DOI: https://doi.org/10.21203/rs.3.rs-1783245/v1
Background: Antihypertensive medication adherence plays a tremendous role in controlling blood pressure, however poor adherence to antihypertensive medication leads to complications of hypertension, wastage of health care resources and workload on health care providers which come up with poor clinical outcome.
Objective: To assess antihypertensive medication adherence and its associated factors among hypertensive patients’ in North Showa Zone public hospitals’, Oromia regional state, North Ethiopia. 2021.
Methods and Materials: An institutional cross-sectional study was conducted from February to March 2021 in three public hospitals of North Showa Zone. Out of four public hospitals, three hospitals were selected through simple random sampling techniques and final sample size 348 was selected systematic random sampling. Data were checked, cleaned and entered into Epi data software version 3.1 and imported to SPSS version 25 software for analysis. Descriptive analysis was done and presented in frequencies and percentages. The association between independent and dependent variables was analyzed using bivariate and multivariate analysis, and variables that show P-values of < 0.05 with 95% CI will be considered statistically significant.
Result: Out of the sample size, 339 patients were participated with 97.4% response rate. Out of this, 56.9% adhere to their medication. Marital status (AOR=3.24,95%CI=1.12-9.60), residence (AOR=2.59,95%CI=1.42-4.75), Controlled BP (AOR=4.26, 95%CI=2.37-7.67), number of medication (AOR= 2.59,95%CI=1.35-4.97), comorbidity (AOR=2.24, 95%CI=1.33-4.67), knowledge (AOR=5.01,95%CI=2.69-9.35) and insurance coverage user (AOR=2.20,95%CI=1.23-3.95) were found statistically significant.
Conclusion: The adherence status of this study was sub-optimal and knowledge status of patients about the disease and their treatments was the most associated factor. Give health education about disease and its treatments, management of comorbidity and adherence counseling plays a vital role in improving the adherence status of clients.
Hypertension (HTN) is a condition in which the blood vessels’ pressure has raised or when a systolic blood pressure is ≥ 140 mm Hg and diastolic blood pressure is ≥ 90 mm. It is accountable for 51% of deaths due to stroke and 45% of deaths due to heart disease, and disproportional in low- and middle-income countries(1)
Hypertension is the most leading risk factor for cardiovascular disease (CVD) and accountable for more than 10 million preventable deaths worldwide each year and expected to increase by 29% to reach 1.56 billion by 2025 (2, 3). According to the global burden of diseases reported in India shows that, HTN leads to 1.6 million deaths and 33.9 million disability-adjusted life(4).
The prevalence of HTN has been continuously increasing at an alarming rate and put a great impact on health care providers(5). In United States 29%, Turkey, approximately 45–50%population over 50years old, Kenya 24.5% and 19.6% in Ethiopia has HTN (6–9)
Adherence is the extent to which a person behavior taking medication and making healthy lifestyle changes corresponds with recommendations from health care providers (10). Therefore adherence to prescribed medical regimens plays an incredible role in sustaining the health and well-being of individuals with HTN (11). Besides, antihypertensive medication adherence (MA) is an important predictor of optimal blood pressure (BP) control and its complication prevention (2, 12, 13)
Blood pressure control remains unsatisfactory in the worldwide and several factors have been identified which hinder BP control. As various studies have been done show that lack of persistence follow-up by a poor day -to -day execution of the prescribed drugs is the major challenge in controlling BP (14). Only the availability of effective antihypertensive medications does not bring a good outcome in controlling BP, rather than adhering to their medication (15)
World health organization (WHO) recommends proper attention to be given towards adherence issues more than the development of new treatment regimens. On average, the overall adherence to long-term therapies for chronic illnesses is 50% among the developed countries and lower in developing countries(16). The proportion of hypertensive patients, whose disease treated effectively with medication remains disproportional. Northern Vietnam 49.8%, Romania (68.9%), Sir Lanka (71.8%), Korea (81.7%), Kenya 62.4% and 31.4% West Ethiopia had good adherence to antihypertensive medication (17–23)
Poor adherence is a global public health concern with substantial health and cost implications worldwide. It is responsible for 125,000 deaths and $US100 billion annually cost in USA (24). It also increases the health care resources expenditure, bad outcome of diseases, an avoidable hospital admission and increases the adverse effects that leads the patients’ hopeless, low satisfaction and come up with the patient decided to stop their treatments, which come up with complication of HTN or death (25–28). Besides, it reduces the effectiveness of medication and the efficiency of the health care system (29–31)
Poor adherence and poor HTN control still unresolved big challenges for health care providers (26). There are various factors that affect a hypertensive patient’s behavior regarding adherence towards antihypertensive medications. Knowledge about HTN and its treatments (22, 26, 32–35), socio-demographic, types of medication prescribed, social support, patient health provider relationship and cost of the health care are the predictors of antihypertensive MA (15, 36, 37).
