Prevalence of pharmacological adherence in patients with coronary artery disease and associated factors

Objective: to assess the prevalence of pharmacological adherence in patients with coronary artery disease and to identify factors associated with adherence. Method: a crosssectional, correlational study, including 198 patients with a previous diagnosis of coronary artery disease. Pharmacological adherence was assessed by the four-item Morisky Green test, and the factors that potentially interfere with adherence were considered independent variables. The association between the variables was determined by the Cox model, with a 5% significance level. Results: 43% of the patients adhered to the treatment. Fatigue and palpitation, never having consumed alcohol and being served by medical insurance were associated with adherence. Lack of adherence was associated with considering the treatment complex, consumption of alcohol and being served by the public health care system. In the multiple analysis, the patients with fatigue and palpitations had a prevalence of adherence around three times higher and alcohol consumption was associated with a 2.88 times greater chance of non-adherence. Conclusion: more than half of the patients were classified as non-adherent. Interventions can be directed to some factors associated with lack of adherence.


Introduction
According to the American Heart Association (AHA) (1) , 18.2 million 20-year-old or older Americans had coronary artery disease (CAD) between 2013 and 2016. In 2020, it was estimated that 720,000 Americans had some coronary event, including hospitalization for acute myocardial infarction (AMI) or death due to CAD. In Brazil, 242,858 hospitalizations occurred in 2019 due to ischemic heart disease. These data have a direct impact on economy and society (2) .
The treatment of CAD includes continuous use of medications, such as antihypertensives, antiplatelet agents, anticoagulants and statins. Invasive treatments can also be implemented, including coronary artery bypass grafting or percutaneous coronary intervention, in addition to the implementation of non-pharmacological measures, represented by incorporating a healthy lifestyle (3)(4)(5) .
Adherence to pharmacological and nonpharmacological treatment plays a crucial role in achieving satisfactory clinical outcomes in patients with CAD, including preventing ischemia, reducing symptoms, improving quality of life, decreasing readmissions and morbidity and mortality due to cardiovascular diseases (5) .
The World Health Organization (WHO) defines adherence as the extent to which a person's behaviour -taking medication, following a diet, and/or executing lifestyle changes, corresponds with agreed recommendations from a health care provider. There are five dimensions that interfere with this adherence: related to the patient, to the disease, the treatment, to the health care system and team, and to the socioeconomic factors (6) .
In developed countries, only 50% of the patients with chronic diseases continue their treatment, including pharmacological and non-pharmacological measures. In developing countries, such as Brazil, this number has varied from 51% to 56.5% (6)(7)(8) when considering only pharmacological adherence. Thus, in order to contribute to the primary and secondary prevention of CAD, the multidisciplinary team must identify the factors that interfere with adherence to treatment, in order to implement interventions that may minimize these barriers (5) .
A Brazilian study identified that only 26% of the hospitalized patients with CAD adhere to the pharmacological treatment (9) . Another study found that 49.3% of the patients had low adherence to statins and antiplatelet agents or potential for such 30 days after hospital discharge for Acute Coronary Syndrome (8) . Lack of adherence to pharmacological treatment is a complicating factor for readmissions (10) due to decompensations. A meta-analysis including 10 studies with 106,002 patients demonstrated that adequate pharmacological adherence reduced the risk of global and cardiovascular mortality, in addition to hospitalization for cardiovascular disease and AMI (11) . Therefore, during the delivery of nursing care for inpatients or outpatients, nurses must assess the factors associated with pharmacological non-adherence, so that nursing interventions can be established individually and focused on these variables. To the best of the authors' knowledge, only three Brazilian studies that evaluated the factors associated with pharmacological non-adherence in patients with CAD were identified (7,9,12) . Thus, new studies are needed, with population samples from different locations and addressing other possible non-adherence factors. This study aimed to assess the prevalence of pharmacological adherence in patients with CAD and to identify factors associated with adherence. Eligible patients were those hospitalized with a medical diagnosis of CAD, aged over 18 years.

Method
The exclusion criteria were chest pain, dyspnea or symptomatic hypotension during data collection, because these symptoms would make it impossible to interview the patients.
The sample size was calculated (13) using the R 3.4.1 software, based on data from a previous study, which identified adequate pharmacological adherence in 56.5% (7) of the patients with CAD. Considering an 80% power and a 10% accuracy, a minimum sample size of 198 patients was obtained.
Patients were selected by one of the authors of the study, a nurse specialized in Cardiology and Hemodynamics, who daily analyzed the patients' medical records in the selected units and identified those who met the eligibility criteria. The nurse explained the objectives of the study and invited them to participate, by offering the free and informed consent form.
Data were collected by means of three questionnaires, in the following sequence: Morisky Green test, to assess pharmacological adherence (dependent variable); an instrument to assess the variables that interfere with patient adherence and a questionnaire to assess patient knowledge about CAD.   Table 1).
Rev. Latino-Am. Enfermagem 2021;29:e3464. With regard to socioeconomic support, most patients were retired/inactive, had a family income of 1 to 3 minimum wages, had their own homes and needed family financial supplementation for the purchase of medications (Table 2). interrupted it when they felt well and five (4.4%) reported neglecting the time to take their medications.
In the univariate analysis of the factors related to the patient and the socioeconomic situation, no significant association was found (Tables 1 and 2).    In the multiple analysis, it was found that patients with fatigue and palpitation had a three-fold increase in the prevalence of medication adherence. In contrast, alcohol consumption was associated with decreased adherence, so that patients who drank had a 2.88 times greater chance of non-adherence than those who did not drink (Table 5).

