Evolution of self-care in patients with heart failure at the first outpatient return and three months after hospital discharge

Objective: to analyze the evolution of self-care in hospitalized patients with decompensated heart failure, between the first return after hospital discharge (T0) and three months after this assessment (T1). Method: an observational, analytical and longitudinal study carried out in the cardiology outpatient clinics of two public hospitals in Ribeirão Preto, São Paulo. The sociodemographic and clinical data were collected through interviews and consultation of medical records. Self-care was assessed using the Brazilian version of the Self-Care of Heart Failure Index-SCHFI instrument. The data were analyzed by means of the Student’s t test and paired distribution (McNemar) with a significance level of 0.05. Results: we verified an increase in the mean scores of the three subscales of SCHFI (Maintenance, Management and Confidence), when comparing the values of T0 and T1, these differences being statistically significant (p<0.001). When comparing the positive changes in self-care actions over these months, we found statistically significant changes in the Maintenance (6 out of 10 items), Management (5 out of 6 items) and Confidence (4 out of 6 items) subscales. Conclusion: self-care for heart failure improved in the period between the first return after discharge and the end of three months of follow-up. Further studies are needed to verify the variables associated with improved self-care after hospitalization.


Introduction
With the increase in the number of older adults in the world population, new chronic diseases started to appear in society, among which we can mention heart failure (HF) (1)(2) . In Brazil, the diseases affecting the circulatory system were responsible for 1,151,522 deaths in 2018, with 200,760 deaths caused by HF (3) . HF is a complex and multi-factorial chronic syndrome and can occur due to cardiac, structural and functional abnormalities resulting from several other diseases, mainly those of cardiovascular origin, resulting in reduced cardiac output. Its treatment is complex and involves measures of a pharmacological and non-pharmacological nature (4)(5)(6) . Failure to adhere to treatment can lead to episodes of decompensation, characterized by the onset, either sudden or gradual, of typical symptoms such as dyspnea, edema in the lower limbs, fatigue, and also by the presence of jugular stasis and lung crackles (6)(7) . Non-pharmacological treatment is usually carried out in the home environment, as it involves lifestyle-related measures such as daily weight control, reduced sodium, fluid and alcohol consumption, increased physical activity, and immunization, in addition to the use of the medications prescribed (5)(6) . The complexity of the clinical condition and the deficiency of self-care can justify the high rates of hospital readmissions due to the decompensation of HF (1)(2)4,7) . Although the focus of the self-care concept is predominantly linked to health promotion and disease prevention (8) , self-care actions are also developed by people in the face of the appearance of signs and symptoms of cardiovascular diseases, for example (9) . In the present study, self-care was investigated according to the definition of North American researchers who consider it as a decision-making process that involves behaviors aimed at maintaining physiological stability, monitoring and managing the symptoms of heart failure.
It is an active and planned process that permeates the following stages: recognition of a symptom (for example, gaining body weight without changing the eating pattern and onset of dyspnea), assessing the change in health status, the decision to implement a treatment strategy (for example, taking an extra dose of diuretic, as prescribed) and evaluating the treatment/action performed (weight loss and improved breathing pattern) (10)(11) . Based on this definition, the authors developed an instrument for assessing self-care in HF that assesses three dimensions of self-care: Maintenance (it involves monitoring, adherence to treatment, and recognizing the signs and symptoms of decompensation); Management (it covers the assessment of signs and symptoms by the patient, their attitudes towards them and the reevaluation seeking to improve these signs and symptoms of HF) and Confidence (it encompasses both maintenance and management of HF). It is expected that, when obtaining positive results of adherence to treatments, both pharmacological and non-pharmacological, the patient will feel more confident in the management of HF (10) .
The patients' knowledge about their health condition, self-care and adherence to therapy reduces the morbidity and mortality caused by the disease. Teaching strategies focused on pharmacological measures, such as adherence to treatment and monitoring of symptoms, and nonpharmacological measures, such as changes in diet, physical activity and daily weight have been shown to be effective in improving the quality of life and prognosis of these patients (6,12) . Nursing is one of the professions of the health team that monitors the treatment of patients with HF, focusing on strategies and activities to improve self-care (13)(14) . In Brazil, the investigation of the self-care of patients with HF has occurred mainly through crosssectional studies. Thus, our study is important because it evaluated how self-care evolves after hospitalization due to decompensation of the syndrome.
The present study aimed to analyze the evolution of self-care in patients who have been hospitalized with decompensated heart failure, in the period between the first return after hospital discharge from the last hospitalization and the third month after the first contact.  (6) were collected, as well as those related to HF decompensation. At the time of hospitalization due to decompensation, it is important to consider the clinical-hemodynamic profile, using the congestion and perfusion parameters of the organs (15) , to support the therapeutic decision. Upon being hospitalized for decompensation, the patients will be characterized regarding the clinical profile of HF, according to the following order: characteristic of tissue perfusion (positive/ hot or negative/cold) and level of congestion (positive/ wet or negative/dry). After this evaluation, they will be classified between profiles A, B, C or L. The most common profile is B (hot and humid), in which perfusion is adequate, although congestion is present (4,15) . To assess the severity of the symptoms and the tolerance to physical activities of the patients, in the two outpatient assessments, we used the self-reported functional class (16) , adapted from the system proposed by the New York Heart