Therefore identifying factors that hinder the adherence status of HTN patients towards their medications was a fundamental step for future improvements that tailor specific interventions to improve MA (38). However in Ethiopia, there is a scarcity and inconsistence results on antihypertensive MA and its associated factors (12, 29), and no study have been done in this study area before. So, this study will be aimed to assess antihypertensive medication adherence and its associated factors among hypertensive patients in North Showa Zone Public Hospitals, North Ethiopia 2021.
A hospital-based cross-sectional study among adult hypertensive patients was conducted at North Showa Zonal public hospitals from November 2020 to May 2021. The Zone’s capital city is Fiche, located at distance of 112 km from Addis Ababa in the north direction. It has four governmental hospitals and the total number of HTNs follow up in this public hospital within a quarter is 8110. Fiche general hospital has 2626, Muke Turi hospital 1144, Kuyu hospital 1980 and 1337at Gundo Meskel hospital.
All hypertensive patients who had a follow up at North Showa Zone public hospitals, were involved in the study with a source of the study, all hypertensive patients found in the public hospitals whose age were 18-years and above, on antihypertensive medications at least for 6-month duration and available during the data collection. Those who had cognitive impaired and seriously ill, were excluded from the study.
A single population proportion formula was used to calculate the total sample size by considering 1.96 for the standard normal variable with 5% level of significance (α-value), 95% confidence interval, 5% margin of error, 10% non-response rate and a correction formula was used. A prevalence of 31.4% antihypertensive MA was taken from a study done in West Ethiopia (23). Therefore the initial sample size will be equal to 331 and 316 after substituting the correction formula. By taking 10% non-response rate, the final sample size used for the study was 348.
Out of four public hospitals, three hospitals were selected using simple random sampling (SRS) techniques for the study and the number of study units for each hospital was proportionally allocated based on the number of patients coming per-month and those who were a part of the final sample size were selected using systematic random sampling. The 1st interviewer was selected by lottery methods from chronic illness clinical registration books coded with the“K”value and continues until sample size required was full filled by adding the“K”value to the sub-sequent selected one.
Socio-demographic (sex, age, marital status, number of children, monthly income, educational status and residence), clinical (comorbidity, BP status, BMI, time of diagnosis, duration of treatment and number of drugs used), personal knowledge, organizational (health care provider-patient relation, insurance coverage, drugs availability and drugs changed by Dr.) and social support were the predicators of antihypertensive MA among hypertensive follow up.
A structured and pretested interviewer-administered questionnaire was used on the following listed parts:- Socio-demographic factors, around ten (10) questions, clinical factors, personal factors, organizational factors, contains 7 questions, social supporting factors that contains 12 questions and antihypertensive MA which contains 8 questions.
