that initially only 28% of patients understood the disease, but 55% understood the disease after one year of follow-up ( p <0.001). Awareness of more than three symptoms
43 increased from 66.5% to 85.5% ( p <0.001). Medication knowledge increased from 33% to 44% ( p <0.001). Weight monitoring co mpliance increased from 21% to 39% (p <0.001), and weekly blood pressure m onitoring increased from 28.5% to 43% (p <0.001). Results reported that only 30% understood HF and 56% understood the disease at one year ( p <0.001, N = 298). Knowing signs and symptoms of HF increased from 66.5% to 86.5% ( p <0.001). Knowledge of medications increased from 33% to 44% ( p <0.001). Initially, 63% monitored their weight only at clinic visit, and 21% monito red their weight at least once a week. After one year, these percentages were 16% and 39% respectively ( p <0.001). Initially, only 45% were monitoring their blood pressure and 28% checked it once a week. At one year these percentages were 12% and 43% respectively ( p <0.001). Kodiath et al. (2005) conducted a one-group design study. The purpose of the study was to find out if behavioral self-management enhances health-related quality of life (HRQL). The behavioral intervention was implemented among a sample of HF patients (n = 58) to hel p participants establis h healthy behaviors that would improve their quality of life. The intervention consisted of 2-hour group classes and telephone call follow-up over a period of 15 weeks. Patients were to change one of the follo wing behaviors: diet, exercise, smoking, sodium, or alcohol consumption. The method incorporat ed the components of the Information Motivation Behavioral (IMB) Skills Model that included information, motivation, and behavioral skills. Feedback forms were given to all who attended at the end of all classes. Major themes were described such as depression, satisfaction with
44 the intervention, participants lack of understanding of HF, denial or disbelief concerning the diagnosis, the influence of age, confusion regarding access to care, lifestyle changes, and depression. The st udy concluded that in this patient population (CHF patients), choosing to change a health behavior was difficult because the physical indicators (such as weight gain) following the change were not obvious (Kodiath et al., 2005). Effect of CHF on Hospitalization Heart failure is significantly related to increased hospitalization (AHA, 2005). In the data from 1980 to 2002, reported from the Center for Disease Control and Prevention (CDC) on heart failu re as a first-listed diagnosis, the average length of stay (LOS ) of hospital discharges decreased by 6.6 days (from 11.9 to 5.3 days). The dat a reported a trend in the decreased average LOS among all ages, regardless of sex and race. Hospital discharges with heart failure as first diagnosis accounted for more than 1. 8 million days of hospital stay, for an average LOS of 5.3 days (Center for Disease Control (CDC), 2005). The agespecific rates of hospital discharges for heart failure patients when listed as first diagnosis for all age groups increas ed during the 1980s and mid 1990s. From 1998 to 2002, age-specific rates leveled o ff in the 0-64 age group and declined in people aged 75 and older (CDC, 2005). Age-adjusted hospital discharge rates were similar for men and women throughout the time period of 1980-2002. Role of B-Type Natriuretic Pept ide (BNP) in CHF Hospitalization Managing CHF costs $56 billion a year, with 70% due to hospitalization (Bhalla et al., 2004). Increased predischarge levels of BNP predicts hospital LOS
45 after decompensated CHF (Cheng et al., 2 001; Mueller et al. 2004). Conversely, BNP levels provide a measurable guide for the estimation of timely diagnosis of heart failure (Prahash & Lynch, 2004; Maisel et al., 2001; Wieczorek et al., 2002; Troughton et. al., 2000). Thus early intervention can alleviate symptoms and delay or halt disease progression; r egular clinic visits and management can potentially decrease hospital LOS by 3 days (Mueller et al., 2004). Clinical studies indicating relati onship between BNP levels and CHF hospitalization Hogenhuis et al. (2006) conducted a study investigating the prevalence and characteristics of HF patients (n = 601) with BNP levels <100 pg per ml and categorized using NYHA Class II-IV. Patient s were enrolled in the study following hospital discharge with diagnosed heart failure. Results showed that patients with BNP levels < 100 pg per ml had higher left ventricular ejection fraction compared with those with BNP levels > or = 100 pg per ml (0.41 0.14 vs. 0.33 0.13, p < .001). The authors concluded that clinically stabl e patients with a recent admission for decompensated HF, and with low BNP levels seemed to have less severe HF and preserved systolic functi on as compared to those patients who had BNP levels > or = 100 pg per ml. Valle et al. (2005) conducted a st udy that measured BNP levels in ambulatory patients with HF with preserved left ventricular ejection fraction (LVEF), including how BNP levels can pr edict the occurrence of cardiovascular events in 6 months. BNP levels were drawn on HF outpatients (n = 233). Outcome measures included cardiova scular death (n = 15) or hospital
46 readmission (n = 33). Results revealed that BNP levels were a strong predictor for subsequent events (ROC, area under curve = 0.84; CLI = 0.78-0.88). BNP cut-off levels of 200 pg per ml found in 67% of patients (HR = 2.2, p < 0.4) predicted 9% event rate within 6 months. BNP levels of 500 pg per ml or more found in 10% of the patients (HR = 5.8, p < .001) predicted 74% of unf avorable events. The study concluded that BNP levels are strong and accurate predictors of cardiovascular mortality and early readmission in patients with HF. The authors recommended that BNP levels might be used successfully for patient follow-up after an event of HF decompensation. Maisel et al. (2004) conducted the Rapid Emergency Department HeartFailure Outpatient Trial (REDHOT), and examined the association between BNP levels, perceived severity, clinical decision-making, and outcomes (hospital LOS and mortality) in CHF patients presenti ng to the emergency department (ED). Physicians were blinded to the actual BNP level and subsequent BNP measurements, and patients were followed to 90 days after discharge. Of the 464 patients enrolled, 90% were hospitalized. The overall hospitalization rate was 90.3% although physicians intended to admit 68.3%. The ED doctors intention to admit or discharge a patient had no in fluence on 90-day outcomes, but the BNP level was a strong predictor of 90-day outcome (22%). Fi ndings indicated a disconnect between ED physicians perceiv ed severity of CHF and severity as determined by BNP levels. Discharged patients were likely to die or be readmitted (19 out of 45 or 42.2%) compared to admi tted patients (110 of 425 or 25.9%) within the 90-day follow-up ( p = 0.02). The study concluded that BNP
47 levels predicted outcomes and was a va luable tool in making the decision whether to admit or di scharge a CHF patient. Heidenreich et al. (2004) studied how BNP-guided treatment is associated with decreased cost of care. This study evaluated the cost effectiveness of screening patients with BNP to classify those with LVD. The study screened 1,000 asymptomatic CHF patients with abnormal BNP levels. Those with abnormal BNP levels were followed up by echocardiography to assess left ventricular function. Results indicated that such screening increased lifetime cost of care ($176,000 for men; $101,000 fo r women) and improved outcome (7.9 quality-adjusted life years fo r men (QALY); 1.3 QALYs for women), resulting in a cost /QALY of $22,300 for m en, $77,700 for women. T he study concluded that screening populations with 1% decr eased LVD with BNP followed by echocardiography may provide a health benef it cost equal to or less than other accepted interventions. Mueller et al. (2004) conducted a study on how BNP-guided decision making is associated with decreased hospital LOS. In this study 75% of the BNP group were hospitalized as compared to 85% of the standard group ( p = 0.0008). A cumulative frequency distribution was used to track for time to discharge of patients in the BNP group as compared wit h the control group. Hospital LOS was reduced from 11 days in the standard group (n = 227) to 8 days in the BNP group (n = 225). A total cost of care decreased from $ 7,264 to $ 5,410 ( p = 0.006). Troughton et al. (2000) found that the BNP group had fewer hospital admissions as compared to the clinical group (n = 69, 17 vs. 46, p = 0.02).
48 Latini et al. (2002) studied the long-term effects of angiotensin receptor blockers (Valsartan) on BNP and norepi nephrine (NE). The Valsartan Heart Failure Trial randomized 4,284 CHF patients to either Valsartan or placebo. BNP levels were measured at baseline, 4, 12, and 24 months. In the placebo group, BNP levels rose over time; in the Valsartan group, BNP levels showed a sustained decrease. Results show ed that Valsartan significantly reduced the risk for both mortality and morbidity by 13.2%, for hospitalizations for HF by 27.5%, but not for mortality alone. BNP was the strongest predictor of outcome (hospitalization, death) when compar ed with other clinical markers ( p = .0001). The findings of Richards et al. (2 002) supported the relationship of BNP levels and CHF hospitalization. The study enrolled 69 symptom atic CHF patients (NYHA II-IV) randomized to receive dr ug treatment guided by BNP or standard clinical assessment. Treatment was guided by the use of a treatment target score according to modified Framingham criteria of <2. When the target score was < 2, medical treatments were intensified until scores were met. Using the t-test statistic, the study reported that using BNP levels for decision-making (19 vs. 54 clinical events) reduced time of firs t cardiovascular event hospitalization ( p = .034) and death ( p = .049). Cheng et al. (2001) studied a conveni ence sample of 72 veteran patients admitted from March to December 1999. Participants were enrolled to determine if BNP level predicts outcomes of pat ients admitted with decompensated CHF. The authors also examined whether an a ssociation existed between initial BNP level measurement, pre-discharge BNP measurement and outcomes of death,
49 and a 30-day readmission rate. A t-test stat istic was used to compare two groups; the BNP group and the control group. BNP levels were measured daily. Daily BNP levels were drawn within 24 hours after admission and within 24 hours prior to discharge or death (in case of death, last drawn BNP levels were considered). Of 13 deaths and 9 readmissions, the last measured BNP level was the single variable most strongly associated wit h patients experiencing one of the prespecified end points of death in hospital, death within 30 days after discharge, or hospital readmission within 30 days. Mean BNP levels were significantly greater in patients experiencing end points (1, 801 273 pg per ml standard error of the mean [SEM] vs. 690 103 pg per ml SEM) compared to patient s with successful treatment of CHF ( p < 0.001). Patients who died or were readmitted had higher BNP levels (mean increase = 233 pg per ml, p <0.001) than those who lived or were not readmitted (mean decrease = 215 pg per ml). The subgroup of patients surviving to discharge, using the NYHA cl assification, was the most significant predictor of readmission ( p = 0.0002); discharge BNP was associated with readmission within 30 days (AUC (C-statistic) = 0.72, p = 0.02). The last measure of BNP was strongly associated with both death and hospital readmission (area under the receiver operator curve of 0.73) and therefore suggested that BNP levels could be successfully used to guide treatment for patients with decompensated heart failure. It is important, therefore, to understand the role of BNP in CHF hospitalization for the early detection and identification of CHF symptoms and for
50 potentially reducing hospital admissions, decreasing ho spital LOS, and lowering healthcare costs. Knowledge of BNP Levels Predicting Treatment Preclinical recognition of increased BNP levels improv es prognosis compared to treatment after onset of severe sy mptoms (Hennekens, 1987). In August 2003, the Food and Drug Administration (FDA) approved two indications for BNP as a point-of-care rapid assay for the diagnosis of CHF. These included (1) distinguishing cardiac cause of acut e dyspnea from pulmonary or other noncardiac causes and (2) distinguishing decompensated CHF from exacerbated chronic obstructive pulmonary disease (C OPD). However, Medicare does not reimburse routine use of BNP assays to assess the effectiveness of CHF therapy, for the titration of therapy of heart failure, or for prognostic uses (Center for Medicare & Medicaid Services, 2003). The ACC/AHA 2005 Pr actice guidelines indicated that high BNP levels are pr edictive of HF but should be used in conjunction with other clinical assessment. The Hear t Failure Society of America (HFSA) 2006 Practice Guide lines state that BNP levels are not recommended as a routine part of HF evaluation for pat ients at-risk but without signs and symptoms of HF. This means that BNP levels ar e useful in the diagnosis of HF, but physicians are cautioned not to depend sole ly on this laboratory assay but to use it as an additive tool in the diagnosis of HF.