Discussion
This study found that most patients with CAD did not adhere to drug treatment. In addition, the majority believed that they used the prescribed medications correctly. This result can be related to the fact that the patients did not consider it lack of adherence when they forget to take their medications at the prescribed times. Other studies have also shown that a large number of patients reported neglecting the medication schedule (8,10,20) .
Regarding the pharmacological treatment, the mean duration of this treatment was less than the time since the diagnosis. This can occur because, at some point, many patients interrupt treatment due to lack of financial resources (8) , the belief that treatment would be unnecessary while they are asymptomatic or to the complex therapeutic scheme, with associated side effects (21) .
The proportion of patients who adhered to pharmacological treatment in this study (43%) was lower than that found in another Brazilian state, in which 56.5% of the patients with CAD were adherent to treatment (7) . However, patient adherence in our study was greater than that of another Brazilian study on adherence to treatment by patients with CAD (26%) (9) or other chronic diseases (30.8%) (18) . These discrepancies reinforce the importance of further studies on the factors that can interfere with adherence to treatment by patients with CAD in Brazil, whose results may contribute to explain the differential prevalence.
Among the variables related to the disease, fatigue and palpitation were significantly associated with adherence in the univariate analysis and remained associated in the multiple analysis. Fatigue is a prevalent, disabling and persistent symptom in patients with CAD (21) . In a study conducted with patients undergoing percutaneous coronary intervention, this symptom was associated with the side effects of the medications (22) . Fatigue has also been identified as a predictor of worsening quality of life in patients newly diagnosed with CAD and in patients with chronic CAD (4) . This symptom also impairs psychosocial and physiological functionality (23)(24) . Other symptoms, such as palpitation, can occur both in the initial stage of the disease, due to arrhythmias related to a recent AMI and in advanced stages of chronic CAD, due to increased areas of ischemia and consequent fibrosis (3) . Both symptoms generate physical discomfort, which imposes restrictions on routine habits. Thus, the individuals tend Padilha JC, Santos VB, Lopes CT, Lopes JL.
to adhere more to the treatment in order to avoid these discomforts (25) . On the other hand, asymptomatic patients who do not adhere to treatment report that, due to the absence of symptoms, they interrupt the medications without consulting a professional, as they feel healthy (11) .
Patients who had never ingested it were almost three times more likely to have drug adherence than those who had ingested it, a finding also identified previously (8,26) . A study (26) showed that alcoholism is associated with lack of Regarding the access to the health care services, it was found that the patients treated by the public health care system were less compliant than those treated by health insurance companies. A study carried out with outpatients showed that those who did not have a health care plan had a 30% greater chance of not adhering to the treatment (p=0.03) (16) . Another study found a relationship between low adherence to treatment and assistance by the public health service (p=0.027) (17) .
This result can be explained by the fact that the intervals between medical appointments of public health care services are often over six months, in addition to the low professional bond caused by the turnover of professionals. A study conducted with patients with hypertension showed that the chance of low pharmacological adherence follow-up with more than one physician increased by three times (30) . Other studies also emphasize that the difficulty of physical access (due to distance or limited means of transportation) and the difficulties in accessing the medications also contribute to lack of adherence (8,17) . The individuals who have a health insurance tend to use the services more and, in turn, attend to medical appointments more often, thereby increasing the opportunity to access information that can support adherence (10) . One of the main benefits of easy access to the health care services is the possibility of therapeutic adjustments and monitoring (17) . consultations, in addition to implementing technologies, such as sending messages to reinforce the importance of medication and/or implementing software programs with alarm sensors for medication schedules (30)(31)(32) . A randomized clinical trial evaluated adherence to drug treatment in three different groups (Group 1: patients in usual care;

Multi-professional interventions and
Group 2: patients who used an application software without interacting with health care professionals; Group 3: patients who used an application software and interacted with the professionals). The results showed that the patients in the groups that used the application had increased medication adherence after three months of intervention, demonstrating that technology helps the patient remember the use of the medications and is effective in increasing medication adherence (31) .
In the context of multi-professional programs, knowing the factors associated with lack of medication adherence helps direct educational interventions by hospital and outpatient health care professionals, with a view to adapting adherence to the pharmacological treatment of CAD and, consequently, delaying the progression of the disease, reducing new cardiovascular events and improving the patient's quality of life.
The data in this study must be considered in the light of some limitations. First, it was carried out in a single center, which hinders the generalization of the results. In addition, the prevalence of adherence can be overestimated, since social desirability may have influenced the patients' self-report in the Morisky Green Test. Multicenter studies in the country must be performed using objective measures of medication adherence, such as serum dosage or vials with electronic dose monitoring.
Despite its limitations, this study evaluated several variables related to adherence, as recommended by the WHO, unlike other, which assessed only a few factors that interfere with adherence. From the data identified, the need is reasserted for nurses to assess the patient globally, in the multidisciplinary context, so that they may be aware of detailed aspects related to medication adherence and establish interventions to address such factors.

Conclusion
More than half of the hospitalized patients with CAD did not adhere to the pharmacological treatment.
The factors associated with adequate pharmacological adherence were the following: fatigue and palpitation, never having consumed alcohol and being served by health care insurance. The factors associated with lack of pharmacological adherence were the following: considering the treatment complex, using or having used alcoholic beverages and being served in the public health care service. The presence of fatigue and palpitation remained as factors associated with pharmacological adherence in the multiple analysis and alcohol consumption remained as a factor associated with lack of pharmacological adherence.