HIV self-testing has become a cornerstone strategy for reaching people who face barriers to conventional clinic-based services. In Cameroon, both same-sex practices and sex work are criminalised, and the resulting stigma and fear of disclosure deter many men who have sex with men and female sex workers from visiting health facilities for testing. Self-testing sidesteps that obstacle: participants collect a small blood sample themselves, apply it to a rapid diagnostic strip, and read the result privately. Crucially, the test functions as a triage tool rather than a definitive diagnosis, meaning any reactive result must be confirmed through standard laboratory algorithms before treatment can begin.

The study enrolled 789 participants through 17 community-based organisations, ten serving men who have sex with men and seven serving female sex workers, in the two largest cities. All participants were aged 21 or older and were recruited through voluntary sampling. After counselling, each person performed an unassisted self-test and completed a structured questionnaire covering demographic characteristics, sexual behaviour, and medical history. The researchers then used modified Poisson regression, a statistical approach suited to common outcomes in cross-sectional data, to calculate adjusted prevalence ratios that quantify how each factor relates to the probability of a reactive test while controlling for other variables.

The headline finding is sobering: 23.3 percent of all participants had a reactive self-test, with a 95 percent confidence interval of 20.5 to 26.4 percent. The proportions were nearly identical across the two populations, at 23.2 percent among female sex workers and 23.4 percent among men who have sex with men, suggesting that both groups carry a comparably heavy and previously underappreciated burden of presumed infection. Because reactive self-tests require confirmatory testing, these figures represent an upper signal of infection prevalence rather than confirmed diagnoses, but the sheer scale of reactivity underscores how much undiagnosed infection may be circulating within networks that clinics rarely reach.

Perhaps the most dramatic signal in the data is geographic. Reactivity was more than three times higher in Yaoundé, the capital, than in Douala, the country’s economic hub, at 38.3 percent versus 12.2 percent. After statistical adjustment, living in Douala was associated with a 78 percent lower probability of reactivity among female sex workers and a 69 percent lower probability among men who have sex with men. This concentration in a single city is epidemiologically significant because it implies that tailored interventions, from intensified testing campaigns to partner services and pre-exposure prophylaxis roll-out, could achieve disproportionate impact if prioritised in Yaoundé rather than spread evenly across both metropolises.

Education emerged as a powerful and consistent protective correlate, particularly among women. Relative to female sex workers with no formal schooling, those with primary education had a 72 percent lower adjusted probability of a reactive test, those with secondary education a 66 percent lower probability, and those with higher education a 56 percent lower probability. A similar gradient appeared among men who have sex with men, where secondary education was associated with a 63 percent reduction. The pattern likely reflects the ways literacy and schooling shape access to health information, negotiating power in commercial sex, economic alternatives, and familiarity with preventive technologies, though the cross-sectional design cannot disentangle the underlying mechanisms.

Behavioural and clinical history markers also carried important signals. Among female sex workers, a history of symptoms suggestive of sexually transmitted infections nearly tripled the adjusted probability of reactivity, tripling the odds relative to women without such a history. This finding aligns with well-established virology: genital ulceration and inflammation amplify both acquisition and transmission of HIV by recruiting activated CD4-bearing target cells to mucosal surfaces and disrupting epithelial barriers. It also suggests that syndromic STI screening, when integrated with self-testing programmes, could serve as an efficient triage layer identifying women at highest risk.

Among men who have sex with men, each additional sexual partner over the preceding six months was associated with a modest but statistically reliable 3 percent increase in the probability of reactivity, a dose-response relationship consistent with cumulative exposure risk. Interestingly, alcohol consumption was associated with a 35 percent lower adjusted probability of reactivity in this group. The authors caution that such counterintuitive associations in cross-sectional data may reflect residual confounding or complex social patterns rather than any protective biological effect of alcohol itself.

The methodological strength of the study lies in its scale, its reliance on trained community-based organisations to recruit criminalised populations that formal health systems typically miss, and its use of regression techniques appropriate for high-prevalence outcomes. The ethical scaffolding was equally careful: the protocol received approval from Cameroon’s National Ethics Committee for Human Health Research and administrative authorisation from the Ministry of Health, and every participant provided written informed consent, with data handled confidentially given the acute legal risks these populations face.

The implications for public health strategy are direct. The authors conclude that reactive self-tests were frequent among both populations, with a marked concentration in Yaoundé and among unschooled female sex workers, and they argue that geographically prioritised, differentiated prevention, combined with STI screening and robust linkage to confirmatory testing, would meaningfully reduce this burden. As Cameroon and its regional neighbours work toward UNAIDS testing and treatment targets, the study demonstrates that self-testing is not merely a convenience but a surveillance instrument, capable of mapping where infection concentrates and guiding resources to the people and places that need them most.