All the above-stated questionnaire was adopted from similar study done before and some standard guidelines tools, used to assess antihypertensive MA and its associated factors. Such as the WHO STEPwise approach to chronic disease risk factor surveillance (39). Adherence status of antihypertensive medication contains eight items Morisky Medication Adherence Scale (MMAS-8) was used to assess the study participant’s MA. A scoring scheme of “Yes” = 1 and “No” = 0 for the first seven, except the 5th question which reversed and items in the last a five-point linkert response, was used with options “never”, “once ”, “sometimes”, “usually”, and “always”, low < 6, 6–8 medium and 8 high adherents (40)
The Duke Social Support and Stress Scale contain 12items that were used to assess social support gained from family, friends or significant others. The responses were coded as follows: “none” =0, “some” =1, “a lot” =2, “yes” =2, “no” =0 and “there was no such person” =0. A blank response was considered as “0”, the support score was calculated by summing the six responses in both sections (family and non-family support); based on the reply to the last question, 2 was added to either the family or non-family support. The total result was divided by 22 and multiplied by 100 to give a 0 to 100 score (41). All question was first prepared in English and translated to the local language of the community “Afan Oromo and Amharic” for data collection and translated back to English by independent language expert in order to ensure its consistency.
Data was collected through face-to -face interviewers by three BSc nurse those assigned as data collector separately to each hospitals and supervised by two experienced BSc nurse. Training was given by the principal investigators for two days on overview of the study topics, interview technique, ethical issues and the rights of participants for the sake of keeping consistence in all of the data collectors and to minimize bias through the data collection. Five 5% pretest was done out of the total sample size at Chanco hospital to assess whether the checklist item was easily understood by data collectors and to take a corrective measurement if it has error.
Close supervisions at the end of every data collection was made and all the interviewed questionnaires were checked visually by the principal investigator. Data were coded, cleaned and entered Epi. Data version 3.1 software. Double entry was made to cross-check the data for completeness before analysis. The entered data was be exported and analyzed with Statistical Package for Social Science (SPSS) version 23 software.
A bivariate analysis was computed to test whether there is an association between dependent and selected independent variables, respectively. Factors associated with MA on bivariate analysis were identified with P-values of < 0.2 were taken to multivariable logistic regression and variable with 95% CI which had a P-value < 0.05 were declared as having a significant relationship with the outcome variables. Multicollinearity and fitness of the model was checked before multivariate logistic regression, therefore a Hosmer-Lemeshow goodness of fit test was done and the model was adequately fit with a P-value > 0.05.
In this study, a total of 348 chronic follow-up hypertensive patients were interviewed from three public hospitals in North Showa Zone those fulfill the inclusion criteria with a response rate of 97.4%. As shown in Table 1, from the participants more than half (56.3%) were female. The mean age of the respondents was 53.26 ± 11.086 years and more than half of the respondents 197(58.1%) were in the 40–59 age group. More than three quarter 280(82.6%) of the participants were married and 55.5% of them had four and above four children. Two hundred twenty-seven (67%) of respondents were live in urban and 109(32.2%) can’t read and write. More than half (63.2%) of them were less than 2500 ETB monthly income and 143(42.2%) farmers.
| Variables | Frequency | Percent | ||
|---|---|---|---|---|
| Sex | Male | 148 | 43.7 | |
| Female | 191 | 56.3 | ||
| Age | 18–39 | 45 | 13.3 | |
| 40–59 | 197 | 58.1 | ||
| ≥ 60 | 97 | 28.6 | ||
| Religion | Orthodox | 225 | 66.4 | |
| Muslim | 73 | 21.5 | ||
| Protestant | 41 | 12.1 | ||
| Marital status | Single | 15 | 4.4 | |
| Married | 280 | 82.6 | ||
| Divorced | 16 | 4.7 | ||
| Widowed | 28 | 8.3 | ||
| Number of children | ≤ 3 | 151 | 44.5 | |
| > 4 | 188 | 55.5 | ||
| Residence | Urban | 227 | 67 | |
| Rural | 112 | 33 | ||
| Educational level | Can’t read and write | 109 | 32.2 | |
| Read and write | 77 | 22.7 | ||
| Primary | 68 | 20.1 | ||
| Secondary | 36 | 10.6 | ||
| College/University | 49 | 14.5 | ||
| Occupation | Farmer | 143 | 42.2 | |
| Merchant | 110 | 32.4 | ||
| Gov’t employee | 58 | 17.2 | ||
| Daily labour | 28 | 8.3 | ||
| Monthly income | < 2500 | 214 | 63.2 | |
| ≥ 2500 | 125 | 36.9 | ||
In this study, more than one -third (39.1%) of the respondents’ BP were ≥ 140/90mmHg, and 84.4% of them had a normal range of BMI. More than three-quarters (78.2%) of them was a 5years history of HTN and half (50.4%) of them were on treatments for 2–4 years. The mean number of pills taken per day was 1.95 ± 0.84 and 30.1% of them taking three and above drugs (Table 2). Out of the participants more than half (52.5%) of them had no comorbidities (Fig. 1) and 62.8% had good knowledge about HTN and its treatments’. The total mean score of respondents‟ social support on the Duke’s social support and stress scale” was 49.96 ± 14.63 and more than half (53.4%) of them got support from family or non-family members (Table 2).
