Association between Spirituality and Adherence to Management in Outpatients with Heart Failure

Background Spirituality may influence how patients cope with their illness. Objectives We assessed whether spirituality may influence adherence to management of outpatients with heart failure. Methods Cross sectional study enrolling consecutive ambulatory heart failure patients in whom adherence to multidisciplinary treatment was evaluated. Patients were assessed for quality of life, depression, religiosity and spirituality utilizing validated questionnaires. Correlations between adherence and psychosocial variables of interest were obtained. Logistic regression models explored independent predictors of adherence. Results One hundred and thirty patients (age 60 ± 13 years; 67% male) were interviewed. Adequate adherence score was observed in 38.5% of the patients. Neither depression nor religiosity was correlated to adherence, when assessed separately. Interestingly, spirituality, when assessed by both total score sum (r = 0.26; p = 0.003) and by all specific domains, was positively correlated to adherence. Finally, the combination of spirituality, religiosity and personal beliefs was an independent predictor of adherence when adjusted for demographics, clinical characteristics and psychosocial instruments. Conclusion Spirituality, religiosity and personal beliefs were the only variables consistently associated with compliance to medication in a cohort of outpatients with heart failure. Our data suggest that adequately addressing these aspects on patient’s care may lead to an improvement in adherence patterns in the complex heart failure management.


Introduction
Heart failure (HF) continues to challenge multidisciplinary health care teams. 1 Its prevalence remains elevated and its management usually requires poly-pharmacy along with satisfactory self-awareness of the disease. 2,3 The course of HF, in its chronicity and frequently inexorable outcomes, shares similarities with many cancer diseases and resembles their impact on poor quality of life standards. 4 Patients face important limitations to adequately adhere to the complexity of HF management. 5 Adherence appears as an important aspect in the course of HF. It influences patients' pattern of decompensation and subsequent hospital re-admissions. 6 In addition, adequate adherence standards may help to improve quality of life. 7 demonstrate a positive association between spirituality and degree of compliance in HF patients. 14 Nevertheless, for many other chronic disorders there is sustained evidence that spirituality could improve compliance. 15 In this study, we examined associations between spirituality and adherence to management in outpatients with HF, independently of psychosocial and educational background.

Study design
This is a cross-sectional study which enrolled HF outpatients followed at a tertiary care University Hospital in Porto Alegre, Brazil, from August, 2012 to June, 2013. The study protocol was approved by the Institutional Research and Ethics Committee and all enrolled participants signed a written informed consent prior to study entry.

Participants
Consecutive ambulatory patients (mainly composed by patients with newly diagnosed left ventricular dysfunction, post HF hospital admission and/or refractory symptoms) being followed for a minimum of six months in the HF Clinic at the Hospital de Clínicas de Porto Alegre were invited to participate. Patients in any New York Heart Association functional class, regardless of HF etiology, were eligible. Exclusion criteria were inability to understand the study protocol and to answer the questions without assistance due to cognitive impairment or auditory deficit.

