Prevalence and predictors of poor self-care behaviors in patients with chronic heart failure

Despite the indispensable role of self-care behavior in managing heart failure, the practice of self-care behavior remains poor, especially in developing countries. There is a scarcity of research focusing on poor self-care behavior and its determinants within our specific context. Therefore, the objective of this study was to investigate the prevalence and predictors of poor self-care behavior among ambulatory heart failure patients. A facility-based cross-sectional study was conducted at a tertiary care hospital in Ethiopia, involving patients with heart failure. We utilized the European Heart Failure Self-Care Behavior Scale (EHFScBS-9) to evaluate adherence to self-care behaviors. Data were gathered through patient interviews and a review of medical records. A binary logistic regression analysis was performed to identify predictors of poor self-care behavior in heart failure patients. We included a total of 343 participants in the final analysis of this study. The findings revealed that a majority of the patients (73.8%) demonstrated poor overall self-care behavior. Specifically, the majority of patients did not engage in regular exercise (76.1%), failed to consult doctors in case of rapid weight gain (75.6%), did not monitor weight daily (71.5%), did not restrict fluid intake (69.9%), and did not contact doctors in case of experiencing fatigue (68.6%). Additionally, 32.4% of patients did not reach out to doctors when experiencing shortness of breath, 30% did not restrict salt intake, 29% did not adhere to prescribed medication, and only 7% did not consult doctors if edema occurred. Our findings indicated that rural residence (AOR: 5.76, 95% CI: 2.47–13.43), illiteracy (AOR: 2.64, 95% CI: 1.52–6.31), prior hospitalization (AOR: 2.09, 95% CI: 1.21–3.61), and taking five or more medications (AOR: 1.83, 1.01–3.33) were significant predictors of poor self-care behavior. In conclusion, a majority of the participants in our study demonstrated poor self-care behavior. Risk factors for this behavior included rural residence, illiteracy, prior hospitalization, and taking five or more medications. Therefore, it is crucial to prioritize these high-risk patients and implement interventional programs aimed at improving self-care behaviors and overall treatment outcomes in heart failure patients.


Study participants
The source population for this study consisted of patients with HF who received follow-up care at the Ayder Comprehensive Specialized Hospital (ACSH) cardiac clinic.Patients diagnosed with heart failure, aged 18 years or older, and who had a minimum follow-up duration of 6 months with at least one prescribed medication were included in the study.Patients who were too ill to complete the interview, declined to provide consent, or had incomplete medical records were excluded.The participants were randomly selected during their scheduled follow-up appointments.

Variables and measurements
The study assessed self-care behavior (SCB) as the dependent variable.The independent variables encompassed sociodemographic characteristics (including age, sex, educational level, marital status, residence, and social drug use), as well as clinical and treatment-related factors (such as prior hospitalization, comorbidity, disease severity, treatment duration, and number of medications).
To assess patients' adherence to self-care behaviors (SCBs), we employed the European Heart Failure Self-Care Behavior Scale (EHFScBS-9) questionnaire, which has been validated in numerous studies [30][31][32] .This self-reported questionnaire consists of nine items, each rated on a 5-point Likert scale ranging from 1 (''completely agree'') to 5 (''completely disagree'').The nine items can be grouped into two dimensions: consulting behaviors and adherence to the regimen.The consulting behaviors dimension explores how frequently individuals with heart failure contact their doctor or nurse when experiencing symptoms such as shortness of breath, swelling (edema), weight gain, and fatigue.On the other hand, the adherence to the recommended regimen dimension evaluates the frequency of self-care activities such as weighing themselves, restricting fluid intake, following a low-salt diet, taking medications regularly, and engaging in exercise.The total possible score on the EHFScBS-9 is 9 to 45, with lower scores indicating better self-care.To facilitate interpretation, the responses were standardized to a possible score range of 0 to 100.In the standardized score of the EHFScBS-9 (0-100), a higher score signifies better self-care.To categorize the level of self-care, patients with an EHFScBS-9 score of ≥ 70 were classified as having good self-care behavior, while those with a score of < 70 were considered to have poor self-care behavior 33 .

