Selective serotonin reuptake inhibitor and serotonin-norepinephrine reuptake inhibitor use and sexual dysfunction: a pharmacovigilance analysis
Mengting Shen; Luyao He; Pei Chen; Lei Zhang; Duan Zeng; Yan Li; Huafang Li
Sexual Medicine2026
In plain terms
Researchers looked at about 1.1 million reports of suspected side effects sent to the US medicines regulator from 2021 to 2023. For each of six common antidepressants, reports mentioning a sexual problem made up a bigger share of that drug's reports than across all medicines: for sertraline about 1 in 52 of its reports, compared with about 1 in 380 overall. The kinds of sexual problem reported differed between drugs: sertraline had the widest range, including genital numbness, and duloxetine the narrowest.
What it doesn’t show: It can't show how often these medicines cause sexual problems, or whether problems lasted after stopping: it counted voluntary reports, many side effects are never reported, and the study did not look at whether problems continued once the medicine was stopped.
Summary (paraphrased)
Using the FDA Adverse Event Reporting System (FAERS), the authors compared disproportionality signals for sexual-dysfunction adverse events across individual SSRIs and SNRIs (sertraline, citalopram/escitalopram, paroxetine, fluoxetine, venlafaxine, duloxetine), applying Bayesian shrinkage to stabilize estimates. All studied agents showed associations with sexual dysfunction reports, with distinct symptom profiles emerging by drug. The analysis demonstrates how spontaneous-reporting data can differentiate drug-specific sexual adverse-event patterns within the antidepressant class.
Evidence
Extracted from the full text; page numbers refer to the paper. The library’s own notes are labelled as such.
- Design
- Pharmacovigilance studyPharmacovigilance disproportionality (case/non-case) analysis of spontaneous adverse-event reports, comparing each drug with all other drugs in the database, with Bayesian shrinkage
- Setting
- US FDA Adverse Event Reporting System (FAERS), reports from Q1 2021 to Q4 2023, mainly from the USA, UK and Canada
- Population
- Spontaneous adverse-event reports naming fluoxetine, paroxetine, citalopram/escitalopram, sertraline, venlafaxine or duloxetine as the main suspect drug, after removing duplicates and reports lacking drug information. Fluvoxamine was excluded for too few reports.
- Size
- 1,144,969 · 1,144,969 valid reports (all drugs), 3,021 of them sexual dysfunction; the six drugs account for 41,793 reports, 541 of them sexual dysfunction (summed from p. 3). Units are reports, not people
- Exposure
- Each of the six drugs as the main suspect drug
- Compared with
- All other reports in FAERS (all other drugs), not other antidepressants
- Outcome
- Reports of sexual dysfunction, overall and by MedDRA preferred term (e.g. anorgasmia, erectile dysfunction, loss of libido, genital hypoaesthesia). A signal required IC025 above 0, "ROR" above 1 and more than 3 reports.
- Follow-up
- Not applicable (reports, not followed patients)
Key results
- Share of each drug's reports that were sexual dysfunction (percentages computed here), with the paper's "ROR" and IC025: sertraline 195 of 10,114 (1.9%; 7.19; 2.61), paroxetine 53 of 2,917 (1.8%; 6.53; 2.25), fluoxetine 73 of 5,558 (1.3%; 4.85; 1.89), citalopram/escitalopram 86 of 7,203 (1.2%; 4.43; 1.79), venlafaxine 83 of 7,655 (1.1%; 4.03; 1.65), duloxetine 51 of 8,346 (0.6%; 2.29; 0.73), against 3,021 of 1,144,969 (0.26%) for all reports. All six met the signal criteria · pp. 2–3, Table 1
- The values labelled ROR equal the shrunk observed-to-expected ratio in the formula on p. 2 (2 to the power of the IC), not a reporting odds ratio from the 2×2 table. Conventional RORs computed from the paper's counts are slightly higher, in the same order: sertraline 7.88, paroxetine 7.10, fluoxetine 5.13, citalopram/escitalopram 4.67, venlafaxine 4.23, duloxetine 2.35 · pp. 2–3, Table 1
- Breadth of symptom signals: sertraline widest (the text says 15 signals; Figure 1 shows 16 positive IC025 values, from 1.32 for orgasmic sensation decreased to 3.06 for loss of libido), citalopram/escitalopram 14 positive values in Figure 1, fluoxetine 8, venlafaxine 5, paroxetine 4, duloxetine 1 (sexual dysfunction, IC025 0.64) · p. 3, Fig. 1 (p. 4)
- Anorgasmia, erectile dysfunction and decreased libido gave signals for every drug except duloxetine; citalopram/escitalopram was the only drug with a female orgasmic disorder signal (IC025 0.57) · p. 3, Fig. 1 (p. 4)
