Applying machine learning and predictive modeling to retention and viral suppression in South African HIV treatment cohorts.

HIV treatment programs face challenges in identifying patients at risk for loss-to-follow-up and uncontrolled viremia. We applied predictive machine learning algorithms to anonymised, patient-level HIV programmatic data from two districts in South Africa, 2016-2018. We developed patient risk scores for two outcomes: (1) visit attendance ≤ 28 days of the next scheduled clinic visit and (2) suppression of the next HIV viral load (VL). Demographic, clinical, behavioral and laboratory data were investigated in multiple models as predictor variables of attending the next scheduled visit and VL results at the next test. Three classification algorithms (logistical regression, random forest and AdaBoost) were evaluated for building predictive models. Data were randomly sampled on a 70/30 split into a training and test set. The training set included a balanced set of positive and negative examples from which the classification algorithm could learn. The predictor variable data from the unseen test set were given to the model, and each predicted outcome was scored against known outcomes. Finally, we estimated performance metrics for each model in terms of sensitivity, specificity, positive and negative predictive value and area under the curve (AUC). In total, 445,636 patients were included in the retention model and 363,977 in the VL model. The predictive metric (AUC) ranged from 0.69 for attendance at the next scheduled visit to 0.76 for VL suppression, suggesting that the model correctly classified whether a scheduled visit would be attended in 2 of 3 patients and whether the VL result at the next test would be suppressed in approximately 3 of 4 patients. Variables that were important predictors of both outcomes included prior late visits, number of prior VL tests, time since their last visit, number of visits on their current regimen, age, and treatment duration. For retention, the number of visits at the current facility and the details of the next appointment date were also predictors, while for VL suppression, other predictors included the range of the previous VL value. Machine learning can identify HIV patients at risk for disengagement and unsuppressed VL. Predictive modeling can improve the targeting of interventions through differentiated models of care before patients disengage from treatment programmes, increasing cost-effectiveness and improving patient outcomes.

Cost evaluation of reproductive and primary health care mobile service delivery for women in two rural districts in South Africa.

BACKGROUND: Cervical cancer screening is a critical health service that is often unavailable to women in under-resourced settings. In order to expand access to this and other reproductive and primary health care services, a South African non-governmental organization established a van-based mobile clinic in two rural districts in South Africa. To inform policy and budgeting, we conducted a cost evaluation of this service delivery model. METHODS: The evaluation was retrospective (October 2012-September 2013 for one district and April-September 2013 for the second district) and conducted from a provider cost perspective. Services evaluated included cervical cancer screening, HIV counselling and testing, syndromic management of sexually transmitted infections (STIs), breast exams, provision of condoms, contraceptives, and general health education. Fixed costs, including vehicle purchase and conversion, equipment, operating costs and mobile clinic staffing, were collected from program records and public sector pricing information. The number of women accessing different services was multiplied by ingredients-based variable costs, reflecting the consumables required. All costs are reported in 2013 USD. RESULTS: Fixed costs accounted for most of the total annual costs of the mobile clinics (85% and 94% for the two districts); the largest contributor to annual fixed costs was staff salaries. Average costs per patient were driven by the total number of patients seen, at $46.09 and $76.03 for the two districts. Variable costs for Pap smears were higher than for other services provided, and some services, such as breast exams and STI and tuberculosis symptoms screening, had no marginal cost. CONCLUSIONS: Staffing costs are the largest component of providing mobile health services to rural communities. Yet, in remote areas where patient volumes do not exceed nursing staff capacity, incorporating multiple services within a cervical cancer screening program is an approach to potentially expand access to health care without added costs.