Subject of Research: Factors associated with reactive HIV self-test results among female sex workers and men who have sex with men in Cameroon

Article Title: Factors associated with a reactive HIV blood self-test result among female sex workers and men who have sex with men in Yaoundé and Douala, Cameroon: a cross-sectional study

Article References: Ndié, J., Noah, J.-P. Y. A., Ndongo, F. A., Dongmo, R. K., Yongwa, O., Tchikangni, G., Palisson, T. A. N., Ketchaji, A., Yepndo, C. D., Ndindjock, C. B., IV, Tabala, F. N., Bonyohe, M., Moulioum, F., Nzaddi, A. R. D., Monteh, C. T., Mboussam, P. H., Nguemkam, G., Seukam, B., Anaba, E. D. M., … Bissek, A. C. Z.-K. (2026). Factors associated with a reactive HIV blood self-test result among female sex workers and men who have sex with men in Yaoundé and Douala, Cameroon: a cross-sectional study. BMC Infectious Diseases. https://doi.org/10.1186/s12879-026-14598-4

Image Credits: AI Generated

DOI: 10.1186/s12879-026-14598-4

Keywords: HIV self-testing, HIV, Cameroon, men who have sex with men, female sex workers, key populations, prevalence, Yaoundé, Douala, sexually transmitted infections, public health, cross-sectional study

The stakes could hardly be higher. Yaoundé currently receives only 120,000 cubic meters of water per day from the Nyong River, far short of what its expanding population demands. To close the gap, the government built a water intake on the Sanaga at Nachtigal, aiming to lift drinking water production from 300,000 to 400,000 cubic meters per day. But the intake sits downstream of a gauntlet of potential contamination sources: the sprawling sugarcane plantations of the Cameroon Sugar Company, known as SOSUCAM, which cover thousands of hectares and rely heavily on fertilizers and pesticides, and the artisanal gold mining zones of the Lom basin upstream, where mercury is used intensively to concentrate ore. Mercury exposure is linked to neurological, renal, and respiratory disorders, and mining districts upstream have already recorded some of the highest heavy metal pollution indices in the country.

To map where contamination risks are greatest, the research team, led by Henri Zobo Mbele of the University of Yaoundé I and colleagues, adapted a French multicriteria method known as DKPR, originally developed by the Bureau de Recherches Géologiques et Minières in 2010. The approach breaks vulnerability down into four measurable factors: the accessibility of the aquatic environment, which captures how quickly polluted runoff can reach a watercourse; the functioning of soil and subsoil water, which governs infiltration versus runoff; watershed physiography, including slope steepness and curvature derived from a 30-meter digital terrain model; and rainfall erosivity, calculated using equations from the Revised Universal Soil Loss Equation. Each factor was scored, classified, and then weighted using Saaty’s Analytic Hierarchy Process, a decision-science technique that converts expert judgments into numerical priorities. The resulting comparison matrix passed the standard consistency test with a ratio of 3.5 percent, well below the 10 percent threshold, lending statistical credibility to the weighting scheme.

The weights themselves tell a striking story. Aquatic accessibility emerged as the dominant driver of vulnerability, receiving a weighting coefficient of 0.46, more than the other three factors combined. Soil and subsoil water functioning followed at 0.26, watershed physiography at 0.20, and rainfall erosivity at just 0.08. When the weighted layers were superimposed across the 78,000-square-kilometer Nachtigal catchment, the resulting map revealed that 49 percent of the area falls into the low vulnerability class, largely thanks to dense forest cover that slows runoff and poor drainage accessibility. But 23 percent of the basin, concentrated within a kilometer of the river network and in the zones of intense agro-industrial activity, was classified as highly or very highly vulnerable. In these buffer zones, the combination of easy contaminant transport and intensive fertilizer and pesticide use poses what the authors describe as a major threat to water quality.

Rigor demanded that the map be tested against its own uncertainties. The team calculated error margins for every parameter, finding that uncertainties ranged from a negligible ±0.0001 for land cover to ±7.64 for flow length, the distance water travels along its path. The final vulnerability map carries an uncertainty of ±0.716 with a 68 percent confidence level, while individual components such as slope curvature and the network development and persistence index achieved 99 percent confidence. These figures, the researchers note, demonstrate that errors were minimized during map production, and the approach mirrors uncertainty analyses used in other multicriteria studies across West Africa.

