Journal of Cardiovascular Nursing

Vol. 29, No. 6, pp 555Y564 x Copyright B 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

Is Health-Related Quality of Life a Predictor of Hospitalization or Mortality Among Women or Men With Atrial Fibrillation? Deborah W. Chapa, PhD, ACNP-BC; Bimbola Akintade, PhD, ACNP-BC, MBA, MHA; Eleanor Schron, PhD, RN, FAAN, FSCT; Erika Friedmann, PhD; Sue A. Thomas, PhD, RN, FAAN Background: Little is known about predictors of mortality or hospitalization in women compared with men in patients with atrial fibrillation (AF). Although there are established gender differences in patients with coronary artery disease (CAD), differences have not been established in AF. Objectives: The aim of this study was to examine clinical and health-related quality of life (HRQOL) predictors of mortality and 1-year hospitalization in women compared with men with AF. Methods: Limited-use data from the National Institutes of Health/National Heart, Lung, and Blood Institute Atrial Fibrillation Follow-up Investigation of Rhythm Management clinical trial provided the sample of 693 patients with AF, 262 women and 431 men. Clinical predictors examined were heart failure (HF), CAD, left ventricular ejection fraction, diabetes, stroke, and age. Predictors of HRQOL included overall HRQOL (Medical Outcomes Study Short Form-36 physical [PCS] and mental component scores) and cardiovascular HRQOL using Quality of Life IndexYCardiac Version. Results: Mortality did not differ (women, 11.4%; men, 14.5%; 221 = 0.437, P = .509) according to gender, with mean 3.5-year follow-up. Different variables independently predicted mortality for women and men. For women, diabetes (hazard ratio [HR], 3.415; P = .003), HF (HR, 2.346; P = .027), stroke (HR, 2.41; P = .032), and age (HR, 1.117; P = .002), and for men, CAD (HR, 1.914; P = 02), age (HR, 1.103, P = G .001), worse PCS (HR, 1.089, P = .001), and worse Quality of Life IndexYCardiac Version score (HR, 1.402, P = .025) independently predicted mortality. One-year hospitalization (women, 38.9%; men, 36.4%) did not differ by gender (2 21 = 0.914, P = .339). Different variables independently predicted 1-year hospitalizationVfor women: diabetes (odds ratio [OR], 2.359; P = .022), worse PCS (OR, 1.070; P = .003), and rhythm control trial arm (OR, 2.111; P = .006); for men: HF (OR, 2.072; P = .007), worse PCS (OR, 1.045; P = .019), living alone (OR, 1.913; P = .036), and rhythm control trial arm (OR, 2.113; P G .001). Conclusion: Only clinical status predicted mortality among women; HRQOL and clinical status predicted mortality among men. Both clinical and HRQOL variables predicted hospitalization for women and men. Increased monitoring of HRQOL and interventions designed to target the clinical and HRQOL predictors could impact mortality and hospitalization. Nursing interventions may prove effective for modifying most of the predictors of mortality and hospitalization for women and men with AF. KEY WORDS:

arrhythmia, QLI-CV, quality of life, SF-36

Background and Significance Atrial fibrillation (AF) is a major global health problem that is increasing in prevalence and affects more

than 3 million Americans.1 It is also associated with an increased risk of stroke, dementia, heart failure (HF), and mortality.2Y4 Estimated annual costs of managing AF are $6.65 billion, which includes $4.88 billion

Deborah W. Chapa, PhD, ACNP-BC Assistant Professor, School of Nursing, George Washington University, Washington, DC.

Bimbola Akintade, PhD, ACNP-BC, MBA, MHA Assistant Professor, School of Nursing, University of Maryland, Baltimore.

Eleanor Schron, PhD, RN, FAAN, FSCT Director, Clinical Applications, Vision Research Program, Division of Extramural Research, National Eye Institute/National Institutes of Health, Bethesda, Maryland.

Erika Friedmann, PhD

The Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) is conducted and supported by the National Heart, Lung, and Blood Institute (NHLBI) in collaboration with the AFFIRM Investigators. This manuscript was prepared using a limited-access data set obtained by the NHLBI and does not necessarily reflect the opinions or views of the AFFIRM Study or the NHLBI. The authors have no conflicts of interest to disclose.

Correspondence

Sue A. Thomas, PhD, RN, FAAN

Deborah W. Chapa, PhD, ACNP-BC, School of Nursing, George Washington University, 2030 M Street NW, Suite 300, Washington, DC 20036 ([email protected]).

Professor, School of Nursing, University of Maryland, Baltimore.

DOI: 10.1097/JCN.0000000000000095

Professor, School of Nursing, University of Maryland, Baltimore.

