Determinant factors of treatment adherence of hypertensive patients in a rural area of Indonesia

Background: The surging prevalence of hypertension due to lifestyle brought forth an increase in degenerative diseases. Adherence is important in achieving the effectiveness of therapy. This study aims to analyze the factors affecting medication adherence in patients with hypertension. Methods: A cross-sectional study was conducted between 20-24 February 2023 at Turirejo Village, Lawang District, Malang, Indonesia. Participants were hypertensive patients currently receiving treatment. The MMAS-8 questionnaire was used to classify adherence to medication for hypertension. Analysis with binary logistic regression test was performed on variables to display odds ratio values was conducted at a significance level of p<0.05. Results: A total of 37 adults diagnosed with hypertension participated in this study. Patients were dominantly consisting of elderly (≥ 56 years old; 73.0%) and females (86.0%). Seventy-six percent of patients have low adherence, twenty-four percent of patients have moderate adherence, and no patient has high adherence. Duration of illness, knowledge, and attitude variables significantly affect adherence (low and moderate) to medication for hypertension. Multivariate analysis showed that patients with a duration of illness above 10 years (Adjusted OR 18.27; 95% CI 1.72-194.47; p-value 0.016) and positive attitude towards treatment (Adjusted OR 12.76; 95% CI 1.25-130.40; p-value 0.032, respectively) increase the possibility of moderate adherence towards the medication of hypertension. Conclusion: Improvements in factors that affect adherence to hypertension treatment are needed to increase the success of the Non-Communicable Disease Prevention Program. Further research in identifying factors of economic capacity and access to health service providers is needed to validate the results of this study.


Background
Indonesia is currently facing a shift in disease patterns, from communicable diseases to non-communicable diseases.The increase in the prevalence of non-communicable diseases occurs as a result of unhealthy lifestyles triggered by urbanization, modernization, and globalization.Increasing life expectancy in line with socio-economic improvements and health services, has the consequence of increasing degenerative diseases [1].Hypertension is one of the most common and most common cardiovascular diseases in society and is the biggest cause of premature death in the world.Around 1.28 billion adults aged 30-79 years in the world suffer from hypertension [2].As many as 8.4% of the population aged ≥18 years in Indonesia suffer from hypertension [3].Patient compliance in carrying out the recommended therapy is a common problem.Several factors such as the length of time a person suffers from hypertension or the many types of antihypertensive drugs used will reduce the level of adherence to treatment [4][5][6].Through discussions with Turirejo Village officials and a preliminary study on February 20, 2023, the researchers were able to collect survey results from the 31 respondents.
The results of the preliminary study found that the most common diseases encountered were hypertension (12 people; 38.7%), diabetes (4 people; 12.9%), and acute respiratory ___________________________________________________ atika@fk.unair.ac.id 2 Department of Public Health, Faculty of Medicine Universitas Airlangga, Surabaya -6013, Indonesia A full list of author information is available at the end of the article infections (5 people; 16.1%), and six of 12 patients with hypertension (50%) patients felt that they were not taking antihypertensive drugs regularly.Knowledge and awareness of a person's illness are important factors in medication adherence [6].Lack of knowledge regarding the course of the disease, hypertension therapy, and a healthy lifestyle will worsen the effectiveness of therapy in patients [7].
Other factors, such as gender, age, place of residence, medical expenses, and socioeconomic status were found to affect medication adherence in hypertension patients [8].It is necessary to identify factors of adherence of hypertensive patients in using drugs, in an effort to plan a more comprehensive therapeutic strategy in order to increase the effectiveness of therapy.Therefore, this study aims to analyze the factors that influence adherence to hypertension treatment in hypertension sufferers in Turirejo Village.

Study design and setting
This research is an observational analytic study with a crosssectional study design.The research took place between 20-24 February 2023 at Turirejo Village, Lawang District, Malang, Indonesia.The research used primary data collected using a questionnaire instrument which was answered by respondents through face-to-face interviews with five medical students.Each interviewer spread in a region of Turirejo Village accompanied by local health cadres to each patient's home.The sampling technique was carried out using the consecutive sampling method, collecting one by one respondent for one day.

