A blood-pressure problem hidden inside routine care
High blood pressure rarely announces itself with a dramatic symptom.
That is precisely why health systems need to find it, treat it and keep checking whether treatment is working.
In South Africa, this task overlaps with the long-established infrastructure for HIV care.
Many patients visit public clinics regularly, creating opportunities to detect hypertension before it contributes to heart disease, stroke or kidney damage.
Yet repeated contact with a clinic does not guarantee that a second chronic condition will be diagnosed or controlled.
A new longitudinal analysis of Johannesburg clinic records offers a more demanding picture than a single screening survey.
Rather than simply asking how many people had high blood pressure on one date, the researchers tracked movement through successive stages of care.
Patients could progress from unrecognised hypertension to diagnosis, medication and control.
They could also move backwards: treatment could stop, or blood pressure could become uncontrolled after an earlier successful measurement.
The distinction between progress and regression is important for policymakers. A programme can report many diagnoses while losing people between subsequent appointments.
It can also report prescriptions without knowing whether blood pressure remains controlled.
These are not interchangeable measures of success. A care pathway is only as dependable as the continuity it provides over time.
What the Johannesburg study actually examined
Siphamandla Bonga Gumede and five colleagues conducted a secondary longitudinal analysis of data from the iHEART-SA trial.
Their paper appeared in PLOS Medicine on 11 September 2026.
The analysis covered nine public primary healthcare clinics in Johannesburg, with data collection from August 2022 to May 2024 and follow-up extending to May 2025.
The authors included adults aged at least 18 who were receiving chronic care, had a known HIV status and had undergone a completed medical-file review.
The dataset contained 23,855 participants.
Of these, 78.5% were people living with HIV.
The median age among that group was 42 years, with an interquartile range from 36 to 49 years, and 69% were female.
Across the sample, the median systolic blood pressure was 132 millimetres of mercury, with an interquartile range from 120 to 145.
These details help describe the clinical population but should not be mistaken for a representative sample of every South African adult.
Clinic-based sampling is useful for examining service delivery because it observes people already interacting with a healthcare system.
It also creates a boundary around the findings.
People who seldom attend clinics, cannot access care or receive services in different settings may face different obstacles.
The researchers were studying the performance of a particular care pathway among people in contact with nine facilities, not measuring national hypertension prevalence.
The cascade reveals more than a screening rate
A hypertension care cascade is a sequence of linked states.
First, high blood pressure must be recognised.
Next, a diagnosis needs to be documented and appropriate treatment initiated.
Finally, blood pressure must reach and remain within the study’s control criteria.
Each transition can fail for different reasons, from missed screening and incomplete records to medication interruptions or changes in clinical response.
The investigators compared participants living with HIV with those not living with HIV.
This comparison matters because HIV services often involve repeat appointments and medication monitoring.
One might expect that regular clinical contact would create an advantage for detecting another long-term condition.
The findings show why that expectation cannot simply be assumed.
After adjustment in the study’s statistical model, participants living with HIV were more likely to remain undiagnosed when they had hypertension.
The adjusted relative risk was 3.40, with a 95% confidence interval of 2.45 to 4.73.
This is a relative comparison between groups, not a statement that three out of four patients were undiagnosed.
It also does not prove that HIV itself caused the difference.
Among people receiving hypertension treatment, those living with HIV were less likely to achieve blood-pressure control.
The adjusted relative risk was 0.85, with a 95% confidence interval of 0.75 to 0.97. A value below one indicates lower relative likelihood of control in the comparison as defined by the authors.
The result points to a gap in sustained management rather than merely a shortage of initial diagnoses.
What changed during follow-up
The longitudinal component adds a particularly practical warning.
People living with HIV experienced more regression from treated hypertension to an untreated state: 22% compared with 11% among those not living with HIV.
The adjusted relative risk for that regression was 1.51, with a 95% confidence interval of 1.03 to 2.23.
These figures describe observed transitions in the analysed records, subject to the study’s definitions and documentation limits.
The researchers also found a greater adjusted likelihood of moving from controlled to uncontrolled blood pressure among participants living with HIV.
The adjusted relative risk was 1.10, with a 95% confidence interval of 1.02 to 1.18.
The numerical difference may appear smaller than the diagnosis gap, but it concerns an especially consequential point in care: someone who had achieved control later no longer met that standard.
A patient may have a prescription recorded yet encounter problems obtaining refills, returning for monitoring or responding to treatment.
The study does not establish which of these explanations accounts for each transition.
Its strength lies in identifying where continuity breaks down.
That is more actionable than treating every patient with a documented prescription as a permanently successful case.
Research Today has previously covered hypertension detection among younger South Africans.
Detection tools are valuable, but this newer study illustrates the separate challenge of maintaining control after diagnosis. A screening improvement alone cannot repair interruptions later in the pathway.
