Dataset Information

Statistical models for incorporating data from routine HIV testing of pregnant women at antenatal clinics into HIV/AIDS epidemic estimates.

ABSTRACT: HIV prevalence data collected from routine HIV testing of pregnant women at antenatal clinics (ANC-RT) are potentially available from all facilities that offer testing services to pregnant women and can be used to improve estimates of national and subnational HIV prevalence trends. We develop methods to incorporate these new data source into the Joint United Nations Programme on AIDS Estimation and Projection Package in Spectrum 2017.We develop a new statistical model for incorporating ANC-RT HIV prevalence data, aggregated either to the health facility level (site-level) or regionally (census-level), to estimate HIV prevalence alongside existing sources of HIV prevalence data from ANC unlinked anonymous testing (ANC-UAT) and household-based national population surveys. Synthetic data are generated to understand how the availability of ANC-RT data affects the accuracy of various parameter estimates.We estimate HIV prevalence and additional parameters using both ANC-RT and other existing data. Fitting HIV prevalence using synthetic data generally gives precise estimates of the underlying trend and other parameters. More years of ANC-RT data should improve prevalence estimates. More ANC-RT sites and continuation with existing ANC-UAT sites may improve the estimate of calibration between ANC-UAT and ANC-RT sites.We have proposed methods to incorporate ANC-RT data into Spectrum to obtain more precise estimates of prevalence and other measures of the epidemic. Many assumptions about the accuracy, consistency, and representativeness of ANC-RT prevalence underlie the use of these data for monitoring HIV epidemic trends and should be tested as more data become available from national ANC-RT programs.

SUBMITTER: Sheng B

PROVIDER: S-EPMC5356494 | biostudies-literature | 2017 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Statistical models for incorporating data from routine HIV testing of pregnant women at antenatal clinics into HIV/AIDS epidemic estimates.

Sheng Ben B Marsh Kimberly K Slavkovic Aleksandra B AB Gregson Simon S Eaton Jeffrey W JW Bao Le L

AIDS (London, England) 20170401

<h4>Objective</h4>HIV prevalence data collected from routine HIV testing of pregnant women at antenatal clinics (ANC-RT) are potentially available from all facilities that offer testing services to pregnant women and can be used to improve estimates of national and subnational HIV prevalence trends. We develop methods to incorporate these new data source into the Joint United Nations Programme on AIDS Estimation and Projection Package in Spectrum 2017.<h4>Methods</h4>We develop a new statistical ...[more]

PMID: 28296804

Similar Datasets

Project description:ObjectivesThe aim of the study was to propose and demonstrate an approach to allow additional nonsampling uncertainty about HIV prevalence measured at antenatal clinic sentinel surveillance (ANC-SS) in model-based inferences about trends in HIV incidence and prevalence.DesignMathematical model fitted to surveillance data with Bayesian inference.MethodsWe introduce a variance inflation parameter (Equation is included in full-text article.)that accounts for the uncertainty of nonsampling errors in ANC-SS prevalence. It is additive to the sampling error variance. Three approaches are tested for estimating (Equation is included in full-text article.)using ANC-SS and household survey data from 40 subnational regions in nine countries in sub-Saharan, as defined in UNAIDS 2016 estimates. Methods were compared using in-sample fit and out-of-sample prediction of ANC-SS data, fit to household survey prevalence data, and the computational implications.ResultsIntroducing the additional variance parameter (Equation is included in full-text article.)increased the error variance around ANC-SS prevalence observations by a median of 2.7 times (interquartile range 1.9-3.8). Using only sampling error in ANC-SS prevalence (Equation is included in full-text article.), coverage of 95% prediction intervals was 69% in out-of-sample prediction tests. This increased to 90% after introducing the additional variance parameter (Equation is included in full-text article.). The revised probabilistic model improved model fit to household survey prevalence and increased epidemic uncertainty intervals most during the early epidemic period before 2005. Estimating (Equation is included in full-text article.)did not increase the computational cost of model fitting.ConclusionsWe recommend estimating nonsampling error in ANC-SS as an additional parameter in Bayesian inference using the Estimation and Projection Package model. This approach may prove useful for incorporating other data sources such as routine prevalence from Prevention of mother-to-child transmission testing into future epidemic estimates.