| Variables | Frequency | Percent | |
|---|---|---|---|
| Blood pressure | less than 120/80mmHg | 30 | 8.8 |
| 120–129/<80mmHg | 57 | 16.8 | |
| 130–139/80-89mmHg | 119 | 35.1 | |
| ≥ 140/90mmHg | 126 | 39.2 | |
| BMI | Less than 18.5 | 14 | 4.1 |
| 18.5 to < 25 | 286 | 84.4 | |
| ≥ 25 | 39 | 11.5 | |
| Time of diagnosed in year(s) | < 5 | 265 | 78.2 |
| ≥ 5 | 74 | 21.8 | |
| Duration of treatment | 0.5-1 | 52 | 15.3 |
| 2–4 | 171 | 50.4 | |
| ≥ 4 | 116 | 34.2 | |
| Types of medication | 1 | 125 | 36.9 |
| 2 | 112 | 33.0 | |
| ≥ 3 | 102 | 30.1 | |
| Knowledge status | Good | 213 | 62.8 |
| Poor | 126 | 37.2 | |
| Social supporting status | Supported | 181 | 53.4 |
| Unsupported | 158 | 46.6 | |
More than half (57.8%) of them used insurance coverage and got their medication free from the hospitals, however around 43.4% of the patients respond that, drugs didn’t available in the hospitals. Almost all (95.6%) of the respondents had good relationships with their health care providers. Nearly half (48.1%) of the respondents’ drugs changed by their Doctors, 15.6% had changed three and above three respectively (Table 3).
| Variables | Frequency | Percent | |
|---|---|---|---|
| Insurance coverage user | Yes | 196 | 57.8 |
| No | 143 | 42.2 | |
| Drug available in hospital | Yes | 192 | 56.6 |
| No | 147 | 43.4 | |
| Relationships with health care providers | Yes | 324 | 95.6 |
| No | 15 | 4.4 | |
| Drugs Changed by Dr. | Yes | 163 | 48.1 |
| No | 176 | 51.9 | |
| If yes, how many | 1 | 57 | 16.8 |
| 2 | 53 | 15.6 | |
| ≥ 3 | 53 | 15.6 | |
Medication Adherence Characteristics of the respondents.
Through using MMAS-8 scale individuals who respond 6 and above were considered as adherent and the opposite was true for non-adherence. Out of the 339 chronic follow-up hypertensive patients, more than half (56.9%) of them adhere to their medication (Fig. 2). More than one-third (34.5%) hadn’t taken their medication due to forgetfulness (Table 4).