555 Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

556 Journal of Cardiovascular Nursing x November/December 2014 in hospitalization expenses and $1.53 billion in outpatient management costs.5 The average annual cost of AF hospitalization per patient with AF in the United States is approximately $8700, with additional costs for other hospitalizations.6 Atrial fibrillation is primarily a disease of the elderly population, with the prevalence doubling with each decade of life after the age of 60 years. It occurs in 10% of the population older than 80 years.7,8 Women comprise more than 60% of the population in older age cohorts. The high prevalence in the older population is anticipated to increase the burden of AF on the healthcare system, with an estimation of a 2.5-fold increase in AF patients over the next 50 years.9 The burden of cardiovascular disease is increasing more among middle-aged and older women than among men.10 A great deal of attention has been focused on gender differences in coronary artery disease (CAD). In CAD, different presentation and management strategies have been investigated for women and men.10 For example, in stable outpatients with CAD, clinical presentation of a myocardial infarction (MI) differs. Women may present with fatigue and/or shortness of breath, whereas men may typically present with pain in a location specific to the area of the infarct. Effective treatment is similar, with dosages adjusted based on weight, renal function, and ideal body weight.11 Recent research examining gender difference in HF found similar age-adjusted mortality rates but different predictors of mortality depending on gender. Among women, aortic stenosis, pulmonary hypertension, and malignancy, and among men, diastolic dysfunction and chronic renal failure predicted mortality.12 Much less attention has been focused on gender differences in AF, the most common sustained cardiac arrhythmia.13 In patients with AF, little is known about differences in rates of or factors influencing mortality and hospitalization in women as compared with men. Most of the studies examining predictors of mortality or hospitalization in AF included largely male populations despite the larger proportions of women among older age cohorts. Five studies compared rates of mortality or hospitalization between genders in patients with AF. Four large studies examined differences between women and men with regard to mortality rates among people with AF, with inconsistent conclusions.3,14Y16 In 2 studies,3,14 men had greater all-cause mortality than women did, and in 2 studies, there was no difference in mortality15 or cardiovascular mortality16 between women and men. Baseline differences between women and men did not appear to be related to mortality in this limited number of studies. Women were significantly older in all studies and had more risk factors than men did in 3 of the 4 studies of mortality.14Y16 Only 1 large study17 either included data that provided differences between women and men in rates of hospi-

talization for AF or provided the data in a way that these rates could be computed. The rate of 1-year hospitalization for AF was higher for men than for women. The possibility of different contributions of healthrelated quality of life (HRQOL) to mortality or hospitalization in women compared with men with AF has not been established. Women with AF report significantly greater impairment in HRQOL, especially physical functioning, when compared with men.18,19 Quality-oflife impairment is not associated with heart disease severity in AF.19 None of the studies comparing rates of mortality or hospitalization between men and women examined HRQOL. The current study was designed to explore whether there are differences in clinical and HRQOL predictors of mortality or hospitalization for women and men with AF.

Methods Subjects This secondary data analysis used the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) Limited Access Database.20 The AFFIRM study was a National Heart, Lung, and Blood Institute international clinical trial designed to compare the effectiveness of 2 approaches to controlling AF on mortality and morbidity of patients with AF. The 2 approaches were (1) cardioversion and treatment with antiarrhythmic drugs to maintain sinus rhythm (rhythm control) and (2) the use of heart rate controlling drugs that allow the arrhythmia to persist (rate control). Patients were randomized to the treatment groups. Complete criteria and protocol are available in the AFFIRM methods paper.21 Patients were enrolled in the HRQOL substudy of AFFIRM from 56 clinical sites, which were randomly selected to be included in the study from the 213 AFFIRM clinical sites throughout the United States and Canada. Patients at these sites were invited to participate in the substudy and signed a separate informed consent. A total of 716 AFFIRM patients were enrolled in the HRQOL substudy. Of the 7401 patients approached to participate in the main AFFIRM trial, 4060 patients (54.9%) were enrolled.22 Of these, 845 were recruited from the sites randomized to the HRQOL substudy and 716 (84.7%) completed the HRQOL forms at baseline.23 Some patients (n = 23) potentially could have been identified based on information in the data file and were thus removed from the data set during preparation of the Limited Access Database.20 The AFFIRM limited-use data set provided the sample of 693 patients with AF. Women represented almost 38% (n = 262) of the study participants. Ages of both male and female participants ranged from 49 years to older than 80 years. Ages older than 80 years were top

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Hospitalization and Mortality in Patients With AF 557

coded as 80 years for inclusion in the limited-access data set to preclude identification of participants. Enrollment characteristics of the study participants according to gender are included in Table 1. Women in the study were significantly older and more likely to have a pacemaker but less likely to have a history of CAD or MI or left ventricular ejection fraction (LVEF) less than 50% than men did. Women reported significantly worse physical component score (PCS) and health and functioning subscale scores and better family subscales scores than men did.

invited to participate in the AFFIRM HRQOL study. The main results of the HRQOL substudy of the AFFIRM clinical trial have been reported.23 It did not examine gender differences in outcome. Clinical predictors examined included CAD, low LVEF, diabetes, stroke, HF, and age. Quality of life predictors included overall HRQOL assessed with the Medical Outcomes Study Short Form-36 as physical and mental function and cardiovascular health-related HRQOL assessed with the Quality of Life IndexYCardiac Version (QLI-CV). Details about these scales are available elsewhere.23

Study Design Patients with AF and at least 1 other risk factor for mortality or stroke, including age 65 years or older, history of MI, systemic hypertension, transient ischemic attack, stroke, HF, or diabetes or LVEF less than 40, were enrolled in AFFIRM. Complete inclusion criteria for the AFFIRM study have been published previously.21 Of the 213 sites participating in the AFFIRM study, 56 were randomly assigned to be included in the HRQOL substudy of the AFFIRM study. All patients who were randomized to either treatment from those sites were TABLE 1