Study Location
Turirejo Village is one of ten villages and two sub-districts located in Lawang District, Malang Regency.Topographically, Turirejo Village is plain with an average elevation of approximately 491 m above sea level.The condition of the land in Turirejo Village is classified as hilly, the soil surface is brown with a slope of less than 15.0%.The average temperature is 22°C to 32°C, with a tropical climate and an average rainfall of 160 mm/year.The livelihoods of the residents of the Turirejo Village area range from civil servants, police, private employees, farmers, farm laborers, traders, breeders, transportation services, entrepreneurs, and others.In terms of facilities and infrastructure, Turirejo Village has health facilities, namely the Ponkesdes of Turirejo Village, Lawang District [9].

Inclusion and exclusion criteria
The population in this study were adult hypertensive patients (over 18 years) registered at the Turirejo Ponkesdes in 2022, totaling 112 people.The inclusion criteria were hypertensive patients who were registered in the data and coverage of the Ponkesdes (Village Health Board) of Turirejo Village who were currently in Turirejo Village.Exclusion criteria were participants who did not give consent to participate in the study, patients who could not be found, or patients who had never received treatment for hypertension.

Sample size
The sample size was obtained by five to ten rules of thumb for determining sample size Roscoe [10].Since there are six variables included for regression analysis, by the rule of five, there are a minimum of 30 samples included in this study.Roscoe suggested that a sample size greater than 30 and less than 500 is suitable for most behavioral studies [10].

Study tool and the variables
The dependent variable in this study was adherence to treatment.Compliance was measured using a questionnaire that was compiled based on a previous study [11].This questionnaire has been tested for construct validity and reliability by previous studies [12].The results of the analysis showed that the MMAS-8 questionnaire used was valid with the r count of all questions > r table (0.355) at a significance of 0.05.The results of the analysis also show that the instrument is reliable with a reliability coefficient of 0.729 [12] Assessment of adherence to hypertension treatment was carried out using the MMAS-8 which included eight questions (Table 2).
Questions number 1-7 use "yes" and "no" answer choices, while question number 8 has 5 answer choices, namely always, usually, sometimes, occasionally, and never.Respondents are said to have high compliance if they have a score of 8, moderate compliance with a score of 6-7, and low compliance if a score < 6 [11].The independent variables in this study were age, education level, length of illness with hypertension, participation with health insurance, knowledge of hypertension treatment, and attitude towards hypertension treatment.Knowledge of hypertension treatment was measured using a questionnaire that was prepared based on questions regarding the basic knowledge that hypertension sufferers need to know regarding hypertension treatment (Table 3).
The questionnaire consists of 8 questions, with each correct answer given a value of 1.The validity and reliability test found that all questions were valid with a Cronbach alpha value = 0.619.If the total score is correct ≥7 then the respondent's knowledge is categorized as good, whereas if the total score is correct <7 then the respondent's knowledge is categorized as lacking.Attitudes toward treatment were measured using a questionnaire compiled based on statements that should be made by hypertensive patients (Table 4).
The questionnaire consists of 8 statements with answers in the form of a Likert scale (1)(2)(3)(4).The maximum total score from the accumulation of the Likert scale is 32.The validity and reliability tests found that all questions were valid with a Cronbach alpha value = 0.691.If the total score is ≥25 then the attitude of the respondent is categorized as positive, whereas if the total score is <25 then the attitude of the respondent is categorized as negative.

Statistical analysis
Analysis was performed using IBM SPSS statistics ver.software.23.Variables with nominal and ordinal data scales are presented in the form of amounts and percentages.Each independent variable was subjected to a bivariate logistic regression test on the dependent variable to display the odds ratio (OR) value.The independent variable with the results of the bivariate logistic regression test that has a p-value <0.25 will be subjected to a multivariate logistic regression test with the backward entry method based on the likelihood ratio.All analyzes were performed at a significance level of p < 0.05 with 95% confidence intervals.

