Metabolic & Cardiometabolic
PCOS causes insulin resistance
Part of: • PCOS
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
How the studies fall
What the evidence shows
PCOS involves an intrinsic insulin-resistance defect, independent of obesity — IR is a primary driver, not a downstream comorbidity. ⭐ The biological core of the 'PCOS is a metabolic disease' reclassification.
The evidence (7)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Alataş H, Arslan N, Keskin L, Ağralı C 2026 · Diabetes, metabolic syndrome and obesity : targets and therapy | observational | supports | moderate | Cross-sectional, 129 PCOS vs 135 age-matched controls (BMI-matching not stated). HOMA-IR: PCOS 3.52 vs control 1.98 (~1.8x higher, 'nearly twice as high'). |
| szkodziak-2025 2025 · Scientific reports | observational | supports | low | PCOS-only retrospective cohort, no control group: 'Insulin resistance was present in 57.5% of participants' (n=200), IR defined by an imported cut-off ('Participants were stratified by HOMA-IR into two groups: non-insulin-resistant (HOMA-IR <= 2.00) and insulin-resistant (HOMA-IR >= 2.01)') whose published range the paper itself gives as 'Reported cut-offs range from 1.55 in the Thai population to >= 3.8 in the French population'. 'Elevated across all phenotypes' is a statement about means (A 3.59, B 2.59, C 2.05, D 2.73): in phenotype C 'most women ... were insulin-sensitive (73.91%)'. Hospitalised referral sample (46% phenotype A vs 19% in unselected populations by the paper's own cited meta-analysis); no clamp, no HbA1c or lipids. |
| zhu-2019 2019 · Fertility and sterility | meta-analysis | supports | moderate | Metabolic dysfunction in NON-obese PCOS vs non-obese controls. |
| Ayonrinde OT, Mori TA, Adams LA, Beilin LJ, Olynyk JK, Hart R 2026 · The Journal of clinical endocrinology and metabolism | observational | contradicts | moderate | Raine Study, longitudinal adolescents (n=199) followed to age 27. 'PCOS without MASLD or obesity in adolescence did not predict future insulin resistance'; only PCOS+MASLD (linked to obesity) predicted later IR. |
| Márquez-Arrico CF, Pelechá-Salvador M, López-Domènech S, Fernández-Reyes M, Silvestre FJ, Silvestre-Rangil J, Montiel-Company JM, Rocha M 2026 · BMC oral health | observational | supports | moderate | Case-control, PCOS vs age+BMI-matched healthy controls (Rotterdam criteria). PCOS group had 'elevated insulin levels, insulin resistance' vs controls; no exact HOMA-IR values/CI reported in abstract. |
| Serin S, Celikaslan H, Cangi S, Dokuyucu R 2026 · Sci Rep | animal | supports | low | Controlled rodent experiment, not a human study — 'Sixty female Wistar Albino rats were randomly divided into six groups' — so grade corrected from observational to animal, matching the source note's study_type. Finding: 'Letrozole administration induced persistent diestrus, hyperandrogenemia, increased ovarian weight, elevated HOMA-IR (6.61 ± 1.18 vs. 2.24 ± 0.42, p < 0.001)'. Weak analogue vote only: the exposure actually manipulated is letrozole, an aromatase inhibitor that imposes both androgen excess and oestrogen deficiency, and oestrogen deficiency alone causes insulin resistance in female rodents — so the design cannot separate PCOS-ness from the inducing drug. HOMA-IR is a surrogate (no clamp), assessment is unblinded, and body weight is not reported, so the claim's actual scope question (intrinsic, independent of obesity) is untouched. |
| Dunaif A 1995 · Am J Med | observational | supports | high | Supplement narrative review (Am J Med 1995;98(1A):33S-39S) summarising prior studies: 'Studies assessing whether abnormalities in insulin action are intrinsic or secondary to the hormonal milieu have found that insulin-induced receptor autophosphorylation is markedly diminished in approximately 50% of polycystic ovary syndrome women.' 'This distinctive post-insulin-binding defect appears to be genetic, since it is present in cells removed from the in vivo environment for generations.' Re-describes earlier primary work — zero weight as review; the underlying primaries are not yet in the vault. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.