Data Analysis Descriptive analysis included means and frequencies. Continuous variables were normalized before multivariable analysis. Initial bivariate analyses were conducted to examine the contributions of each clinical and psychosocial or HRQOL variable to mortality in Cox regression and to 1-year hospitalization within the first year in logistic regression analyses. Cox regression analyses for time to death and logistic regression analyses for 1-year hospitalization were used to examine

Enrollment Characteristics of Study Participants Women (n = 262)

Characteristic Minority Lives alone CAD Obese (BMI 930 kg/m2) MI HF LVEF e50 Current AF Hypertension Stroke Diabetes Smoker CABG Pacemaker "-Blocker Digoxin Rhythm control arm Died Hospitalized in year 1

Age MCS PCS SES Family Health PSP

Men (n = 431)

n

%

n

%

#2

P

21 87 72 54 33 42 27 92 190 32 41 18 17 21 88 98 120 31 102

8.0 33.6 27.5 20.6 12.6 16.0 10.3 35.1 72.5 12.2 15.6 6.9 6.5 8.0 33.6 37.4 45.8 11.8 38.9

23 54 185 107 83 74 101 204 299 62 88 61 78 19 132 148 223 62 157

5.3 12.6 42.9 24.8 19.3 17.2 23.4 47.3 69.4 14.4 20.4 14.2 18.1 4.4 30.6 34.3 51.7 14.4 36.4

1.967 43.31 16.654 1.623 5.189 0.152 15.595 9.940 0.776 0.655 2.446 8.557 18.564 3.898 0.467 0.426 2.299 0.914 0.437

.161 G.001 G.001 .203 .023 .697 G.001 .002 .373 .418 .118 .003 G.001 .048 .494 .514 .129 .339 .509

Mean

SD

Mean

SD

t

P

72.02 41.18 36.48 24.72 24.66 20.92 23.79

7.46 5.85 6.18 4.62 4.72 5.16 5.46

68.49 40.62 39.59 24.32 23.51 22.37 23.97

8.34 4.90 5.76 4.52 4.67 4.88 5.16

j5.757 j1.275 6.085 j1.098 j3.082 3.706 0.444

G.001 .205 G.001 .272 .002 G.001 .657

Health refers to the QLI health and functioning subscale score; Family, the QLI family subscale score. Abbreviations: AF, atrial fibrillation; BMI, body mass index; CABG, coronary artery bypass graft; CAD, coronary heart disease; HF, heart failure; LVEF, left ventricular ejection fraction; MCS, SF-36 mental component score; MI, myocardial infarction; PCS, SF-36 physical component score; PSP, QLI psychological-spiritual subscale score; QLI, Quality of Life Index; SES, QLI socioeconomic subscale score; SF-36, Medical Outcomes Study Short Form-36.

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

558 Journal of Cardiovascular Nursing x November/December 2014 interactions between gender and selected predictors. Based on significant interactions, analyses were stratified according to gender. Simultaneous analyses that included all clinical variables that predicted the respective outcomes at P G .20 were used to evaluate the independent contributions of these variables to mortality and to 1-year hospitalization. Parallel analyses were conducted for the psychosocial and HRQOL variables. For each outcome, final models were created by combining the significant and potentially influential (P G .20) clinical and psychosocial/HRQOL variables and then sequentially eliminating predictors that did not meet these criteria for inclusion.

Results Mortality Overall, mortality was 14.3% (93/693), with a mean follow-up of 3.5 years. There was no significant difference (2 21 = .437, P = .509) in mortality between women (11.4%, 31/262) and men (14.4%; 62/431). In the first year of participation in the study, 62.2% (426/693) were hospitalized within the first year and 3 died without hospitalization within the first year. They were eliminated from analysis of 1-year hospitalization. This case, a woman, was eliminated from further analyses of 1-year hospitalization. There was no significant difference (2 21 =.914, P = .339) in 1-year hospitalization between women (39.1%, 102/261) and men (36.4%, 156/431). Treatment Arm The effect of treatment arm on mortality did not differ according to gender (hazard ratio [HR], 0.718; P = .381), and as in the parent AFFIRM study, treatment arm did not predict mortality (HR, 1.032; P = .900). The effect of treatment arm was not related to gender (interaction HR, 1.388; P = .348). When gender differences in age and diabetes, key predictors of mortality in AFFIRM, were examined, the contribution of diabetes to mortality differed according to gender (gender HR, 0.668, P = .140; diabetes HR, 0.977, P = .942; gender  diabetes HR, 2.874, P = .035). On the basis of the interaction, women with diabetes had the highest risk of mortality and women without diabetes had the lowest risk of mortality. Men with diabetes had approximately a 2% greater risk of mortality than did men without diabetes. Compared with men without diabetes, women with diabetes had 88% greater risk and women without diabetes had 33% less risk of mortality. Compared with women without diabetes, men with diabetes had 36% greater risk, men without diabetes had 64% greater risk, and women with diabetes had 140% greater risk of mortality. The contribution of age to mortality did not differ according