51 Studies on Knowledge of BNP Levels Predicting Treatment Aspromonte et al. (2006) evaluat ed whether BNP levels associated with echocardiography would effectively strati fy patients with new symptoms as a part of a cost-effectiveness progr am. The study enrolled patients suspected of CHF (n = 357) referred to the cardiology clinic by primary physicians. All patients were clinically examined. Blood was drawn for BNP levels and a transthoracic echocardiography was performed. Findings reported high BNP levels at 469-505 pg per ml (n = 240) on those diagnosed wit h HF, compared to those without HF, 43-105 pg per ml (n = 117, p = 0.001). BNP cut-off level was at 80 pg per ml (sensitivity 84%, specificity 91%). The authors findings indicated that cost analysis at this cut-off level might pr ovide a cost savings of 31%. The study concluded that BNP levels drawn from patients suspected of CHF are cost effective and are helpful in stratifying CHF patients. Daniels et al. (2006) conducted a mult i-center site study to determine if there was a disconnect between perceived se verity of HF by physicians and the severity of HF as determined by BNP leve ls. Patients (n = 151) were enrolled if they were seen, treated, or admitt ed through the emergency department (ED). BNP levels were drawn and treating physicians were blinded to BNP levels. After clinical assessment, ED physicians classifi ed the patients according to the NYHA functional classifications I-IV. ED physicians were asked whether the initial disposition of these patient s would warrant hospitalizat ion. Patients were followed up after 90 days. Results showed that of 90% of those hospitalized, 32.5% were white and 63.4% were black. African Am ericans discharged from the ED had a
52 higher median of BNP levels co mpared to white (1,295 vs. 533, p = 0.004). African-American HF patients who we re discharged had a higher mean BNP levels as compared to African-American HF patients who were admitted (1,293 pg per ml vs. 769 pg per ml, p = .04). This finding was not the same for whites (692 pg per ml vs. 533 pg per ml, p = .09). The authors concluded that a disconnect in perceived severity of HF and in severity using BNP levels is more evident in African Americans. This means t hat the perceived severity of HF by ED physicians did not often correlate highly with BNP levels. In the original REDHOT study, patients who were sent home fr om the ED had higher BNP levels than those who were admitted. In this study, BNP levels were stronger predictors of outcome than was perceived severity of CHF (Maisel et al., 2004). Mueller et al. (2004) studied the us efulness of BNP-gui ded treatment in ED patients. Participants (n = 452) were randomized to a diagnostic strategy of measuring BNP levels (n = 225) or to standardized treatment (n = 227). Of the BNP group, 15% required intensive treat ment; of the standard group, 24% ( p = 0.01). The 30-day mortality rates were lower ( p = 0.45) in the BNP group (10%) than in the standard group (12%). The study concluded that BNP levels used with other clinical information improv es evaluation and treatment of ED CHF patients with acute dyspnea. The mean cost of treatment in the BNP group was $5,410 compared to $7,264 in the control group ( p = .006). Using BNP values in decision-making decreased treatment cost and reduced hospitalization in the treatment group ( p = .001). Another purpose of t he same study (Mueller et al., 2006) was to find out the cost effectiv eness of BNP-guided testing in the ED
53 patients. Results showed that total treatment cost was si gnificantly reduced in the BNP group ($7,930 vs. $10,503 in the control group; p <0.004). The authors concluded that BNP guidance is cost e ffective in patients with acute dyspnea. Ishii et al. (2003) determined whether cardiac troponin T and BNP would stratify CHF patients (NYHA Classes III and IV) after initiation of treatment. Enrolling 100 consecutively admitted pati ents, the study included 54 CHF patients categorized as NYHA Class III and 46 CHF patients categorized as Class IV. Serum cardiac troponin T (cTnT) and plasma BNP were measured on admission and then 2 months later when the number of patients classified as NYHA Class III decreased to 40 and those classified as Cla ss IV decreased to 3; and 54 patients decreased to NYHA Class II. Using t he Mann Whitney U test for continuous variables, the results showed a decrease in cTnT (0.023 vs. 0.063) and BNP levels (249 vs. 753), significantly improv ed NYHA functional class (2.5 vs. 3.5), and improved left ventricular ejection fraction (13% vs. 12%, p < 0.01) two months after treatment as compared to admission. Thus, a combination of both cardiac troponin T and BNP-guided level treatments may prove to be highly effective in risk stratification. Troughton et al., (2000) reported that pharmacotherapy guided by BNP would produce better outcomes than t herapy guided by standard clinical assessment. Patients were recrui ted after hospital admission with decompensated CHF. During clinic visits, patients were double-blind randomized to treatment guided by BNP measurement s or standard clinical assessment. An objective scoring system (with a score of 2 or more indicating decompensated
54 heart failure) was used. If target scores (less than 2) were not achieved, early treatments were intensified according to a rigid predetermined protocol at 2-week intervals until target scores were achiev ed. Utilizing the Mann Whitney U test, the study found a 50% reduction in total cardio vascular events in comparison to the control group. Patients not receiving BNP-guided treatment experienced more cardiovascular-event hospitalizations and deaths (78%) than those receiving BNP-guided treatment (22%). Death ra tes and hospital admissions were less in the BNP-guided group than in t he clinical group (19 vs. 54, p = .02). Thus, BNP became a preventive strategy targeting a more intensive pharmacological treatment, allowing for tailo red therapy and follow-up of patients (Troughton et al., 2000). Summary In summary, the literature review re vealed that QOL is influenced by how individuals perceive their life situations QOL of patients wit h heart failure is affected directly or indirectly by thei r functional limitations and psychological needs. Studies show that physical symptoms of HF such as shortness of breath, peripheral edema, and easy fatigability cause functional and social limitations that affect the QOL in HF patient s. The review of literatur e suggests that the presence of symptoms, severity, and symptom burden such as insomnia and sleep disordered breathing are also associated with poor QOL. Clinical studies demonstrate that treating t he psychological symptoms of depression, having a strong social support, finding meaning in life, achieving perceived control, and having spiritual faith and beliefs are an integral part of the psychosocial
55 adjustment to any disease condition. According to the AHA (2005), HF pat ients frequent and regular hospital admissions have a tremendous impact on the economic burden of the healthcare system in the United States. The review of the literature suppor ts the view that CHF is a chronic debilitating disease that is associat ed with increased hospitalization. Reduction in HF admissi ons and hospital LOS, therefore, may result in lowered social and eco nomic costs and decreased healthcare expenditures. Numerous studies indica te that when physicians are aware of patients BNP levels, HF patients have shorter hospital LOS and fewer hospital admissions. BNP is a biomarker that pr ovides a measurable guide in the diagnoses of HF. The literature review supports the co ncept that recognition and knowledge of BNP levels during ED visits may be critical for patients to receive timely diagnosis and early relief of the sy mptoms of HF. Current research shows that the drawing of blood for BNP levels in HF clinics is not routine, and evidence does not support its general use. HF pati ents may benefit from having BNP levels drawn as part of a clinic visit assessment. Therefore, there is a need to examine whether knowledge of BNP levels influenc es physicians treatment of HF. This study will examine whether physicians kno wledge of BNP levels and treatment of HF at time of clinic visit influences CHF patients QOL and hospital LOS.
56 Chapter 3 Methods Introduction Chapter 3 outlines the research methods and procedures for this study. The discussion of the research design is followed by a description of the sample and its inclusion and ex clusion criteria. Next comes a description of the setting, instruments used, procedur es, institutional review board (IRB) approvals, and informed consent. Finally the data analysis procedures are presented. Design The research used a randomized contro lled trial assigning participants into two clinic groups. The participants were randomly allocated to a group, one having the physician informed and the ot her, not informed of patients BNP levels. The experimental group was com posed of subjects whose BNP levels were disclosed to the physician. The control group included those subjects whose BNP levels were not disclosed to the physician. Subjects were not informed of their BNP levels. The specific aims of the study were: (1) to determine if physicians knowledge of BNP levels would make any difference in the quality of life scores between the ex perimental and control group at 90 days and (2) to determine if physicians knowledg e or lack of knowledge of BNP levels at time of CHF clinic visit would affe ct hospital LOS on all hospital admissions, regardless of how many hospital admissions occurred in 90 days.
Figure 2 illustrates the hypothesized logic model of the relationship of BNP levels and hospital length of stay and quality of life. The logic model illustrates that knowledge of BNP was the independent variable, and CHF hospital LOS and QOL were the two dependent variables of the study. BNP levels were measured using the Triage BNP Immunoassay Kit (Biosite, San Diego, CA), and QOL was measured by the Minnesota Living With Heart Failure (MLWHF) questionnaire. Hospital LOS was documented on the medical information form, and hospital charts were reviewed to verify hospital admission and diagnosis. Input Intervention Outcome 57 Setting CHF/BNP Demographic data Hospital length of stay and QOL at 90 days Physician knowledge of BNP leading to early treatment Withholding physician knowledge of BNP leading to standard care QOL Figure 2. Logic model: Relationship between disclosed BNP levels to hospital LOS and quality of life in patients treated for heart failure. Setting The research took place at Shands Jacksonville Cardiovascular Center located at Jacksonville, Florida. The heart failure clinic is located on the fifth floor of the Ambulatory Care Center. This Center offers numerous
58 national and international clinical trials and state-of-the-art diagnostic, therapeutic, and rehabilitative cardiac services. The cardiovascular center includes: examining rooms, stress te sting labs, electrocardiographic and quantitative 2-D echocardiographic l abs, patient and family education libraries, catheterization laborat ories, a technologically advanced observation unit for patients undergoing outpatient heart ca theterization, and electrophysiology laboratories. The 24,000-square-foot Center annually serves approximately 500 to 600 heart failure patients per year, whose estimated ethnic/racial composit ion is 54% Caucasian, 45% AfricanAmerican, and 1% other minorities. The heart failure clinic has a triage/laboratory room, physician offices, and nine examining rooms. The clinic schedule for heart failure patients was Monday and Friday each week At times, HF clinics were rescheduled randomly according to physician availability. The average number of patients seen wa s approximately 30 to 40 patients a week. The clinic is managed by two University of Florida cardiologists, nursing and clinical staff, unit secretaries, re gistration/accounting staff, and a receptionist. A total of 15 personnel staff the heart failure clinic. The heart failure research director is highly in volved with the daily operations of the clinic. He is a trained cardiologi st specializing in the diagnosis and management of heart failure.
59 Population and Sample A total of 108 participants were enrolled in the study. The power analysis was based upon the effect size of 0.5, which wa s reported in the literature. According to Troughton et al. (2000), the effect of BNP-guided treatment of HF had an effect size of 0. 5. Combining this effect size with an alpha of 0.05 and a power of .80, the projected sample size was 128. The target population included patients from the clinic diagnosed with heart failure. Participants were included in the study if the fo llowing conditions were met: at least 21 years old; able to read, write, and speak English; able to give voluntary consent; with Ne w York Heart Association (NYHA) classifications of II-IV; and with an ejection fraction (EF) of less than 40%. Participants were excluded from the study if the following conditions were present: co-morbid conditions limiting life expectancy to less than one year as determined by the attending cardiologist and a histor y of acute or chronic renal failure as evidenced by serum creatinine of over 2.0. Instruments The instruments used for the study included the BNP Immunoassay Kit, a medical in formation form, and the MLWHF questionnaire. The independent variable, BNP level, was measured using BNP Triage Immunoassay Kit (Biosite Co., 2004). Demographic data that included age, gender, ethni c background, education, occupation, marital status, and insurance payers were documented on the demographic information form (Appendix A). Quality of life was assessed by using the
60 MLWHF survey (Appendix B). Hospital LOS was documented utilizing the medical information form (Appendix C). Independent Variable BNP Rapid Assay B-type natriuretic peptide levels ar e elevated in cardiac disease and are sensitive to increased ventricu lar stretch (Mark & Felker, 2004). BNP levels are reflective of left v entricular diastolic filling pressure and therefore correlate with pulmonary capill ary wedge pressure (Jiang, et al., 2001; Maisel, et al., 2002; Cheng et al ., 2001; Ishii et al., 2003; Tabbizar et al., 2002; Anand et al., 2002). Unlike card iac enzymes that are ordered in series, BNP assays are performed on an as-needed basis in hospitals, emergency rooms, or clinics. According to the American College of Cardiology and American Heart Association Task Force on Heart Failure Guidelines (2005), BNP levels great er than 100 pg per ml predict the diagnosis of symptomatic heart failure. The BNP test, therefore, meas ures the presence of b-type natriuretic peptide levels present in t he circulating bloodstream. The test is called a rapid assay because it is a simple blood test that can be done at the bedside or clinic and takes 15 minutes to complete. BNP testing in the outpatient clinic is feasible because BNP testing is not affected by food or exercise. It is used when there is a need to have immediate results. BNP levels may help diuretic adjustment after discharge, reflect CHF exacerbation, or may reflect successful treatment or titration (Maisel, 2002).
61 The two most commonly used criteria for diagnosing heart failure are those of the National Health and Nutrition Examination Survey (NHANES) and the Frami ngham criteria (AHA, 2005) When compared to the NHANES and the Framingham criter ia for diagnosing heart failure, the BNP screening test is more accurate at 83%; the National Health and Nutrition Examination Survey (NHAN ES) showed an accuracy of 67%, and the Framingham criterion showed an accuracy of 73% in confirming the diagnosis of heart failure (Maisel, 2002). The Triage BNP Test (Biosite, San Diego, CA) uses a fluorescence immunoassay that measures B-type natriuretic peptide (BNP) in whole blood and plasma specimens using trisodium ethylenediam inetetraacetate trihydrate (EDTA) as the anticoagulant (Biosite, 2004). In order to satisfy Clinical Laboratory Im provement Amendments (CLIA) and Joint Commission on Accreditation of Healthcare Organizations (JCAHO) standards, three levels of assayed liqui d controls were used to verify the calibration of the Triage BNP Test throughout the reportable range every six months. There is no time to firs t-result since the BNP immunoassay is not an analyzer and not run in batch m odes. Calibration of the test was done by using the provided assayed cont rols supplied by Biosite. There were five assay controls, and CLIA r equired doing at least three out of the five controls. The lowest and highest controls were used to perform the calibration. Controls used should result in numbers not less than 5 pg per ml or higher than 5000 pg per ml.
62 The BNP Triage Immunoassay Kit has a reportable range of 5 pg per ml to 5000 pg per ml (Biosite, 2001). BNP results less than 100 pg per ml are representative of normal va lues in patients without CHF. BNP results higher than 100 pg per ml ar e considered abnormal and suggestive of CHF. According to Maisel (2001) the mean BNP values for NYHA Class I are 152 pg/ml; for NYHA Class II, 332 25 pg per ml; NYHA Class III, 590 31 pg per ml; and 960 34 pg per ml for NYHA Class IV. BNP results higher than 5000 pg per ml ar e considered very high values and exceed the upper limits of the BNP test. Higher BNP concentrations measured in the first 72 hours after an acute coronary syndrome are associated with an increased risk of death, myocardial infarction, and CHF (Biosite, 2004). The BNP Immunoassay Kit uses a sample type of either whole blood or plasma drawn in plastic tubes. Sample collection and storage for whole blood and plasma is up to 24 hours at room temperature or 2-8 centigrade degrees in a refrigerat or. Reagent stability is good until expiration date on the box or up to 14 days at room tem perature. The BNP analysis is based on the amount of fluor escence the meter detects within a measurement zone on the device. A greater amount of fluorescence detected by the meter indicates a higher BNP value (Biosite, 2004). A daily quality control (QC) proc edure was performed on the BNP machine to maintain consistently a ccurate readings. This was done on on heart failure clinic days (Mondays and Fr idays) before any participants
63 blood was drawn for the BNP test. To run the daily QC, a stimulator chip code was inserted into the meter and prompts on the screen were followed. A test was run, and a pass or fail result was displayed/printed when testing was completed. Biosite provided QC mate rials containing plasma to run the daily QC procedure (Biosite, 2004). Performance Characteristics of the Triage BNP Test Linearity of the BNP Triage Immunoassay Kit Plasma specimens anticoagulat ed with EDTA were spiked with purified BNP to a final concentrati on of 5000 pg per ml. Each spiked plasma specimen was diluted gravim etrically with unspiked plasma to obtain BNP values throughout the r ange of the Triage BNP Test. Linear regression analysis of the data indicat ed that the assay is linear throughout the measurable range of the test (Biosite, 2004). Interfering substances Hemoglobin up to 10,000 mg per dL, cholesterol up to 1,000 mg per dL, triglycerides up to 1,000 mg per dL and bilirubin up to 20 mg per dl added to plasma concentrations contai ning BNP did not in terfere with the recovery of BNP. Hematocrit varied between 27% and 51% with no significant effect on the reco very of BNP (Biosite, 2004). Analyti cal Sensitivity Analytical sensitivity differs from clinical sensitivity; analytical sensitivity refers to the test and not to the patient populat ion. Analytical sensitivity refers to the lowest value t hat the test can read that distinguishes
64 from zero (Biosite, 2004). The av erage 95% confidenc e limit of the analytical sensitivity of the Triage BNP Test was less than 5 pg/mL (95% confidence interval 0.2 pg/L to 4.8 pg/ L). Thus, the test cannot be exactly zero. Analytical sensitivity or the lo west detectable concentration that is distinguishable from zero for the BNP test was determined by testing a zero calibrator 20 times each using three lo ts of reagents and five meters on 5 days (Biosite, 2004). Analytical specificity Analytical specificity refers to the accuracy with which the test is able to detect the correct molecule, BNP (Biosite, 2004). Precision of the BNP machine, the use of various pha rmaceuticals and the use of blood and plasma for drawing BNP levels are discussed as they relate to the analytical specificity of the BNP assay. Precision The average within-day and total prec ision of the BNP assay was determined using the ANOVA model by testing control materials that had BNP added at concentrations near t he decision points of the assay and throughout the range of the standard curv e. This study was done over 12 days, testing each control ten times a day. Each device was read on five Triage meters (Biosite, 2004). It is not ed that the use of different Triage meters does not significantly affect the test precision. Coefficient of variation (CV) is equal to the standar d deviation multiplied by 100 divided by the mean (Hulley & Cummings, 1988). This means that the higher the
65 CV, the less precise the test. As a point-of-care test and the use of an immunoassay, the BNP test is cons idered precise at 10% within the National Laboratory Guidelines (Biosite, 2004). Table 2 illustrates the coefficient of variation (CV) measures of the BNP test as stated by the Biosite instruction manual. Table 2 Coefficient of Variation Measures for BNP Test (Total CV %) Mean (pg per ml) SD (pg per ml) CV (%) 71. 3 7.0 9.9% 629.9 75.5 12.0% 4087.9 500.1 12.2% Biosite Inc., 2004 Test-retest reliability A test-retest reliability of the Triage BNP rapid assay was performed for this study. The purpose of performing the test-retest reliability was to make sure that the BNP ma chine produced the same BNP results when a participant took the same test twice. BNP-levels results from the first and second tests should fall within the coefficient of variation percent (CV%) in order to be precise. A te st-retest method of the Triage BNP Immunoassay Kit was done on 10% (n = 14) of the partici pants. The CV% for this study was 9.3%. Usi ng systematic sampling, every 10 th subject had
66 blood tested for the second BNP levels. Only one cubic centimeter (cc) of whole blood was drawn; this was suffi cient for both the initial BNP test and for the test-retest reliability proc edure. A detailed description of the procedure for performing BNP test is described under baseline data collection. Pharmaceuticals Fifty-four drugs, ranging from common acetaminophen to angiotensin-converting enzyme inhibitors, diuretics, beta blockers, statin drugs, and antibiotics, were evaluated for potential cross-reactivity and interference of the Triage BNP Tes t. Results showed none of the drugs interfered with the recovery of BNP; neither did they produce a significant response when tested in a specimen not containing BNP. There was no significant interference with the BNP measurement and no assay crossreactivity (Biosite, 2004). Use of whole or plasma correlation A study comparison performed on EDTA whole blood versus plasma showed the correlation data as r 2 = .9878, y = 0.925 x +13.439 (Biosite, 2004). Plasma is indicated by y and whole blood by x (Biosite, 2004). This means that there is a sli ght difference in using plasma and whole blood, but the correlation is high enough that it does not make any difference when the test is run by ei ther whole blood or plasma. For the study, whole blood was used to test for BNP levels.
