Bayesian inference for prevalence and diagnostic test accuracy based on dual-pooled screening.

Timothy E Hanson; Wesley O Johnson; Joseph L Gastwirth

doi:10.1093/biostatistics/kxi039

Home/Archive/PubMed

PubMed · 15947009

Bayesian inference for prevalence and diagnostic test accuracy based on dual-pooled screening.

Timothy E Hanson·Wesley O Johnson·Joseph L Gastwirth

View original source ↗Export metadata ↓

Abstract

We propose a useful protocol for the problem of screening populations for low-prevalence characteristics such as HIV or drugs. Current HIV screening of blood that has been donated for transfusion involves the testing of individual blood units with an inexpensive enzyme-linked immunosorbent assay test and follow-up with a more accurate and more expensive western blot test for only those units that tested positive. Our cost-effective pooling strategy would enhance current methods by making it possible to accurately estimate the sensitivity and specificity of the initial screening test, and the proportion of defective units that have passed through the system. We also provide a method of estimating the distribution of prevalences for the characteristic throughout the population or subpopulations of interest.

Explore related subjects

AIDS Serodiagnosis Acquired Immunodeficiency Syndrome Bayes Theorem Blood Donors Humans Midwestern United States Monte Carlo Method Predictive Value of Tests Prevalence Sampling Studies Sensitivity and Specificity Seroepidemiologic Studies Southeastern United States Southwestern United States Comparative Study Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Timothy E Hanson, Wesley O Johnson, Joseph L Gastwirth. 2005-06-09. Bayesian inference for prevalence and diagnostic test accuracy based on dual-pooled screening.. https://doi.org/10.1093/biostatistics%2Fkxi039

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Can increasing awareness of HIV seropositivity reduce infections by 50% in the United States?

BACKGROUND: In 2003, US Centers for Disease Control and Prevention (CDC) announced an HIV prevention initiative that emphasized increasing the proportion of persons living with HIV who are aware of their positive serostatus. Previously, CDC set a national HIV prevention goal of reducing new infections by 50% by 2005, with increasing serostatus awareness a central goal. The objectives of this article are to examine the national epidemiologic consequences of increasing serostatus awareness and the important question of how far increasing awareness could lead toward achieving CDC's goal of a 50% reduction in incident infections. METHODS: Scenario analysis methods were used to project incident HIV infections over 1- and 10-year time horizons at varying levels of HIV seropositivity awareness. Because HIV transmission rates are several times higher for those who are unaware of their HIV seropositivity than for those who are aware, we modeled the effects of increasing awareness via HIV counseling and testing and moving people from the higher to the lower transmission category. RESULTS: Increasing awareness of HIV seropositivity via HIV counseling and testing in the United States from a baseline 75% to 95% has the potential of reducing new infections by as much as 32% in a single year. Achieving the national goal of 50% reduction seems out of reach, however. CONCLUSIONS: Reduction of new infections by 50% in the United States requires a combination of prevention services to reduce further the already low HIV transmission rate from persons aware of their HIV seropositivity and HIV risk reduction interventions for at-risk persons. Counseling and testing alone seems insufficient to meet the 50% incidence reduction goal.

AIDS Serodiagnosis↗

PubMed2007

Prozone effect with the VIDAS HIV DUO Ultra assay.

Explore the source record for details and available documents.

AIDS Serodiagnosis↗

PubMed2007

Estimating HIV incidence in voluntary counseling and testing clients in Uganda (1992-2003).

OBJECTIVES: To estimate HIV incidence from first-time testers among voluntary counseling and testing (VCT) clients in Uganda. METHODS: Data on 203,000 VCT clients tested from 1992 through 2003 were adjusted for temporal changes in the testing population. Differential mortality rates by HIV status were used to derive expected prevalence at future times from baseline prevalence within 5-year birth cohorts. Incidence was computed as the proportion of HIV-uninfected persons who seroconverted divided by the proportion negative at baseline. RESULTS: Annual HIV incidence per 100 uninfected persons increased from 0.9 (95% confidence interval [CI]: 0.8 to 1.1) in 1993 to 2.3 (95% CI: 2.2 to 2.5) in 2003 (chi test for trend, P < 0.001). Prevalence decreased from 23% to 13% in 1999 to 2000 and increased to 15% in 2003. Women had a higher incidence. Peak incidence shifted to older age groups over time. CONCLUSIONS: Estimating incidence from routine data presents a practical way of tracking HIV incidence and is a useful tool in targeting and evaluating the impact of prevention programs. Our analysis reveals a new phase of the HIV epidemic in Uganda: decreasing prevalence and increasing incidence, especially among middle-aged persons. These findings support the need for intensified prevention interventions among middle-aged persons in Uganda.

AIDS Serodiagnosis↗