Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Problems in interpreting HIV sentinel seroprevalence studies

H Strickler1, D R Hoover, R Dersimonian

  • 1Viral Epidemiology Branch, Division of Cancer Etiology, National Cancer Institute, National Institutes of Health, Rockville, MD 20852, USA.

Annals of Epidemiology
|November 1, 1995
PubMed
Summary

Estimating human immunodeficiency virus (HIV) prevalence using sentinel serosurveys is challenging due to biased sampling and time-varying factors. A meta-analysis approach is proposed to improve the accuracy of HIV prevalence estimates.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Frailty-Related Factors among Women Living with and without HIV Aged 40 Years and Older. The Women's Interagency HIV Study.

The Journal of frailty & aging·2024
Same author

Simple Power and Sample Size Estimation for Non-Randomized Longitudinal Difference in Differences Studies.

Journal of biometrics & biostatistics·2019
Same author

Longitudinal evaluation of markers of inflammation in HIV-positive and HIV-negative Rwandan women.

HIV medicine·2018
Same author

Bone quality assessed using quantitative ultrasound at the distal radius does not differ in antiretroviral therapy-naïve HIV-positive and HIV-negative Rwandan women.

HIV medicine·2016
Same author

Risk Factors for the Development of BK Virus Nephropathy in Renal Transplant Recipients.

Transplantation proceedings·2015
Same author

Depression in the first year of stay for elderly long-term nursing home residents in the USA.

International psychogeriatrics·2010

Area of Science:

  • Epidemiology
  • Public Health
  • Biostatistics

Background:

  • Sentinel seroprevalence studies are commonly used to estimate human immunodeficiency virus (HIV) prevalence.
  • These studies often involve incompletely defined cohorts and nonprobability samples, posing challenges for accurate estimation.
  • Time-averaged prevalence data are susceptible to various time-dependent biases, including migration, mortality, and incidence changes.

Purpose of the Study:

  • To characterize the methodologies and limitations of HIV sentinel seroprevalence studies.
  • To identify and explain sources of bias affecting prevalence estimates.
  • To propose an improved methodology for more reliable HIV prevalence estimation.

Main Methods:

  • Characterization of typical HIV sentinel seroprevalence study designs.

Related Experiment Videos

  • Analysis of time-dependent biases and their impact on prevalence data.
  • Evaluation of traditional measures of reliability and validity in these studies.

    • Main Results:

    • Sentinel serosurveys often utilize hypothetical cohorts and nonprobability samples.
    • Prevalence estimates are time-averaged and vulnerable to migration, deaths, and incidence shifts.
    • Standard error is an inadequate measure of reliability; internal validity suffers from self-selection and lab issues; generalizability is limited.

    Conclusions:

    • Existing HIV sentinel seroprevalence studies have significant methodological limitations.
    • Assumptions about time-dependent biases are often necessary but potentially flawed.
    • A formal meta-analysis approach, incorporating data from multiple studies reproducibly, is needed to improve HIV prevalence estimation.