Data collection instrument and procedure
We developed a comprehensive data collection tool consisting of a questionnaire and a data abstraction checklist to gather all necessary information.The questionnaire included socio-demographic factors and SCB, while the data abstraction checklist focused on clinical and treatment-related characteristics.To ensure accuracy and consistency, the questionnaire was translated into the local language (Tigrigna) and then back-translated into English.This process guaranteed the preservation of meaning across translations.Additionally, the Tigrignalanguage version of the questionnaire underwent content validity evaluation by a panel of four experts in the field, including two experts from the clinical pharmacy department, one expert from the internal medicine department, and one expert from the public health department.The Tigrigna-language version questionnaire demonstrated good reliability with a Cronbach's α of 0.86.Before the actual data collection, a pre-test was conducted on 5% of the sample population to assess the face validity of the drafted questionnaire.To carry out the data collection for this study, we employed fifth-year clerkship pharmacy students who received comprehensive training and orientation.

Bias
To mitigate any potential biases, we implemented various measures during the participant enrollment process.Firstly, we selected participants randomly in order to minimize selection bias.Additionally, in order to address participant bias, we extensively explained the study's objectives and reassured participants about the confidentiality of their responses.To overcome recall bias, we referred to patients' medical records for accurate information.Furthermore, a pilot test was conducted to identify and rectify any potential sources of bias or ambiguity in the questionnaire.The questionnaire was evaluated for its content validity by a panel of four experts in the field.As a result, a well-designed and validated questionnaire, incorporating clear and unbiased language, was developed.Finally, interviewers received comprehensive training to ensure consistency in data collection techniques, thus minimizing any potential interviewer bias.

Study size
The sample size for this study was determined using the single population proportion formula for the prevalence of SCB.Accordingly, a sample of 385 subjects was calculated, assuming a prevalence of SCB of 39.8% based on a previous study 23 , a 95% confidence level, a 5% margin of error, and a 5% contingency for nonresponse rate.Out of the 385 participants approached, 343 patients were included in the final analysis of the study.A simple random sampling technique was employed to recruit subjects for the study.

Data analysis
The data were checked, cleared, and entered into EPI data management (version 4.2.0).Subsequently, the data were exported to the Statistical Package for Social Science (SPSS version 24.0) for analysis.Descriptive statistics were used to determine the frequency of categorical variables and the mean (standard deviation) of continuous variables.A logistic regression model was used assess factors associated with poor SCB.We checked the model fitness using the Hosmer-Lemeshow test.Multicollinearity was assessed to test the correlation among predictor variables using the variance inflation factor (VIF), and no collinearity was found.The association of each independent variable with poor SCB was determined using bivariate logistic regression analysis.Furthermore, the variables with a p-value < 0.25 in the bivariate analysis were re-entered into a multivariate binary logistic regression model to control for possible effects of the confounders.Finally, the results were reported in terms of an adjusted odds ratio (AOR) with a 95% confidence interval (CI), and a p-value of < 0.05 was considered statistically significant in all analyses.

Ethics
Our study obtained ethical approval from the Ethics Review Committee (ERC1227/2018) of the School of Pharmacy, College of Health Sciences, Mekelle University.We duly informed all study participants about the study's objectives and protocol, and written informed consent was obtained from each participant.The confidentiality of personal information was rigorously maintained throughout the study.All procedures were conducted in accordance with the approved institutional guideline.

Patient and public involvement
Patients or members of the public were not directly involved in the design or planning of this study as the study focused on the HF-SCB.

Result Socio-demographic characteristics
Among the 385 participants approached, 42 patients were excluded from the study due to critical illness preventing them from giving a response 11 , unwillingness to give consent 18 , and incomplete medical records 13 .Therefore, a total of 343 patients were included in this study.The participants had a mean age of 54.16 (SD = 16.42), and more than half (56.3%) were males.The majority of the participants were urban residents (68.5%) and married (74.6%), while 41% were illiterate.In terms of social drug use, 19.8% were alcohol consumers, and 6.7% were cigarette smokers.The findings also showed that 20.4% of the participants used traditional medicine for their illnesses (Table 1).

Clinical characteristics
Among the participants, nearly two-thirds (60.1%) had been hospitalized once or more in the past year, and 74.1% had one or more comorbidities.Around 44.3% of the participants had been receiving treatment for more than two years, and 34.1% were taking five or more medications.The majority of the participants fell under NYHA class II (46.4%) or NYHA class III (43.1%).The most commonly identified comorbidities in patients with HF were valvular heart diseases (21.3%), ischemic heart diseases (20.4%), diabetes (19.2%), and hypertension (17.2%) (Table 2).