| Variables | Frequency and percent | Frequency and percent |
|---|---|---|
| Yes | No | |
| Do you ever forget to take your medicine? | 117 (34.5) | 222(65.5) |
| In the last two weeks, is there any day when you did not take your high blood pressure medication? | 82(24.2) | 257(75.8) |
| Have you ever stopped taking your medications or decreased the dose without your doctor order, because you felt worse when you took them? | 82(24.2) | 276(78.8) |
| Do you forget to take your medications, when you travel or leave the house? | 111(32.7) | 228 (67.3) |
| Did you take your high blood pressure medication yesterday? | 297 (87.6) | 42(12.4) |
| Do you stop taking your medications, when you feel your blood pressure is controlled? | 38 (11.2) | 301(88.8) |
| Have you ever felt distressed for strictly following your high blood pressure treatment? | 49 (14.5) | 290 (85.5) |
| How often do you have difficulty to remember taking all your blood pressure medications? | ||
| Never 146(43.1) | ||
| Once 49(14.5) | ||
| Sometimes 110(32.4) | ||
| Usually 33(9.7) | ||
| Always 1(0.3) | ||
| Adherent 193 56.9 | ||
| Non-adherent 146 43.1 | ||
A total of nineteen (19) variables were found to be candidates for multivariate analysis with a p-value less than 0.2 (Table 5). In order to find out the predictors of antihypertensive MA, multivariate logistic regression analysis was done. Hosmer-Lemeshow goodness of fit (P = 0.245) and multicollinearity (VIF = 1.04–1.58) were tested and the model was adequately fit and no multicollinearity among the variables. From the entered variables in multivariate logistic regression seven variables, namely: marital status, residence, BP status, number of medications used, co-morbidity, knowledge status about HTN and its treatments, and insurance user showed a significant association with antihypertensive MA among HTN patients follow-up at P-value < 0.05.
| Variables | Adherence | COR (95%CI | P-value | AOR (95%CI) | P-value | ||
|---|---|---|---|---|---|---|---|
| Adherent | Non-adherent | ||||||
| Sex | Male | 79 | 69 | 1 | |||
| Female | 124 | 67 | 2.12(1.37–3.29) | 0.001* | |||
| Age | 18–39 | 23 | 23 | 1.21(0.06–2.43 | 0.604 | ||
| 40–59 | 126 | 70 | 2.17(1.32–3.56) | 0.002* | |||
| ≥ 60 | 53 | 44 | 1 | ||||
| Marital status | Single | 3 | 12 | 0.63(0.14–2.82 | 0.54 | 0.46(0.08–2.64) | 0.380 |
| Married | 178 | 102 | 4.36(1.86–10.26) | 0.001* | 3.28(1.12–9.60) | 0.030** | |
| Divorced | 4 | 12 | 0.83(0.21–3.37) | 0.798 | 0.59(0.11–3.17) | 0.539 | |
| Widowed | 8 | 20 | 1 | 1 | |||
| No. children | ≤ 3 | 99 | 52 | 1.90(2.07–5.31) | 0.004* | ||
| > 4 | 94 | 94 | 1 | ||||
| Residence | Urban | 152 | 76 | 3.31(2.07–5.31) | 0.002* | 2.59(1.42–4.75) | 0.002** |
| Rural | 42 | 70 | 1 | 1 | |||
| Educational level | Informal | 88 | 103 | 1 | |||
| Formal | 105 | 43 | 2.86(1.81–4.50) | < 0.001* | |||
