Diagnoses and medications associated with delayed ejaculation

Abstract

Background: Delayed ejaculation (DE) is a disorder that can cause significant distress for sexually active men. The etiology of DE is largely idiopathic, with even less being known about clinical factors associated with the condition.

Aim: We sought to use data mining techniques to examine a broad group of health conditions and pharmaceutical treatments to identify factors associated with DE.

Methods: Using an insurance claims database, we evaluated all men with a diagnosis of DE and matched them to a cohort (1:1) of men with other male sexual disorders of urologic origin (ie, erectile dysfunction [ED] and Peyronie's disease [PD]). Given the low prevalence of DE, we incorporated the random forest approach for classification of DE vs controls, with a plethora of predictors and cross-validation with the least absolute shrinkage and selection operator (LASSO). We used both a high-performance generalized linear model and a multivariate logistic model. The area under the curve was reported to demonstrate classifier performance, and odds ratios were used to indicate risks of each predictor. We also evaluated for differences in the prevalence of conditions in DE by race/ethnicity.

Outcomes: Clinical factors (ie, diagnoses and medications) associated with DE were identified.

Results: In total, 11 602 men with DE were matched to a cohort of men with PD and ED. We focused on the 20 factors with the strongest association with DE across all models. The factors demonstrating positive associations with DE compared to other disorders of male sexual dysfunction (ie, ED and PD) included male infertility, testicular dysfunction, anxiety, disorders of lipid metabolism, alpha adrenergic blocker use, anemia, antidepressant use, and psychoses such as schizophrenia or schizoaffective disorder. In addition, the prevalence of several conditions varied by race/ethnicity. For example, male infertility was present in 5% of Asian men compared to <2% of men of other races.

Clinical implications: Several medical conditions and pharmacologic treatments are associated with DE, findings that may provide insight into the etiology of DE and offer treatment options.

Strengths and limitations: This study is to our knowledge the first to use using data mining techniques to investigate the association between medical conditions/pharmacologic agents and the development of subsequent DE. The generalizability of our findings is limited given that all men were commercially insured.

Conclusion: DE is associated with multiple medical conditions, a finding that may help identify the etiology for this disorder.

Keywords: Peyronie’s disease; data mining; ejaculation; erectile dysfunction.