Blood pressure control status of patients with hypertension on treatment in Dessie City Northeast Ethiopia

A cross-sectional study was done from March to May 2021 among adults aged 18 or more patients with hypertension for three months or more on treatment [14] in Dessie City. Patients were treated in eight health facilities with different health care system levels. One Comprehensive Specialized Hospital, one Generalized Specialty Hospital, and six health centers within the district of the city. There were more than 2109 patients with hypertension on treatment in the city at the end of January 2021. The national guideline recommends for patients with hypertension visit the health facility on monthly basis by checking their blood pressure status at each visit and modification of drugs according to the recommended guidelines [13]. But the guideline was violated in the era of COVID-19 and amended as patients visit the health facility every three months or more. Adult patients with hypertension active on treatment were considered for inclusion in this study. Since chronic hypertension is a risk of high blood pressure during pregnancy [14], pregnant women were excluded from this study.

The largest sample size was obtained using a single population proportion formula from the study conducted in Southwest Ethiopia with 49.7% of uncontrolled blood pressure including an assumption of 95% level of Confidence and 5% of errors [9]. So, the final sample size of this study was 380 patients with hypertension. Participants were recruited into the study using simple random sampling methods. Study participants were listed from the hypertension register book with serial numbers for each facility then a computer-generated lottery method was employed from serial numbers.

Most of the data were collected using structured interviewer-administered questionnaires, which were prepared to address all the important variables with supportive document review using patients’ charts after obtaining informed consent to immediately exit from their clinical care. The questionnaires were adapted from different kinds of literature of similar studies [8‐10, 15]. The perceived social support-related questionnaire is adopted from the Multidimensional Scale of Perceived Social Support (MSPSS) -12 item check [16] and it was validated with a similar setting [17]. Each of these items was scored from 1 to 5 on a response scale, which agreed as a five-point Likert scale. Assessment of Anthropometric Measurements for Height (cm) and weight (kg) were measured for all participants using standard methods [18] Additionally, body mass index (BMI) was estimated by dividing weight (kg) by the square of height in meters. The BP measurement was taken from the left arm (mmHg) using a stethoscope and mercury sphygmomanometer at a sitting position with 90-degree back support. The arm is bare and the middle of the cuff at the heart level with uncrossed legs without talking during and between measurements. An average level of the two consecutive measurements (Average 2^nd-3^rd measurement) was taken as the accurate measurement [14]. In this study, Uncontrolled BP was defined according to the Eighth Joint National Committee (JNC-8) definition, as a systolic BP equal to or above 140 mmHg and/or diastolic BP equal to or above 90 mmHg after taking treatment of three months or more for patients with diabetes chronic kidney diseases and adults aged less than 60 years old. Whereas, for patients older than 60 years uncontrolled blood pressure is defined as above 150/90 mmHg of their systolic/diastolic blood pressure [19]. Adherence means accepting, agreeing, and correctly following a prescribed treatment [13]. The adherence status of the respondents was assessed using patients' self-report. So, poor adherence is missing 20% or more of the monthly prescribed doses based on patient reports.

Data were checked for completeness, then entered into Epi Data version 3.1 and exported to STATA/SE 14.0 for further analysis. Also, the data were coded, cleaned, and explored to identify missing values, outliers, and inconsistencies through tabulation and graphical display.

The blood pressure measurement was taken as a continuous variable, but for the sake of understanding it compute as controlled when both systolic and diastolic blood pressure amounts were less than the limit of uncontrolled blood pressure by considering age and diabetes mellitus. Uncontrolled blood pressure was also computed either the systolic or diastolic blood pressure was greater than the limit of uncontrolled blood pressure with regarding the age and diabetic status of patients.