| Occupation | Farmer | 79 | 64 | 2.22(0.96–5.15) | 0.063* | ||
| Merchant | 66 | 44 | 2.70(1.14–6.39) | 0.024* | |||
| Gov’t employe | 38 | 20 | 3.42(1.33–8.79) | 0.011* | |||
| Daily worker | 10 | 18 | 1 | ||||
| Monthly income | < 2500 | 117 | 97 | 1 | |||
| ≥ 2500 | 76 | 49 | 1.29(0.82–2.01) | 0.27 | |||
| BP status | Controlled | 145 | 61 | 4.21(2.65–6.70) | < 0.001* | 4.26(2.37–7.67) | < 0.001** |
| Uncontrolled | 48 | 85 | 1 | 1 | |||
| BMI | Less than 18.5 | 6 | 8 | 1.08(0.33–3.50) | 0.898 | ||
| 18.5-24.99 | 162 | 102 | 2.29(1.30–4.03) | 0.004* | |||
| ≥ 25 | 25 | 36 | 1 | ||||
| Duration of diagnose | < 5years | 163 | 102 | 2.34(1.39–3.97) | 0.002* | ||
| ≥ 5years | 30 | 44 | 1 | ||||
| Duration of treatmen | < 5years | 163 | 102 | 2.34(1.39–3.97) | 0.002* | ||
| ≥ 5years | 30 | 44 | 1 | ||||
| No. of drugs | ≤ 2 | 91 | 40 | 2.36(1.49–3.75) | < 0.001* | 2.59(1.35–4.97) | 0.004** |
| ≥ 3 | 102 | 106 | 1 | 1 | |||
| Comorbidity | Absence | 131 | 46 | 4.59(2.90–7.29) | < 0.001* | 2.48(1.33–4.61) | 0.004** |
| Presence | 62 | 100 | 1 | 1 | |||
| Knowledge status | Poor | 41 | 86 | 1 | 1 | ||
| Good | 152 | 61 | 5.17(3.21–8.32) | < 0.001* | 5.01(2.69–9.35) | < 0.001** | |
| Insurance user | Yes | 134 | 62 | 3.08(1.96–4.82) | < 0.001* | 2.20(1.23–3.95) | 0.008** |
| No | 84 | 59 | 1 | 1 | |||
| Drugs in hospital | Yes | 113 | 79 | 1.37(0.89–2.16) | 0.104* | ||
| No | 80 | 67 | 1 | ||||
| Relation with HCPs | Yes | 188 | 136 | 2.77(0.92–8.27) | 0.069* | 3.73(0.98–14.22) | 0.054 |
| No | 5 | 10 | 1 | 1 | |||
| Drugs changed by Dr. | Yes | 86 | 77 | 1 | 1 | ||
| No | 107 | 69 | 1.39(0.90–2.14) | 0.136* | 0.56(0.30–1.07) | 0.078 | |
| If yes how many | 1 | 32 | 25 | 1.55(0.73–3.28) | 0.256 | ||
| 2 | 30 | 23 | 1.58(0.73–3.39) | 0.245 | |||
| ≥ 3 | 24 | 29 | 1 | ||||
| Social support status | Unsupported | 76 | 82 | 1 | 1 | ||
| Supported | 117 | 64 | 0.51(0.33–0.78) | 0.002* | 1.78(0.99–3.20 | 0.055 |
|
Note: “*” Variables with P-value<0.2, “**”variables with p value <0.05, P- value =0.000 considered as P<0.001, AOR adjusted odds ratio, CI confidence interval, COR crude odds ratio,” BMI” body mass index, “1” reference category, “BP” blood pressure, “HCPs” health care providers
Table 5 showed that, the odd of those who got married were 3 times (AOR = 3.28, 95%CI = 1.12–9.60) more adhere than the odd of widowed. Individuals who live in the urban area were three times (AOR = 2.59, 95%CI = 1.42–4.75) more adhere than the counterpart. The odd of those who have controlled BP had four (AOR = 4.26, 95% CI = 2.37–7.67) higher than the odds of uncontrolled BP. The odds of adherence of those who took less than or equal to two drugs were three times (AOR = 2.59, 95%CI = 1.35–4.97) higher than the odd of adherence among hypertensive patients who took three and more drugs (Table 5).