Socio-demographic characteristics of study participants
A total of 37 adult hypertensive patients out of 112 patients diagnosed with hypertension at the Turirejo Ponkesdes participated in this study (Table 1).Most of the participants were over 56 years old (73.0%) and had less than 9 years of education (73.0%).There were 29 patients (73.0%) who had been sick for less than 10 years and the rest (27.0%) had had hypertension for more than 10 years.More than half of the participants (57.0%) took part in a health insurance program.
The average knowledge score of all participants was 5.83 ± 1.8 with the distribution of the number of participants with a score below 7 of 19 people (51.0%) and a score above or equal to 7 of 18 people (49.0%).The average attitude score of all participants was 24.14 ± 4.2 with the distribution of the number of participants with a score below 25 of 28 people (76.0%) and a score above or equal to 25 of 9 people (24.0%).From the results of adherence calculated by the MMAS-8 score, there were no patients who had high compliance or a score of 8. Most of the participants had low adherence (76.0%) and the rest had moderate adherence (24.0%).Table 2-4 described the distribution of answers to the knowledge, attitude, and MMAS-8 questionnaire.We found that most patients were disobedient because they forgot to take their medication (76.0%) and stopped their medication on their own because they felt their condition was improving (controlled blood pressure) (76.0%).
The statement that most patients obeyed was not forgetting to bring medicine when traveling (59.0%).

Bivariate analysis
Bivariate analysis shows the effect of the independent variables on the dependent variable (Table 5).There was no significant effect of age, education, and health insurance participation on participant compliance in treating hypertension.The length of illness variable, knowledge score, and attitude score had a significant effect on participant adherence in carrying out hypertension treatment.

Multivariate analysis
Multivariate analysis using multiple logistic regression was performed to look for factors that simultaneously affect adherence to hypertension treatment.There are three steps in the multivariate logistic regression analysis test; the first step was followed by four independent variables (age, length of illness, knowledge, and attitude).The final step leaves two variables, namely length of illness and attitude (Table 6).
The Omnibus Tests and the Hosmer and Lemeshow tests show that the analytical model formed is fit to the data.Participants with an illness of more than 10 years were eighteen times more obedient than participants with an illness of less than 10 years (Adjusted OR 18.27; 95% CI 1.72-194.47;p=0.02).Participants with an attitude score of more than or equal to 25 were twelve times more obedient than participants with an attitude score below 25 (Adjusted OR 12.76; 95% CI 1.25-130.40;p=0.03).It can be concluded that the longer the illness and the higher the attitude score, the more likely the participants are to comply with hypertension treatment.The constants of the multivariate binary logistic regression mathematical model are negative.This means that before the addition of the variable length of illness or attitudes toward hypertension treatment, there were already other factors that lead to low adherence.Supported by Nagelkerke R Square of 46.3%, there are still many factors that have not been successfully included in this analysis model.