Study end-points and procedures
Patients were assessed for adherence to therapy, quality of life, depression, religiosity and spirituality utilizing validated questionnaires. All utilized instruments were previously validated to Brazilian Portuguese language. [16][17][18][19][20][21] Interviews were performed following the clinic appointment by research staff previously trained in questionnaire application. Time required for answering all the instruments ranged from 50 to 70 minutes. Patients answered questions orally and staff filled questionnaires as requested. Demographics and clinical characteristics were obtained from electronic chart review and clinical data were acquired during the clinic visit by a researcher who was unaware of questionnaire results. Definitions of psychosocial variables of interest are detailed below.
Adherence to therapy. The adherence to pharmacologic and non-pharmacologic therapy was assessed according to the Repetitive Education and Monitoring For Adherence for Heart Failure (REMADHE) study protocol, which has been adapted and is currently used in clinical practice in our HF Clinic. 17,22 The questionnaire is composed by ten-questions involving four domains: use of medications (one question); food and fluids (seven questions); alcohol consumption (one question); and medical appointments (one question). The score ranges between 0 and 26 points, with higher scores indicating better patient's adherence. A REMADHE score equal to or higher than 18 points indicates adequate level of adherence. 22 Quality of life: Two instruments were used to assess quality of life: generic and disease-specific questionnaires. Generic quality of life assessment was performed with the utilization of the World Health Organization Quality of Life (WHOQoL-Bref) while disease-specific was assessed by the Minnesota Living With Heart Failure Questionnaire (MLHFQ). 23,24 The WHOQoL-Bref is an abbreviated version of the WHOQOL-100 which is composed of 26 questions: a question about quality of life in general, a question about satisfaction with one's self health status, and 24 questions divided into four domains -physical, psychological, social relations and environment. The MLHFQ evaluates quality of life related to HF symptomatology within the previous month and correlates proportionally to functional class. 25 Higher WHOQoL-Bref scores indicate better quality of life in general, whereas lower MLHFQ represent better HF-related quality of life.
Depression: Depression was evaluated by the Patient Health Questionnaire (PHQ-9), which is a screening tool for detection of depression, based on symptom occurrence within the previous two weeks. It comprehends nine questions based on the major criteria for the diagnosis of major depression according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV). 26,27 Depression is classified, according to the score, as moderate depressive symptoms (total score between 10 and 14), moderate major depression (score between 15 and 19) and severe major depression (score equal or higher than 20). 26 Religiosity, Spirituality and Personal Beliefs: Two instruments were used to evaluate these dimensions.
1) The Duke University Religion Index (DUREL) scale is a tool for assessment of spirituality that is focused on religious aspects. 28 Its transcultural adaptation was developed and validated by Moreira-Almeida. 18 The DUREL scale has five items that describe three dimensions of religiosity, known to best correlate with health-related outcomes: organizational (ORA); non-organizational (NORA); and intrinsic religiosity (IR). The score ranges from 1 to 30 points and higher scores indicate elevated levels of religiosity.

Statistical analyses
Normally distributed (according to Shapiro-Wilks testing) continuous variables were expressed as mean ± standard deviation, while non-normally distributed ones were expressed as median and interquartile ranges. Categorical variables were reported as absolute numbers and percentages. Normally distributed continuous variables were analyzed by unpaired t-test. Non-normally distributed continuous variables were analyzed using Mann Whitney U test. Chi-square test (or exact Fisher test when appropriate) was used to compare categorical variables. Spearman coefficients were used for evaluation of correlations between adherence and psychosocial variables of interest. Kruskal Wallis was used to compare scores of spirituality across REMADHE quartiles. Logistic regression models were used to explore the association of spirituality to an adequate level of adherence (REMADHE ≥ 18 points). Adjusting covariates for multivariable models were tested for colinearity and selected among demographic, clinical and psychosocial variables of either clinical or statistical significance. The report by Black and co-workers, that correlated spirituality and adherence utilizing different instruments, was used to estimate a sample size of 130 subjects in the current study (α = 5%, β = 80%; effect size 25%). 14 All analyses were performed using the SPSS 20.0 statistical package (SPSS Inc., Chicago, IL, USA). A p value lower than 0.05 was considered of statistical significance.

Results
One hundred and thirty patients were interviewed between August, 2012 and June, 2013. Demographic and clinical characteristics of the studied population are detailed in Table 1. A description of the average scores obtained through the study instruments is provided in Table 2.
Overall, there was a low level of adherence, according to REMADHE scores. Adequate adherence was observed in 38.5% of the population.
The correlations between the adherence score with clinical characteristics and psychosocial scores are demonstrated in Table 3. A description of associations of demographic and clinical variables with the adherence score is also described in the Supplemental Table. Among demographics, REMADHE score differed only according to marital status. Clinical characteristics associated to higher adherence scores were ischemic HF etiology, presence of an implantable cardiac defibrillator and chronic kidney impairment. Adherence was positively correlated to the generic quality-of-life measure, but not to the disease-specific HF score. Neither depression nor religiosity was correlated to adherence. Interestingly, spirituality, when assessed by both total WHOQoL-SRPB score sum and by many specific domains, was positively correlated to adherence. Although significantly correlated to adherence score, the magnitude of spirituality association was relatively weak (Figure 1). Notably, there was a trend towards higher spirituality scores across quartiles of the adherence score ( Figure 2). Spirituality was also found to be significantly correlated to other psychosocial variables evaluated. WHOQoL-SRPB was moderately correlated to both generic There was an inverse correlation between WHOQoL-SRPB and depression classification by PHQ-9 (r = -0.49; p < 0.0001). Of note, there was a positive correlation between WHOQoL-SRPB and religiosity assessed by DUREL (r= 0.55; p= 0.0001), which was also observed within domains of both instruments (Table 4).  Among multivariable models to identify clinical and psychosocial variables associated to the presence of adequate adherence, WHOQoL-SRPB was an independent predictor when adjusted for demographics, clinical characteristics and psychosocial instruments (Table 5). Aside from WHOQoL-SRBP, DUREL was the only additional psychosocial instrument to demonstrate borderline significance for association to adequate adherence.