Prevalence of poor self-care behavior in heart failure patients
Our findings revealed that a majority of the patients (73.8%) exhibited poor overall self-care behavior.With regard to the specific SCBs, the majority of patients did not engage in regular exercise (76.1%), did not consult doctors in case of rapid weight gain (75.6%), failed to monitor weight daily (71.5%), did not restrict fluid intake (69.9%), and did not contact doctors in case of experiencing fatigue (68.6%).Moreover, a considerable percentage of patients (32.4%) did not reach out to doctors when experiencing shortness of breath, 30% did not limit salt www.nature.com/scientificreports/intake, and 29% did not adhere to prescribed medication.Only 7% failed to consult doctors if edema occurred (Table 3).Furthermore, variables with a p-value less than 0.25 in the bivariate analysis underwent further analysis in the multivariate logistic regression model.The overall model, consisting of all predictors, was statistically significant (Chi-square = 62.350, df = 12, p < 0.001).In the multivariate analysis, rural residence (AOR: 5.76, 95% CI: 2.47-13.43),illiteracy (AOR: 2.64, 95% CI: 1.52-6.31),prior hospitalization within one year (AOR: 2.09, 95% CI: 1.21-3.61),and having five or more medications (AOR: 1.83, 95% CI: 1.01-3.33)were identified as independent predictors of poor SCB (Table 4).