Discussion
The health behavior of an individual is influenced by several factors.This study measured the influence of factors on medication adherence, including the level of knowledge, attitudes, and demographic factors.In this study, most of the participants had low adherence according to the results of the preliminary studies conducted.Based on data from the results of the distribution of answers to the MMAS-8 questionnaire, it was found that most patients did not comply because they forgot to take their medication and stopped their own medication because they felt that their condition was improving (controlled blood pressure).Patients with hypertension in general are often accompanied by other comorbid diseases [12].This causes a large number of drugs to be taken together with hypertension drugs so that many patients are confused or even forget to take their medication [12].The study by Vrijens et al. [13] stated that the burden of more pills makes the habit of taking daily medication more complicated, often a barrier to optimal medication adherence.Respondents who stopped selfmedication could be due to a misunderstanding about the absence of symptoms as a cure [14,15].The wrong perception that hypertension drugs are ineffective in controlling blood pressure also plays a role in non-compliance [16,17].Ineffective communication between patients and health workers can hinder the delivery of disease-related education, including the importance of taking antihypertensive drug therapy in asymptomatic conditions [18].The results of this study found that older respondents (≥56 years) had better adherence (22%) than respondents aged <56 years (3%), although there was no significant difference.The insignificant effect of age can occur due to factors that both encourage and hinder adherence in older and younger groups [18,19].The factors that encourage adherence in the older age group are better health awareness and the presence of cadres who often conduct counseling and encourage people to comply with taking medication [20,21].
Obstacles are decreased cognitive ability, physical function, difficulty swallowing medication, ability to care for oneself, and a lower likelihood of living with other people [22][23][24].The results showed that there was no significant relationship between recent education and adherence to taking antihypertensive medication [25].Even though a higher level of education is considered to have more potential to maintain health, people with low levels of education are also considered to be able to access health information just as well [26].The existence of counseling for various groups of people can be a reason for the balance of information between groups of community education.Insurance participation has no significant effect on adherence to hypertension treatment [27,28].This is due to the small number of people using insurance, namely those living in rural areas.
People tend to choose to buy their own medicines at pharmacies compared to coming to take advantage of health insurance [29].
Another factor is the government's policy regarding the price of drugs, which refers to drug price margins that can be reached by the public, making it easier to buy drugs independently [30].In this study, the patient's length of time with hypertension affects adherence to hypertension treatment.This is similar to previous studies [31][32][33][34].The longer a person suffers from a disease, the more knowledge and understanding he will have about the disease [5].Greater concern about symptoms due to hypertension can also increase adherence [34][35].Health knowledge is defined as all the results of finding out about an object, therefore health knowledge is the result of finding out with all the senses about health [36] Patients with a good level of knowledge generally have better adherence [37,38].This is because these patients understand more about how to treat hypertension and the dangers if not routinely treated [37].Patients with more knowledge can also discuss with health workers to determine the most suitable treatment method [38][39][40].Similar to knowledge, there is a significant relationship between patient attitudes and adherence to hypertension treatment [41].This is because attitude statements show an individual's tendency to act [42].
Both attitude and duration of illness can interact to influence adherence.Patients who are sick longer generally gain more experience so that they have a positive attitude [31].In addition, patients who were ill for longer were found to have a good doctor-patient relationship and have greater confidence in the doctor's advice [31].Patients with long-standing hypertension are more prone to complications, thus adding to the patient's consideration to continue to comply to avoid complications [43].This research has several drawbacks.First, the number of hypertensive patients participating in this study is still limited, this is because the population of hypertensive patients is limited and the public's willingness to be interviewed is quite low.Second, there are still many factors that have not been identified, such as the economy, distance to health access, availability of companions, and places to buy medicines.Third, the questionnaire used in this study was in the form of a closed questionnaire.This resulted in not conducting in-depth interviews for every answer regarding the factors that influence compliance.In-depth interviews are expected to help identify the reasons for their decision.

Conclusion
Length of hypertension illness, knowledge of hypertension treatment, and attitudes regarding hypertension treatment affect adherence to hypertension treatment in hypertensive patients in Turirejo Village.This research supports an increase in service and assistance for cadres, especially for the elderly, most of AbbreviationMMAS-8: Morisky Medication Adherence Scale-8; OR: Odds RatioDeclaration AcknowledgmentWe acknowledged the contribution of the Turirejo Village Government and Health Cadre in providing the demographical data and access for conducting this study.

Table 1 :
Demographics and other characteristics of the sample (n=37)

Table 2 :
Distribution of Answers to the MMAS-8 Questionnaire

Table 3 :
Distribution of Answers for Knowledge Regarding Hypertension Treatment Questionnaire

Table 4 :
Distribution of Answers for Attitude Regarding Hypertension Treatment Questionnaire

Table 6 :
Results of multivariate binary logistic regression (N=37) *significant p<0.05.Only significant variables are included in the table.