Metabolic & Cardiometabolic
PCOS increases mood disorder
Strong support Metabolic & Cardiometabolic
Part of: • PCOS
RefutedContestedStrong support
consensus score 1.00
📅 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)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
7 support 0 contradict 0 tested null 0 mixed · 7 sources, 7 independent groups
What the evidence shows
PCOS markedly increases depression (~37%) and anxiety (~48%) prevalence.
The evidence (9)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Humeniuk 2025 · Ann Agric Environ Med | meta-analysis | supports | low | PRISMA meta-analysis, PROSPERO CRD42023462618, of 19 cross-sectional studies (4,002 PCOS patients) all scoring anxiety with HADS-A. COMPARATIVE RESULT — the part that tests 'increases': pooled RR of anxiety caseness (HADS-A >=8) 1.91 (95% CI 1.52-2.38, I2=51%) — but from only FIVE studies, 'resulting in 290 women deemed as suffering from PCOS and 113 women from the control group'; and pooled SMD in HADS-A score from only SIX studies ('The group of patients suffering from PCOS consisted of 503 women, while 350 women belonged to the control group'), where the 'cumulative standardized mean difference between the mean score of anxiety in the women suffering from PCOS and controls was 0.27 with 95% confidence interval, 0.12 - 0.43 and I2=18%, proving a low overall effects size without heterogeneity'. The 19-study / 4,002-patient figure is the whole review, NOT the RR base. SINGLE-GROUP PREVALENCE: 'The prevalence of anxiety ranged from 15% - 85.9% (mean 48%)' over 13 studies, with 'The random effects model showed significant heterogeneity in study outcomes (I2 = 97%; p<0.001)' — the headline 48% is near-uninterpretable as a point estimate. Effect is SHRINKING against prior work: 'In the previous meta-analyses, the overall effect size was higher (SMD = 0.49)'; the authors also note Brutocao OR 2.75 and Cooney OR 5.62. LIMITS: 'Every research project used in this meta-analysis was cross-sectional, limiting causal inference regarding the relationship between PCOS diagnosis and depression/anxiety'; risk of bias was scored with a randomized-trial tool ('which were performed using the Cochrane Risk of Bias 2 (RoB 2) tool'); minor publication-bias asymmetry on the RR ('A small asymmetry was found in the funnel plot (Fig. 6b), and confirmed by the LFK=1.31'); the Discussion is unproofed against its own subject, still reporting 'probabilities of depression' inside an anxiety meta-analysis. DEPENDENCY: the same authors' companion depression MA used the identical corpus — 'For both of these metanalysis, the same studies were reviewed' — and co-author Dybciak's own 2022 study is one of the five RR studies. Supports the direction ('women with PCOS may have a significantly increased risk of anxiety compared to those without PCOS') on a thin and non-independent comparative base; quality low. Full text read 2026-08-19. |
| Senturk S et al 2025 · J Clin Med 2025;14(19):7068 | observational | supports | low | Cross-sectional case-control, n=70 (35 Rotterdam-criteria PCOS vs 35 clinic controls, Turkiye): continuous BAI 16 (11-24) vs 12 (2.5-16.5), p=0.006 and BDI 16 (10-22) vs 9 (4-15.5), p=0.007, both higher in PCOS; moderate-to-severe anxiety (BAI >= 16) 57.1% vs 28.6%, p=0.030. But the categorical depression comparison was NULL - depression (BDI >= 17) 40.0% vs 22.9%, p=0.198 - and despite the methods stating the groups were 'matched for age and BMI', BMI differed (26.11 vs 22.25 kg/m2, p=0.004) with no multivariable adjustment, while the paper's own correlations show BDI and BAI tracking BMI (r=0.479 and r=0.372) and HOMA-IR (r=0.532 and r=0.501). Directionally supportive but small, unadjusted, single-centre and confounded by adiposity; the authors concede the design 'is insufficient to establish causality'. |
| Infante-Cano M, García-Muñoz C, Matias-Soto J, Pineda-Escobar S, Villar-Alises O, Martinez-Calderon J 2025 · Arch Womens Ment Health 2025 Jun;28(3):475-489 (e-pub 2024 Oct 25) | meta-analysis | supports | moderate | Overview of 10 systematic reviews (AMSTAR-2 appraised, overlap quantified by corrected covered area; searched to 22 Apr 2024). ON-SCOPE FOR THE PREDICATE 'increases' — unlike the prevalence-only PCOS overviews, this paper reports a non-PCOS comparator: 'Meta-analyses also showed women with PCOS had a higher risk of having anxiety disorders, and depression symptoms than women without PCOS.' Its own pooled figures are single-group prevalence — depressive disorders 34.8%; anxiety disorders 16.9%, panic 4%, social phobia 5% — and are strongly instrument-dependent ('the Hamilton Anxiety Scale 69.4%, the Hospital Anxiety and Depression Scale 41.5%, and the Self-rating Anxiety Scale 32.4%'; depression 46.0% by BDI vs 31-33.8% by HADS). LIMIT: the comparative statement carries no effect estimate in the retrievable text and is inherited narratively from the included reviews (Dokras 2011/2012, Cooney 2017, Brutocao 2018, Veltman-Verhulst 2012) rather than freshly pooled here — so this source confirms the direction but adds no new comparative number, and it overlaps the same systematic-review pool as the Fornaro 2026 umbrella. Abstract-grade: not open access, no full text available. |
