A Simplified Formula for Inferring HIV Incidence from Cross-Sectional Surveys Using a Test for Recent Infection

Abstract BackgroundMany surveys have attempted to estimate HIV incidence from cross-sectional data which includes ascertainment of ‘recent infection’, but the inevitable age and time structure of this data has never been systematically explored – no doubt partly because statistical precision in such estimates is often insufficient to allow for satisfactory disaggregation. Given the non-trivial age structure of HIV incidence and prevalence, and the enormous investments that have been made in such data sets, it is important to understand effective ways to extract valid age structure from these precious data sets. MethodsUsing a comprehensive demographic/epidemiological simulation platform developed for this, and some wider, purposes (documented in more detail separately) we simulated a complex ‘South Africa inspired’ HIV epidemic, with explicitly specified 1) age/time dependent incidence, 2) age/time dependent mortality for uninfected individuals, and 3) age/time/time-since-infection dependent mortality for infected individuals. In this simulated world, we conducted cross-sectional surveys at various times, and applied variants of the recent infection based incidence estimation methodology of Kassanjee et al. We analysed in considerable detail how to smooth, and average over, the age structure in these surveys to produce the incidence estimates, paying attention to the fundamental trade-off between bias and statistical error.ResultsWe summarise our detailed observations about incidence estimates, generated by various age smoothing or age disaggregation procedures, into a straightforward fully specified ‘one size fits most’ algorithm for processing the survey data into age-specific incidence estimates: 1) generalised linear regression to turn observations into ‘prevalence’ of ‘infection’ and ‘recent infection’ (logit, and complementary log log, link functions, respectively; fitting coefficients of up to cubic terms in age/time); 2) a ‘moving window’ data inclusion recipe which handles each age/time point of interest separately; 3) post hoc age averaging of resulting pseudo continuously fitted incidence; 4) bootstrapping as a generic variance/significance estimation procedure.ConclusionsAs far as we are aware, this is the first analysis of several fine details of how age structure in cross-sectional surveys interacts with recency-based incidence estimation. Our proposed default estimation procedure generates incidence estimates with negligible bias and near-optimal precision, and can be readily applied to complex survey data sets by any group in possession of such data. Our code is available, in part freely through the R computing platform, and in part upon request.

The added value of recent-infection testing in population-based HIV surveys

10.21203/rs.3.rs-996585/v2 ◽

2021 ◽

Author(s):

Laurette Mhlanga ◽

Eduard Grebe ◽

Alex Welte

Keyword(s):

Sexual Debut ◽

Survey Design ◽

Value Added ◽

Early Years ◽

Added Value ◽

Hiv Incidence ◽

Recent Infection ◽

Cross Sectional ◽

Marginal Benefit ◽

Incidence Trends

Abstract Background: There is no clear consensus on how best to use increasingly available data derived from large populationbased surveys featuring HIV infection status ascertainment. In particular, for the purpose of estimating HIV incidence, there is considerable scope for better elucidation of the benefit of adding ‘recent infection’ ascertainment, which adds considerable additional cost and complexity to surveys which are already costly and complex. Methods: Using an epidemic/survey simulation tool developed for this and some closely related investigations, we explore the value added by ‘recent infection’ data from population surveys, to support HIV incidence estimation. This directly piggy-backs on to two companion pieces which have explored, independently, the use of the ‘synthetic cohort’ paradigm of Mahiane et al (analysing age/time structure of prevalence, in conjunction with estimates of mortality) and the paradigm of Kassanjee et al (focusing on ‘recent infection’ data). Results: Our headline findings are that: 1) Recent infection data adds marginal benefit to surveillance focused on the early years after sexual debut, which can reasonably be taken to be a core sentinel group in which surveillance is significantly more efficient than attempts to cover all ages; and 2) by contrast, recent infection data is crucial for the reliable estimation of incidence trends when only two cross sectional surveys are available. We detail numerous components of a general and robust approach to analysing data when both the Mahiane and Kassanjee analyses are in play. Conclusion: Our main results present non-trivial dilemmas for survey design, as recency data is crucial for stabilising the more timely estimates, but of marginal benefit for the most important sentinel group. We hope that adaptation of our analysis, to simulated scenarios closely aligned to specific contexts facing expensive choices, will support rational investments in, and use of, precious surveillance opportunities and data sets.