Discussion
The main finding of the present study is that SRPB were consistently associated with adherence to treatment in a cohort of HF patients followed in a tertiary care clinic. Importantly, in our study, this association was independent of relevant demographic and clinical data known to influence adherence to HF management.
This is the first study to show a clear association of spirituality and adherence to treatment in HF. However, our study cannot determine if there is a direct effect of spirituality in adherence or if spirituality is only a marker of broader and more complex    effect. For example, someone who is spiritualized is probably more prone to follow recommendations coming from someone he has a close relationship with (e.g., a physician). We identified three previous studies addressing the possible interactions between HF and spirituality. 14,30,31 Black et al. 14 sent a package of instruments (Spiritual Assessment Scale and the Heart Compliance Questionnaire) by mail to a convenience sample of 213 patients with a return rate of 45%. The authors did not find a significant correlation between spirituality and compliance. The study by Thomas 30 using a convenience sample of 97 patients showed a positive result with moral-ethical-spiritual self which accounted for 10.8% of the variance in adherence. Dickson et al. 31 studying socio-cultural influences on HF self-care in an ethnic minority, black population, using a mixed-methods strategy, found that spirituality was linked to self-care. More recently, issues related to wellbeing were shown to impact positively in patients with stage B asymptomatic HF -spirituality apparently played a role in mediating these effects. 32 Although religiosity and spirituality have been associated with better healthcare practices, such observations failed to translate into better cardiovascular disease outcome in an adequately-powered study. [33][34][35] The hypothesis that SRPB could affect compliance in chronic diseases and, particularly, in HF has also been raised by different authors. 14,30,36 There are some possible models proposed to explain such relationship. Black et al. 14 suggested that spiritual beliefs influence health beliefs which could lead to the practice of health-related activities such as the use of medications, control of weight, and diet compliance. Thomas 30 applied the Roy's Self Concept model to identify several potential predictors of medical compliance. 30,37 In this model, any stimulus is perceived either as a threat or a challenge to one's self-concept of body image, body sensation, self-consistency and moral-ethical-spiritual self. Briefly, stimuli perceived as a threat are reacted to in a negative way and consequently avoided, while stimuli perceived as a challenge are reacted to in a positive way and consequently followed. Thomas 30 found that patients who perceived the HF regimen as a threat either to body image, self-consistency, body sensation or self-ideal were less likely to adhere to it. On the contrary, those to whom the regimen was perceived as a challenge to moral-ethical-spiritual self were most likely to adhere to medical therapy. Lastly, a recent survey conducted on HF patients showed that they would have welcomed spiritual care in their management. 38 Our study has some limitations. First, as we used a crosssectional design we can only conclude about association between spirituality and adherence, but not a causal relationship. Second, our sample was obtained in Brazil, a country where spirituality and religion are very notoriously important values. Additional studies are necessary to assess if these findings are replicable in different cultural and religious backgrounds. Finally, the effect of spirituality in adherence to different aspects of the HF management -pharmacological and non-pharmacological therapy -was not individually assessed. The REMADHE tool used in our study does not discriminate between the various components of the HF management in depth. If available, such information could be useful to better allocate the role of multidisciplinary care, vis a vis the spirituality of patients, and be taken into consideration accordingly, to improve patients´ adherence.

Conclusions
Our study highlights that spirituality could be an important variable associated with adherence to treatment in the setting of outpatients with HF, suggesting that physicians and health professionals should be aware of its importance to improve clinical practice outcomes and implement measures to address the spiritual needs of patients. Further studies are warranted to better determine whether pharmacologic and non-pharmacologic measures in the management of HF are equally influenced by spirituality-related behavior.