Improved HIV-1 incidence estimates using the BED capture enzyme immunoassay.
ABSTRACT To validate the BED capture enzyme immunoassay for HIV-1 subtype C and to derive adjustments facilitating estimation of HIV-1 incidence from cross-sectional surveys.
Laboratory analysis of archived plasma samples collected in Zimbabwe.
Serial plasma samples from 85 women who seroconverted to HIV-1 during the postpartum year were assayed by BED and used to estimate the window period between seroconversion and the attainment of a specified BED absorbance. HIV-1 incidences for the year prior to recruitment and for the postpartum year were calculated by applying the BED technique to HIV-1-positive samples collected at baseline and at 12 months.
The mean window for an absorbance cut-off of 0.8 was 187 days. Among women who were HIV-1 positive at baseline and retested at 12 months, a proportion (epsilon) 5.2% (142/2749) had a BED absorbance < 0.8 at 12 months and were falsely identified as recent seroconverters. Consequently, the estimated BED annual incidence at 12 months postpartum (7.6%) was 2.2 times the contemporary prospective estimate. BED incidence adjusted for epsilon was 3.5% [95% confidence interval (CI), 2.6-4.5], close to the 3.4% estimated prospectively. Adjusted BED incidence at baseline was 6.0% (95% CI, 5.2-6.9) and, like the prospective estimates, declined with maternal age. Unadjusted BED incidence estimates were largely independent of age; the pooled estimate was 58% higher than adjusted incidence.
The BED method can be used in an African setting, but further estimates of epsilon and of the window period are required, using large samples in a variety of circumstances, before its general utility can be gauged.
- SourceAvailable from: Nurhayati Hamim Kawi
[Show abstract] [Hide abstract]
- "False incident tests occur when HIV IgG decreases below the test threshold as the result of late-stage HIV disease, immune reconstitution, viral suppression among patients receiving anti-retroviral therapy (ART) and concurrent infections (Killian et al, 2006; McDougal et al, 2006; Chawla et al, 2007; Hargrove et al, 2008). About 5% of individuals, fail to progress above the test threshold, and are thus misclassified as recent cases (McDougal et al, 2006; Kartikeyan et al, 2007; Bärnighausen et al, 2008; Hargrove et al, 2008). After reviewing the available evidence, UNAIDS recommended the BED-CEIA method not be used for surveillance purposes due to its tendency to overestimate HIV incidence (UNAIDS, 2006). "
ABSTRACT: Although the BED capture enzyme immunoassay (BED-CEIA) tends to over-estimate HIV incidence in general population epidemics, its limitations may be less relevant to some sub-populations in concentrated epidemics. This study assesses the plausibility of BED-CEIA estimates for female sex workers (FSWs) in Indonesia. Data were derived from a cross-sectional anonymous linked behavioral and biological surveillance survey. Independent samples of 2,917 direct and indirect FSWs, were gathered from seven and five cities, respectively, via three-stage time-location sampling. Participants provided behavioral information, venous blood and vaginal swabs. Specimens testing positive for HIV were subjected to BED-CEIA to identify recent infections. The median duration of sex work was 12 months. The estimated HIV prevalence was 8.2% and the incidence was 4.1 per 100 person years, slightly lower than an Asian Epidemic Model (AEM) estimate. HIV incidence was higher among: direct FSWs (p<0.001), those reporting genital ulcers in the past year (p<0.001), those with active syphilis (p=0.017), and those not receiving periodic presumptive treatment for STIs during the previous 6 months (p=0.045). Low general population HIV prevalence, short durations of sex work and low ART coverage of those eligible for treatment make it unlikely that HIV incidence estimates for FSWs in Indonesia are distorted by long-standing infections and viral suppression. External consistency with model-based estimates and internal consistency in regard to known risk factors for HIV infection add to the plausibility of the estimates. Pending advances in methods for estimating HIV incidence, it may be premature to summarily dismiss the BED-CEIA in concentrated HIV epidemics.The Southeast Asian journal of tropical medicine and public health 05/2011; 42(3):634-42. · 0.55 Impact Factor
[Show abstract] [Hide abstract]
- "This leads to only very minor modifications of the previous expression for weighted incidence, namely a systematic 'subtraction' of over-counted 'not recently infected' individuals which are included in the experimental category 'under threshold'. This subtraction is similar, but not identical, to that proposed in . Section 3 explores the consequences of designing a cross-sectional survey with a sample size N based on the relations derived in Section 2. Using a systematic expansion of the incidence estimator in powers of 1/ √ N (derived in the Appendix), we note consistency of the estimator (no bias in the limit of large N ) and derive an approximate expression for its relative variance. "
ABSTRACT: We present a new analysis of relationships between disease incidence and the prevalence of an experimentally defined state of 'recent infection'. This leads to a clean separation between biological parameters (properties of disease progression as reflected in a test for recent infection), which need to be calibrated, and epidemiological state variables, which are estimated in a cross-sectional survey. The framework takes into account the possibility that details of the assay and host/pathogen chemistry leave a (knowable) fraction of the population in the recent category for all times. This systematically addresses an issue which is the source of some controversy about the appropriate use of the BED assay for defining recent HIV infection. The analysis is, however, applicable to any assay that forms the basis of a test for recent infection. Analysis of relative contributions of error arising variously from statistical considerations and simplifications of general expressions indicate that statistical error dominates heavily over methodological bias for realistic epidemiological and biological scenarios.Journal of Mathematical Biology 08/2009; 60(5):687-710. DOI:10.1007/s00285-009-0282-7 · 2.39 Impact Factor