Method
Association (17) . For the assessment of HF self-care, we used the Self Care of Heart Failure Index v. 6.2 (SCHFI) (14) in its version adapted to Portuguese (18) . Permission to use the scale in the Brazilian version was requested and granted by the authors. This instrument assesses self-care in three subscales: Maintenance (10 items), Management (6 items) and Confidence (6 items). In the Maintenance domain, the answers for each item vary from "never/rarely" (value 1) to "always/daily" (value 4); for the Management domain, they range from "unlikely" (value 1) to "very likely" (value 4); and, for the Confidence domain, from "not confident" (value 1) to "extremely confident" (value 4). The values of the sum of the scores in each domain are transformed into a standardized scale from 0 to 100. Higher values indicate better self-care according to each subscale (14) . In the validation for Portuguese, the hospitalized consistency evaluated by Cronbach's alpha was 0.43 for Maintenance, 0.76 for Management and 0.94 for Confidence (18) .
In this study, we used the measures obtained in the domains as quantitative variables and, in addition, the self-care measures according to the categories so that the evolution of the participants was observed, comparing the alternatives chosen at T0 with those indicated at T1.
Our categorization proposal was carried out according to the degree of similarity between the response scales, for example, in the Maintenance subscale, the responses ranged from 1 to 4 and were grouped into two categories (example: grouping "Never or Rarely/Sometimes" or "Frequently/Always or Daily"). When the subscale had more than four answer options, a new category was created (example: item 1 of the Management subscale has six answer categories, which were grouped into four categories: "I did not have such symptoms", "I did not recognize", "it took me a short/long time to recognize" and "I recognized quickly/immediately"). We categorized the answer scales respecting the similarity pattern of the alternatives in order not to compromise the interpretation of the results.

Results
During the recruitment period, 245 patients were hospitalized due to HF decompensation; of this total, 186 (76%) patients attended the first outpatient visit and were invited to participate in the study. Eight individuals did not agree to participate and 21 were excluded for presenting one or more of the criteria for exclusion. Thus, www.eerp.usp.br/rlae    We did not find any statistically significant results when we analyzed, in a bivariate way, the measures of the subscales of the self-care instrument with the sociodemographic and clinical variables, shown in Table 1.
The comparisons of the means of the three subscales of SCHFI, assessed during the first outpatient return after hospital discharge (T0) and at three months after the first contact (T1), revealed statistically significant differences with an improvement in the self-care actions over the evaluated period (