to gender (gender HR, 2.668, P = .736; age HR, 1.109, P G .001; gender  age HR, 0.981, P = .622). A set of stratified Cox regression analyses were conducted to explore predictors of mortality among women and men with AF. For women and men with AF, trial arm did not predict mortality. For women, the clinical variables CAD, low LVEF, diabetes, stroke, HF, and age significantly predicted mortality in bivariate analyses (Figure 1). In bivariate analysis, none of the psychosocial or HRQOL variables predicted mortality among women with AF. However, PCS was a potentially influential variable and was included in further analyses. In multivariable analysis, diabetes, stroke, HF, and age were significant independent predictors of mortality among women (Table 2). Female AF patients with diabetes were 3.4 times as likely to die as those without diabetes, those with stroke were 2.3 times as likely to die as those without a stroke, and those with HF were 2.3 times as likely to die as those without HF. For every year older, women’s risk of dying increased by 4% after taking into consideration their diabetes, stroke, and HF status. No psychosocial or HRQOL variables made a significant independent contribution to women’s mortality. For the clinical variables, CAD and age significantly predicted mortality in bivariate analyses; hypertension was a potentially influential predictor (see Figure 1). In bivariate analyses, PCS was a significant predictor of mortality among men. Other potentially influential psychosocial or HRQOL predictors of mortality among men with AF included living alone, QLI socioeconomic subscale score, mental component score, and QLI psychological-spiritual subscale score (PSP). In multivariable analysis, age, CAD, PCS, and PSP were significant independent predictors of mortality among men (see Table 2). Male AF patients with CAD were 94% more likely to die than those without CAD. For every year older, a man’s risk of dying increased by 10% after taking the other predictors into account. Worse PCS and worse PSP were also associated with mortality among men with AF beyond the effects of CAD, age, and each other. Predictors of mortality in women and men differed. Predictors in women were diabetes, HF, and stroke. Predictors of mortality in men were CAD, worse PCS, and worse PSP. Hospitalization Within 1 Year The effect of treatment arm on hospitalization did not differ according to gender (odds ratio [OR], 1.023; P = .891), and as in the main AFFIRM study, treatment arm predicted hospitalization (OR, 1.75; P = .006). Hospitalization was higher in the rhythm control arm, where there were more adverse drug effects and hospitalization was required to assess QTC changes.24 Patients randomly assigned to the rhythm control arm were 75% more likely to be hospitalized than those in

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Hospitalization and Mortality in Patients With AF 559

FIGURE 1. Summary of Cox regression analyses examining individual predictors of mortality for women (n = 262) and for men (n = 431). Hazard ratios for mortality with an increase of 1 unit in each individual predictor are presented. Trial arm: rate control is reference category. CAD indicates coronary heart disease; CI, confidence interval; Family, QLI family subscale score; Health, QLI health and functioning subscale score; HF, heart failure; HTN, hypertension; LVEF, left ventricular ejection fraction; MCS, SF-36 mental component score; PCS, SF-36 physical component score; PSP, QLI psychological-spiritual subscale score; QLI, Quality of Life Index; SES, QLI socioeconomic subscale score; SF-36, Medical Outcomes Study Short Form-36. ySquare root transformation. zReflected.

the rate control arm. If strokes differed in the trial arms, because of suboptimal anticoagulation, hospitalizations due to these strokes would have been explained by differences between the trial arms. This confirms the importance of including trial arm as a predictor when examining the contributions of other variables to hospitalization. The effect of treatment arm was not related to gender (interaction OR, 1.261; P = .480).

In AFFIRM, the key predictors of hospitalization, independent of trial arm, were age and diabetes. When gender differences were examined, using these key predictors, the contribution of diabetes to hospitalization differed according to gender when controlling for treatment arm (gender OR, 1.014, P = .638; diabetes OR, 1.12, P = .638; gender  diabetes OR, 2.350, P = .047; treatment arm OR, 1.93, P G .001). On the

TABLE 2 Cox Proportional Hazards Regression Analysis for the Final Models for Predictors of Mortality for Women (n = 261) and for Men (n = 431) 95.0% CI for HR B Women Diabetes Stroke HF Age Men CAD Age PCS PSP

SE

Wald

P

HR

Lower

Upper

1.228 0.854 0.853 0.110

0.409 0.416 0.387 0.035

9.013 4.208 4.865 9.676

.003 .040 .027 .002

3.415 2.349 2.346 1.117

1.532 1.039 1.100 1.042

7.613 5.311 5.006 1.197

0.649 0.098 j0.085 j0.339

0.278 0.023 0.025 0.151

5.438 18.158 11.577 5.034

.020 .000 .001 .025

1.914 1.103 0.918 0.713

1.109 1.055 0.874 0.530

3.304 1.154 0.965 0.958

Trial arm: rate control is the reference group. Abbreviations: CAD, coronary artery disease; CI, confidence interval; HF, heart failure; HR, hazard ratio; PCS, physical component score; PSP, psychological-spiritual subscale score.