Using tests for recent infection to estimate incidence: problems and prospects for HIV

Eurosurveillance ◽

10.2807/ese.15.24.19589-en ◽

2010 ◽

Vol 15 (24) ◽

Author(s):

A Welte ◽

T A McWalter ◽

O Laeyendecker ◽

T B Hallett

Keyword(s):

Future Development ◽

Theoretical Work ◽

Incidence Rates ◽

Hiv Incidence ◽

Unified Framework ◽

Cross Sectional Survey ◽

Recent Infection ◽

Cross Sectional ◽

Recent Theoretical Work ◽

Survey Sample

Tests for recent infection (TRIs), such as the BED assay, provide a convenient way to estimate HIV incidence rates from cross-sectional survey data. Controversy has arisen over how the imperfect performance of a TRI should be characterised and taken into account. Recent theoretical work is providing a unified framework within which to work with a variety of TRI- and epidemic-specific assumptions in order to estimate incidence using imperfect TRIs, but suggests that larger survey sample sizes will be required than previously thought. This paper reviews the framework qualitatively and provides examples of estimator performance, identifying the characteristics required by a TRI to estimate incidence reliably that should guide the future development of TRIs.

Recent HIV infection among pregnant women in the 2017 antenatal sentinel cross–sectional survey, South Africa: Assay–based incidence measurement

PLoS ONE ◽

10.1371/journal.pone.0249953 ◽

2021 ◽

Vol 16 (4) ◽

pp. e0249953

Author(s):

Selamawit Woldesenbet ◽

Tendesayi Kufa-Chakezha ◽

Carl Lombard ◽

Samuel Manda ◽

Mireille Cheyip ◽

...

Keyword(s):

South Africa ◽

Pregnant Women ◽

Young Women ◽

Reference Group ◽

Hiv Positive ◽

Hiv Incidence ◽

Cross Sectional Survey ◽

Recent Infection ◽

Cross Sectional ◽

One Year

Introduction New HIV infection during pre-conception and pregnancy is a significant contributor of mother–to–child transmission of HIV in South Africa. This study estimated HIV incidence (defined as new infection within the last one year from the time of the survey which included both new infections occurred during pregnancy or just before pregnancy) among pregnant women and described the characteristics of recently infected pregnant women at national level. Methods Between 1 October and 15 November 2017, we conducted a national cross–sectional survey among pregnant women aged 15–49 years old attending antenatal care at 1,595 public facilities. Blood specimens were collected from pregnant women and tested for HIV in a centralised laboratory. Plasma viral load and Limiting Antigen Avidity Enzyme Immunosorbent Assay (LAg) tests were further performed on HIV positive specimens to differentiate between recent and long–term infections. Recent infection was defined as infection that occurred within one year from the date of collection of blood specimen for the survey. Data on age, age of partner, and marital status were collected through interviews. Women whose specimens were classified as recent by LAg assay and with viral loads >1,000 copies/mL were considered as recently infected. The calculated proportion of HIV positive women with recent infection was adjusted for assay–specific parameters to estimate annual incidence. Survey multinomial logistic regression was used to examine factors associated with being recently infected using HIV negative women as a reference group. Age–disparate relationship was defined as having a partner 5 or more years older. Results Of 10,049 HIV positive participants with LAg and viral load data, 1.4% (136) were identified as recently infected. The annual HIV incidence was 1.5% (95% confidence interval (CI): 1.2–1.7). In multivariable analyses, being single (adjusted odds ratio, aOR: 3.4, 95% CI: 1.8–6.2) or cohabiting (aOR: 3.8, 95% CI: 1.8–7.7), compared to being married as well as being in an age–disparate relationship among young women (aOR: 3.1, 95% CI: 2.0–4.7; reference group: young women (15–24years) whose partners were not 5 years or more older) were associated with higher odds of recent infection. Conclusions Compared to previous studies among pregnant women, the incidence estimated in this study was substantially lower. However, the UNAIDS target to reduce incidence by 75% by 2020 (which is equivalent to reducing incidence to <1%) has not been met. The implementation of HIV prevention and treatment interventions should be intensified, targeting young women engaged in age–disparate relationship and unmarried women to fast track progress towards the UNAIDS target.