| Fornaro M, et al. 2026 · Eur Neuropsychopharmacol | meta-analysis | supports | high | Umbrella review of meta-analyses of observational studies: pooled PREVALENCE in PCOS of major depressive disorder 37.0% (95% CI 29.0-44.0) and anxiety 48.0% (95% CI 37.0-59.0), both among the 10 of 64 (6.4%) prevalence estimates reaching strong credibility. These are single-group prevalence estimates, not comparative — no non-PCOS comparator is reported, and none of the comparative outcome estimates named in the paper concerns PCOS — so this source establishes mood-disorder burden in PCOS, not excess risk versus women without PCOS. |
| Talukder A et al 2025 · PLOS Global Public Health | observational | supports | low | PCOS-positive women had significantly greater prevalence of depression/anxiety/stress/insomnia (DASS-21+ISI-7) vs PCOS-negative women (chi-square + logistic regression) |
| alswh-mood-2019 2019 | observational | supports | moderate | ALSWH cohort (n=5239, PCOS self-reported): antenatal depression 8.9% vs 4.4%, antenatal anxiety 11.7% vs 5.6%, postnatal depression 26.8% vs 18.6% (all p<0.001). After adjustment (incl. pre-existing depression/anxiety) PCOS remained associated with ANTENATAL depression/anxiety (aOR 1.8, 95% CI 1.2-2.6) but NOT postnatal. Cross-sectional, self-reported PCOS. |
| Stanczyk, Lopacinska, Kedzia, Gawlik-Kotelnicka 2026 · J Clin Med | meta-analysis | supports | low | SR+MA of PROPORTIONS (GLMM, logit link) over 35 observational studies, 2006-2024: 'The analysis was conducted on a group of 5857 women (study group: 3610; control group: 2247) obtained from 35 studies, which met the inclusion criteria.' COMPARATIVE RESULT - the part that tests 'increases': BDI depressive-symptom prevalence 'higher in the PCOS group at 42.1% (95% CI: 32.6% to 52.2%) compared to 13.6% (95% CI: 8.4% to 21.5%) in the control group', subgroup test p<0.001; HADS-D 'higher in the PCOS group at 28.9% (95% CI: 20.7% to 38.8%) compared to 15.8% (95% CI: 11.3% to 21.7%) in the control group', p=0.010. BUT THE OTHER HALF IS NULL, which the prior extract omitted: Mini-NPI was 59.1% (95% CI: 30.9% to 82.4%) in PCOS vs 65.9% (95% CI: 24.2% to 92.1%) in controls - direction REVERSED - with 'The test for subgroup differences was not statistically significant (Q = 0.07, df = 1, p = 0.792)'; PHQ 26.5% vs 9.1%, p=0.255; and ANXIETY was null - HADS-A 48.25% (95% CI: 36.1% to 60.6%) vs 31.4% (95% CI: 18.8% to 47.4%) with 'The test for subgroup differences was not statistically significant (Q = 2.74, df = 1, p = 0.098)'. So this source supports the DEPRESSION half of the claim and is tested-null on the ANXIETY half; the claim's '~48% anxiety' figure is a single-group prevalence, not an excess-risk finding. QUALITY LIMITS: no risk-of-bias or methodological-quality appraisal of any included study is performed anywhere (only publication-bias funnel/Peters tests); 'The study protocol was not registered in the International Prospective Register of Systematic Reviews (PROSPERO).'; heterogeneity is extreme in every analysis (I 2 = 92.1% (95% CI: 89.8% to 93.8%) for BDI, up to 95.5% for PHQ); the PCOS-vs-control contrast is a BETWEEN-SUBGROUP comparison of unequal study sets (BDI k=17 PCOS arms vs k=15 control arms; HADS-A k=15 vs k=7), not a paired within-study effect, so between-study variance is never differenced out; no BMI adjustment despite PCOS median BMI 27.2 vs control 24.4 kg/m2 and Table S2 recording matching variables as 'None' for many included studies. The authors concede 'the broad methodological range of the studies included in the anxiety symptoms analysis resulted in the inability to reach definitive conclusions, as evidenced by the heterogeneity of the sample size' and close with 'Due to the limited number and quality of included studies, these conclusions require further confirmation by larger-sample, multi-centre clinical trials.' Their own conclusion is nonetheless directional: 'The meta-analysis demonstrates that polycystic ovary syndrome significantly increases the prevalence of depressive symptoms.' INDEPENDENCE: primary-study pool (Dybciak, Hollinrake 2007, Tariq 2021, Zueff 2015, Asik 2015, Batool 2016 et al.) overlaps the corpora behind s40586494, fornaro-2026-mood and infante-cano-2025-pcos-mood on this same claim. Funding: 'This research received no external funding.'; 'The authors declare no conflicts of interest.' Full text + Supplementary Tables S2/S3 read 2026-08-19. |
| Atinga A, Bashiru HA, Solomon AO, et al. 2025 · Front Glob Womens Health | meta-analysis | supports | moderate | SR+MA (CRD420251069068) pooling PREVALENCE of depression and anxiety among women with PCOS in LMICs, 2005-2025, with within-PCOS risk-factor ORs (menstrual irregularity, infertility, hirsutism). No PCOS-vs-non-PCOS comparison is analysed, so the paper establishes burden in PCOS, not excess risk versus women without PCOS. |
| Vela AM et al 2026 · Am J Prev Cardiol | observational | supports | low | Adjusted for BMI, age, race, ethnicity, education and hormonal contraception, PCOS was independently associated with higher depression (CES-D +1.95, 95% CI 0.62-3.27, p=0.004) and anxiety (GAD-7 +1.38, 95% CI 0.78-1.99, p<0.001); the authors call these 'very small to small effects'. Cross-sectional online survey, n=1574 (881 PCOS), PCOS status and BMI both self-reported and pooling suspected with confirmed cases. |
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.