67 Clinical sensitivity Clinic al sensitivity, or sensitivit y, refers to the proportion of subjects with the disease who have a positive test; it also indicates how well a test identifies the diseas e (Hennekens, 1987). In a study conducted by Maisel et al. (2002), using the BNP Triage Immunoassay Kit, the BNP cutoff value of 100 pg per mL had a sensitivity of 90%, meaning that approximately 90 heart failure patients wil l test positive for heart failure, and 10 patients will test false negative. Clinical specificity Clinic al specificity, or specificit y, refers to the proportion of subjects without the disease who have a negative test; it indicates how well a test identifies the non-diseased s ubjects (Hulley & Cummings, 1988). In the Breathing Not Properly (BNP) study conducted by Maisel (2002), using the BNP triage Immunoassay Kit, the s pecificity was 76% with a predictive negative value of 90% (cut-off value of 100 pg / mL). Relative risk Re lative risk compares the incidence of disease among exposed people with the incidence of dis ease among non-exposed people by means of a ratio (Hennekens, 1987). A BNP leve l of 230 pg per ml is correlated with a relative risk of 7.0 (Maisel, 2002) A relative risk of 7.0 means that the incidence of heart failure is sev en times as high in patients with heart failure as in those without heart failure. Table 3 illustrates comparison
68 decision statistics using the BNP Triage Immunoassay Kit as reported from several studies. Table 3 Summary of Studies Utilizing the BNP Triage Immunoassay Kit Study BNP cut-off (pg per ml) Sensitivity (%) Specificity (%) PPV a (%) NPV b (%) Diagnostic Accuracy (%) Maisel et al.(2002) n = 1,586 100 50 90 76 96 83 Wiezorek et al. (2002) n = 1050 100 82 97 93 Dao et (2001) n = 250 100 94 94 92 96 94 Morrison et al. (2002) n = 321 94 86 98 98 83 91 Note a PPV = positive predictive value b NPV = negative predictive value Demographic Data Form The demographic form ( Appendix A) created for this study was a ten-item survey used to collect info rmation that could be related to the outcome variables. These variables included the following: age, gender, marital status, ethnic background, educ ation, occupation, and insurance payers. These factors were document ed to determine the relationship
69 between these variables and BNP levels CHF hospital LOS, and quality of life. Dependent Variables The Minnesota Living with Heart Fa ilure Questionnaire (Appendix B) was utilized to measur e the dependent variable of quality of life. The medical information form wa s used to abstract patients hospital length of stay. Description of the Minnesota Living With Heart Failure Questionnaire The dependent variable quality of lif e (QOL) was measured by the Minnesota Living With Heart Failu re (MLWHF) questionnaire (Rector, Kubo, & Cohn, 1987). This was a 21-it em, self-administered questionnaire with subscales covering physical, socioeconomic, and psychological impairments among CHF patients. T he instrument was developed to systematically and thoroughly evaluate pati ents perceptions of the effects on daily life of heart failure and its treatment. Patients ranked specific impairments, on a scale from 0 (best) to 5 (worst), according to how CHF prevents them living as they would lik e, in other words, ranking each impairments significance or difficu lty. The MLWHF questionnaire was scored by adding all the numbers circled by the participant. The higher the patients score, the greater the limitations, with the worst possible score being 105 (Rector, Kubo, & Cohn, 1987). The questionnaire contained total, physical, and emotional subscales. These subscales included groups of questions that contained
70 similar information. Responses to questions 2 (rest during the day), 3 (walking and climbing stairs), 4 (w orking around the house), 5 (going away from home), 6 (sleeping), 7 (doing things with others), 12 (dyspnea), and 13 (fatigue) were highly correlated to the physical dimension. Questions 17 (feeling burdensome), 18 (feeling a loss of self control), 19 (worry), 20 (difficulty concentrating and remem bering), and 21 (feeling depressed) were highly correlated to the emoti onal factor (Rector et al., 1992). The physical subscale had 8 item s, the emotional subsca le had 5 items, and a total of 21 items were used in the study. Reliability and Validity of the MLWHF A study conducted by Riegel (2002) reported that total QOL scores among discharged HF patients improved after receiving intensive interventions. After 1 month, 3 m onths, and 6 months of intensive treatment, the study showed significant differences in quality of life scores among treatment dose groups ( F = 3.43, df = 9,579, p <.001 ; F = 7.45, df = 9,579, p <0.001; F = 4.86, df = 9,768, p <.001, respectively). The trial concluded that the MLWHF questi onnaire was sensitive to major differences in symptom severity but not to subtle signs and symptoms of heart failure. A recommendation of the study was to caution researchers that the instrument was best used with a control or comparison group (Riegel, 2002). The alpha coefficients for total QOL scores at each time period ranged from 0.92 at baseline to 0.96 at one month. The alpha coefficient for the physical subscale was .92 at baseline and .95 at one
71 month; the alpha coefficient for the em otional subscale wa s .87 at baseline to .92 at one month (n = 1,136). In another study by Rector et al. (1987), 83 patients with left ventricular dysfunction completed t he MLWHF. Baseline variability was assessed by a second administration a fter 21 days. Statistical analysis on the differences and weighted kappas provided evidence for reliability, and internal consistency of the MLWH F questionnaire was examined using Spearman rank order. Validity was assessed by correlating the MLWHF scores with the response to, Overall, how much did your heart failure prevent you from living as you wanted the past month? The correlation between the NYHA classification and the patients rating on the MLWHF questionnaire was statistically significant ( r = 0.80, p <0.01). This association suggests that the questionnaire is a suitable representation of functional impairment. For this study, the MLWHF score reliably measured QOL among CHF patients with a weighted kappa of 0.84 between the individual items and the total score Thus, the MLWHF questionnaire showed potential to increase knowledge of CHF symptoms and effects of medical interventions. In 1992, Rector and Cohn evaluat ed the quality of life of 198 ambulatory patients using the ML WHF questionnaire. The inter-item correlations of the questi onnaire identified categorie s of questions for both physical and emotional score s. Testretest reliability was high for the total
72 score: r = 0.93, physical: r = 0.89; and emotional: r = 88. Cronbachs alpha was 0.94 (total), 0.94 (physi cal), and 0.90 (emotional). In a study conducted by Gorkin et al. (1993), psychometric properties of the baseline measures us ed in the Studies of Left Ventricular Dysfunction (SOLVD) trial were analyze d. The measures included the 6minute walk test, dyspnea scale, MLWHF, physical limitations, psychological distress, and health perc eptions. Researchers concluded that the internal consistencies of the self -report instruments we re high, with the exception of the health perceptions of NYHA Class II or III patients. The MLWHF questionnaire revealed a Cronbach alpha of 0.95 for 135 NYHA Class I patients and an alpha of 0.94 for 12 3 NYHA class II and III patients. Table 4 illustrates the re liability of the MLWHF questionnaire as presented from several studies. Table 4 Reliabilities of the MLWHF Questionnaire Study Sample Si ze NYHA Class Reliability Rector et al. (1992) 198 III = 0.94 Rector et al. (1987) 83 I-III Weighted kappa 0.84 Gorkin et al. (1993) 135 123 I II-III = 0.95 = 0.94 Reigel et al. (2002) 1136 -= 0.92 Note NYHA = New York Heart Association
73 Hospital Length of Stay Hospital length of stay (LOS) wa s tallied for each admission after the baseline clinic visit. The medical records of all participants who reported a hospitalization were imm ediately reviewed by the primary investigator (PI) to see whether the primary discharge diagnosis was reported as CHF. The PI reviewed hospital charts to examine whether patients were admitted because of CHF diagnostic symptoms and/or other causes. Only the CHF admissions were of interest to this study. The frequency of hospital admissions were monitored and documented in the medical information form fo r up to 3 months and were reviewed for the number of days patients stay ed in the hospital. Portions of the day were considered as one day of hospital stay. Medical Information Form The medical information form ( A ppendix C)developed by the (PI), is a three-item survey used to doc ument NYHA classifications, hospital admissions and dates (if any), diagnoses, and number of days participants stayed in the hospital.
74 Summary of Reliabilities of Instruments Table 5 Summary of Reliabilities of In struments Used in the Study Variables Number of items Instruments Reliabilities Range of scores BNP a 1 Triage BNP Sensitivity = 95% Specificity = 98% PV= 96% Standard Error = 4% 5 to 5000 pg / ml QOL b 21 MLWHF c = 0.95 = 0.94 0-105 Hospital LOS d 10 3 Demographic Form Medical Information Form NA NA Note. a BNP = B-type natriuretic peptide b QOL = quality of life c MLWHF = Minnesota Living With Heart Failure d LOS = Length of stay; Procedures Approvals Permission to use the Minnesota Living With Heart Failure Questionnaire was obtained from the University of Minnesota, which provided a waiver for research purposes for one year Approval to conduct the study and recruit participants was obtained from both the Institutional Review Boards (IRBs) of the University of South Florida and the University of Florida, Shands Jacksonville A letter of support from Shands Heart
75 Failure Clinic Medical Director wa s also provided (Appendix D). The informed consent form (ICF) and any changes made thereafter had to be approved by both IRB institutions. A c ontinuing review report was filed and had to be approved by both IRB institutions prior to the expiration date of the study. Recruitment Upon approval of the IRBs and Health Insurance Patient Portability Act (HIPPA) compliance Board, a res earch assistant employed by Shands Jacksonville reviewed the CHF clinic schedule and patient charts and then screened eligible participants for the study. During the clinic visit, eligible participants were referred to the Princi pal Investigator (P I). Prior to being seen by the physician, the participant was approached by the PI in the waiting room and taken to a consultati on room for privacy. The purpose of the study was explained thoroughl y and ample time given for any questions regarding the study to be answered. Potential risks and benefits were explained to the pat ient. If the patient agreed to participate, the participant signed the informed consent fo rm that incorporated the HIPPA authorization form (Appendix E). The average number of heart failure patients scheduled on Mondays and Fridays varied from 16 to 20 patients. The PI saw participants on both days. Data collect ion for each participant took no less than 30 to 45 minutes. The participant signed two copies of the informed consent, which included a contact number for the PI. A leaflet of frequently
76 asked questions (FAQs) regarding the study was given to each patient (Appendix F). One copy of the in formed consent was given to the participant, and the other copy remained in the participants clinic chart. The PI explained to the participant s the importance of completing the study. Participants were informed that once the study was completed, the results were available to them if they so desired. Additionally, participants were assured that participation in the study would not increase their medical costs. The participant, or the participants family member, was asked to contact the PI in the event of any hospitalization within 90 days from time of clinic visit. Once the PI had in formation on participants hospitalization via telephone or mail, the PI reviewed the hospital chart to verify if the hospitalization was due to CHF. Subjects were informed that participation was voluntary, that they could wit hdraw from the study at any time, and that all data collected were kept confidential. Random Assignment Participants were randomly assigned to either the experimental (BNP) or control group after agreeing to take part in the study. It is important that the PI and physician not k now in advance whether a patient is assigned to the BNP or the control group. Participants were randomized by the PI using randomization with concealment using the following process: To prevent bias, patient treatment allocation was done by the Statistical Analysis System (SAS) ( SAS Institute Incorpor ated, 2001) based
77 on a computer generated table of random numbers, using a syntax to allow random allocation. Numbers were pr inted, cut apart, and placed inside identical opaque envelopes and then sealed. The envelopes were numbered from 1 to 140 and stacked in num erical order. The original table of random numbers was destr oyed. The PI was not abl e to anticipate into which group the next subject would be assigned. This randomization with concealment provided initial blinding of group assignments to the researcher. After eligibility was assessed and consent was obtained, each participant took an envelope from t he top of the st ack. Following assignment to treatment condition, a ll the envelopes had subjects names written on the outside and were kept in a locked cabinet. Baseline Data Collection The PI used a consultation room near the waiting room as a private area to talk to the parti cipant regarding the study, have the informed consent signed, and the survey completed by the patient. The PI assisted with the completion of the QOL survey if the patient verbalized the need for help. The consultation room was equipped with a table and chair so that participants had space to co mplete the questionnaire. Blood for BNP levels was drawn by the PI and anal yzed at point-of-care prior to the patient being seen by the physician. Syringes, tubes, pipettes and the BNP machine were available on site for drawing BNP levels.