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

560 Journal of Cardiovascular Nursing x November/December 2014 basis of the interaction, women with diabetes had the highest odds of hospitalization and men without diabetes had the lowest odds of hospitalization. Men with diabetes had approximately 12% greater odds of hospitalization than did men without diabetes. Compared with men without diabetes, women without diabetes had 1% greater odds and women with diabetes had 168% greater odds of hospitalization. The contribution of age to hospitalization also differed according to gender when controlling for treatment arm (gender OR, 1.19, P = .308; age [mean centered] OR, 1.03, P = .019; gender  age [mean centered] OR, 0.946, P = .010; treatment arm OR, 1.956, P G .001). On the basis of the interaction, women at lower ages had the highest odds of hospitalization and men at lower ages had the lowest odds of hospitalization. For a man, every year above the mean age of 69.8 years increased the odds of hospitalization by approximately 3% compared with a man of average age. For women of the mean age, the risk of hospitalization was 19% above that for men of the same age. A woman whose age is 1 year more than the mean had odds of hospitalization that are 16% above those of a man of mean age. A set of separate stratified logistic regression analyses were conducted to explore individual predictors

of 1-year hospitalization controlling for treatment arm among women and men with AF. Among women with AF, trial arm was a significant predictor of 1-year hospitalization (Figure 2). Hospitalization was more likely for rhythm control. Diabetes was the only clinical variable that was a significant predictor of 1-year hospitalization for women after controlling for trial arm in bivariate analysis. Heart failure and older age were potentially influential predictors that were included in multivariable analysis. Among the psychosocial predictors, QLI socioeconomic subscale score, PCS, health and functioning, and QLI-CV were independent predictors of 1-year hospitalization after controlling for trial arm among women with AF. In simultaneous logistic regression analysis, diabetes and PCS were significant independent predictors of 1-year hospitalization among women with AF after controlling for treatment arm and PCS score (Table 3). Female AF patients with diabetes had 2.4 times the odds of hospitalization within 1 year compared with those without diabetes. Worse PCS scores were associated with increased odds of 1-year hospitalization among women with AF beyond the effects of trial arm and diabetes. Among men with AF, trial arm was a significant predictor of 1-year hospitalization (see Figure 2). Age,

FIGURE 2. Summary of logistic regression analyses examining the predictors of 1-year hospitalization in the first year for women

(n = 261) and for men (n = 431) after controlling for trial arm. Odds of hospitalization with an increase of 1 unit in each individual predictor are presented. Trial arm, with rate control as the reference category, was included as an additional predictor in all other analyses in this figure. CAD indicates coronary heart disease; CI, confidence interval; Family, QLI family subscale score; Health, QLI health and functioning subscale score; HF, heart failure; HTN, hypertension; LVEF, left ventricular ejection fraction; MCS, SF-36 mental component score; PCS, SF-36 physical component score; PSP, QLI psychological-spiritual subscale score; QLI, Quality of Life Index; SES, QLI socioeconomic subscale score; SF-36, Medical Outcomes Study Short Form-36. ySquare root transformation. zReflected.

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Hospitalization and Mortality in Patients With AF 561 TABLE 3

Logistic Regression Analysis for the Final Models for Predictors of 1-Year Hospitalization for Women (n = 261) and for Men (n = 431) 95% CI for OR

Women Trial arm Diabetes PCS Constant Men Trial arm Heart failure PCS Lives alone Constant

B

SE

Wald

P

OR

Lower

Upper

0.747 0.858 j0.067 1.491

0.272 0.375 0.023 0.843

7.542 5.250 8.680 3.126

.006 .022 .003 .077

2.111 2.359 .935 4.440

1.239 1.132 .894

3.599 4.915 .978

0.748 0.729 j0.044 0.649 0.570

0.217 0.271 0.019 0.310 0.745

11.862 7.201 5.482 4.375 0.586

.001 .007 .019 .036 .444

2.113 2.072 0.957 1.913 1.769

1.381 1.217 0.922 1.042

3.235 3.528 0.993 3.513

Trial arm: rate control is the reference group. Abbreviations: CI, confidence interval; OR, odds ratio; PCS, Medical Outcomes Study Short Form-36 physical component score.

CAD, stroke, and HF were significant clinical variable predictors of 1-year hospitalization after controlling for trial arm in bivariate analysis. Smoking was a potentially influential clinical predictor that was included in multivariable analysis. Among men with AF, PCS was an independent psychosocial predictor of 1-year hospitalization after controlling for trial arm. Living alone and QLI-CV were potentially influential psychosocial predictors of 1-year hospitalization among men and were included in further analyses. In simultaneous logistic regression analysis, HF, PCS, and living alone were significant independent predictors of 1-year hospitalization among men with AF (see Table 3). Male AF patients with HF were 2.07 times as likely to be hospitalized or die within 1 year compared with those without HF. Worse PCS scores were associated with 1-year hospitalization among men with AF beyond the effects of trial arm. Men who live alone were 91% more likely to be hospitalized or die within 1 year compared with those who do not live alone, after controlling for trial arm and the other variables.