Optimal accounting for age and time structure of HIV incidence estimates based on cross-sectional survey data with ascertainment of ‘recent infection’

10.21203/rs.3.rs-871044/v1 ◽

2021 ◽

Author(s):

Laurette Mhlanga ◽

Grebe Eduard ◽

Alex Welte

Keyword(s):

Survey Data ◽

Age Structure ◽

Estimation Procedure ◽

Time Dependent ◽

Data Sets ◽

Hiv Incidence ◽

Recent Infection ◽

Cross Sectional ◽

Incidence Estimation ◽

Dependent Mortality

Abstract BackgroundMany surveys have attempted to estimate HIV incidence from cross-sectional data which includes ascertainment of ‘recent infection’, but the inevitable age and time structure of this data has never been systematically explored – no doubt partly because statistical precision in such estimates is often insufficient to allow for satisfactory disaggregation. Given the non-trivial age structure of HIV incidence and prevalence, and the enormous investments that have been made in such data sets, it is important to understand effective ways to extract valid age structure from these precious data sets. MethodsUsing a comprehensive demographic/epidemiological simulation platform developed for this, and some wider, purposes (documented in more detail separately) we simulated a complex ‘South Africa inspired’ HIV epidemic, with explicitly specified 1) age/time dependent incidence, 2) age/time dependent mortality for uninfected individuals, and 3) age/time/time-since-infection dependent mortality for infected individuals. In this simulated world, we conducted cross-sectional surveys at various times, and applied variants of the recent infection based incidence estimation methodology of Kassanjee et al. We analysed in considerable detail how to smooth, and average over, the age structure in these surveys to produce the incidence estimates, paying attention to the fundamental trade-off between bias and statistical error.ResultsWe summarise our detailed observations about incidence estimates, generated by various age smoothing or age disaggregation procedures, into a straightforward fully specified ‘one size fits most’ algorithm for processing the survey data into age-specific incidence estimates: 1) generalised linear regression to turn observations into ‘prevalence’ of ‘infection’ and ‘recent infection’ (logit, and complementary log log, link functions, respectively; fitting coefficients of up to cubic terms in age/time); 2) a ‘moving window’ data inclusion recipe which handles each age/time point of interest separately; 3) post hoc age averaging of resulting pseudo continuously fitted incidence; 4) bootstrapping as a generic variance/significance estimation procedure.ConclusionsAs far as we are aware, this is the first analysis of several fine details of how age structure in cross-sectional surveys interacts with recency-based incidence estimation. Our proposed default estimation procedure generates incidence estimates with negligible bias and near-optimal precision, and can be readily applied to complex survey data sets by any group in possession of such data. Our code is available, in part freely through the R computing platform, and in part upon request.

Cross-Sectional HIV Incidence Surveillance: A Benchmarking of Approaches for Estimating the ‘Mean Duration of Recent Infection’

Statistical Communications in Infectious Diseases ◽

10.1515/scid-2016-0002 ◽

2017 ◽

Vol 9 (1) ◽

Cited By ~ 1

Author(s):

Reshma Kassanjee ◽

Daniela De Angelis ◽

Marian Farah ◽

Debra Hanson ◽

Jan Phillipus Lourens Labuschagne ◽

...

Keyword(s):

Random Effects ◽

Hiv Incidence ◽

Recent Infection ◽

Cross Sectional ◽

Loss To Follow Up ◽

Missed Visits ◽

The Mean ◽

Substantial Bias ◽

Similar Accuracy ◽

Study Designs

AbstractThe application of biomarkers for ‘recent’ infection in cross-sectional HIV incidence surveillance requires the estimation of critical biomarker characteristics. Various approaches have been employed for using longitudinal data to estimate the Mean Duration of Recent Infection (MDRI) – the average time in the ‘recent’ state. In this systematic benchmarking of MDRI estimation approaches, a simulation platform was used to measure accuracy and precision of over twenty approaches, in thirty scenarios capturing various study designs, subject behaviors and test dynamics that may be encountered in practice. Results highlight that assuming a single continuous sojourn in the ‘recent’ state can produce substantial bias. Simple interpolation provides useful MDRI estimates provided subjects are tested at regular intervals. Regression performs the best – while ‘random effects’ describe the subject-clustering in the data, regression models without random effects proved easy to implement, stable, and of similar accuracy in scenarios considered; robustness to parametric assumptions was improved by regressing ‘recent’/‘non-recent’ classifications rather than continuous biomarker readings. All approaches were vulnerable to incorrect assumptions about subjects’ (unobserved) infection times. Results provided show the relationships between MDRI estimation performance and the number of subjects, inter-visit intervals, missed visits, loss to follow-up, and aspects of biomarker signal and noise.

Comparison of cross-sectional HIV incidence assay results from dried blood spots and plasma

PLoS ONE ◽

10.1371/journal.pone.0172283 ◽

2017 ◽

Vol 12 (2) ◽

pp. e0172283 ◽

Cited By ~ 11

Author(s):

Katherine E. Schlusser ◽

Christopher Pilcher ◽

Esper G. Kallas ◽

Breno R. Santos ◽

Steven G. Deeks ◽

...

