Prevalence threshold of screening tests in obstetrics and gynecology

Liora Elfassy; Ariane Lasry; Yaron Gil; Jacques Balayla

doi:10.1016/j.ejogrb.2021.02.015

Prevalence threshold of screening tests in obstetrics and gynecology

Eur J Obstet Gynecol Reprod Biol. 2021 Apr:259:191-195. doi: 10.1016/j.ejogrb.2021.02.015. Epub 2021 Feb 18.

Authors

Liora Elfassy¹, Ariane Lasry², Yaron Gil³, Jacques Balayla⁴

Affiliations

¹ Faculty of Medicine, Universite De Sherbrooke, Sherbrooke, Quebec, Canada.
² Faculty of Medicine, McGill University, Montreal, Quebec, Canada.
³ Department of Obstetrics and Gynecology, McGill University, Montreal, Quebec, Canada; Department of Obstetrics and Gynecology, Lis Maternity Hospital for Women, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel.
⁴ Faculty of Medicine, McGill University, Montreal, Quebec, Canada; Department of Obstetrics and Gynecology, McGill University, Montreal, Quebec, Canada. Electronic address: jacques.balayla@mcgill.ca.

PMID: 33711775
DOI: 10.1016/j.ejogrb.2021.02.015

Abstract

Objective: We define the prevalence threshold as the prevalence level below which a test's positive predictive value (PPV) declines most sharply relative to disease prevalence - and thus the rate of false positive results/false discovery rate increases most rapidly. The objective of this study is to determine the prevalence threshold of various screening tests used in obstetrics and gynecology among low-risk women in modern clinical practice.

Methods: We searched Medline, EMBASE, Google Scholar, Scopus, ISI Web of Science, Cochrane database, and PubMed to obtain the sensitivity and specificity estimates for the following screening tests: 50 g-oral glucose tolerance test (GDM-50 g), non-invasive prenatal testing (NIPT), combined first trimester screening (FTS), vagino-rectal swab for group B streptococcus (GBS) in pregnancy, cervical cytology (Pap) and HPV testing, mammography and manual breast exam, urinary PCR and cervical-vaginal swab testing for gonorrhoea and chlamydia as well as AMH for the diagnosis of PCOS. We used these estimates to calculate disease-specific prevalence thresholds, comparing them to the actual estimates of disease prevalence.

Results: The prevalence thresholds and average estimates of disease prevalence (shown in brackets) are as follows: GDM-50 g 31 % (6%), NIPT 7% (0.2 %), combined FTS 19.5 % (0.2 %), GBS swab 18 % (15-45 %), Pap 21 % (0.2 %), HPV 27 % (0.2 %), mammography 25 % (12.5 %), breast exam 25 % (12.5 %), gonorrhoea -chlamydia 6-13 % (4.2-4.7 %), AMH for PCOS 32 % (10 %).

Conclusion: The prevalence thresholds of various screening tests used in obstetrics and gynecology are well above the estimated disease prevalence. This implies that when undertaking population-level screening a significant proportion of positive screening tests obtained are likely false-positives. Attempts at individualizing pre-test probability when undertaking population-level screening are needed in order to best interpret the results of screening tests.

Keywords: Bayes’ theorem; Prevalence; Prevalence threshold; Screening.

MeSH terms

Female
Gynecology*
Humans
Mass Screening
Obstetrics*
Pregnancy
Prevalence
Sensitivity and Specificity
Vaginal Smears