Discussion In the current study, gender predicted mortality but not hospitalization. Higher mortality among men was consistent with findings in 2 other studies of mortality

according to gender in patients with AF.3,14 One study was similar to ours in that women were significantly older and had more risk factors than men did.14 The lack of differences in hospitalization between the genders contrasts with the only previous study examining hospitalization in AF; it found higher age-adjusted hospitalization for men than for women.17 Different variables predict mortality and 1-year hospitalization in women and men with AF (Table 4). Older age predicted mortality in both men and women with AF. The mean age for women in this study was 72 years, and for men, mean age was 68 years. This is consistent with gender differences in onset of AF.13 The relationship of age to mortality did not differ for men and women. Every year older was associated with a 12% increase in the likelihood of dying for women and a 10% increase in the likelihood of dying for men. The higher annual increase for women may be attributed to the fact that women were older than men at the onset of AF. The additional clinical predictors of diabetes, stroke, and HF predicted mortality among women. Diabetes predicted the development of HF in patients with AF.25 In patients with diabetes, the number and severity of complications have been linked to increased mortality and increased hospitalizations.26 There were no significant differences in prevalence of HF between women and men, but LVEF less than 50 was twice as common in men as in women at the beginning of the study. In

TABLE 4 Summary of Independent Predictors of Mortality (Cox Regression) and 1-Year Hospitalization (Logistic Regression) in Women and Men With Atrial Fibrillation in the Atrial Fibrillation Follow-up Investigation of Rhythm Management Study

Women Men

Mortality

1-y Hospitalization (Controlling for Trial Arm)

Diabetes, stroke, heart failure, age Coronary artery disease, age, PCS, PSP

Diabetes, PCS Heart Failure, PCS, lives alone

Abbreviations: PCS, physical component score; PSP, psychological-spiritual subscale score.

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

562 Journal of Cardiovascular Nursing x November/December 2014 the Cardiovascular Health Study, women had 15% to 20% lower adjusted risk of total and cardiovascular mortality compared with men.27 Stroke predicted mortality among women, but not men, with AF in the current study. With increasing age, women are at higher risk for stroke then men are.35 In the general population, over an entire lifetime, about 16% of women but only 8% of men will die of stroke.28 In the current study, women were older than men. Although the lifetime risk of stroke is higher in men, women are more likely to die of stroke, probably because of their older age at its occurrence and their longer life expectancy. The women with AF in the study were less likely than the men to have CAD. This is consistent with literature reporting prevalence rates in the United States and the United Kingdom.29,30 Women with CAD are more likely to be older than men and to experience more comorbidities.31 Two of the HRQOL predictors, poor PCS and poor QLI-CV indicator PSP, predicted mortality for men but not for women. One of these predictors was generally poor HRQOL, where the other was specifically related to limitations due to cardiac disease. Having poor physical functioning at younger age is consistent with mortality at a younger age. The median age for men in the study is 68 years, and one would expect relatively good physical functioning for a healthy man 68 years of age. Older patients have been shown to be more accepting of physical decline.32 Because women in the study were older, they may be more accepting of a decline in physical function. Women report lower HRQOL in CAD and HF.33 The impact of this requires further study. Predictors of hospitalization within 1 year also differed for women and men with AF. Among clinical predictors, diabetes, predicted hospitalization for women, and HF predicted hospitalization for men. These predictors for hospitalization are consistent with predictors of hospitalization in older patients. Patients with diabetes34 and HF35 experience increased hospitalization. Our findings of different relationships of clinical and HRQOL variables to mortality and hospitalization are consistent with differences in other heart disease populations. For example, the relationship of marital quality to mortality differed significantly according to gender among patients with HF.5,36 These differences also may be related to gender differences in perceptions of symptoms. There are differences in the relationship of HRQOL to symptoms in men and women with HF.37 Poor PCS predicted hospitalization for both women and men. Patients with poor physical function may require more in home assistance or facility placement to prevent hospitalization. In addition, men (12.6%) who lived alone were more likely to be hospitalized in the first year than were women (33.6%). Marital status and physical function have been shown to impact hos-

pitalization in general medicine patients.38 Transitional care programs designed to improve HRQOL and reduce hospital readmissions by providing needed social support may also be appropriate for this population.39 Our findings suggest that periodic assessment of HRQOL for patients with AF is indicated. The importance of HRQOL in cardiovascular disease is gaining recognition in the professional community. The recent American Heart Association Scientific Statement on Quality of Life recommendations that HRQOL be assessed on an ongoing basis in patients with heart disease33 support the findings of this study. On the basis of these findings, nurses could identify members of high-risk groups, such as men who live alone and have poorer HRQOL, and target them for more frequent monitoring. Limitations The AFFIRM study was conducted approximately 10 years ago. Despite the age of these data, the population represents approximately 70% of patients with AF.40 Because of the rigor of the AFFIRM study, its applicability continues today. The generalizability of both the current study and AFFIRM is limited because of the small number of minorities and selection criteria. Minorities may be more reluctant to provide psychosocial and HRQOL data than other potential participants.23 The AFFIRM investigators were concerned about this possibility and documented a significantly higher rate of refusal to participate in the HRQOL substudy in minority than in nonminority patients enrolled in the AFFIRM study.21 Ethnic differences have not been adequately examined to evaluate clinical and HRQOL predictors in patients with AF. More research is required to determine if the same predictors for women and men in the current study are the same for patients of other ethnic backgrounds. Availability of data from clinical trials enhances opportunities for analyses of complex issues requiring large samples with extensive data.20,28 As typical of clinical trials, the AFFIRM databases include data obtained with a documented process of instrument selection, data collection procedures, data management, and data quality assurance measures.21 However, existing data sets limit analysis to data obtained in the study. Analyses included only those variables assessed in the AFFIRM study. Data for additional variables, including cognitive function, sexuality, anxiety, or concerns about the need for anticoagulation, which may be important to this population, were not obtained in AFFIRM and thus could not be investigated.