78 Procedure for Completing MLWHF Questionnaire The participant self-administe red the MLWHF questionnaire, unless assistance was requested. The PI explained to the participant the importance of completing the MLWHF questionnaire again in 90 days. The participants read and responded to all 21 questions and was asked to rank, on a Likert-type scale (0-5), each s pecific impairment according to how heart failure had affected their lives dur ing the past month. If a participant was not sure an item appl ied or if an item was not related to heart failure, then the participant was directed to circle 0 (No). If an item did apply, then the patient was directed to circle a num ber from 1 (very little) to 5 (very much). The total scores indicated the difficulty or importance of the impairment. Participants completed the questionnaire prior to any assessment or physician interaction. Procedure for Performing BNP Test Prior to enrolling participants in the study, the PI was certified in obtaining BNP levels utilizing the Triage BNP Immunoassay Kit. Using sterile technique, the PI drew one cc of whole blood by venipuncture collected into a 1-cc tube containing potassium EDTA and measured with a fluorescence immunoassay (BNP Tri age Immunoassay Kit). The blood was mixed gently by inverting the tube several times before transferring the blood to the test device. The specimen was added to the sample port of
79 the test device with a transfer pipette designed to deliver 250 ul to the test device. After the specimen was added, the device was inserted into the Triage meter. The meter was programmed to automatically perform the BNP analysis after the sample had reacted with the reagents within the BNP device. The PI was responsible fo r collection and disposal of all data and blood drawn. Used needles, tubes, and syringes were disposed of in a sharps container located inside the laboratory. Blood levels for BNP were drawn following completion of the survey. BNP blood levels were drawn only once. The physician involved in the study was not given the BNP results of participants in the control group. For those in the BNP group, the physician was informed of BNP levels by the PI handing him the BNP results generated from the Triage BNP Immunoassay Kit when the experimental participant entered the examining room. Results were documented on the demographic data in formation form and destroyed. Results of BNP levels were not docum ented in the patient records. On any given clinic day, only the PI and one ph ysician exposed to the experimental subject had access to the results of the BNP levels. The PI documented the BNP level results on the medical information form. Participants were not informed of their BNP levels. The PI did not inform the physician of BNP levels if he was seeing a patient in the control group even if an extremely high level was obtained. Treatment and management of CHF patients was guided at this facility
80 using HF guidelines. Currently, BNP te sting is not part of routine heart failure guidelines (HFSA, 2006). Ther efore, the withholdi ng of extremely high BNP levels was ethical. Follow-Up Following the clinic visit and examination of the participant, the clinic staff scheduled the next study visit appointment in approximately 90 days. If the physician requested to see the patient prior to 90 days, this was considered a clinic visitnot a part of the study. Every effort was made to ensure that the par ticipant came back in 90 days to complete the MLWHF survey. The PI contacted t he participants via telephone once a month, for 3 months to ensure t hat every CHF hospital admission and hospital LOS was known. Three monthly le tters were sent inquiring if there had been any hospitalizations within the pas t month, and if so, the place of admission, date of admission, and number of days the patient stayed in the hospital. Once the PI had information on participants hospitalization via telephone or mail, the PI reviewed the hospital chart to verify whether the hospitalization was due to CHF. Ninety days after enrollment in the study, the participants again completed the MLWHF questionnaire at the clinic, and the PI completed the medical information form to verify whether participants had any previous hospital admissions. If unable to reach the participants, the PI contacted the alternative contact numbers. In the ev ent that the PI was not able to contact anyone, a telephone message was left for the participant to
81 call the PI regarding the survey. Within 90 days plus minus 7 days after the baseline clinic visit, if a ttempts to contact the par ticipant and to do the 90day follow-up interview were unsuccessful, then the patients were considered lost to follow-up. Data Management Demographic and medical informa tion used computer generated random codes to protect personal identif iers. A master file with names was kept and locked in a secure location in the physicians office. Only the PI and supervisors had access to this information. To ensure patient confidentiality, the PI entered all completed surveys, the demographic and medical information form, and BNP resu lts into an Excel program. After data collection was completed all ident ifiers were deleted and destroyed. Per IRB protocol, the quality of life questionnaire, demographic, and medical information forms are to be kept for 5 years, and stud y findings will be available to participants upon request. Data Analysis Data were analyzed using Statisti cal Analysis System (SAS). A confidential, password-secured database was used for data entry, management, and data analysis. Frequency dist ributions of the data were used to check for missing values, outlie rs, inconsistencies. Descriptive statistics of the data were provided An independent t-test was used to compare means of QOL scores between the experimental group and control group at baseline as well as at 90 days. Bivariate correlations were
82 done among the independent and dependent va riables. BNP data were log transformed as necessary. Results of the study were reported as aggregate numbers, and particip ants were not identified. Hypothesis 1: It was hypothesized that clinicians knowledge or lack of knowledge of BNP levels at the time of the baseline clinic visit would make a difference in the quality of life scores between the experimental and the control group at 90 days. An independent t-test was used to examine the relationship between experimental and control groups and qualit y of life scores at 90 days. Hypothesis 2: It was hypothesized that clinicians knowledge or lack of knowledge of BNP levels at time of CHF clinic visit would affect hospital LOS on all hospital admissions of CHF patients within 90 days. An independent t-test was used to examine the difference in the mean LOS between the experimental and control groups
83 Chapter Four Results This chapter presents the results of this study on the effects of clinic physician knowledge of BNP levels on participants quality of life at the 90-day follow-up. Preliminary analyses are discussed. These included bivariate correlations, checking normality distributio ns, checking for errors in data entry, missing data or outliers. This is follow ed by the discussion of the demographic characteristics and clinical pr ofile of the participants. Next, the statistical analyses are discussed in depth, and outcomes fo r each hypothesis are presented. Procedure Participants were recruited by the PI from a heart failure clinic in Northeast Florida, where 112 subjects were assessed for eligibility. Two potential participants were excluded because of high serum creatinine levels, and two refused to participate. Enrollment spanned from October 2005 to August 2006, and the 90-day follow-up period was co mpleted inn November 2006. The intended number of participants to be enrol led was 62 per group for a total of 124 for both groups. Because of staffing and clinic scheduling issues, such as decrease in the frequency of clinic days, fewer patients were enrolled in the study than intended. After enro lling patients with due process, questionnaires were distributed by the PI. Completion of the questionnaire took an average of 15 minutes. The survey consisted of 10 demographic questions and 21 QOL survey questions.
84 The next step in the procedure was fo r the PI to draw blood. The blood was processed according to the procedure described in Chapter 3 (Procedure for Performing the BNP Test). After blood wa s drawn, a physician examined each participant. To minimize error, only one physician performed all the patient assessments. The participants were then allocated to either the control group or the experimental group. Randomization wa s done by a random number generator. The participants and the physician were blinded to the randomization status. After randomization, the BNP level result s for those in the experimental group were given to the physician. The physician did not receive the BNP results for the control group. Sample A total of 108 participants completed the Minnesota Living with Heart Failure questionnaire and had their BNP leve ls blood drawn at baseline. Of the 108, 57 (53%) participants were assigne d to the experimental group and 51 (47%) participants to the control group at ti me one (baseline). At time two (followup), 16 participants did not complete the st udy. Of these, 10 (15%) left the study and 6 (5%) died. Of the 10 who left the study, 6 were from the experimental group and 4 were from the control group. This left 50 participants in the experimental group and 42 participants in the control group, a total of 92 participants who completed the study at 90 days. Due to the concerns about the effects of attrition, data were examined for those who did not complete the study. T-test analyses between completers and
85 non-completers showed that there were no significant differences in age, BNP, ejection fraction, QOL or the physical and emotional subscales at baseline. Therefore, it appears that t he loss of these patients di d not affect the overall outcome. Figure 3 illustrates t he trial sample flow diagram.
Figure 3. The flow of participants through the trial. Enrollment (n = 108) Excluded (n = 4) Not meeting inclusion criteria (n = 2) Refused to participate (n = 2) Total allocated (n = 108) Screenedfor eligibility (n = 112) Allocated to Control (n = 51) 86 Allocated to Experimental (n = 57) Allocation Lost to follow-up (n = 4) Reasons = unable to reach at 90 days. Discontinued intervention (n = 5) Reasons = Death Lost to follow-up (n = 6) Reasons = unable to reach at 90 days. Discontinued intervention (n = 1) Reasons = Death Follow-Up Total Analyzed Control (n = 42) Analysis Total Analyzed Experimental (n = 50) Description of Baseline Demographic Characteristics The results for the demographic data are discussed next for the total sample (N = 92). Demographic data were collected to include the following characteristics: [a] age, [b] ethnicity, [c] gender, [d] education level, [e] insurance
87 payers, [f] marital status, and [g] occupat ion. All participants completed their demographic forms at baseline. Data analysis was conducted, and frequency distribution of values of each variabl e was performed. Utilizing the frequency distributions was helpful in examining missing data or revealing the number of missing values for each variable that may be due to problems in data collection. Patient demographics and clinical prof ile at baseline were compared for equivalence between the two gro ups by an independent t-test and X 2 as appropriate (SAS Institute, Version 8. 2, Gary, North Carolina, 2001). A X 2 was used to analyze the categorical/nominal variables of gender, ethnicity, marital status, occupation, education, insur ance payers and NYHA classification. The independent t-test was used to anal yze the continuous variables of age, ejection fraction, BNP levels and QOL scores. The four assumptions of the t-test which include independence, scale of measurement used, normality, and homogeneity were used to validate whether t he data met the criteria for using the independent t-test. The BNP values were highly skewed. They ranged from 9 pg per ml to 5000 pg per ml at baseline. The mean was 458 pg per ml (SD = 759). However, the mode was 101 pg per ml; this showed the effects of the outliers of the mean. Of the total sample, 31% had BNP values of less than or equal to 101 pg per ml, and 59% had values ranging from greater than 101 pg per ml to 1000 pg per ml. The remaining 10% (n = 10) had BNP values ranging from 1110 pg per ml to 5000 pg per ml. These last ten data drastically affected the mean; therefore, the mode has been reported. Because of the skewness, it was
88 deemed necessary to do a log transformation of the BNP value before using it in statistical analyses. The demographic and clinical characteristics of the expe rimental group (n = 50) and the control group (n = 42) at baseline are summarized in Table 6. There were no significant differences by group in gender, ethnicity, marital status, education, occupation, or insurance payers. Additional ly, there were no significant differences on the clinical profile between groups with regard to age, ejection fraction, and BNP levels. Ther efore, random assignment of study participants appeared to successfully minimize differences between the groups at baseline.
Table 6 Baseline Demographic and Clinical Characteristics of Patients by Group Note. *SD = Standard Deviation Variables Experimental Group n = 50 Control Group n = 42 P value Age Mean SD* 63.3.5 65.4.4 .410 Gender (%) Female Male 24 (48%) 26 (52%) 18 (43%) 24 (57%) .621 Ethnicity (%) African American Caucasian Asian 26 (52%) 23 (46%) 1 (2%) 23 (55%) 19 (45%) 0 (0%) .645 Marital Status (%) Single Married Divorced Widow/Widower 3 (6%) 19 (38%) 14 (28%) 14 (28%) 7 (17%) 16 (38%) 8 (19%) 11 (26%) .364 Occupation (%) Retired Disabled Working 36 (72%) 12 (24%) 2 (4%) 27 (64%) 10 (24%) 5 (12%) .354 Education (%)
Table 6 (continued) Variables Experimental Group n = 50 Control Group n = 42 P value NYHA Classification a NYHA II NYHA III 18 (36%) 32 (64%) 12 (29%) 29 (71%) .496 Ejection Fraction (%) Mean SD* Serum Creatinine (mg/dl) b Mean SD 27.5 7.3 1.2 0.37 25.9 9.4 1.0 0.40 .385 .054 BNP levels(pg/ml) c LogBNP levels (pg/ml) Mean SD* 407 713 4.9 1.44 452 598 5.4 1.30 .744 .131 Note. *SD = Standard Deviation a NYHA = New York Heart Association Classification b mg/dl = milligrams per deciliter c BNP = b-type natriuretic peptide; measured in picograms per milliliter (pg/ml) The mean age for the total sample was 64.5 years (range 24-94 years, SD = 12.4, N = 92). By racial identity, 49% were African American, 42% were Caucasian, and 1% was Asian. There were more married participants in both the experimental and control groups (n = 35, 38%) than single (n = 10, 11%), divorced (n = 22, 24%), or widowed/widower (n = 25, 27%). More of the total sample completed high school (39%) than did not (25%). Additionally, 36% had some college. Of the total study sample, 68% were retired, 24% were disabled, and 8% were employed. The majority of the study participants from both groups had Medicare and Medicaid insurance coverage (n = 75, 82%), private insurance (n = 11, 12%), military insurance (n = 2, 2%) and Shands Clinic card (n = 4, 4%). 90
91 Description of Baseline Clinical Prof ile of the Participants (N = 92) For both study groups, the majority ( 67%) were classified as NYHA III and, 33% were classified as NYHA II. T he median and the mode for the total distribution were both 3.0. The NYHA IIIB were coded as NYHA III for statistical purposes. Of particular interest, the NYHA classification was only weakly correlated with BNP levels ( r = .25, p = .009) and the log BNP levels ( r = .22, p = 0.03). Data on both the ejection fraction and serum creatinine levels were obtained within one year from office visit. The overall mean ejection fraction for this sample of HF patients was 26.7% (SD = 8.3, median = 28.0, mode = 30). The overall mean serum creatinine levels for the total sample was 1.1 milligrams per deciliter (mg/dl) (SD = .39, median = 1.0, mode = 0.8). Again, the mean BNP was 458 pg per ml, and the mode was 101 pg per ml. In su mmary, participants had a poor ejection fraction and normal seru m creatinine levels as required by the inclusion criteria. However, they also had BNP levels that barely crossed the threshold for heart failure. Description of Dependent Variables at Baseline Quality of life was measured using the Minnesota Living With Heart Failure questionnaire. The questionnaire had a total of 21 questions and two subscale scores. The physical subscale dimension score consisted of items 2-7, 12, and 13. The emotional subscale dimension score consisted of items 17-21. The total QOL score ranged from 0-105; t he total physical subscale score ranged from 0 to 40; and the total emotional subscale sco re ranged from 0 to 25 (Rector et al.,
92 2002). The remaining 8 questions (range of 0-40) consisted of general questions. An independent t-test of the QOL variables dem onstrated equivalence at baseline. The mean scores for QOL (o verall, physical, and emotional) are summarized in table 7.
