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.

Factors associated with an interruption in treatment of people living with HIV in USAID-supported states in Nigeria: a retrospective study from 2000-2020.

BACKGROUND: Patient interruption of antiretroviral therapy (ART) continues to limit HIV programs' progress toward epidemic control. Multiple factors have been associated with client interruption in treatment (IIT)- including age, gender, CD4 count, and education level. In this paper, we explore the factors associated with IIT in people living with HIV (PLHIV) in United States Agency for International Development (USAID)-supported facilities under the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) program in Nigeria. METHODS: We conducted cross-sectional analyses on data obtained from Nigeria's National Data Repository (NDR), representing a summarized record of 573 630 ART clients that received care at 484 PEPFAR/USAID-supported facilities in 16 states from 2000-2020. IIT was defined as no clinical contact for 28 days or more after the last expected clinical contact. Univariate and multivariate logistic regression models were computed to explore the factors associated with IIT. The variables included in the analysis were sex, age group, zone, facility level, regimen line, multi-month dispensing (MMD), and viral load category. RESULTS: Of the 573 630 clients analysed in this study, 32% have been recorded as having interrupted treatment. Of the clients investigated, 66% were female (32% had interrupted treatment), 39% were aged 25-34 at their last ART pick-up date (with 32% of them interrupted treatment), 59% received care at secondary level facilities (37% interrupted treatment) and 38% were last receiving between three- to five-month MMD (with 10% of these interrupted treatment). Those less likely to interrupt ART were males (aOR = 0.91), clients on six-month MMD (aOR = 0.01), adults on 2nd line regimen (aOR = 0.09), and paediatrics on salvage regimen (aOR = 0.02). Clients most likely to interrupt ART were located in the South West Zone (aOR = 1.99), received treatment at a tertiary level (aOR = 12.34) or secondary level facilities (aOR = 4.01), and had no viral load (VL) on record (aOR =10.02). Age group was not significantly associated with IIT. CONCLUSIONS: Sex, zone, facility level, regimen line, MMD, and VL were significantly associated with IIT. MMD of three months and longer (especially six months) had better retention on ART than those on shorter MMD. Not having a VL on record was associated with a considerable risk of IIT.

Norovirus disease surveillance using Google Internet query share data.

Google Internet query share (IQS) data for gastroenteritis-related search terms correlated strongly with contemporaneous national (R(2) = 0.70) and regional (R(2) = 0.74) norovirus surveillance data in the United States. IQS data may facilitate rapid identification of norovirus season onset, elevated peak activity, and potential emergence of novel strains.