Project description:The SD BIOLINE HIV/Syphilis Duo assay is the first World Health Organization prequalified dual rapid diagnostic test for simultaneous detection of HIV and Treponema pallidum antibodies in human blood. Prior to introducing the test into antenatal clinics across South Sudan, a field evaluation of its clinical performance in diagnosing both HIV and syphilis in pregnant women was conducted. SD Bioline test performance on venous blood samples was compared with (i) Vironostika HIV1/2 Uniform II Ag/Ab reference standard and Alere Determine HIV 1/2 non-reference standard for HIV diagnosis, and (ii) Treponema pallidum hemagglutination reference standard and Rapid plasma reagin non-reference standard for syphilis. Sensitivity, specificity, positive predictive value (PPN), negative predictive value (NPV) and kappa (κ) value were calculated for each component against the reference standards within 95% confidence intervals (CIs); agreements between Determine HIV 1/2 and SD Bioline HIV tests were also calculated. Of 442 pregnant women recruited, eight (1.8%) were HIV positive, 22 (5.0%) had evidence of syphilis exposure; 14 (3.2%) had active infection. For HIV diagnosis, the sensitivity, specificity, PPV and NPV were 100% (95% CI: 63.1-100), 100% (95% CI: 99.2-100), 100% (95% CI: 63.1-100) and 100% (95% CI: 99.2-100) respectively with κ value of 1 (95% CI: 0.992-1.000). Overall agreement of the Duo HIV component and Determine test was 99.1% (95% CI: 0.977-0.998) with 66.7% (95% CI: 34.9-90.1) positive and 100% (95% CI: 0.992-1.000) negative percent agreements. For syphilis, the Duo assay sensitivity was 86.4% (95% CI: 65.1-97.1) and specificity 100% (95% CI: 99.1-100) with PPV 100% (95% CI: 82.4-100), NPV 99.2% (95% CI: 97.9-99.9) and κ value 0.92 (95% CI: 0.980-0.999). Our findings suggest the SD Bioline HIV/Syphilis Duo Assay could be suitable for HIV and syphilis testing in women attending antenatal services across South Sudan. Women with positive syphilis results should receive treatment immediately, whereas HIV positive women should undergo confirmatory testing following national HIV testing guidelines.

Project description:BACKGROUND: Since 1998 the Joint United Nations Programme on HIV/AIDS and the World Health Organization has provided estimates on the magnitude of the HIV epidemic for individual countries. Starting with the 2003 estimates, plausibility bounds about the estimates were also reported. The bounds are intended to serve as a guide as to what reasonable or plausible ranges are for the uncertainty in HIV incidence, prevalence, and mortality. METHODS: Plausibility bounds were developed for three situations: for countries with generalised epidemics, for countries with low level or concentrated epidemics (LLC), and for regions. The techniques used build on those developed for the previous reporting round. However the current bounds are based on the available surveillance and survey data from each individual country rather than on data from a few prototypical countries. RESULTS: The uncertainty around the HIV estimates depends on the quality of the surveillance system in the country. Countries with population based HIV seroprevalence surveys have the tightest plausibility bounds (average relative range about the adult HIV prevalence (ARR) of -18% to +19%.) Generalised epidemic countries without a survey have the next tightest ranges (average ARR of -46% to +59%). Those LLC countries which have conducted multiple surveys over time for HIV among the populations most at risk have the bounds similar to those in generalised epidemic countries (ARR -40% to +67%). As the number and quality of the studies in LLC countries goes down, the plausibility bounds increase (ARR of -38% to +102% for countries with medium quality data and ARR of -53% to +183% for countries with poor quality data). The plausibility bounds for regions directly reflect the bounds for the countries in those regions. CONCLUSIONS: Although scientific, the plausibility bounds do not represent and should not be interpreted as formal statistical confidence intervals. However in order to make the bounds as meaningful as possible the authors have tried to apply reasonable statistical approaches and assumptions to their derivation. An understanding of the uncertainty in the HIV estimates may help policy makers take better informed decisions to address the epidemic in their respective countries.

Dataset Information

Statistical models for incorporating data from routine HIV testing of pregnant women at antenatal clinics into HIV/AIDS epidemic estimates.

Publications

Statistical models for incorporating data from routine HIV testing of pregnant women at antenatal clinics into HIV/AIDS epidemic estimates.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets