BackgroundTreatment with effective antiretroviral therapy (ART) lowers morbidity and mortality among HIV positive individuals. Effective highly active antiretroviral therapy (HAART) should lead to undetectable viral load within 6 months of initiation of therapy. Failure to achieve and maintain viral suppression may lead to development of resistance and increase the risk of viral transmission. In this paper three logistic regression based machine learning approaches are developed to predict early virological outcomes using easily measurable baseline demographic and clinical variables (age, body weight, sex, TB disease status, ART regimen, viral load, CD4 count). The predictive performance and generalizability of the approaches are compared.MethodsThe multitask temporal logistic regression (MTLR), patient specific survival prediction (PSSP) and simple logistic regression (SLR) models were developed and validated using the IDI research cohort data and predictive performance tested on an external dataset from the EFV cohort. The model calibration and discrimination plots, discriminatory measures (AUROC, F1) and overall predictive performance (brier score) were assessed.ResultsThe MTLR model outperformed the PSSP and SLR models in terms of goodness of fit (RMSE = 0.053, 0.1, and 0.14 respectively), discrimination (AUROC = 0.92, 0.75 and 0.53 respectively) and general predictive performance (Brier score= 0.08, 0.19, 0.11 respectively). The predictive importance of variables varied with time after initiation of ART. The final MTLR model accurately (accuracy = 92.9%) predicted outcomes in the external (EFV cohort) dataset with satisfactory discrimination (0.878) and a low (6.9%) false positive rate.ConclusionMultitask Logistic regression based models are capable of accurately predicting early virological suppression using readily available baseline demographic and clinical variables and could be used to derive a risk score for use in resource limited settings.
BackgroundThe WHO recommends supervised administration of sulphadoxine-pyrimethamine (SP) as intermittent preventive treatment for malaria (IPTp) during pregnancy. Logistical constraints have however favoured unsupervised intake of SP-IPTp, casting doubts whether recent guidelines requiring more frequent intake can be effectively implemented. To propose strategies for enhancing compliance under limited supervision, this study sought to identify pregnant women’s knowledge and practices gaps as well as determine predictors of compliance with SP-IPTp, given under limited supervision.MethodsA cross-sectional study of 700 women used exit interviews at an urban clinic in Uganda to obtain a descriptive summary of demographic and obstetric characteristics, including knowledge, practice and experiences with SP. Predictors of compliance with SP intake instructions were explored using logistic regression.ResultsMedian age of respondents was 25 (IQR 22–28) and median parity was two (IQR one to three) while median number of antenatal clinic (ANC) visits was 3.0 (IQR three to four). Most women had completed primary (36%) or ordinary secondary education (25.6%) while 16.1% had not completed primary education. Awareness about SP was high (99.4%) although correct knowledge regarding its use in pregnancy was low (57%), with 15.4% thinking it was used to treat malaria and 26.7% lacking any idea about its use. Correct knowledge on SP use during pregnancy significantly predicted compliance with SP-IPTp instructions (OR 1.98, C.I. 1.12-3.55), while age, education level, parity, number of ANC visits, or history of unwanted effects with SP did not. SP was mostly accessed from hospitals (64.4%) followed by private clinics (16.9%) both for preventive and treatment purposes. SP was considered safe by most women, who were willing to take it again in future, without supervision.ConclusionDespite high awareness, knowledge of SP as an intervention for malaria prevention in pregnancy was low. Correct knowledge on use of SP predicted compliance with SP-IPTp intake instructions. Focused malaria-related education during ANC visits may improve compliance with SP intake amidst limited supervision.Electronic supplementary materialThe online version of this article (doi:10.1186/1475-2875-13-399) contains supplementary material, which is available to authorized users.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.