Conclusion Atrial fibrillation poses a significant global public health challenge. Different clinical and HRQOL variables

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Hospitalization and Mortality in Patients With AF 563

What’s New and Important h Predictors of mortality and hospitalization are different for women and men with AF. h Biomedical predictors of mortality and hospitalization are different for women and men with AF. h Health-related quality of life (Medical Outcomes Study Short Form-36 PCS) predicts hospitalization for both men and women but mortality only among men beyond the effect of biomedical predictors in this patient population.

predict mortality and 1-year hospitalization for women and for men. Increased monitoring of HRQOL and nursing interventions specifically targeting the predictors of mortality and 1-year hospitalization with regard to gender differences could significantly impact mortality, HRQOL, and costs in patients with AF. Therefore, it could be postulated that different interventions should be designed for men and for women. More genderspecific cardiovascular research is needed to adapt current guidelines to improve health in women with AF. Acknowledgments Thank you to the investigators, coordinators, staff, and patients of the AFFIRM study. REFERENCES 1. Naccarelli GV, Panaccio MP, Cummins G, Tu N. CHADS2 and CHA2DS2-VASc risk factors to predict first cardiovascular hospitalization among atrial fibrillation/atrial flutter patients. Am J Cardiol. 2012;109(10):1526Y1533. 2. Amin AN, Jhaveri M, Lin J. Temporal pattern and costs of rehospitalization in atrial fibrillation/atrial flutter patients with one or more additional risk factors. J Med Econ. 2012; 15(3):548Y555. 3. Asbach S, Olschewski M, Faber TS, Zehender M, Bode C, Brunner M. Mortality in patients with atrial fibrillation has significantly decreased during the last three decades: 35 years of follow-up in 1627 pacemaker patients. Europace. 2008; 10(4):391Y394. 4. Avgil Tsadok M, Jackevicius CA, Rahme E, Humphries KH, Behlouli H, Pilote L. Sex differences in stroke risk among older patients with recently diagnosed atrial fibrillation. JAMA. 2012;307(18):1952Y1958. 5. Coyne KS, Paramore C, Grandy S, Mercader M, Reynolds M, Zimetbaum P. Assessing the direct costs of treating nonvalvular atrial fibrillation in the United States. Value Health. 2006;9(5):348Y356. 6. Kim MH, Johnston SS, Chu BC, Dalal MR, Schulman KL. Estimation of total incremental health care costs in patients with atrial fibrillation in the United States. Circulation. 2011;4(3):313Y320. 7. Wolf PA, Mitchell JB, Baker CS, Kannel WB, D’Agostino RB. Impact of atrial fibrillation on mortality, stroke, and medical costs. Arch Intern Med. 1998;158(3):229. 8. Kannel WB, Benjamin EJ. Status of the epidemiology of atrial fibrillation. Med Clin North Am. 2008;92(1):17Y40. 9. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults. JAMA. 2001;285(18): 2370Y2375.