Table 7 Baseline Descriptives and Reliabilities of QOL Scores by Group Note. a MLHFQ = Minnesota Living With Heart Failure Questionnaire b min-max = Minimum and maximum scores Variables MLHFQ a Overall MLHFQ, Physical MLHFQ, Emotional Experimental Group (n = 50) mean SD median mode min-max b 45.9 29.8 48.0 13.0 3.0-101 19.7 13.0 20.5 0.0 0.0 40 9.76 7.91 9.5 0.0 0.0 -25 Control Group (n = 42) mean SD median mode min-max 48.6 29.3 53.0 6.0 11.0-93 21.1 14.2 23.5 0.0 0.0 40 11.2 8.37 10.5 0.0 0.0 -24 Cronbachs alpha 0.94 0.95 0.89 t value 0.60 0.49 0.90 P value 0.54 0.62 0.37 Bivariate correlation was examined next. As expected, the data revealed an inverse relationship between ejection fraction and BNP levels (r = -0.28, p = .006). However, the magnitude was not as strong as expected. That is, participants with the lower ejection fractions demonstrated higher BNP levels. There was a direct relationship between NYHA classification and QOL. As NYHA increased, the QOL score increased; this demonstrated a corresponding deterioration in the quality of life. There was a weak but significant correlation between NYHA and BNP levels (r = .23, p = 0.02). Based upon literature findings 93
94 from larger sample sizes, it was antici pated that the correlation would be stronger. Table 8 summarizes the bivariate correlations among the variables. Table 8 Bivariate Correlational Analysis of Baseline Variables Bivariate Variables (N = 92) QOL Baseline BNPL Age Ejection Fraction Serum Creatinine NYHA QOL Baseline 1.00 BNPL 0.25** 1.00 Age -0.09 0.21* 1.00 Ejection Fraction -0.14 -0.28*** 0.23* 1.00 Serum Creatinine 0.10 0.22* 0.16 -0.07 1.00 NYHA 0.50*** 0.23* 0.00 -0.12 0.16 1.00 Note P value = *0.05, **<0.01, ***<0.0001 Discussion of Results by Hypo theses at the 90-Day Follow-Up Research Hypotheses: Hypothesis 1: It was hypothesized that clinicians knowledge or lack of knowledge of BNP levels at time of clin ic visit would make a difference in the quality of life scores between the experimen tal group and the control group at 90 days. An independent t-test between experim ental and control groups was used to compare mean QOL scores at 90 days. Based on the results of the t-test,
95 there was no significant differenc e in the QOL scores at 90 days ( t = -0.79, df = 90, p = 0.43). Table 9 summarizes the re sults of the two groups at 90 days. Table 9 Comparison of QOL Score Means by Group at 90 Days Variables Cronbachs alpha Experimental Group 90 days n = 52 (mean SD) Control Group 90 days n = 40 (mean SD) t Value p Value MLHFQ, overall 0.94 37.5 29.8 32.7 27.9 -0.79 0.43 Note MLHFQ = Minnesota Living With Heart Failure Questionnaire Therefore, physician knowledge of BNP levels at time of clinic visit did not make a difference in QOL scores at 90 day s for either group. It was noted that QOL scores went down for each group; this indicated an impr ovement of their overall QOL at 90 days. However, this improvement was equal for both groups. The pre and post paired t-test for the experimental group was significant ( t = .5.07, df = 90, p = .001). For the control group the pre and post t-test was significant as well ( t = 4.62, df = 41, p = 0.00). These findings will be discussed in depth in the next chapter. Hy pothesis one was not supported. Hypothesis 2: It was hypothesized that clinicians knowledge or lack of knowledge of BNP levels at the time of CHF clinic visit would affect hospital length of stay (LOS) on all hospital admissions of CHF patients within 90 days. Portions of the day were considered as one hospital day. The experimental group had a total of 9 participants who were admitted for heart fa ilure for a total
96 of 31 hospitalization days within the 90-day follow-up period. The control group had a total of 2 participants who were admi tted for heart failure for a total of 6 hospitalization days within the 90-day fo llow-up period. An i ndependent t-test showed that there was no signific ant difference in the mean LOS ( t = 1.10, df = 90, p = ns). Table 10 displays the number of hospital days per group assignment. Table 10 Hospital Length of Stay by Group Assign ment Group Number of Patients Admitted (n = 11) Admission Rate Percent (%) Total Days In Hospital Mean Hospital LOS Experimental Group (n=50) 9 18 31 3.4 Control Group (n=42) 2 5 6 3.0 Note LOS = length of stay Ancillary Findings Mortality There were a total of six deaths between enrollment and the 90-day followup. By the 90-day follow-up, 12% (n = 5) in the control group had died, and 2% (n = 1) from the experimental group had died. There was a significant difference in the mortality rate with the control group having more deaths than the experimental group ( t = 1.99, df = 90, p = .04). These deaths were attributed to complications of heart failure. Table 11 displays the number of deaths by group assignment.
97 Table 11 Total Number and Percent of Deaths per Group Assignment Sample Deaths Number/ Percent by group Experimental Group (n = 50) 1 (2%) Control Group (n = 42) Total (N = 92) 5 (10%) 6 (5%) To summarize, hypotheses one and two were not supported, but an ancillary finding showed a significant differ ence in the mortality rate. Additionally, BNP levels and QOL scores were significantly correlated but with a weak magnitude for both groups at baseline. A more in-depth discussion of findings, limitations, implications for nursing, and recommendations for future study will be described in chapter five.
98 Chapter 5 Discussion, Conclusions, and Recommendations Introduction This chapter presents a broad discu ssion of the findings, conclusions, implications, limitations, and recommendat ions for future research. In the discussion section, hypotheses one and two are discussed in detail. Several plausible explanations are offered for the findings of each hypothesis. The factors that limit the generalizability of the st udy are discussed. Finally recommendations for future research are included. Discussion Hypothesis 1 Previous research on QOL and its re lation to BNP levels in HF has been widely studied. However, little has been known about whether a physicians knowledge or lack of knowledge of BNP levels affects QOL scores and hospital LOS at 90 days. It was hypothesized that if physicians were aware of BNP levels, then appropriate treatment would circum vent the development and progression of HF. Additionally, timely treatment w ould help alleviate and ease the symptoms of HF and might improve the patients quality of life. In the current study, the physicians knowledge of BNP levels at t he time of the clinic visit did not have any significant effect on the QOL scores fo r either the experimen tal or the control group at 90 days. Despite the fact that no significant association was observed between the experimental and control gr oup at 90 days, the data indicated a decrease in the mean QOL scores at 90 days (37.46 28.67) as compared to
99 the mean QOL scores at base line (46.87 29.63) for bo th groups. Because the QOL scale is reversed, this indicated that there was a pos itive change in QOL scores during the 90-day interval. There are several plausible explanati ons for this outcome First, the data showed a QOL mean of 45.9 29.8 and 48.6 29.3 for the experimental and control groups, respectively. Although, these mean QOL scores at baseline demonstrated an impaired QOL (based on the range of 0-105), they may have had little room for improvement. The patients QOL might already have been quite positive. It is possible that thes e HF patients may have adapted to their chronic disease and adjusted their perception of their QOL. This lack of room for improvement reflects a ceiling effect. Second, another possible explanation for failing to detect a significant difference between the experimental and c ontrol groups on QOL scores at 90 days could have been that the same ph ysician was treating both groups. The patients were already well managed, as evidenced by the BNP level of 101 pg per ml as mode. The clinician continued to provide optim um care to each group, and knowing current BNP val ues at the clinic visit may not have altered the clinicians overall treatment plan for either group. Third, it is possible that if t he participants were newly diagnosed HF patients at baseline, t he finding might have been different since new patients might not have adapted to their functional limitations and QOL Perhaps at this point, quality of life is a not a stable c onstruct in newly diagnosed HF patients.
100 Fourth, sample size ma y not have been large enough to detect a significant difference in QOL scores between the two groups. Although hypothesis one was not supported, the lack of association between physician knowledge of BNP levels is an unusual finding, since previous studies suggested that c linicians knowledge of BNPguided treatment provided alleviation of symptoms and impr ovement of QOL (Troughton, 2000). Hypothesis 2 The data revealed no significant di fference on the hospital LOS when comparing groups. In this study, the ex perimental group had a higher frequency of admission when compared to the control group; however, this was not statistically significant. Perhaps the ph ysicians awareness of the BNP levels for those participants in the experimental gr oup prompted closer monitoring; this could have accounted for the increased number of hospitalizations. However, it does not explain the non-significant differ ence in the length of stay across the two groups. Three possible reasons are proposed; an obvious one is that the sample size of those admitted was quite limited. A second possible explanation is that the event rate of hospitalization wa s too low, and the study was not powered for this outcome. Expanding the interv al beyond 90 days might have allowed for more events to occur. Another possibl e explanation could be that the admitted HF patients were treated by different attending physicians with differing modalities of treatment. The results of the second hypothe sis was an unexpected finding since it has been suggested by Cheng et al. (2001) and Mueller et al. (2004) that
101 physician knowledge of BNP levels predict s shorter hospital LOS. Additionally, findings from a previous study conclude d that patients who did not receive BNPguided treatment experienced more card iovascular event hospitalizations and deaths (78%) than those receiving BNPguided treatment ( 22%) (Troughton et al., 2000). BNP levels should not be interpret ed in isolation, but as part of the whole clinical assessm ent of the patient. Ancillary Finding Additionally, the study revealed an ancillary finding, demonstrating that there was a statistically significant asso ciation between BNP at baseline and the physical subscale scores at baseline (r = .24, p = 0.01). This corroborated the idea that increased BNP levels indicated more severe HF symptoms as reflected in the QOL scores. Regardless of w hether patients were randomized to the experimental or control groups, the data reflected that as BNP level increased, patients experienced more impaired qualit y of life and physical health. These findings were consistent with previous research concerning elevated BNP level and its association with bad prognosis and poor quality of life (Morrison et al. 2002; Ninuma et al., 1998; Wieczorek et al., 2002; Lubarsky & Mandell, 2004; Hirata et al., 2001; Teboul et al., 2 004; Sagnella, 1998; Vanderheyden et al., 2004; Heidenreich et al., 2004; Mair et al., 1999; Valle et al., 2005). Findings of this study are similar to those of Mue ller et al. (2004), Steg et al. (2005), and Maisel et al. (2002), who concluded that clinicians knowledge of BNP levels as an additive diagnostic tool along with ot her clinical information is useful in assessing QOL. According to Heidenrich et al. (2004), Maisel et al. (2001), and
102 MacMahon et al. (2002), timely detection of BNP levels may lead to early diagnosis and treatment of CHF, thereby decreasing readmissions and improving patients quality of life. The results of this study extend these same findings to this clinic sample: increased BNP level directly correlated with QOL scores; an increased QOL score indicated more functional impairment. Another incidental finding was that t he mean of NYHA II HF patients for this sample was higher (460 pg per ml) than reported by Maisel et al. (2001) (332 pg per ml). These results provided evidence that the participants in the NYHA II classification were sicker than t hose in the same NYHA II classification reported by Maisel et al. (2001). The mean for the NYHA III patients of this study was 569 pg/ml, similar to the mean of 590 pg per ml for NYHA III as reported by Maisel et al. (2001). Although the NYHA cl assification was made within 6 months of assessments, the discrepancy between t he mean BNP of this sample when compared to the mean of BNP sample report ed by Maisel et al. (2001) highlights the subjective validity of the classification. So even though the t-test showed no statistical difference at baseline betw een the two groups in the current study, there may have been a subjective difference in the clinical picture that affected the treatment by the physician and therefore, the results. Additionally, data from th is study showed a higher number of deaths in the control group as compared to the experimental group at 90 days ( t = 1.99, df = 90, p = .04). A possible explanation is the physicians lack of awareness of BNP values; this may have distorted the assessment of the HF patient. Troughton et al. (2000) concluded that death rates and hospital admissions were
103 fewer in the BNP-guided group than in the clinical group (19 vs. 54, p = .02). Ishii et al. (2003) concluded that BNP levels were strong predictors of cardiac events and were associated with mortality rates. Th erefore, further st udy is warranted to determine whether physicians lack of kno wledge of BNP levels is related to increased heart failure mortality. Consequently, the use of BNP rapid assay and its modest cost ($26/ test), is crucial in comparison to heart failures enormous economic healthcare burden to society. Mueller et al. (2006) concluded that the cost effectiveness of BNP-guided testi ng in heart failure patients showed that total treatment cost of $7,930 for the BNP-guided group was si gnificantly reduced from the $10,503 incurred by the control group ( p <0.004). Limitations Several limitations of this study warr ant caution in interpreting the results. One major limitation is that the PI did not collect data on medications used for this cohort or track any change in medi cations from baseline to 90 days. A retrospective preliminary pilot study conducted at the same HF clinic indicated that these patients were receiving the optimum pharmacological treatment for HF. It was the researchers assumption that knowledge of BNP levels would somehow prompt the physici an to improve their care Perhaps the assumption that physician knowledge of BNP leve ls would have affected the physicians decision on the treatment rendered to the patient was incorrect. In retrospect, there were several variables that s hould have been controlled and analyzed. One such variable was the presence of cardiac devices like the biventricular pacemakers. As noted earlier, medication hi story was not obtained for all patients
104 at baseline or 90 days. These patients we re likely receiving optimum medications for heart failure; however, changes in the me dications were not tracked at the 90day interval. Such changes could have in fluenced any hospital admissions, LOS, and QOL. Another variable that could have been monitored was presence of comorbidities in these HF patients. Thes e comorbidities, such as arthritis or emphysema, could have significantly impacted QOL. Another limitation to the study was its setting. The study sample was recruited from one clinic in the Southeastern United States, limiting generalization of findings to other clinics and geographical areas. This was a unique clinic population consisting largely of elderly, African-American men who were relatively unwell. This limited heterogeneity amongst subjects. This sample may be similar to those who live in the nor theastern region of Fl orida yet different from those who live in the southeastern region of Florida. Additionally, the study might have benefited from a larger sample si ze in order to better address attrition and be powered for a low base rate of hos pitalizations. Given the low base rate for hospitalizations, a larger sample si ze might have permitted detection of a statistically significant difference bet ween groups in the number of hospital admissions and LOS. Furthermore, there is a likelihood that participants of the study may have been enrolled in other heart failure studi es in the past and may have been exposed to the same MLWH F questionnaire. Also, there is a possibility of excessive systematic e rror in the MLWHF questionnaire and a possibility of random error occurring in the BNP rapid assay machine. Lastly,
105 there exists the possibility of investigat or bias since the PI was not blinded to group assignment. Implications for Nursing The implications drawn from this ex perimental study are presented in this section. As evidenced by the results, increased BNP levels are negatively associated with QOL in patients with hear t failure. Based on this information, clinical nurses need to understand that BNP levels may be directly related to volume overload. Increased volume overload begets increased symptoms and increased functional limitati ons. Understanding the relati onship of BNP levels in relation to these symptoms may help del ay the progression of the disease, alleviate deterioration of functional status, and improve QOL. Thus, nurses need to keenly observe signs and symptoms of hear t failure exacerbation; they need to know that BNP values are an additive diagno stic tool to be used in the context of all other available clinical information. Furthermore, knowledge of BNP leve ls empowers patients to make informed consumer decisions regarding their health conditions, such as importance of daily weights, complianc e with medications, exercise, and proper diet. Physical indicators of heart failure ar e not as obvious as indicators of other disorders. For example, although HF pat ients can do daily weights, no objective at-home blood tests are currently availabl e to assess volume status, but patients with diabetes have immediate access to point-of-care tests for blood glucose. Awareness of BNP level and its relations hip to the pathophysiology of the
106 disease process would allow the patient to have better control and management of this chronic and debilitating condition. Recommendations for Future Study Based upon the results of this study as well as the review of literature, the following recommendations are made for futu re research. It is recommended that [a] this study be replicated using a larger sa mple size in order to detect any effect of knowledge of BNP levels to QOL and to detect low event rate such as hospitalizations; [b] the follow-up peri od be extended for more than 90 days to allow for the occurrence of more cardiovascular ev ents; [c] correlations be retested between BNP and current NYHA classification to validate the use of BNP as a clinical marker, [d] other physiol ogic variables that heart failure patients experience (such as sleep disorders, impl antation of cardiac devices, and effects of life-sustaining medications on pat ients QOL) be examined, [e] adding a psychological component such as a depression questionnaire may provide a more holistic view of how heart failure patie nts experience their quality of life; and [f] conducting a structural equation modeling procedure would provide a more comprehensive statistical analysis when attempting to evaluate the physical, psychological, and emotional dimensions when assessing the quality of life of heart failure patients.