The odd of adherence of those who had no comorbidities were two times (AOR = 2.48, 95%CI = 1.33–4.67) more adhere than odd of the counterpart. The odds of adherence to anti-HTN medication among those who had good knowledge was five times (AOR = 5.01, 95%CI = 2.69–9.35) higher than the odds of adherence among HTN patients who had poor knowledge. Besides, the odd of those whose drug cost covered by insurance was two (AOR = 2.20, 95%CI = 1.23–3.95) times higher than the odd of those paid a cost (Table 5)
The adherence status of HTN patients to anti-hypertensive medications remains a major challenge for public health, especially in developing countries. Poor adherence to antihypertensive medication is the main reason for uncontrolled HTN, serious complications, and wastage of health care resources. Drug adherence is a crucial issue in the management of HTN diseases. As different studies show the major challenge in controlling BP is lack of persistence follow up with poor day to day execution of the prescribed drugs (14)
In this study, more than half (56.9%) of the respondents were found to be adherent to their antihypertensive medication treatment. This finding consistent with the study done in Kenya (62.4%) (22) and JUSH (61.7%)(30). The observed 56.9% is higher than the study reported in India (23.7%) (14), North Vitienema (49.8%) (18), Cameron (43.9%) (5) and Nedjo general hospital (3.14%) (23). The possible discrepancy might be explained, in the current study more than half (57.8%) of HTN patients receive free medical care and drugs, whereas, in the other study, patients pay for their treatment. In Nedjo general hospital small (172) sample size was used.
However the present finding is lower than the studies done in Romania (69.8%) (19), Sri Lanka (71.8% ) (20) and Korea (81.7%) (21), Debre tabor general hospital (75.1%) (12) and University of Gondar hospital (64.6%) (28). This inconsistency might be due to better health care and health facility, better knowledge toward HTN and its management among participants and socio-demographic variation in Romania, Sri Lanka, and Korea, as well as sample size variation (1,523) in Korea than the current study. In Debre tabor general hospital and University of Gondar hospital study, the adherence status was assessed using the MMAS-4.
In the current study, there is a significant association between marital status and antihypertensive MA. Those who got married were three times more adherent than widowed. This finding is in line with the study done in north United Arab Emirates, Ghana and Black Lion hospital (37, 42, 43). This association could be due to, those who married had a person to whom share their stressor as well as one can remind the others, the time to take their medication.
This study shows that, those who live in urban area were three times more adherent than the counterpart. This finding is consistent with a study done in Debra tabor general hospital and Hawass referral hospitals (12, 15). The association might be explained that urban residents lived to have a high chance to access media, too near the health facility, and took less time to reach the health facility for their medications. Besides this, they might have a high chance to remember the time to take their medication.
However, there is an inconsistent study done in India (44), Korea (45), and Northwest Ethiopia (29). The discrepancy might be due to India and Korea is a developed country in which all communities can easily access the media, health facility and reach the health facility on time. While the local discrepancy could be due to the study setting, more than three quarter (78.2%) of them were urban, whereas in the current study (67%) is urban.
BP control remains unsatisfactory in all countries of the world and several factors have been identified which hinder BP control. From them, poor MA plays a tremendous role (14). In this study, those who had controlled BP were four-time more adherent than those who had uncontrolled BP. This finding in line with the study done in Uzbekistan (33), Romania (19), Ghana and Nigeria (46), University of Gondar hospital (28) and Hawassa referral hospital (15).
The association could be explained that controlled BP might be contributing to better outcome of treatment, offer the patient good satisfaction and creates strong motivation of the patients towards the treatment. But, uncontrolled BP could make the patient hopeless and low satisfaction, and come up with stop their treatment.
There is a significant association between the number of antihypertensive drugs taken and the adherence status of patients. In this study, those who took less than or equal to two drugs per day were three times more adherent than those who took three and more. This finding agreed with the study done in Brazil (47), United Arab Emirates (37), Algeria (25), Debre tabor general hospital (12), and JUSH (30). The association might be due to the fact that when the number of drugs taken by patients decreases, their ability to remember and memorize the proper intake of the drugs (right drugs at right time) increased. Besides, fewer drugs have fewer side effects, which ultimately increase the adherence status of patients.
Co-morbidities can worsen the conditions of patient's diseases and the burden of cost that decrease anti-hypertensive MA. This study revealed that individuals with one or more comorbidities were less adherent to their medication than the counterpart. This finding is supported by the study done in Debra tabor general hospitals, Algeria, and Iran (12, 25, 48). The possible reason could be due to a complicated treatment regimen for both HTN and comorbidities could result to poly-pharmacy and create fear on patients about its side effect that hinder the adherence of prescribed medication. Additionally, when patients have another added health problem, their attention could be diverted towards the newly developed disease.