Discussion
The aim of our study was to evaluate the evolution of self-care in patients who were hospitalized due to HF decompensation for three months. The initial assessment designed to measure self-care (10,19) . Self-care actions in HF result from a decision-making process by the patient.
In this process, several factors are involved (personal, environmental and clinical) and influence the patients' daily decisions and, consequently, their self-care actions.
Experience, knowledge, skills and values must be considered when determining these actions (19) .
Other research studies carried out in Brazil (20) or in other countries (18, (18) ; 50.0; S.D.=20.3 (20) ) and in Italy (53.2) (21) . Similar findings to those obtained in our study for the Confidence subscale were found in cross-sectional studies by other authors: 54.1 (21) , 58.0 (S.D.=25.5) (18) and 58.1 (S.D.=18.2) (20) , 70.0 (S.D.=16.2) (14) . In our study, the Confidence subscale also showed better scores in the assessment at three months. These results differ from those presented by other authors who investigated whether greater experience with the disease would be associated with greater confidence in self-care (20)(21) .
Our results lead us to think about the difficulty in adhering to physical activity for patients with HF, considering that the "Practice physical activity" (p=0.31); "Practice physical exercise for at least thirty minutes" and "Request foods with little salt when eating out" items were the ones that showed the least adherence (choice of the never or rarely/sometimes alternative) for, respectively, 83.9%, 83.2% and 71.6% of the participants. Other studies assessed the detailed score of each item in the Maintenance subscale and our data corroborate these findings, considering the low score in relation to physical activity and the non-request for food with less salt when eating out (20)(21) .
In the Management subscale, we found statistically differences for the "Recognize HF symptoms"; "Reduce the salt in your diet"; "Reduce your fluid intake"; "Take a further diuretic" and "Contact your doctor or nurse for guidance" items. The strategies pointed out as "likely/ very likely" to be used by the participants were the items about contacting a professional for guidance However, differently from our interviewees, 60.1% said that they were "sure/absolutely sure" that these self-care measures would help them to control the symptoms of HF. A study carried out in Brazil (20) identified that hiring a professional for guidance (Mean=3.1; S.D.=1.0) and the mean score considering the use of resources to help solving the problem was 1.4 (S.D.=1.6).
In the Confidence subscale, the "Be free of the HF symptoms"; "Recognize the symptoms"; "Do something that can relieve your symptoms" and "Assess whether a drug works" items showed statistically significant differences between the answer groups, changing from "Not confident/A little confident" to "Very confident/Extremely confident". In our study, in the two assessments, the subscale items with the highest percentage of answers for the "Very confident/Extremely confident" option were "Follow the recommended treatment" (52.6%), followed by "Evaluate the importance of your symptoms" (35%), results that corroborate the findings of other studies (20)(21) .
The item with the least confidence on the part of the participants, throughout the study, was "Be free of the HF symptoms". This item was also the worst rated in another Brazilian study (20) .
The sociodemographic profile of the participants with a predominance of men, older adults, married or in a stable relationship, retired, with low income and low schooling is consistent with the profile observed in other national and international surveys (16,(20)(21)(22)(23)(24) . Although we have found no association between the self-care measures and the gender, age, marital status, family income and schooling variables, the personal characteristics are associated with the process of determining self-care actions in HF (19) .
Regarding the clinical variables, according to an adaptation of the NYHA classification (17) by Brazilian authors (16) , the self-reported functional classes II and III were informed by the majority of the participants (77.4%). In the present research, the functional class was self-reported, through reading by the researcher of the alternatives and indicating by the participant the answer that most reflected their situation at that moment. This procedure was adopted based on an international study (25) that pointed out a discrepancy between the NYHA assessment performed by the physician and that reported by the patient. According to the authors, the health professional often does not consider all the manifestations reported by the patients during the consultation, which can lead to impaired diagnosis and prognosis, with consequences for the indication of pharmacological and non-pharmacological therapy for HF (25) . If we compare these results with those obtained with the evaluation, by the NYHA functional classification, traditionally used by health professionals, we verify that they are similar to the data obtained in studies that evaluated inpatients or outpatients (16,20,(26)(27)(28)(29) .
The same is true for the most frequent etiology, which was ischemic (24.8%), followed by chagasic (21.9%) and decompensation profile B (73%) (18,30) . As in other national and international studies (20, self-control the symptoms of this complex syndrome (12) .
Decision-making for self-care is a complex process for patients with HF. A better understanding of the nature of this process will assist the health team in understanding it, opting for the best way to teach the patient with a view to adopting self-care measures for HF. Understanding the causes and reasons that are commonly related to failure in this decision-making process by the patient can help professionals to develop strategies that help patients and their families (9) . Nursing is the profession that is closest to the patient with HF during the entire hospitalization, assuming the monitoring and follow-up of their evolution until hospital discharge. A well-established therapeutic plan focusing on patient education for self-care that considers the participation of the caregiver/family may lead to an improvement in the quality of life related to the health of these individuals, and to a reduction in the number of hospitalizations due to decompensated HF and, consequently, in hospital costs.
We considered the following limitations of our study: reduced number of patients who completed the study and a long interval of days between discharge from the last hospitalization due to HF decompensation and the first outpatient return. As recommended in the guidelines of the American Heart Association (34) , the number of days for the first outpatient return should take place between 7 to 14 days after discharge; in our study, the mean was

Conclusion
We verified a positive evolution in self-care, with better scores from the participants in relation to HF when comparing the measurements of the three domains of the SCHFI instrument between the first assessment at the first outpatient return after hospital discharge and three months after that return. When comparing the two moments of assessment of the SCHFI items, the results also showed positive changes (better scores) in most self-care behaviors, addressed in the three subscales (Maintenance, Management and Confidence).