107 Summary Managing congestive heart failure conti nues to present a challenge for nurses and other clinicians. The impact of heart failure on patients as well as its economic burden to society is undeniable Despite a high level of public awareness of this disease, a majority of the population is unaware of their risks of developing heart failure. Nurses can educat e heart failure patients, so they may empower themselves in circumventing the development of this clinical syndrome. Even though the hypotheses of this st udy were not supported, incidental findings warrant further re search. The limitations, as previously discussed, should be addressed so that future studies in the same area have more power. The inclusion of additional variables woul d further enhance the st udy. Quality of life remains a key area of heart failure re search and a focal point in the treatment and management of heart failure.
108 References Adams, K., Lindenfield, J., Ar nold, J., Massie, B., Baker, D., Mehra, M., et al. (2006). Executive Summary: HFSA 2006 Comprehensive Heart Failure Practice Guidelines. Journal of Cardiac Failure, 12 (1), 10-38. Anand, I., DPhil, O., Florea, V., & Fischer, L. (2002). Surrogate end points in heart failure. Journal of the American College of Cardiology, 39(9), 1414-1421. Journal of American College of Cardiology, 38, 2101-2113. Ancheta, I. (2006). B-type nat riuretic peptide rapid assay: a diagnostic test for heart failure. Dimensions of Critic al Care Nursing, 25 (4), 149-154. American College of Cardiology. ACC/ AHA 2005 Guideline Update for the Diagnosis and Management of Chr onic Heart Failure in the Adult (2005). Circulation, 112e, e154-e235. American College of Cardiology. ACC/ AHA 2005 Guideline Update for the Diagnosis and Management of Chr onic Heart Failure in the Adult Summary Article. A Report of the American College of Cardiology/American Heart Asso ciation Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure) Journal of American College of Cardiology, 46 (6), 1116-1143. American Heart Association. 2005 Hear t Disease and Stroke Statistics 2005 Update. Dallas, Texas. Am erican Heart Association: 2005
109 Artinian, N., Artinian, C., Saunders M. (2004). Identifying and treating depression in patients with heart fa ilure. Journal of Cardiovascular Nursing, 19 (6 Suppl), S47S56. Aspromonte, N., Feola, M., Scardovi, A., Coletta, C., Deri, A., Giovinazzo, P., et al. (2006). Early diagnosis of congestive heart failure: Clinical utility of b-type natri uretic peptide testing associated with Doppler echocardiography. Journal of Card iovascular Medicine (Hagerstown), 7 (6), 406-413. Bennett, S., Pressler, M., Hays, L., Firest ine, L. & Huster, G. (1997). Psychosocial variables and hospitaliz ation in persons with chronic heart failure. Progress in Cardiova scular Nursing, 12 (4), 4-11. Berger, R., Huelsman, M., St recker, K., Bojic, A., Moser, P., Stanek, B. et al. (2002). Btype natriur etic peptide predicts sudden death in patients with chronic heart failure Circulation, 105 (20), 2392. Bhalla, V., Willis, S., & Maisel, A. (2004) B-type natriuretic peptide: the level and the drug-partners in the diagnos is of congestive heart failure. Congestive Heart Failure, 10 (1 Suppl 1), 3-27. Biosite Incorporated (2004). Triage BNP test Product Insert. New dimensions in diagnosis, p 1-21. Borsody, J., Courtney, M., Taylor, K., Ja irah, N. (1999). Usi ng self-efficacy to increase physical activity in pati ents with heart failure. (1999). Home Healthcare Nurse, 17 (2) 113-118.
110 Bosworth, H., Steinhauser, K., Orr, M., Lindquist, J., Grambow, S., Oddone, E. (2004). Congestive heart failure pat ients perceptions of quality of life: the integration of physical and psychological factors. Aging & Mental Health, 8 (1), 83-91. Braunstein, J., Anderson, G., Gerstenb lith, G., Weller, W. Niefeld, M., Herbert, R., et al. (2003). Nonc ardiac comorbidity increases preventable hospitalizations and mortality among Medicare beneficiaries with chronic heart failure. Journal of the American College of Cardiology, 42 (7), 1234-1237. Brenden, C., Hollander, J., Guss, D., Mc Culluogh, P., Nowak, R., Green, G. et al. (2006). Gray zone BNP leve ls in heart failure patients in the emergency department: results from the Rapid Emergency Department Heart Failure Outpatient Trial (REDHOT) multicenter study.American Heart Journal, 151(5), 1006-1111. Brostrom, A., Sromberg, A., Maelst rom, U., Fridlund, B. (2004). Sleep difficulties, daytime sleepiness, and health related quality of life in patients with chronic heart failure Journal of Cardiovascular Nursing, 19 (4), 234 242. Cardarelli, R., & Lumicao, T. (2003). B-type natriuretic peptide: a review of its diagnostic, prognostic, and therapeutic monitoring value in heart failure for primary care physicians. The J ournal of the American Board of Family Practice, 16 (4), 327-333.
111 Carels, R. (2004). The association bet ween disease severity, functional status, depression and daily quality of life in congestive heart failure patients. Quality of Life Research, 13 (1), 63-72. Center for Disease Contro l and Prevention (2005). Heat failure Fact sheet/ CVH. Retrieved January 20, 2006, from Web site: http://www.cdc.gov/cvh/lib rary/fs_heart_failure.htm Center for Disease Control and Prevent ion (2005). National Center for Chronic Disease prevention and Health Promotion. Retrieved March 5, 2006, from Web site: http://www.cdc.gov/od/oc/media/fact/f990305.htm Center for Disease Control and Prevent ion (2005). National Center for Health Statistics. Retrieved fr om January 20, 2006, from Web site: http://www.cdc.gov/nchs/ Center for Medicare and Medicaid Se rvices. Healthcare Financing Review, 2001 Medicare and Medicaid Statistical Supplement. Retrieved December 2, 2005, from Web site: http://www.cms.hhs.gov/HealthCareFinancingReview/ Center for Medicare and Medicaid Se rvices (2003). Medicare Coverage Database. Retrieved Septem ber 13, 2003, from Web site: http://www.cms.hhs.gov/mcd/viewl mrp.asp?lmrp_id = 13097&lmrp_vers n = 3 &show = all
112 Cheng, V., Kazanegra, R., Garcia, A., Lenert, L., Krishnaswamy, P., Gardetto, N., et al (2001). Rapid bedside test for B-type peptide predicts treatment outcomes in patients admitted for decompensated heart failure: a pilot study. Journal of the American College of Cardiology, 37 (2), 386-391. Chiong, J., & Miller, A. (2002). Renin-angiote nsin system antagonism and lipid lowering therapy in card iovascular risk management. Journal of Renin-Angiotensisin Aldos terone System, 3(2), 96-102. Clark, DO., Tu., W., Weiner M. and Murray, M. ( 2003). Correlates of healthrelated quality of life among lowe r-income, urban adults with congestive heart failure. Heart & Lung, 32 (6) 391-401. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2 nd ed). Hillsdale (NJ): Erlbaum. Costelo, J., Boblin S (2004). What is the experience of men and women with congestive heart failure? Canadian J ournal of Cardiovascular Nursing, 14 (3), 9-20. Daniels, L., Ballad, V., Clapton, P., Holl ander, J., Gus, D., McCullough, P., et al. (2006). P-type natriuretic peptide (BNP) levels and ethnic disparities in perceived severity of heart failu re: results from t he Rapid Emergency Department Heart Failure Trial (R EDHOT) multicenter study of BNP levels and emergency department decision making in patients with shortness of breath. Journal of Cardiac Failure, 12 (4) 281-285.
113 Dao, Q., Krishnaswamy, P., Kazanegra, R., Harrison, A., Amirnovin, R., Lenert, L., et al.(2001). Utility of B-type natriur etic peptide in the diagnosis of congestive heart failure in an urgent care setting. Journal of American College of Cardiology, 37 (2), 379-385. Dracup K., Walden J., Stevenson L., Breac ht, M (1992). Qualit y of life in patients with advanced heart failure. Heart Lung Transplant, 11 (2), 273279. Eichhorn, E. & Bristow, M. (2001). The Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS ) trial. Current Control Trials Cardiovascular Medicine, 2(1):20-23. Ekman, I., Fagerberg B., Lundman, B. (2002). Health related quality of life and sense of coherence among elderly patients with severe chronic heart failure in comparison with heal thy controls. Heart & Lung, 31 (2), 94-101. Evangelista, L., Moser, D., Westlake, C ., Hamilton, M., Fonarow, G., Dracup, K. (2006). Impact on o besity and quality of life in patients with heart failure. European Journal of Heart Failure (article in press). Ferrans, C and Powers, M. (1992). Psychom etric assessment of the quality of life index. Research in Nursi ng and Health, 15 (1), 29-38. Ferrier, K., Campbell, A., Yee, B., Ric hards, M., O Meeghan, T., Weatherall, M., et al. (2005). Sleep disordered breathing occurs frequently in stable outpatients with congestive hear t failure. Chest, 128 (40), 2116-2122.
114 Grigoni, F., Carigi, S., Gr andi, S., Potena, L., Cocco lo, F., Bacchi-Reggiani, L., et al. (2003). Distance between patients subjective perceptions and objectively evaluated disease severity in chronic heart failure. Psychotherapy & Psychomatics, 72 (3), 166-170. Gonzalez, B., Lupon, J., Herrero s, J., Urrutia, A., Altimi r, S., Coll, R., et al. (2005). Patients education by nurse: What we really do achieve? European Journal of Cardiovascular Nursing, 4 (2), 107-111. Gorkin, L., Norvell, N., Rosen, R., Charles, E., Shumaker, S. McIntyre, K., et al. (1993). Assessment of quality of life as observed from the baseline data of the Studies of Left Vent ricular Dysfunction (SOLVD) Trial Quality of LIfe Substudy. The American Journal of Cardiology, 71, 1069-1073. Gottlieb, M., Khatta, M., Friedmann, E. Einbinder, L., Katzen, S., Baker, B., et al. (2004). The influence of age, gender, and race on the prevalence of depression in heart failure patient s. Journal of the American College of Cardiology, 43 (9), 1542-1549. Harrison, M., Browne, G., Roberts, J., T ugwell, P., Gafni, A., Graham, I. (2002). Quality of life of individu als with heart failure: a randomized trial of the effectiveness of two model s of hospital to home transition. Medical Care, 40 (4), 271-282. Havranek, E., Ware, M., Lowes, B. ( 1999). Prevalence of depression in congestive heart failure. The American Journal of Cardiology, 84 (3), 348-350.
115 Heart Failure Society of America (2006). Executive summary: HFSA 2006 Comprehensive Heart failure guideline. Journal of Cardiac Failure, 12 (1) 10-38. Heart Failure Society of America ( 2002). The stages of heart failure. NHYA Classification. Retrieved March 6, 2006, from Web Site: http://www.abouthf.org/questions_stages.htm#top Heidenreich, P., Gubens, M., Fonarow, G., Konstam, M., Stevenson, L., & Shekelle, P. (2004). Cost-effe ctiveness of screening with B-type natriuretic peptide to identify patients with reduced left ventricular ejection fraction. Journal of Amer ican College of Cardiology, 43 (6), 1019-1026. Hennekens, C. & Buring, J. (1987). Epidemiology in Medicine (1 st ed). Boston (MA): Little Brown & Company. Hulley, S. & Cummings, S. (1988). De signing Clinical Research. An epidemiological approach. Williams & Wilkins. Hirata, Y., Matsumoto, A., Aoyagi, T., Yamaoki, K., Komuro, I., Suzuki, T., et al. (2001). Measurement of plasma brain natriuretic peptide as a guide for cardiac overload.Cardiovascu lar Research, 51 (3), 585-591. Hogenhuis, J., Voors, A., Jaarsma, T. Hillege, H., Hoes, A., van Veldhuisen, D. (2006). Low preval ence of B-type natriuretic peptide levels < 100 pg/mL in patients with heart fa ilure at hospital discharge. American Heart Journal, 151 (5), 1012, e1-e5.
116 Howie, J., Caldwell, M. & Dracup, K. (2003). The measurem ent of brain natriuretic peptide in heart fa ilure precision, accuracy, and implications for practice. AACN Cli nical Issues: Advanced Practice in Acute & Critical Care. Wound Ca re: Heart Failure, 14(4), 520-531. Hunt S, Baker D, Chin M, et al. ACC/AHA guidelines fo r the evaluation and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Hear t Association task force on practice guidelines. 2001; [ 56 screens]. Available at: http://www.acc.org/clinical/guidelines Accessed May 14, 2004. Hunt, S., Abraham, W., Chin, M., Feldman, A., Francis, G., Ganiats, T. et al. (2005). ACC/AHA 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluati on and Management of Heart Failure): developed in collaboration with the American College of Chest Physicians and the International Society for Heart and Lung Transplantation: endorsed by the Heart Rhythm Society.Circulation, 112 (12), e154-e235. Ishii, J., Cui, W., Kitagawa, F., Kuno, T., Nakamura, Y., Naruse, H., et al. (2003). Prognostic value of combi nation of cardiac Troponin T and Btype natriuretic peptide after initiation of treatment in patients with chronic heart failure. Cli nical Chemistry, 49, 2020-2026.