This study inconsistent with the study done in India (14) and Saudi Arabia (38). The variation might be due to better health facilities accessibility increases the chance of early diagnosis and management of comorbidities, the ability to perceive the seriousness of their health condition and want to prevent further complications. Besides, experiencing more severe symptoms of other conditions increase the patients’ adherence to their medication.
Poor adherence and poor HTN control still unresolved big challenges for health care providers, but having information about the HTN and its treatment is a significant independent determinant of good adherence (32). In this study those who had good knowledge about HTN and its treatments were five-time more adhere than the counterpart. This finding in line with the study done in Debra tabor general hospital, Poland, United Arab Emirates, Nigeria, Kenya, Uzbekistan and India (12, 22, 32–34, 49, 50). The association might be due to good knowledge about HTN and its treatment creates a clear understanding and avoids confusion about the disease condition and its treatment.
In this study, those who used insurance coverage were two times more adherent than those who paid cost per their follow-up. This finding similar to the study done in Algeria (25) and Northwest Ethiopia (29). The association might be explained that insurance users didn’t worry about the cost of drug due to the burden of cost pay was already covered. However the present study contrary to the study done in Korea (45) and United Arab Emirates (37). The discrepancy might be due to Korea and the United Arab Emirates is a country in which the community can afford the cost of drugs and the health care facility.
This study includes the most important variables which have an association with anti-hypertensive MA and incorporates more public hospitals found in North Showa Zone. The MMAS-8 was used with 83% reliability and generalizes all hypertensive patients who had follow-up at North Showa Zone public hospitals. However, it didn’t include the private hospitals and patients who didn’t visit the hospitals during the data collection. Besides, it didn’t determine the cause and effects, and might have a bias of self-response.
The adherence status of this study was sub-optimal and the knowledge status of HTN patients about the disease and its treatments was the most predictors of antihypertensive medication adherence. Give health education for patients about HTN and its treatments, early diagnosis, management of comorbidity and adherence counseling plays a vital role in improving the adherence status of clients.
Adjusted odd ratio
Blood pressure
Confidence intervals
Crude odd ratio
Cardio Vascular Disease
Ethiopian Birr
Hypertension
Jima University Specialized Hospital
Medication Adherence:MMAS:Morisky Medication Adherence Scale
Statistical package for social science
Ethical approval and consent to participate: Ethical clearance was obtained from the research and ethics committee of the department of nursing and midwifery of Addis Ababa University. After permission was granted from each institutions, verbal and written informed consent was obtained from participants’. All methods were carried out in accordance with relevant guidelines and regulation. Participants’ identity was kept anonymous throughout the data collection and analysis process. The study was conducted according to the Declaration of Helsinki
Consent for publication: Not applicable.
Availability of data and materials: - All the data supporting the study findings are within the manuscript. However additional detailed information and raw data are available from the corresponding author on reasonable request.
Competing interest: The authors declare that they have no conflict/competing of interests.
Funding: Addis Ababa University has covered the required fund for this research project
Authors' contribution: - AG, ZM and TA designed the study, drafted the manuscript and contributed to its content. ID, AT, DH, and MM contributed to data collections, analysis and interpretation of the findings. AG analyzed and interpreted the data and wrote the manuscript. All authors revised the overall of manuscript and approved the final for publication.
Acknowledgement; Authors would like to thank Addis Ababa University College of Health Sciences, Department of Nursing and Midwifery, and Salale University for their financial and ideal support. We also thank data collectors and study participants of this research.
Author Information: - 1Department of Nursing, College of Health Science, Salale University, Fiche, Ethiopia, 2School of Nursing and Midwifery, College of Health Science, Addis Ababa University, Addis Ababa, Ethiopia, 3 Department of Medicine, College of Health Science, Salale University, Fiche, Ethiopia