Predictors of chronic pelvic pain in an urban population of women with symptoms and signs of pelvic inflammatory disease

OBJECTIVE: The objective of this study was to assess the risk profile for chronic pelvic pain (CPP) after pelvic inflammatory disease (PID). STUDY: Multivariate logistic regression was used to assess risk factors for CPP in a longitudinal study of 780 predominately black, urban women with clinically suspected PID: complaints of acute pain (<30 days); a clinical finding of pelvic tenderness; and leukorrhea, mucopurulent cervicitis, or untreated gonococcal or chlamydial cervicitis. CPP was defined as pain reported at >or=2 consecutive interviews conducted every 3 to 4 months for 2 to 5 years. RESULTS: Nonblack race (odds ratio OR], 2.17; 95% confidence interval CI], 1.31-3.58), being married (OR, 2.06; 95% CI, 1.02-4.18), a low SF-36 mental health composite score (OR, 2.71; 95% CI, 1.69-4.34), >or=2 prior PID episodes (OR, 2.84; 95% CI, 1.07-7.54), and smoking (OR, 1.65; 95% CI, 1.01-2.71) independently predicted CPP. Histologic endometritis or evidence of endometrial Neisseria gonorrhoeae or Chlamydia trachomatis infection was negatively associated with CPP (OR, 0.69; 95% CI, 0.44-1.10). CONCLUSIONS: A range of demographic, clinical, historical, and behavioral factors predict CPP after PID.