The perceived social support was also transformed into three categories using tertile of their computed raw scores. The lowest, middle and highest tertiles were recoded as low, moderate, and high social supports, respectively [16]. Continuous variables were described in terms of mean with their standard deviation and median with their interquartile range. Categorical data were also described in terms of frequencies and percentages. All necessary assumptions of log binomial regression model were checked and fulfilled. Those variables in the bivariable log binomial model regression analysis with a p-value less than 0.25 were transferred to multivariable log binomial model [20]. Then, a multivariable log binomial regression analysis was done and those variables whose p-value was less than 0.05 at a 95% CI were declared statistically significant.

Ethical clearance was obtained from the Ethical Review Committee of Wollo University College of Medicine and Health Sciences. The respondents were informed about the purpose of the study, and written consent was obtained from each participant. The data were collected in accordance with the Helsinki declaration. The respondents’ right to refuse or withdraw from participation in the interview is fully maintained and the information provided by each respondent was kept strictly confidential.

A total of 360 study participants were included in this study. The median age of the respondents was 55 years with an interquartile range of 17 years old. The minimum and maximum ages of the participants were 20 years old and 88 years old, respectively. One hundred fifty-three (28.3%) of the participants were males. Regarding marital status and residence of the participants 243(67.5%) and 258(71.67%) were married and urban dwellers, respectively. Concerning participants' education and occupational status, 226(62.78%) and 160(44.4%) have no formal education and were government-employed, respectively (Table 1).

The prevalence of uncontrolled blood measures in patients with hypertension with 95% CI was 55.8(50.7, 61.0). The clinical characteristics of the study participant showed 130(36.1%) patients with hypertension had at least one complication of hypertension. Most of the patients 303(84.2%) had one or more hypertension-related symptoms and 235(65.3%) of the participant also took two or more hypertensive drugs. The presence of comorbidity and adherence status 131(36.4%) and 28(7.8%) of patients with hypertension had one or more co-morbidity and poor adherence, respectively. The mean body mass index (BMI) of the participant was 23.6 kg/m² with a standard deviation of 3.4 kg/m². The median stay of the patient with antihypertensive treatment was 6 years with an interquartile range of 7 years (Table 1).

Only fourteen (3.9%) of patients with hypertension were smoking cigarettes. Considering alcoholic consumption and ‘Khat’ chewing practice 21(5.8%) and 34(9.4%) of patients with hypertension were also drinking alcohol and Chewing ‘Khat’, respectively. For the habits of physical exercise and salt restriction, 252(70%) and 51(14.2%) of the patients were performed regular physical exercise and unable to restrict their salt intake, respectively. The perceived social support status of participants 85(23.6) had a low perceived social support (Table 1).

Due to a high prevalence of uncontrolled blood pressure, we chose PR over POR because POR would have significantly "overestimated" the strength of the association [21]. Some modifiable and non-modifiable factors were assessed for the presence of associations with blood pressure control status in multivariable log binomial model and the following variables were significantly associated.

Male patients with hypertension were 1.35 (1.11,1.64), times more likely to be uncontrolled blood pressure than those female patients with hypertension. The odds of having uncontrolled blood pressure in poor medication adherence patients were 1.86(1.59,2.19) times higher than in patients with good medication adherence. Those patients with hypertension with educational status of secondary and above were 48% [0.52(0.35,0.77)] less likely of having uncontrolled blood pressure than patients with no formal education. Patients with hypertension on treatment, who had low support were 1.24(1.01, 1.54) times more likely of having uncontrolled blood pressure than those patients with hypertension with high social support (Table 2).

Since this was a cross-sectional study, it had the same limitations as other cross-sectional studies. In this study the outcome and exposure variables are measured at the same time, it is relatively difficult to establish causal relationships from a cross-sectional study. Patients may be biased to social desirability bias for doctors since adherence status was measured utilizing a document review.

This study suggests that patients with hypertension on medication are still struggling to achieve target blood pressure control with a high level of low social support in the pandemics of COVID-19, so health care providers better screen patients for social support status, provide a coping mechanism for patients with low social support and refer patients to psychiatric clinics for additional psychological support.