Does the potential for selection bias in semen quality studies depend on study design? Experience from a study conducted within an infertility clinic.

Background: The low participation rates in human semen quality studies raises concern for the potential of differential participation based on semen quality (or a surrogate). To explore the potential for differential participation, we compared semen analysis results from study subjects with those of non-study subjects.

Methods: We obtained semen analysis results from 235 study subjects and retrospectively obtained results from a subset of 235 infertility clinic patients that were not study subjects but met the same eligibility criteria. The study was conducted at the Massachusetts General Hospital Infertility Clinic. All semen samples (study subjects and non-study subjects) were analysed for sperm concentration and motility by computer-aided semen analysis (CASA), and morphology was assessed using strict criteria. Semen analysis parameters for the non-study subjects were compared with the semen analysis results from study subjects.

Results: For all semen characteristics (sperm concentration, total sperm count, sperm motility and morphology), there were only marginal (non-significant) differences between study subjects and non-study subjects.

Conclusions: Among men from an infertility clinic, we found no strong evidence of differential participation based on semen quality. This is reassuring since the potential for selection bias is of concern in semen quality studies. However, the potential for selection bias in other study designs remains unclear.