117 Jaarsma, T., Halfens, R., Tan, F., AbuSaad H., Dracup, K., Diederiks, J. (2000). Self-care and qual ity of life in patients with advanced heart failure: the effect of a supportive educational intervention. Heart & Lung, 29 (5), 319-330. Johansson, P., Dahlstrom., U., Brostr om, A. (2006). The measurement and prevalence of depression in chronic hear failure. Progress in Cardiovascular Nursing, 21 (1), 37-43. Jiang, W., Alexander, J., Christopher, E., Kuchibhat la, M., Gaulden, L., Cuffe, M., et al. (2001). Relati onship of depression to increased mortality and rehospitalization in patients with congestive heart failure. Archives of Internal Medicine, 161 (15), 1849-1856. Karlsson, M., Edner, M., Henricksson, P., Mejheart M, Persson, H., Grut, M. et al. (2005). A nurse-based man agement program in heart failure patients affects female persons with cognitive dysfunction most. Patient Education Counseling, 58 (2) 146-153. Kodiath, M., Kelly, A., Shivel y, M. (2005). Improving quality of life in patients with heart failure: An innovative behavio ral intervention. Journal of Cardiovascular Nursing, 20 (1) 43-48. Koenig, H. (1998). Depression in hospita lized older patient s with congestive heart failure. General Hosp ital Psychiatry, 20, 29-43.
118 Konstam, V., Salem, D., Pouler, H., et al (1996). SOLVD In vestigators. Baseline quality of life as a predictor of mortality and hospitalization in 5025 patients with congestive heart failure. The American Journal of Cardiology, 78 (8), 890. Konstam, V., Moser, D., De Jong, M. (2005). Depression and anxiety in heart failure. Journal of Cardiac Failure, 11 (6), 455-463. Kutzleb, J. & Reiner, D. (2006). The impact of nurse-directed patient education on quality of life and functi onal capacity in people with heart failure. Journal of the American Ac ademy of Nurse Practitioners, 18 (3), 116-123. Latini, R., Masson, S., Anand, I., Judd, D., Maggioni, P., Chiang, Y., et al. (2002). Effects of valsartan on ci rculating brain natriuretic peptide and norepinephrine in symptomatic ch ronic heart failure: the Valsartan Heart Failure Trial (Val-Heft) Circulation, 106 (19), 2454-2458. Lee, T., Yu, D., Woo, J., Thompson, D. (2005). Health-related quality of life in patients with congestive heart fa ilure. European Journal of Heart Failure, 7 (3), 419-422. Lesman-Leegte, I., Jaarsm a, T., Sanderman, R., Linssen, G., van Veldhuisen, D. (2006). Depressi ve symptoms are prominent among elderly hospitalized heart failure patients. European Journal of Heart Failure, (article in press).
119 Lubarsky, L. & Mandell, K. (2004). B-ty pe natriuretic peptide: practical diagnostic use for evaluation of ventricular dysfunction. Congestive Heart Failure, 10 (3), 140-143. Luttik, M, Jaarsma, T, Veeger, N., Ve ldhuisen, V. (2005). For better and for worse: Quality of life impaired in HF patients as well as in their partners.European Journal of Cardiovascular Nursing, 4 (1), 11-14. Luttik, M., Jaarsma, T., Moser, D, Sanderman, R., Veldhuisen, V. (2005). The importance and impact of soci al support on outcomes in patients with heart failure: an overview of the literature. Journal of Cardiovascular Nursing, 20 (3),162-169. MacMahon, K., & Lip, G. ( 2002). Psychological factor s in heart failure: A review of literature. Archives of Internal Medicine, 162 (3), 509-516. Maisel, A. (2001). B-type natriuretic pept ide levels: a potential novel white count for congestive heart failure. J ournal of Cardiac Fa ilure, 7 (2), 183-193. Maisel, A. (2002). B-Type natriuretic peptide m easurements in diagnosing congestive heart failure in the d yspneic emergency department patient. Reviews in Cardiovascular Medicine, 3 Suppl 4, S10-S17.
120 Maisel, A., Hollander, J., Gu ss, D., McCullogh P., Nowa k R., Green, G., et al. (2004). Primary results of the Rapid Emergency Department Heart Failure Outpatient Trial (REDHO T): A multicenter study of B type natriuretic peptide levels, emer gency department decision making, and outcomes in patients presenting with shortness of brea th. Journal of College of Cardiology, 44 (6), 1328-1333. Maisel, A., Krsihnaswamy, P., Nowak, R., et al. (2002). Rapid measurement of B-type natriuretic peptide in t he emergency diagnosis of heart failure. New England Journal of Medicine, 347 (3), 161-168. Mair, J., Friedl, W., Thomas, S., Pusc hendorf, B. (1999). Natriuretic peptide in assessment of left ventricula r dysfunction. Scandinavian Journal of Clinical and Laboratory In vestigation, 230,132-142. Majani, G., Oierobon, A., Giardini A., et al. (1999). Relationship between psychological profile and cardiologi cal variables in chronic heart failure. European Heart Journal, 20, 1579-1586. Mark, D. & Felker, G. (2004). B-type nat riuretic peptide a biomarker for all seasons? The New England Journal of Medicine, 350 (7), 718-720. Martensson, J., Karlsson, J, Fridl und, B. (1997). Male patients with congestive heart failure and their c onception of the life situation. Journal of Advanced Nursing, 25 (3), 579-586. Martensson., J, Karlsson, J. Fridl und B. (1998). Female patients with congestive heart failure: How they c onceive their life situation. Journal of Advanced Nursing, 28 (6) 1216-1224.
121 Martensson, J., Dracup, K., Canary, C., Fridlund, B. (2003). Living with heart failure: depression and quality of life in patients and spouses. The Journal of Heart and Lung Trans plantation, 22 (4), 460-470. Martensson, J., Stromberg, A., Dahlstro m, U., Karlsson, J., Fridlund, B. (2005). Patients with heart failure in primary healthcare: effects of a nurse-led intervention on health-rela ted quality of life and depression. European Journal of Heart failure, 7 (3), 393-403. Masson, S., Latini, R., Anand, I., Vago, T., Angelici, L., BArlera, S., et al. (2006). Direct Comparison of BType Natriuretic Peptide (BNP) and Amino-Terminal proBNP in a Large Population of Patients with Chronic and Symptomatic Heart Failure: The Valsartan Heart Failure (Val HeFT) Data. Clinical Chemistry, 52(8),1528-1538. McCullough, P., Hollander, J. Nowak, R., Storrow, A. Duc, P., Omland, T., et al. (2003). Uncovering heart failure in patients with a history of pulmonary disease: rationale for t he early use of B-type natriuretic peptide in the emergency department. Academic Emergency Medicine, 10(3), 198-204. Mejhert, M., Kahan, T., Persson, H., Edner, M.(2006) Predicting readmissions and cardiovascular ev ents in heart failure patients. International Journal of Cardiology,109 (1),108-113.
122 Morrison, L., Harrison, A., Krishnaswa my, P., Kazanegra, R., Clopton, P., Maisel, A.(2002). Utility of a rapid b-type natriuretic peptide assay in differentiating congestive heart fail ure from lung disease in patients presenting with dyspn ea. Journal of American College of Cardiology, 39 (2), 202-209. Moser, D. (2002). Psychological factor s and their association with clinical outcomes in patient with heart failu re: why clinicians do not seem to care. European Journal of Cardiovascular Nursing, 1 (3)183-188. Moser, D., Doering, L., Chung, M. Vul nerabilities of patients recovering from an exacerbation of chronic heart failure. American Heart Journal, 150 (5), 984 e7984 e13. Mueller, T., Laule-Klian,K., Schindler,L., Klia, T., Rodr iquez, D., Sholer, A et al. (2006). Cost-effectiveness of btype natriuretic peptide testing in patients with acute dypsnea. Archives of internal Medicine, 166(10), 1081-1088. Mueller, T., Gegenhuber, A., Poelz, W ., & Haltmayer, M. (2005). Diagnostic accuracy of Btype natriuretic peptide and amino terminal proBNP in the emergency diagnosis of heart failure. Heart, 91 (5), 606-612. Mueller, C., Scholer, A., LauleKilian, K ., Martina, B., Schi ndler, C., Buser, P., et al. (2004). Use of btype nat riuretic peptide in the evaluation and management of acute dyspnea. New England Journal of Medicine, 350 (7), 647-654.
123 Murberg, T., Bru, E., Svebak, S., Tvet eras, R. & Aarsland, T. (1999). Depressed mood and subjective sympt oms as predictors of mortality in patients with congestive heart failure : a two -year followup study. Internal Journal of Psychiatry in Medicine, 29, 311-326. Murberg, T. & Bru, E. (2001). Social relations hips and mortality in patients with congestive heart failure. Journal of Psychosomatic Research, 51, 521-527. National Center for Health Statisti cs (2004). Heart Failure Fact Sheet. Retrieved January 12, 2006, from Web site: http://www.cdc.gov/cvh/libra ry/pdfs/fs_heart_failure.pdf Ninuma, H., Nakamura, M. & Hiramori, K. (1998). Plasma b-type natriuretic peptide measurement in a multiphasic health screening program. Circulation, 90 (2), 89-94. Prahash, A. & Lynch, T. (2004). B-type natriuretic peptide: a diagnostic, prognostic, and therapeutic tool in heart failure. American Journal of Critical Care, 13 (1), 46-53. Ramakrishnan, S., Kothari, S., Bahl, V. (2005). Heart failure Definition and diagnosis. Indian Heart Journal, 57 (1), 13-20. Rector, T. Kubo, S. & Cohn, J. (1987) Patients self-assessment of their Congestive heart failure: content, reliability and validity of a new measure, the Minnesota Living with Heart Failure Questionnaire. Heart Failure, 3, 181-209
124 Rector, T. & Cohn, J. (1992). Assessment of patient outcome with the Minnesota Living with Heart Failure questionnaire: reliability and validity during a randomized, double-blind, placebo-controlled trial of pimobendan. American Heart Journal, 124 (4), 1017-1025. Rector, T. & Cohn, J. (1992). Assessment of patient outcome with the Minnesota Living with Heart Failure questionnaire: reliability and validity during a randomized, double-blind, placebo-controlled trial of pimobendan. American Heart Journal, 124 (4), 1017-1025. Rector, T., Kubo, S. & Cohn, J. (1993) Validity of the Minnesota Living with heart failure questionnaire as a measure of therapeutic response to enalapril or placebo. The American Jour nal of Cardiology, 71, 1106-1107. Rector, T. (2005). A conceptual model of quality of life in relation to heart failure.Journal of Cardiac Failure, 11 (3), 173-176. Rector, T., Anand, I., Cohn, J. (2006) Relationships between clinical assessments and patients perception of the effects of heart failure on their quality of life. Journal of Cardiac Failure, 12 (2), 87-92. Richards, A., Lainchbury, J., Nicholls M., Troughton, R., & Yandle, T. (2002). BNP in hormone-guide treatm ent of heart failure. Trends in Endocrinology and Metabolism, 13 (40), 151-155. Riegel, B., Dickson, V., Ho ke, L., Mcmahon, J., Reis, B., Sayers, S. (2006). A motivational counseling approach to improving heart failure self-care: mechanisms of effectiveness. Journal of Cardiovascular Nursing, 21 (3) 232-241.
125 Riegel B., Moser, D., Carls on, B., Deaton, C., Armola, R., Sethares, K. et al. (2003).Gender differences in quality of life are minimal in patients with heart failure. Journal of Cardiac Failure, 9 (1), 42-48. Riegel, B., Moser, D., Glas er, D., Carlson, B., Deaton, C., Armola, R., et al. (2002). The Minnesota Living with Heart Failure questionnaire: sensitivity to differences and responsiveness to intervention intensity in a clinical population. Nursing Research, 51(4), 209-218. Sagnella, G. (1998).Measurement and significance of circulating natriuretic peptide in cardiovascular disease. Clinical Science, 95 (5), 519-529. Sarvimki A. & Stenbock-Hult B. (1993). Caring: An Introduction to Health Care from a Humanistic Perspec tive. Almqvist & Wiksell, Stockholm SAS Institute Incorporated (2001). The SAS System (Version 8.2, Windows Workstation) [Computer software ]. Cary, North Carolina : SAS institute Incorporation. Sneed, N., Paul, S., Michel, Y., Van Bakel, A., Hendrix, G. (2001). Evaluation of 3 quality of life measurement tools in patients with chronic heart failure. Heart & Lung, 30, 332-340. Sullivan, M.m Simon, G., Spertus, J. Russo, J. (2002). Depression-related costs in heart failure care. Archiv es of Internal Medicine, 162 (16), 1860-1866. Tabbizar, R., & Maisel, A. (2002). The impact of B-type natriuretic peptide levels on the diagnoses and management of congestive heart failure. Current Opinion in Ca rdiology, 17(4), 340-345.
126 Taniguchi, T., Kawasaki, T., Miyai, N., Kamitani, T., Kawasa ki, S., Sugihara, H. (2006). Brain natriuretic pepti de and QRS duration as a predictor for cardiac events in patients with heart failure. Journal of Cardiology, 47 (6), 277-283. Teboul, A., Gaffinel, A., Meune, C., Greffet, A., Sauval P., & Carli, P. (2004). Management of acute dyspnoea: use and feasibility of brain natriuretic peptide (BNP) assay in prehospita l setting. Resuscitation, 61(10), 91-96. Troughton, R., Frampton, C., Yandle, T., Espiner, E., Nicholls, G., & Richards, A. (2000). Tr eatment of heart failure guided by plasma amino terminal brain natriuretic peptide (N-BNP) concentrations. The Lancet, 355 (9210), 1126-1130. Trupp, R., Hardesty, P., Osborne, J., Shel by, S., Lamba, S., Ali, V., et al. (2004). Prevalence of sleep disor dered breathing in a heart failure program. Congestive Heart Failure, 10 (5), 217-220. Turvey, C., Klein, D., Pies, C. (2006) Depression, physical impairment, and treatment of depression in chr onic heart failure. Journal of Cardiovascular Nursing, 21 (3), 178-185. Valle, R., Aspromonte, N., Feola, M., Milli, M., Canali, C., Giovanazzo, C., et al. (2005). B-type natriuretic pepti de can predict the medium-term risk in patients with acute heart failure and preserved systolic function. Journal of Cardiac Failure, 11 (7), 498-503.