Keyword(s):

Dried Blood Spots ◽

Hiv Incidence ◽

Cross Sectional ◽

Blood Spots

Short Communication: Defining Optimality of a Test for Recent Infection for HIV Incidence Surveillance

AIDS Research and Human Retroviruses ◽

10.1089/aid.2013.0113 ◽

2014 ◽

Vol 30 (1) ◽

pp. 45-49 ◽

Cited By ~ 23

Author(s):

Reshma Kassanjee ◽

Thomas A. McWalter ◽

Alex Welte

Keyword(s):

Short Communication ◽

Hiv Incidence ◽

Recent Infection

A cross sectional study of leptospirosis and fetal death in Yucatan, Mexico

Colombia Medica ◽

10.25100/cm.v47i1.1975 ◽

2016 ◽

pp. 11-14 ◽

Cited By ~ 2

Author(s):

María Fidelia Cárdenas Marrufo ◽

Ignacio Vado Solis ◽

Gaspar Fernando Peniche Lara ◽

Carlos Perez Osorio ◽

José Correa Segura

Keyword(s):

Spontaneous Abortion ◽

Low Income ◽

Fetal Death ◽

Indirect Elisa ◽

First Trimester ◽

Cross Sectional Study ◽

Sectional Study ◽

Serum Samples ◽

Recent Infection ◽

Cross Sectional

Introduction: Leptospirosis is a zoonotic disease affecting mainly to low income human population. Acute leptospiral infection during pregnancy has been associated with spontaneous abortion and fetal death during the first trimester and the abortion may occur as consequence of systemic failure. Objective: To estimate the frequency of Leptospira interrogans infection in women with spontaneous abortion in the state of Yucatan, Mexico. Methods: A cross sectional study on women with spontaneous abortion was conducted. Serum samples were tested for Leptospirosis by the microaglutination test, to estimate the frequency of the infecting serovar. The indirect ELISA IgM was used to detect recent infection by L. interrogans. DNA was extracted from paraffin-embedded tissue of placenta for PCR detection of L. interrogans. Results: Overall frequency of infection with L. interrogans in the 81 women with abortion was 13.6%. Five of the 12 serovars evaluated were found and included. Two of the 11 women with abortion and positive to microaglutination test were also positive to the ELISA IgM test. None samples were positive for PCR Leptospira diagnosis. Conclusion: two women could be associated with spontaneous abortion due to leptospirosis, because they showed antibodies against L. interrogans in the microaglutination test and ELISA IgM assays. Differences between regions were found with respect to the prevalences of lesptospirosis.

Moving towards a reliable HIV incidence test – current status, resources available, future directions and challenges ahead

Epidemiology and Infection ◽

10.1017/s0950268816002910 ◽

2016 ◽

Vol 145 (5) ◽

pp. 925-941 ◽

Cited By ~ 19

Author(s):

G. MURPHY ◽

C. D. PILCHER ◽

S. M. KEATING ◽

R. KASSANJEE ◽

S. N. FACENTE ◽

...

Keyword(s):

Performance Metrics ◽

Critical Path ◽

Population Level ◽

Current Status ◽

Hiv Incidence ◽

Recent Infection ◽

Novel Approach ◽

Product Profile ◽

And Performance ◽

The Impact

SUMMARYIn 2011 the Incidence Assay Critical Path Working Group reviewed the current state of HIV incidence assays and helped to determine a critical path to the introduction of an HIV incidence assay. At that time the Consortium for Evaluation and Performance of HIV Incidence Assays (CEPHIA) was formed to spur progress and raise standards among assay developers, scientists and laboratories involved in HIV incidence measurement and to structure and conduct a direct independent comparative evaluation of the performance of 10 existing HIV incidence assays, to be considered singly and in combinations as recent infection test algorithms. In this paper we report on a new framework for HIV incidence assay evaluation that has emerged from this effort over the past 5 years, which includes a preliminary target product profile for an incidence assay, a consensus around key performance metrics along with analytical tools and deployment of a standardized approach for incidence assay evaluation. The specimen panels for this evaluation have been collected in large volumes, characterized using a novel approach for infection dating rules and assembled into panels designed to assess the impact of important sources of measurement error with incidence assays such as viral subtype, elite host control of viraemia and antiretroviral treatment. We present the specific rationale for several of these innovations, and discuss important resources for assay developers and researchers that have recently become available. Finally, we summarize the key remaining steps on the path to development and implementation of reliable assays for monitoring HIV incidence at a population level.