Abstract Body: Introduction: Endometriosis is a chronic inflammatory condition associated with pelvic pain and dysmenorrhea. Nonsteroidal anti-inflammatory drugs (NSAIDs) are a first-line analgesic therapy, yet real-world NSAID use patterns by endometriosis status remain poorly characterized in surgical cohorts.
Hypothesis: We assessed the hypothesis that endometriosis-positive patients demonstrate higher rates of analgesic NSAID use compared with endometriosis-negative controls after adjustment for age and body mass index (BMI).
Methods: We performed a cross-sectional analysis of 648 subjects (344 endometriosis-positive, 304 endometriosis-negative) from a surgical registry. NSAID exposure was defined at the subject level (any use), collapsing multiple medication entries per subject. The primary definition included analgesic NSAIDs (ibuprofen, naproxen, diclofenac, meloxicam, celecoxib, indomethacin, ketorolac) and excluded aspirin due to its predominant use for cardiovascular prophylaxis. A sensitivity analysis included aspirin. Group differences were assessed using Fisher’s exact test. Logistic regression estimated adjusted odds ratios (ORs) controlling for age and BMI. Complete-case analysis was performed (n=492); missingness was limited to age and BMI and assumed non-differential.
Results: NSAID use occurred in 23.3% of endometriosis-positive patients (80/344) and 17.4% of endometriosis-negative patients (53/304), an absolute difference of 5.9 percentage points (crude OR 1.44; p=0.079). After adjustment, endometriosis was associated with higher odds of NSAID use (OR 1.45, 95% CI 0.93–2.25; p=0.101). Sensitivity analysis including aspirin yielded consistent estimates (24.1% vs 18.4%; adjusted OR 1.40, 95% CI 0.90–2.16).
Conclusions: Endometriosis-positive patients demonstrated higher NSAID use compared with controls, with a consistent effect estimate across analytic approaches. Although not statistically significant, the magnitude and stability of the association suggest a modest difference that may be underpowered in this sample. In conclusion, larger prospective studies with prescription-level medication data are needed to confirm and extend these findings.