Between February and July 2020, a community-based, house-to-house blood pressure screening was carried out in 12 of the 16 villages of Mata Sector in the District of Nyaruguru, a remote, rural sub-Saharan region in the Southern province of Rwanda at an altitude of approximately 1900 m above sea level. The screening identified hypertensive blood pressure values in 12% of approximately 7000 participants [6]. After a pause due to the COVID-19 pandemic, between May and June 2022, individuals previously identified with elevated blood pressure values underwent a second identical screening. Among those confirmed with a diagnosis of arterial hypertension, a first cohort of 229 individuals was invited to enter the subsequent phase of the program.

This third phase was conducted by a committed team composed of two Italian near-graduate medical students, a local nurse with a bachelor’s degree in general nursing, and one local assistant. They were supported by the principal investigators and study supervisor, from whom they received specific training. The project was conducted in accordance with the Declaration of Helsinki of 1964 and its later amendments. The project was approved by the Rwanda National Ethics Committee (RNEC) as “Study for better Blood Pressure and Cardiovascular Risk Control in a rural area of the District of Nyaruguru” (RNEC Approval Nr 752/RNEC/2019). Written and verbal informed consent in Kinyarwanda, the national spoken language, was obtained from all participants.

This third phase was conducted between July and December 2022 at the hypertension outpatient clinic of the Health Care Center of Nyamyumba, one of Mata Sector’s villages. Only adults on no previous antihypertensive treatment were eligible for inclusion. Collected data encompassed socio-demographic characteristics, medical history including self-reported previous diagnosis of diabetes mellitus (with or without treatment), and habits of tobacco smoking, alcohol consumption, and adding extra salt to food. Apart from diabetes mellitus, in order to be included, subjects needed to have no previously known comorbidity. Measurements included body anthropometrics, blood pressure, and resting heart rate. At the following outpatient clinics, i.e., after 4, 8, 12, and 16 weeks, only blood pressure and resting heart rate were recorded.

Weight was measured using a digital Tanita® platform scale (Tanita Corporation, Tokyo, Japan), height using a 220-cm Seca 206® roll-up wall-attached measuring tape (Seca GmbH & Co.KG., Hamburg, Germany). Waist circumference was measured at the narrowest point between the lower costal border and the top of the iliac crest in standing and outbreathing position (using a non-stretching tape measure). The waist-to-height ratio was calculated dividing waist circumference by height in centimeters.

Blood pressure was assessed following the European Society of Hypertension recommendations for blood pressure measurements in the office and at home [7], using a validated, automated, oscillometric OMRON M7 IT-HEM-7322-E blood pressure monitor with Intelli Wrap Cuff (HEM-FL31-E) technology (Omron Healthcare UK Ltd, Milton Keynes, UK) [8, 9]. Resting heart rate was recorded with the same device.

Hypertension was defined as systolic blood pressure ≥ 140 mmHg and/or diastolic blood pressure ≥ 90 mmHg [7]. The diagnosis was established during the first outpatient clinic based on the mean of blood pressure readings recorded during the 2020 and 2022 house-to-house screening visits, and the office blood pressure measurements on that day. Arterial hypertension was classified as grade I (systolic 140–159 and/or diastolic 90–99 mmHg), grade II (systolic 160–179 and/or diastolic 100–109 mmHg), or grade III (systolic ≥ 180mmHg and/or diastolic ≥ 110 mmHg) [7]. After recommending appropriate lifestyle changes, as indicated in the recent ESH Guidelines [7], an antihypertensive, low-dose, single-pill combination of olmesartan, amlodipine, and hydrochlorothiazide (OLM/AML/HCTZ) was prescribed to newly diagnosed patients. This medication was available in two fixed-dosage combinations (20/5/12.5 and 40/10/12.5 mg). A third dosage (10/2.5/6.2 mg) was obtained by halving the former with a pill-splitter. A hypertension specialist determined the dosage of the medication based on age, weight, and the degree of hypertension. Each patient received an envelope with reported medication name, dose, and regimen, containing 1-month therapy, plus some spare pills. Patients were recommended to take the prescribed treatment once daily, early in the morning. They were also provided with culturally adapted notions of health care education related to cardiovascular diseases and hypertension and informed about possible side effects of this therapy, such as dizziness, fatigue, headaches, palpitations, peripheral edema, musculoskeletal pain, nausea, diarrhea, or constipation. Medication tolerance was considered excellent in cases without any of the mentioned side effects.

For each patient, blood pressure and heart rate readings as well as the assigned therapy and the following appointment were recorded into a personal booklet to gather at each subsequent visit along with any remaining pill (drug reconciliation). Patients were received at the outpatient clinic after 4, 8, 12, and 16 weeks from the initial visit to perform a clinical examination, verify drug tolerance, and adjust, if needed, their treatment regimen. When needed, dose was increased in order to attain a target clinic blood pressure of < 140/90 mmHg.

Nominal and dichotomous categorical variables are expressed as counts and were analyzed by means of the χ2 test [10]. The normality D’Agostino-Pearson omnibus test [11] disclosed that age, weight, height, waist-to-height ratio, blood pressure, and resting heart rate did not follow a Gaussian distribution. Consequently, continuous data are presented as median and interquartile range and were analyzed using the non-parametric Mann–Whitney–Wilcoxon U test and the Friedman test with the Dunn post hoc multiple comparison [10]. Blood pressure and resting heart rate are presented also in box-and-whisker plots (the bottom and top of the box represent the 25th and 75th centile, the middle band the median, and the ends of the whiskers the 10th and the 90th centile) [12]. Two-sided P values of < 0.05 were considered significant. GraphPad Prism for Macintosh 10.0.3 (GraphPad Software, San Diego, CA, USA) was used for statistical comparisons.