127 Vanderheyden, M., Bartunek, J., & Goet hals, M. (2004). Brain and other natriuretic peptides: Molecular aspects. European Journal of Heart failure, 6 (3) 261-268. Venugopal, J. (2001). Cardia c natriuretic peptides-hope or hype? Journal of Clinical Pharmacy and T herapeutics, 26, 15-31. Wieczorek, S., Wu, A., Chritenson, R., Krishnaswamy, P., Rosano, T., Hager, D., et al.(2002). A rapid B-type natriuretic peptide assay accurately diagnoses left ventri cular dysfunction and heart failure: A multicenter evaluation. Americ an Heart Journal, 144 (5), 834-839. White, R. (2005). The role of brain natriuretic peptide in systolic heart failure. Dimensions of Critical Ca re Nursing, 24 (4), 171-174. Zambroski, C. (2004). Hospice as an alte rnative model of care for older patients with end stage heart failure. Journal of Cardiovascular Nursing, 19, 76-83. Zambroski, C., Moser, D., Bhat, G., Ziegler, C. (2005). Impact of symptom prevalence and symptom burden on qual ity of life in patients with heart failure. European Journal of Cardiovascular Nursing, 4 (3), 198-206. Zile, M., Baicu, C., Bonnema, D. (2005) Diagnostic heart failure: Definitions and terminology. Progress in Cardio vascular Disease, 47(5), 307-313.
128 Appendices Appendix A Demographic Information Form 1. Data Collection Date_____________ 2. Subject code_____________ 3. Date of clinic visit ________ 4. Age (circle one):  age 21 to 40  age 41 to 60 age 61 to 80  age 90 and older 5. Ethnic Background (circle one):  Asian  African-American  Caucasian  Hispanic  Native American  other _____________________ 6. Education (highest level achieved):  Less than high School  High school Diploma  Some college  4 year Degree  Masters Degree  Doctoral Degree 7. Occupation (circle one):  Management  Bu siness, Finance and Administration [3 ] Natural and Applied Scienc es and related occupations  Health Occupations  Social Science Education, Government Service & Religion  Ar t, Culture, Recreation and Sport  Sales and Service  Occ upations unique to Primary Industry  Occupations unique to Processing, Manufacturing and Utilities 8. Marital Status:  Single  Married  Divorced  Widow/ Widower
129 9. Insurance:  Yes  No 10. Insurance Payers:  Private Insurance  Medicare  Medicaid  Tricare Military
Appendix B Subject Code_______ Date_________ 130
131 Appendix C Medical Information Form Data Collection Date_____________ Subject code___________________ 1. Hospital Admission [A] Yes [B] No 2. If admitted: [A] Date of admission__ ___________ [B ] Hospital_ ____________________ [C] Diagnosis___________________ [C] Date of discharge_____________ 3. NYHA Classification [A] NYHA I [B] NYHA II [C] NYHA III [D] NYHA IV
Appendix D Letter of Support from Medical Director 132
133 Appendix E Informed Consent Form IRB# UFJ-2005-50/ USF 103600 Informed Consent to Pa rticipate in Research The University of Florida Health Science Center Jacksonville, Florida 32209 You are being asked to particip ate in a research study. This form provides you with information about the study. The Principal Investigator (Irma B. Ancheta) will describe this study to you and answer all of your ques tions. Before you decide whether or not to take part, read the information below and ask questions about anything you do not understand. 1. Name of the Participant 2. Title of Research Study The relationship between of b-type natriur etic peptide (BNP) levels and hospital length of stay and quality of life in congestive heart failure patients 3. Principal Investigator(s), Address and Teleph one Number(s) Irma B. Ancheta RN, MSN 4120 Shoal Creek Lane East Jacksonville, Fl. 32225 (904) 645-3862 (904) 629-7923 4. Source of Funding or Other Material Support None
134 IRB# UFJ-2005-50/ USF 103600 5. What is the purpose of this research study? The purpose of the study is to find out if physicians knowledge of your blood levels of the result of a particular blood test will affect the management of your illness, hospital length of stay and qualit y of life. This blood test measures the level of BNP, or B-type natriuretic peptide present in yo ur blood. BNP is a substance secreted from lower chambers of the he art, in response to change s in pressure that occur when heart failure deve lops or worsens. You are being asked if you are interested in taking part of this study because you have been diagnosed with congest ive heart failure. Your doctors and researchers involved with heart failure patients around the country would like to learn more about this condition. Approximately 160 patie nts will participate at our site for this study. 6. What will be done if you ta ke part in this research study? You will be asked to respond to two ques tionnaires that would take about 15 minutes of your time. The questionnaire will ask you how living with congestive heart failure has affected your life. Y ou will be asked to respond to the same questionnaire in 90 days in the privacy of this clinic office. Approximately one cc (about teaspoon) of your blood will be drawn from a vein to test for BNP levels in the clinic. You will be randomly (like a flip of a coin) put in a group who will have the BNP blood test disclosed to the physician or in the group whose BNP levels will not be disclosed to t he physician. You will allow us to follow up and record an y heart failure hospitalization in the next 90 days. Information from your c linic visit and medical record will be collected such as age, gender, ethnicity, education, marital status, insurance payers, the severity of your congestive heart failure, the date of any potential hos pital admission, what the diagnosis was and the date of hospital discharge. You will be contacted by phone or during y our normal clinic visits at monthly intervals for the next three months to fi nd out how you are doing. During these follow-ups we will be inquiring about whether or not you have been in the hospital.
135 IRB# UFJ-2005-50/ USF 103600 7. What are the possible discomforts and risks? Blood Sampling The risks may include discomfort at the puncture site; possible bruising and swelling around the site; and uncommonly, faintness; and rarely, an infection. Potential risks may involve violation of c onfidentiality, embarrassment or discomfort that might arise if the data regarding your diagnosis, hospital length of stay and quality of life survey is unnecessary expos ed. Your answers are confidential and we will take all necessary precautions to ensure your confidentiality. 8a. What are the possib le benefits to you? There are no direct benefits to you for taking part in this study, but the information gained from your participati on may help other patients. 8b. What are the possible benefits to others? As an outcome of this study, the possible benefi ts to others might include improving quality of life and re ducing hospital length of stay fo r future congestive heart failure patients. 9. If you choose to take part in this research study, will it cost you anything? The study will not cost you anything. 10. Will you receive compensation for taki ng part in this research study? You will not receive compensation for taking part in this study.
136 IRB# UFJ-2005-50/ USF 103600 11. What if you are inju red because of the study ? If you experience an injury that is directly caused by this study, only professional care that you receive at the University of Florida Health Scie nce Center will be provided without charge. You may call th e Shands Cardiovascular Center at (904) 244-4198 or your principa l investigator (Irma B. An cheta) at (904) 645-3862. However, hospital expenses wi ll have to be paid by you or your insurance provider. No other compensation is offered. If you are harmed while taking part in the stud y: The state of Florida enjoys what is called "sovereign immunity." This means th at you usually cannot sue the state of Florida. However, the state has waived sovereign immunity (agreed to be sued) in certain situations. One of those situations is if a st ate employee, such as your study doctor or other USF employee is neglig ent in doing his or her job in a way that harms you during the st udy. The money that you mi ght recover from the state of Florida is limited in amount. You can also call the University of South Florida Self Insurance Programs (SIP) at 1-813-974-8008 if you think: You were harmed because you took part in this study. Someone from the study did somethi ng wrong that caused you harm, or didnt do something they should have done. Ask the SIP to look into what happened. 12. What other options or treatments ar e available if you do not want to be in this study? If you do not want to take part in this study, tell the prin cipal Investigator (Irma B. Ancheta) you do not want to participate in the study. 13a. Can you withdraw from this research study? You are free to withdraw your consent and to stop participat ing in this research study at any time. If you do withdraw your co nsent, there will be no penalty and you will not lose any benefits you are entitled to. If you decide to withdraw your consent to pa rticipate in this research study for any reason, you should cont act: Irma Ancheta at (904) 645-3862. If you have any question s regarding your right s as a research subject, you may phone the Un iversity of Florida Institutional Review Board (IRB) office at (904) 244-3136 and the University of South Florida In stitutional Review Board (IRB) office at (813) 974-5638.
137 13b. If you withdraw, can information ab out you still be used and/or collected? If you withdraw, we would like to retain the surveys an d questionnaires that you have responded during the in itial interview. We woul d like your permission to continue to collect information from your physician or your medical records. Please initial: Yes_____ If I choose to wi thdraw from the study you may continue to use my personal data that has been collected. No_______If I choose to with draw from the study, you ma y not continue to use my personal data that has been collected. 13c. Can the Principal Investigator wi thdraw you from this research study? You may be withdrawn from the study for th e following reasons: [A] If knowledge of any unexpected or unexplained side effects that affect your safety becomes known; [B] If on receiving new in formation about the treatment, y our PI (Irma B. Ancheta) might consider it to be in your best intere sts to withdraw you fr om the study without your consent if they judge t hat it would be better for your health. An example of this is when the study shows that an increas ed number of congestive heart failure patients die as a result of failure to report BNP levels. 14. How will your privacy an d the confidentiality of y our research records be protected? Authorized persons from the University of Florida and the Un iversity of South Florida, have the legal right to review your research records and will protect the confidentiality of those records to the ex tent permitted by law. Otherwise, your research records will not be re leased without your consent unless required by law or a court order. If the results of this research are published or presented at scientific meetings, your identity wi ll not be disclosed. 15. If you agree to particip ate in this research stud y, what protected health information about you may be collected used and disclos ed to others? Your diagnosis, number of days in hospit al and results of y our BNP levels may be collected, used and di sclosed to others. IRB# UFJ-2005-50/ USF 103600
138 16. For what study-related purposes will your protected heal th information be collected, used and di sclosed to others? Your protected health information may be co llected, used and disclosed to others to determine eligibility for the st udy. Other protected health information collected, used and disclosed to others woul d include study information su ch as your diagnosis and number of days in hospital. 17. Who will be authorized to collect use and disclose to others your protected health information? Your protected health information may be collected, used, and disclosed to others by the study principa l investigator (Irma B. Ancheta, RN) and ot her employees at the University of South Florida other professionals at the University of Flor ida or Shands Hospital that provide study-related treatment or procedures the University of Florida and University of South Florida Inst itutional Review Boards. 18. Once collected or used, who may yo ur protected health information be disclosed to? Your protected health information may be given to: The study investigator, co-inv estigators, supervisors and the University of Florida and University of Sout h Florida Institutional Review Boards (IRB). 19. If you agree to partic ipate in this research, ho w long will your protected health information be colle cted, used and disclosed? Your information will be maintained in a password prot ected database for 5 years. 20. Why are you being asked to author ize the collection, use and disclosure to others of your protected health information? Under a new Federal Law, researchers cannot collect, use or disclose any of your protected health information for research un less you allow them to by signing this consent and authorization form. IRB# UFJ-2005-50/ USF 103600
139 21. Are you required to sign this consent and authorizat ion and allow the researchers to collect, use and disclose (g ive) to others of your protected health information? No, and your refusal to sign will not affect your treatment, enrollment, or eligibility for any benefits outside this rese arch study. However, you cannot participate in this research unless you allow the collection, us e and disclosure of your protected health information by signi ng this consent/authorization. 22. Can you review or copy your protected heal th information collected, used or disclosed under this authorization? You have the right to review and copy your protected health information. However, you will not be allowed to do so un til after the stud y is finished. 23. Is there a risk that your protected health information could be given to others beyond your authorization? Yes. There is a risk that information received by authorized persons could be given to others beyond your authorization and not covered by the law. To ensure confidentiality all dat a will be recorded in a secure, password-protected data file, the PI (Irma B. Ancheta) and t he Institutional Review Boards (IRB) of the University of Fl orida and University of South Florida will have access to the database. 24. Can you revoke (cance l) your authorization for collection, use and disclosure of your protected health information? Yes. You can revoke your authorization at any time bef ore, during or after your participation in the research. If you revo ke, no new information will be collected about you. However, informati on that was already collect ed may be still be used and disclosed to others if the researchers have relied on it to complete and protect the validity of the research. You can revoke this authorization by givi ng a written request with your signature on it to the Principal Investigator. 25. How will the researcher(s) benefit from your being in this study? The PI (Irma B. Ancheta) wi ll learn more about how to care for cong estive heart failure patients and furt her the science. UFJ-2005-50/ USF 103600
140 26. Signatures As a representative of this study, I, I rma B. Ancheta, have explained to the participant the purpose the procedures, the possible be nefits, and the risks of this research study; the al ternatives to being in the study; and how privacy will be protected: __________________________________ _______ ___________ Signature of Person Ob taining Consent Date You have been informed about this studys purpose, procedures, possible benefits, and risks; the alternatives to being in the study; and how y our privacy will be protected. You will receiv e a copy of this form. You have been given the opportunity to ask questions before you sign and you have been told that you can ask other questions at any time. You voluntarily agree to partici pate in this study. By si gning this form, you are not waiving any of yo ur legal rights. __________________________________ ________ __________ Signature of Person Consenting Date
141 Appendix F Frequently Asks Questions 1. What is the purpose of the study? The purpose of the study is to find out if this blood test can help improve your care and keep your heart failure symptoms under control. 2. What does the study involve? The study will involve having to have a small amount of blood drawn, about 1 ml, from a vein to test your blood levels for heart failure. This will only be done once during your clinic visit. You will also be asked to answer a survey of 21 questions pertaining to the physical, psychological and emotional aspects of having to live with congestive heart failure. This survey will be done once during the clinic visit and again in 90 days in the clinic or via telephone. You will also be asked to report to the Principal Investigator (PI) any hospital admissions wit hin the next 90 days. 3. What happens if I do not want to participate in this study? The study is entirely voluntary. Non-parti cipation will not in any way hinder your standard of care. 4. Can I refrain from the study anytime? You may elect to refrain from the study at any time. You may withdraw from the study at any time by telephoning or writing the Principal Investigator. 5. Will it cost me to participate in the study? You will not have any ad ditional costs because of your involvement in this study. 6. How long is the study? The study will run for 90 days from time of clinic visit. 7. Will I have the chance to k now the results of the study? Once study is completed the results of t he study will be available to you if you so desire.
142 About the Author Irma B. Ancheta, Ph.D., R.N., an assistant professor of nursing at the University of North Florida. She received her Bachelor s Degree in Nursing from Cebu Velez College of Nursing in Cebu City, Philippi nes. She went on to complete a Masters Degree in Nursing at the University of Phoenix at Jacksonville Florida. Her research interest focuses on finding clinic al strategies that may help improve quality of life among congestive heart failu re patients. Dr. Ancheta resides in Jacksonville, Florida where she is actively involved in her community.