- Genital hypoaesthesia (genital numbness) gave signals for sertraline (IC025 2.86) and citalopram/escitalopram (1.81) but not venlafaxine (−0.38); persistent genital arousal disorder gave a signal for sertraline (1.35) · Fig. 1 (p. 4)
- Of the 3,021 sexual-dysfunction reports, 26.02% concerned women and 67.56% men; 1,312 (43.43%) came from the USA, 434 (14.37%) from the UK and 367 (12.15%) from Canada. The text calls these shares "of the total" and "of all valid reports", but they match the 3,021 sexual-dysfunction reports, not the 1,144,969 · pp. 3–4
Limitations the authors note
- Voluntary reporting brings data-quality problems (drug mapping, missing data, duplicates) that may bias results
- With no denominator, true incidence cannot be calculated, and disproportionality cannot establish causation
- Possible protopathic bias, as sexual dysfunction may precede antidepressant use in some depressed patients
- Dose and severity of depression could not be assessed against sexual dysfunction
- Genetic differences in susceptibility were not accounted for
- Generalisability is limited by reporting biases and by reports coming mainly from the USA, Canada and the UK
Also worth weighing (library’s note)
- The analysis cannot tell whether problems began on the drug or persisted after stopping; the therapy-dates file (THER) was not used (p. 2), so it is not direct evidence on PSSD
- Each drug is compared with all other drugs in FAERS, not with other antidepressants; differences between the six drugs are not tested statistically
- The number of symptom signals per drug partly reflects how many reports each drug has (sertraline has the most sexual-dysfunction reports, 195)
- Table 1's upper limits (IC975) cannot be reproduced. The formula on p. 2 is printed with a minus sign, which would put them below the IC; the standard formula gives 1.53–3.03 against 2.81–4.08 reported. The lower limits (IC025), which define the signals, do reproduce
- The discussion describes duloxetine as "least likely to produce" sexual dysfunction (p. 5), which reads reporting ratios as risk
Funding: Shenzhen Medical and Health Three Project (SZSM202011014); Collaborative Innovation Center for Clinical and Translational Science, Ministry of Education & Shanghai (CCTS-202306, CCTS-202409PT); Clinical Research Plan of SHDC (SHDC2020CR2053B); Shanghai clinical research center for mental health (19MC1911100); Shanghai Sailing Program (23YF1438000); Shanghai Mental Health Center (2022024-FX-05). Interests: None declared
What it can support (library’s note): That sexual problems are reported disproportionately often for each of six common SSRIs and SNRIs, with the kinds of problem differing by drug. It cannot give rates, establish causation, or say whether problems persisted after stopping.
Why it’s in the corpus
Recent large-scale pharmacovigilance comparison of sexual-dysfunction signals across individual antidepressants, useful for drug-specific risk discussion in the corpus's epidemiology section. It concerns reported sexual side effects in general and does not separate problems during treatment from problems persisting after stopping (therapy-date data were not used), so it is not direct evidence on PSSD.
Related records
- Peer-reviewed paper
Citizen petition: Sexual side effects of SSRIs and SNRIs
David Healy · International Journal of Risk & Safety in Medicine
This is the published text of a citizen petition asking the US FDA to require boxed warnings and label changes for all SSRI and SNRI products regarding persistent sexual dysfunction after discontinuation. The petition marshals the history of regulatory…
LIT-0202018PSSDLiterature sweep · Oct 2026 - Peer-reviewed paper
Isotretinoin-related adverse sexual and reproductive outcomes: a real-world pharmacovigilance study of the FDA Adverse Event Reporting System (FAERS)
Cansu Altınöz Güney; Huriye Aybüke Koç · Cutaneous and Ocular Toxicology
Using FAERS data from 2004-2024 via the public dashboard and OpenVigil 2.1, the authors pulled isotretinoin-related adverse events and ran disproportionality analyses (PRR, ROR with 95% CIs, Evans' criteria) across four MedDRA system organ classes. From…
LIT-0392025PFSPSSDPRSDLiterature sweep · Oct 2026 - Peer-reviewed paper
Post-SSRI Sexual Dysfunction (PSSD): Biological Plausibility, Symptoms, Diagnosis, and Presumed Risk Factors
Peleg, L. C.; Rabinovitch, D.; Lavie, Y.; et al. · Sexual Medicine Reviews · 10:91-98
Reviews neurobiological hypotheses for PSSD and lists presumed risk factors: prior exposure to certain drugs, genetic predisposition, psychological/chemical stress reactions, and pre-existing conditions affecting neuroplasticity; proposes a future research…
PSSD-0022022PFSPSSDCore corpus