10. Maas AH, van der Schouw YT, Regitz-Zagrosek V, et al. Red alert for women’s heart: the urgent need for more research and knowledge on cardiovascular disease in women. Proceedings of the Workshop held in Brussels on Gender Differences in Cardiovascular Disease, 29 September 2010. Eur Heart J. 2011;32(11):1362Y1368. 11. Ferrari R, Abergel H, Ford I, et al. Gender-and age-related differences in clinical presentation and management of outpatients with stable coronary artery disease. Int J Cardiol. 2012;167(6):2938Y2943. 12. Stein GY, Ben-Gal T, Kremer A, et al. Gender-related differences in hospitalized heart failure patients. Eur J Heart Fail. 2013;15(7):734Y741. 13. Menezes AR, Lavie CJ, DiNicolantonio JJ, et al. Atrial fibrillation in the 21st century: a current understanding of risk factors and primary prevention strategies. Mayo Clin Proc. 2013;88(4):394Y409. 14. Gomberg-Maitland M, Wenger NK, Feyzi J, et al. Anticoagulation in women with non-valvular atrial fibrillation in the Stroke Prevention Using an Oral Thrombin Inhibitor (SPORTIF) trials. Eur Heart J. 2006;27(16):1947Y1953. 15. Dagres N, Nieuwlaat R, Vardas PE, et al. Gender-related differences in presentation, treatment, and outcome of patients with atrial fibrillation in Europe: a report from the Euro Heart Survey on Atrial Fibrillation. J Am Coll Cardiol. 2007;49(5):572Y577. 16. Rienstra M, Van Veldhuisen DJ, Hagens VE, et al. Genderrelated differences in rhythm control treatment in persistent atrial fibrillation: data of the Rate Control Versus Electrical Cardioversion (RACE) study. J Am Coll Cardiol. 2005;46(7): 1298Y1306. 17. Humphries KH, Jackevicius C, Gong Y, et al. Population rates of hospitalization for atrial fibrillation/flutter in Canada. Can J Cardiol. 2004;20(9):869Y876. 18. Paquette M, Roy D, Talajic M, et al. Role of gender and personality on quality-of-life impairment in intermittent atrial fibrillation. Am J Cardiol. 2000;86(7):764Y768. 19. Ong L, Irvine J, Nolan R, et al. Gender differences and quality of life in atrial fibrillation: the mediating role of depression. J Psychosom Res. 2006;61(6):769Y774. 20. Geller NL, Sorlie P, Coady S, Fleg J, Friedman L. Limited access data sets from studies funded by the National Heart, Lung, and Blood Institute. Clin Trials. 2004;1(6):517Y524. 21. Wyse DG, Anderson JL, Antman EM, et al. Atrial fibrillation follow-up investigation of rhythm management-the AFFIRM study design. Am J Cardiol. 1997;79(9):1198Y1202. 22. The AFFIRM Investigators. Baseline characteristics of patients with atrial fibrillation: the AFFIRM Study. Am Heart J. 2002;143:991Y1001. 23. Jenkins LS, Brodsky M, Schron E, et al. Quality of life in atrial fibrillation: the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Am Heart J. 2005;149(1):112. 24. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347(23):1825. 25. Potpara T, Marinkovic-Eric J, Grujic M, RadojkovicCirovic B, Vujisic-Tesic B, Petrovic M. Effect of diabetes mellitus in recovery and maintenance of sinus rhythm in patients with persistent atrial fibrillation. Srp Arh Celok Lek. 2002;130(5Y6):189Y192. 26. Young BA, Lin E, Von Korff M, et al. Diabetes complications severity index and risk of mortality, hospitalization, and healthcare utilization. Am J Manag Care. 2008; 14(1):15. 27. Parashar S, Katz R, Smith NL, et al. Race, gender, and mortality in adults 965 years of age with incident heart failure

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

564 Journal of Cardiovascular Nursing x November/December 2014

28.

29.

30.

31.

32.

33.

(from the Cardiovascular Health Study). Am J Cardiol. 2009; 103(8):1120Y1127. Zeni MB, Kogan MD. Existing population-based health databases: useful resources for nursing research. Nurs Outlook. 2007;55(1):20Y30. Jones DW, Chambless LE, Folsom AR, et al. Risk factors for coronary heart disease in African Americans: the atherosclerosis risk in communities study, 1987Y1997. Arch Intern Med. 2002;162(22):2565. Rothwell PM, Coull AJ, Silver LE, et al. Population-based study of event-rate, incidence, case fatality, and mortality for all acute vascular events in all arterial territories (Oxford Vascular Study). Lancet. 2005;366(9499):1773Y1783. Ghali WA, Faris PD, Galbraith PD, et al. Sex differences in access to coronary revascularization after cardiac catheterization: importance of detailed clinical data. Ann Intern Med. 2002;136(10):723Y732. Gottlieb SS, Khatta M, Friedmann E, et al. The influence of age, gender, and race on the prevalence of depression in heart failure patients. J Am Coll Cardiol. 2004;43(9): 1542Y1549. Rumsfeld JS, Alexander KP, Goff DC Jr, et al. Cardiovascular health: the importance of measuring patient-reported health status: a scientific statement from the American Heart Association. Circulation. 2013;127(22):2233Y2249.

34. Tait RJ, French DJ, Burns RA, Byles JE, Anstey KJ. Alcohol, hospital admissions, and falls in older adults: a longitudinal evaluation. Int Psychogeriatr. 2013;25(06):901Y912. 35. Hall MJ, Levant S, DeFrances CJ. Hospitalization for congestive heart failure: United States, 2000Y2010. NCHS Data Brief. 2012;108(Oct):1Y8. 36. Coyne JC, Rohrbaugh MJ, Shoham V, Sonnega JS, Nicklas JM, Cranford JA. Prognostic importance of marital quality for survival of congestive heart failure. Am J Cardiol. 2001; 88(5):526Y529. 37. Heo S, Moser DK, Widener J. Gender differences in the effects of physical and emotional symptoms on healthrelated quality of life in patients with heart failure. Eur J Cardiovasc Nurs. 2007;6(2):146Y152. 38. Hasan O, Meltzer DO, Shaykevich SA, et al. Hospital readmission in general medicine patients: a prediction model. J Gen Intern Med. 2010;25(3):211Y219. 39. Stamp KD, Machado MA, Allen NA. Transitional care programs improve outcomes for heart failure patients: an integrative review [published online ahead of print January 2013]. J Cardiovasc Nurs. 40. Albert NM, Bozkurt B, Brindis FR, et al. Management of patients with atrial fibrillation (compilation of 2006 ACCF/ AHA/ESC and 2011 ACCF/AHA/HRS Recommendations). J Am Coll Cardiol. 2013;61(18):1935Y1944.

Copyright © 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Is health-related quality of life a predictor of hospitalization or mortality among women or men with atrial fibrillation?

Little is known about predictors of mortality or hospitalization in women compared with men in patients with atrial fibrillation (AF). Although there ...
242KB Sizes 0 Downloads 0 Views