Lizneva · Fertility and sterility 2016 · systematic review and meta-analysis of observational studies · n=41 studies (13,796 participants across 43 populations)

Phenotypes and body mass in women with polycystic ovary syndrome identified in referral versus unselected populations: systematic review and meta-analysis.

Cited 170 times in the scientific literature.

Level 3 - non-randomized controlled study

Systematic review and meta-analysis of observational studies

PubMed 27530062 · doi:10.1016/j.fertnstert.2016.07.1121 · record verified 2026-08-26

What was done

A systematic review and meta-analysis of observational studies from PubMed, EMBASE, and the Cochrane Library (2003–2016) was conducted to compare polycystic ovary syndrome (PCOS) phenotypes and body mass index (BMI) between referral and unselected cohorts. The analysis included 41 studies reporting on 43 populations (13,796 reproductive-age women diagnosed by Rotterdam 2003 criteria). Pooled prevalence estimates were calculated for phenotypes A (hyperandrogenism [HA] + oligo-/anovulation [OA] + polycystic ovarian morphology [PCOM]), B (HA + OA), C (HA + PCOM), and D (OA + PCOM).

What was found

Prevalence estimates differed significantly between referral and unselected populations for phenotypes A, B, and C: - Phenotype A: 50% (95% CI, 46%–54%) referral vs. 19% (95% CI, 13%–27%) unselected. - Phenotype B: 13% (95% CI, 11%–17%) referral vs. 25% (95% CI, 15%–37%) unselected. - Phenotype C: 14% (95% CI, 12%–16%) referral vs. 34% (95% CI, 25%–46%) unselected. - Phenotype D: 17% (95% CI, 13%–22%) referral vs. 19% (95% CI, 14%–25%) unselected (not statistically significant). Referral PCOS patients had a significantly higher mean BMI than local controls, a pattern not found in unselected PCOS cohorts.

Why it matters

This demonstrates marked referral bias in clinical PCOS research, indicating that clinic-based studies disproportionately represent more severe, complete phenotypes and higher BMI compared to community-based populations.

Limits

The analysis relied exclusively on observational studies, introducing potential variability in diagnostic definitions and baseline characteristics. The abstract does not provide exact numerical values for BMI differences or measures of study heterogeneity.

Cited by