Why HIV care and hypertension care intersect
South Africa’s HIV programmes have built substantial experience with long-term medication, scheduled follow-up and laboratory monitoring.
That infrastructure creates an opportunity to integrate cardiovascular risk management into routine visits.
It does not mean that HIV clinics automatically have the staff, equipment, prescribing pathways or reliable records needed to manage every additional chronic disease.
Hypertension can require repeated measurements, medication adjustments and attention to adherence over many years.
Some patients manage several conditions at once, and their treatment plans may involve different clinic workflows. A service organised primarily around one diagnosis can inadvertently overlook another even when patients attend regularly.
The authors’ findings are consistent with that kind of systems challenge, although the observational analysis cannot isolate any single administrative or clinical mechanism.
Another Research Today report examined the cost of monitoring strategies in South African HIV care.
The two studies address different questions, but together they illustrate how programme design affects the use of limited clinical resources.
Better integration should be evaluated for both patient outcomes and the workload it creates for facilities.
The practical implication is not that patients living with HIV are inherently unable to control blood pressure.
The study compares outcomes under observed healthcare conditions.
Its results support closer examination of whether routine visits consistently include blood-pressure assessment, whether elevated readings trigger a documented plan, and whether medication continuity is checked at subsequent appointments.
How much confidence should readers place in the numbers?
The paper was a secondary analysis of a larger trial dataset, not a randomised experiment assigning patients to different hypertension services.
Researchers adjusted comparisons statistically, but differences in age, clinical histories, attendance patterns and other characteristics can still matter.
Adjusted relative risks are informative associations; they are not proof that HIV status independently produces a particular clinical outcome.
The investigators also relied on routinely collected clinical information.
Such records are essential for understanding real-world care, yet they can be incomplete.
Medication use was determined from documentation, which may not perfectly capture whether a patient collected or took treatment.
Follow-up duration varied across participants.
These limitations could affect classification of transitions and the interpretation of differences between groups.
Importantly, the nine clinics were in Johannesburg.
Their patient mix, staffing and service organisation may differ from those of rural facilities, private practices or clinics elsewhere in the country.
The results identify an important local pattern and a testable health-system problem, but they do not provide a national causal estimate.
Similar caution applies to the distinction between measured blood pressure and long-term cardiovascular outcomes.
The study analysed stages of hypertension care.
It did not establish that a particular service intervention reduced strokes or heart attacks.
Demonstrating those downstream benefits would require additional research and, ideally, evaluations of specific changes to clinical practice.
What a stronger care pathway could look like
The most immediate response is to treat hypertension management as a continuing process rather than a one-off achievement.
Clinics could examine how consistently blood pressure is recorded during chronic-care appointments, whether high readings are followed up, and whether patients who previously achieved control are reassessed.
Quality indicators should capture regression as well as initial progress.
Medication continuity deserves particular attention because the analysis observed movement back into untreated hypertension.
That observation does not identify the reason for discontinuation.
Clinics would need to investigate stock availability, prescribing processes, appointment schedules, patient experiences and documentation before deciding which remedy would be effective.
Digital reminders or electronic records might help in some settings, but technology is not a substitute for a reliable care pathway.
As Research Today has reported in research on technology-assisted screening, identifying patients at risk and ensuring follow-up are distinct tasks. A system should be judged by whether it supports sustained clinical decisions, not merely whether it generates more alerts.
Future studies could test integrated hypertension and HIV follow-up models, compare continuity of medication access and examine patient-reported reasons for interrupted care.
Such evaluations would help distinguish between plausible explanations that routine records cannot resolve.
The key question is which interventions actually keep more patients diagnosed, treated and controlled over time.
The Johannesburg findings therefore shift attention from a static picture of high blood pressure to a moving one.
People enter, advance through and sometimes fall backwards along the treatment pathway.
The study does not supply a single cure for that problem.
It makes a strong case that health services should measure continuity and design care around the possibility of regression, especially where multiple chronic conditions overlap.
Source Information
Study title: The longitudinal care cascade for hypertension in people with and without HIV in South Africa: A clinic-based observational study
Authors: Siphamandla Bonga Gumede, Jennifer Manne-Goehler, Elizabeth Kelechi Oladimeji, Nicola Bulled, Alana T. Brennan, and Samanta Tresha Lalla-Edward
Journal: PLOS Medicine, 23(9), e1004957
Publication date: 11 September 2026
Study design: Longitudinal secondary observational analysis of the iHEART-SA trial dataset
Sample or scope: 23,855 adults receiving chronic care at nine Johannesburg public primary healthcare clinics; follow-up through May 2025
Key finding: Participants living with HIV were more likely to remain undiagnosed and to regress from treated or controlled hypertension states