Discussion
Self-care behaviors are essential components of HF management that not only improve the quality of life for HF patients but also reduce morbidity, mortality, and healthcare costs 15 .However, there is limited knowledge regarding the actual adoption of self-care behaviors among HF patients.To address this knowledge gap, it is proposed that assessing an individual's HF-specific self-care behavior (HF-SCB) and its determinants can provide valuable insights to healthcare professionals and practitioners involved in HF management.By evaluating SCBs and the factors that influence them, clinicians can gather crucial information about patients' adherence to SCBs and identify areas that require intervention or improvement.
Despite the proven morbidity and mortality benefits of SCB in HF patients 11,15 , a majority of the patients (73.8%) in our study demonstrated poor overall SCB.This finding is consistent with findings reported from studies conducted in Sudan (72%), Poland (72.2%), and Ethiopia (77.7%) 21,34,35 .Conversely, a survey conducted with black Africans reported a higher prevalence of poor SCB at 88.4% 22 .In contrast, our finding was lower  36 and Vietnam (50.9%) 37 .This discrepancy may be attributed to variations in socio-demographics, disease knowledge, healthcare accessibility, and the method used to assess SCB.Thus, implementing intervention programs that include educational initiatives is crucial to enhance HF self-care practices, leading to improved overall HF outcome 13,38,39 .Engagement in regular physical activity is a class I recommended non-pharmacologic therapy for all stable HF patients 40 .More importantly, there is substantial evidence indicating that regular physical activity is beneficial for stable HF patients 41,42 .Despite the importance of exercise as a crucial component of HF therapy, our study revealed that 76.1% of patients were not regularly engaged in exercise.This finding aligns with similar findings from previous studies conducted in Ethiopia (80.6%) and Sudan (78.7%) 21,34 .The lack of national coordination  www.nature.com/scientificreports/ in promoting physical exercise, absence of physical exercise guidelines, and prevailing misconceptions about exercise among the African population, including Ethiopia, may contribute to this issue [43][44][45] .Non-compliance with SCBs, such as daily weight measurement, is associated with unfavorable clinical outcomes, including increased morbidity and mortality 11,46 .Regular weight monitoring is a recommended aspect of HF-SCB, as it enables patients to track their volume status, detect deterioration, and prevent exacerbations 10,47 .However, our study found that 71.5% of patients did not regularly monitor their weight.Many other studies have also reported similar findings [48][49][50] .Additionally, it is crucial for patients to promptly seek medical assistance if they experience sudden weight gain exceeding 2 kg within a week 51 .Surprisingly, our study discovered that a significant majority (75.6%) of participants failed to reach out to their doctors in such instances of rapid weight gain.This may be attributed to the fact that most patients do not perceive weight gain as a symptom of deteriorating HF and therefore do not associate it with adverse outcomes related to HF 52 .Consequently, implementing intervention programs that emphasize education could prove beneficial in encouraging patients to engage in HF-SCB, including regular weight monitoring, ultimately improving overall HF prognosis 46 .
Fluid overload is associated with adverse outcomes in patients with HF and is responsible for a substantial number of hospital admissions 10,53 .Therefore, patients with HF are often advised to limit their fluid intake to a maximum of 1500 ml/day 54,55 .However, this study found that more than two-thirds (69.9%) of the patients did not restrict their fluid intake, which is consistent with a previous study conducted in Ethiopia 56 .The reason for this might be the misconception among HF patients regarding fluid intake, as revealed by a previous study where more than one-third of the patients believed that they should drink plenty of fluids as part of HF-SCB 52 .
Among the HF-SCBs, medication adherence is the most important activity for effective management of HF 57,58 .However, it remains challenging in HF patients 59,60 .Our study revealed that 29% of the participants were non-compliant with their prescribed medication.This finding aligns with similar results reported in a South African study (29%) 20 and a study conducted in Sudan (25%) 34 .For patients with HF, salt restriction is commonly recommended to prevent fluid retention, symptoms exacerbation, and HF decompensation 10,40 .In our study, approximately 30% of the participants did not adhere to the low-salt diet recommendation, aligning with findings from other similar studies 21,61,62 .On contrary, our finding was lower compared to finding reported in Sudan (72.3%) 34 .
Moreover, early recognition of symptoms and prompt response to worsening symptoms are important aspects of HF self-care behaviors, as they can improve the patients' quality of life and prevent unnecessary www.nature.com/scientificreports/hospitalizations 63,64 .In HF patients, the inability to respond to symptoms and delayed seeking of treatment for worsening symptoms contribute to unnecessary hospitalizations, poor quality of life, and increased morbidity and mortality 63,65 .Several studies reported that HF patients delay for several days before seeking care for HF symptoms [64][65][66] .In our study, while the majority of patients contacted their doctors if they experienced edema and shortness of breath, 68.6% failed to contact their doctors for fatigue.This lack of response might be attributed to insufficient symptom monitoring, waiting to see if the symptoms would improve, or perceiving the symptom as minor 65,66 Therefore, it is crucial to educate HF patients about recognizing these warning symptoms to enable them to take appropriate and timely action 40 .
Several studies have demonstrated that non-adherence SCB is associated with a high risk of hospitalization in HF patients 14,[67][68][69] .Similarly, in our study, prior hospitalization was significantly associated with poor SCB.Patients who have been hospitalized one or more times in the last year were two times more likely to have poor SCB than those who have not been hospitalized.Moreover, patients who took five or more medications were more likely to have poor SCB compared to those who took a lower number of medications.This could be explained by the fact that patients with multiple medications may be reluctant to take their medications appropriately due to pill burden, safety, and cost issues 70 .Additionally, the level of education has been found to be significantly associated with SCB in different studies 19,24,26,71,72 .Similarly, participants who could not read and write were 2.6 times more likely to have poor SCB compared to those who attended college and above in our study.This could be justified that participants with better education have better capability of accessing information related to SCB while illiterate patients may need more time to learn self-care 71 .
Our study also found that rural residents were approximately six times more likely to have poor SCB compared to their urban counterparts.This finding is consistent with a similar study conducted in Iran 73 .The reasons for this disparity can be attributed to various factors, including rural-urban inequalities in socioeconomic resources and access to healthcare 74,75 .Furthermore, inadequate health literacy, limited access to healthcare services, negative perceptions of modern medicine, and a greater reliance on traditional and spiritual healers among rural populations could contribute to this trend 76,77 .Therefore, healthcare providers and policymakers need to be aware of these challenges in the rural population and design specific strategies to meet their needs, helping them develop tailored self-care management plans based on their unique circumstances.
Finally, it is important to acknowledge certain limitations of our study.The cross-sectional nature of our study may not establish causality between non-adherence to SCB and its contributing factors.Patients may understate socially undesirable activities like non-adherence to SCBs due to self-report concerns.Furthermore, the findings of this study should be cautiously extrapolated to other countries, as they depend on factors such as population socioeconomic status, disease distribution, the employed method of SCB assessment, and the healthcare system in place.

Conclusion
Most of the participants in our study had poor overall self-care behavior.Rural residence, illiteracy, prior hospitalization, and the use of five or more medications were identified as predictors of poor self-care behavior in HF patients.Therefore, more emphasis should be given to these high-risk patients.We recommend implementing further interventional programs to enhance SCBs, and improve overall treatment outcomes in heart failure patients.Moreover, we suggest researchers to conduct longitudinal studies with a more robust design to establish a cause-and-effect relationship between SCB and their determinants.

Table 2 .
Clinical and treatment related characteristics of patients with heart failure (n = 343).

Table 3 .
Self-care behavior among patients with heart failure (n = 343).

Table 4 .
Factors associated with self-care behavior among patients with heart failure (n = 343).