High blood pressure and associated factors among HIV-infected young persons aged 13 to 25 years at selected health facilities in Rwenzori region, western Uganda, September–October 2021

Our study utilized secondary data routinely collected in antiretroviral therapy (ART) clinics. The BP measurements were part of routine good clinical practice at the ART clinics. We obtained written informed consent for persons aged 18 to 25 years and written informed assent for those aged 12 to 17 years before taking their BP measurements. In addition, the parents/guardians of the minors (< 18 years) provided written informed consent before participation. We did not capture any personal identifiers. We obtained permission to use the secondary data from the district health officers of Kasese and Kabarole districts and the directors of the respective health facilities. In addition, the US Centers for Disease Control and Prevention’s Center for Global Health determined our study was nonresearch whose primary intention was to address public health problems. The data were stored in password-protected computers and data was not shared with anyone outside the investigation team.

We conducted a cross-sectional study among PLHIV aged 13 to 25 years at selected clinics supported by Baylor College of Medicine Children’s Foundation-Uganda (Baylor Uganda) in the Rwenzori region (Kasese and Kabarole districts). Baylor Uganda was the implementing partner supporting HIV and AIDS prevention, care, and treatment services in the region, funded by President’s Emergency Plan for AIDS Relief (PEPFAR), and also strengthened routing BP measurements in clinics providing ART in the region. The study utilized both primary and secondary data. We obtained laboratory, demographic, and clinical data through retrospective review of medical records and took BP measurements (see procedure below) for the participants at a single clinic visit at each of the nine clinics during the study period (September 16–October 15, 2021).

We purposively selected nine health facilities in Kasese and Kabarole districts that had scheduled their adolescent clinic days within our study period (Fig. 1). Facilities selected in Kabarole District were Bukuuku Health Center IV, Virika Hospital, and Fort Portal Regional Referral Hospital. Facilities selected in Kasese District were Kagando Hospital, Kilembe Mines Hospital, Bwera Hospital, Rwesande HCIV, Kasese Municipal Health Center III, and St. Paul Medical Centre. We enrolled all 1,045 treatment-experienced young persons aged 13 to 25 years who came to refill antiretroviral drugs (ARV) within the study period.

Clinic nurses trained in the use of a standardized data collection tool abstracted data from study participants’ medical records on demographics (age and sex) and clinical data (weight, height, BP, mode of transmission and duration of HIV infection, ARVs received over the previous 2 years, current medications, viral load suppression status, past medical history, family history of CVD, and tobacco exposure). The participants were categorized into two groups: adolescents (individuals aged 13–17 years) and young adults (individuals aged 18–25 years). BP and resting heart rate were measured with an automated sphygmomanometer (Omron HEM 705 LP; Omron Healthcare Inc., Bannockburn, IL, USA) when the study participants came for refills of their ARVs during September to October 2021. BP was measured in the left arm after resting for at least 5 min in a sitting position, legs uncrossed, with the arm resting on a table and the antecubital fossa at the level of the lower sternum. We selected the appropriate cuff size (small, medium, or large) based on the mid-upper arm circumference. All the nurses involved in measurements of BP underwent a standardized training. Before taking BP measurements, we obtained written informed consent from all participants. Additionally, prior to participation, parents/guardians of the minors (< 18 years) gave written informed consent.

We obtained data on viral load suppression status within 6 months before BP readings. Unsuppressed viral load was defined as viral load > 1,000 copies/mL [13]. Height and weight of participants were measured to the nearest 0.5 cm and 0.1 kg, respectively. We computed body mass index (BMI) as weight (kg) divided by the square of height (m²). We adopted categorization of BMI as normal, underweight, overweight, and obese, as per recognized criteria [17]. Tobacco use was defined as any current or previous use. Family history of CVD was defined as a first-degree relative with a history of heart attack, stroke, or HBP.

The assessment of HBP was based on the updated American Academy of Pediatrics definitions of BP categories and stages for individuals aged ≥ 13 years [18]. Accordingly, BP categories were defined as normal BP (< 120/ < 80 mmHg), elevated BP (120/ < 80 to 129/ < 80 mmHg), stage 1 HTN (130/80 to 139/89 mmHg), and stage 2 HTN (≥ 140/90 mmHg). Accordingly, participants with elevated BP or HTN (systolic BP ≥ 120 mmHg or diastolic BP ≥ 80 mmHg) were considered to have HBP. Participants on antihypertensive medication at the time of the survey were also considered to have HBP, regardless of measured BP.

Data were entered by the data collectors using Open Data Kit enabled database. Data were exported to Microsoft Excel (Microsoft Corp., Redmond, WA, USA) and all statistical analyses were performed using Stata ver. 15.0 (Stata Corp., College Station, TX, USA).

We categorized all variables and summarized them using frequencies. Additionally, normally distributed continuous variables, including age, resting heart rate, and BMI were summarized using means ± standard deviation, while non-normally distributed variables (e.g., duration of HIV infection) were summarized using medians and interquartile range (IQR). Associations between HBP and potential associated factors were assessed using modified Poisson regression, reporting prevalence ratios as our measures of association. We chose modified Poisson regression rather than logistic regression due to the high prevalence of our outcome, HBP [19]. All variables demonstrating an association at a P < 0.2 significance level were included in the multivariable models. We also adjusted the final model for known confounders, including duration of HIV, HIV viral load, sex, and self-reported family history of CVD. Variables in the final model with a P < 0.05 were considered statistically significant.

We reviewed medical records of 1,045 HIV-infected persons aged 13 to 25 years from nine health facilities in the Rwenzori region, western Uganda. Of the 1,045 participants enrolled, 68% were female (Table 1). The mean age of the participants was 20 ± 3.8 years and 29% were adolescents. The median duration of HIV infection was 6 years (IQR, 3–13 years). Most participants (83%) had suppressed or undetectable viral loads. Nearly all participants (97%) had never smoked. Two-thirds (64%) had normal BMI, and 84% had never consumed alcohol.

Half of the participants (n = 530, 51%) had normal BP and 515 (49%) had HBP. In total, 229 (22%) had elevated BP, 220 (21%) had stage 1 HTN, and 66 (6%) had stage 2 HTN (Fig. 2). Only three participants were already on medications for HBP.

In multivariable analysis, older age, history of tobacco smoking, and higher resting pulse rate were significantly associated with HBP (Table 2). The prevalence of HBP was 21% higher in the age group of 18 to 25 years compared to 13 to 17 years (adjusted prevalence ratio [aPR], 1.21; 95% confidence interval [CI],1.01–1.44), 41% higher among those with a current or past history of smoking compared to nonsmokers, and 15% higher among those with high resting heart rates (aPR, 1.15; 95% CI, 1.01–1.32; heart rate, > 76 beats/min vs. ≤ 76 beats/min).