The vulnerability modeling was only half the story. Since 2021, in collaboration with the PAEPYS drinking water project, the team collected forty-one water samples from two sites: one 500 meters downstream of the new Nachtigal dam, and another at the drinking water intake itself. The samples were filtered through 0.45-micrometer membranes, acidified for cation analysis, and run through ion chromatography systems at the Pasteur Center in Yaoundé. Trace metals and cyanides were measured photometrically, and results were only accepted when the charge balance error stayed below 5 percent, a strict quality control that guards against analytical drift.

The chemistry revealed a river in reasonably good health, with important caveats. The water is dilute, with electrical conductivity averaging just 35 microsiemens per centimeter, a signature of the ancient plutonic and metamorphic bedrock of southern Cameroon, whose inert granitic minerals release few dissolved ions. Major ions all fell comfortably within World Health Organization guidelines, and the heavy metal pollution index came out at 89.59, below the critical threshold of 100. Yet two elements stood out: iron, averaging 1.58 milligrams per liter against a WHO limit of 0.30, and aluminum, with peaks reaching 1 milligram per liter against a 0.20 limit. The researchers attribute these exceedances mainly to the leaching of the region’s deeply weathered ferrallitic and lateritic soils, whose iron-rich chemistry and humic acid complexation naturally release metals into runoff, though upstream deforestation, mining, and agro-industrial land disturbance amplify the problem.

Physical parameters told a more worrying tale. Turbidity averaged 20.3 NTU, more than four times the WHO guideline of 5, and color reached an average of 117 Hazen units against a recommended 15. Total suspended solids, averaging 11.7 milligrams per liter, tracked turbidity closely, and both reflect the erosion of bare land driven by deforestation and urbanization. Encouragingly, all these values were lower than those recorded at the Bétaré Oya gold mining district upstream, pointing to a powerful dilution and sedimentation effect as the Sanaga gathers volume. When the team computed the Water Quality Index, results ranged from 16 to 101: 32 percent of samples were rated excellent, 44 percent good, but 24 percent fell into the poor, very poor, or non-potable categories. Crucially, quality deteriorated moving upstream, with the worst value recorded at the dam site rather than the intake, confirming that human pressure concentrates in the upper catchment while the river’s sheer size cleans up downstream.

The convergence between the modeled vulnerability and the measured chemistry is perhaps the study’s most compelling result. The low contamination levels align with the low vulnerability scores produced by the DKPR model, suggesting that limited accessibility between pollution sources and the watercourse, combined with the Sanaga’s enormous dilution capacity, currently shields the intake from the worst upstream impacts. But the authors are careful to frame this as a snapshot of a dynamic system. SOSUCAM’s plantations may expand significantly to meet production targets, artisanal gold mining continues to attract migrants and spread mercury contamination, and dam construction upstream has already flooded former mining areas, fragmenting habitats and accelerating erosion.

The vulnerability map, the researchers argue, should serve as a practical decision-making tool, allowing Cameroonian authorities to prioritize the 24 percent of the basin where intervention matters most. Their recommendations are concrete: strict regulation of pesticide use, the establishment of formal protection zones around the intake, continuous water quality monitoring, community education on water management, and integrated land use planning supported by scenario modeling. For a capital city betting its water future on a single river intake, the message is clear: the Sanaga is holding its own for now, but the margin of safety depends on decisions being made today, in the sugarcane fields and mining camps far upstream of every glass of water poured in Yaoundé.

Subject of Research: Assessment of surface water contamination vulnerability in the Sanaga drinking water supply basin at Nachtigal, Cameroon, using the DKPR multicriteria method and water quality indices.

Article Title: Assessment of the vulnerability to surface water contamination using a multicriteria decision-making approach: case of the Sanaga drinking water supply basin in Nachtigal (central Cameroon)

Article References: Zobo Mbele, H., Bella Atangana, M. S., Fouépé Takounjou, A., & Ndam Ngoupayou, J. R. (2026). Assessment of the vulnerability to surface water contamination using a multicriteria decision-making approach: case of the Sanaga drinking water supply basin in Nachtigal (central Cameroon). Heliyon, 12(15), Article e45562. https://doi.org/10.1016/j.heliyon.2026.e45562

Image Credits: AI Generated

DOI: 10.1016/j.heliyon.2026.e45562

Keywords: Sanaga River, Cameroon, water vulnerability, DKPR method, multicriteria analysis, drinking water, heavy metal pollution, water quality index, gold mining, agro-industry, Yaoundé, GIS mapping