TABLE 3.
Adjusted difference of semen quality parameters per 5 cm increase in male waist circumference within categories of BMI1
| BMI, kg/m2 | P value for interaction | ||
|---|---|---|---|
| Semen quality parameters | 18.5 ≤ BMI < 25 | ≥25 | |
| Number of semen samples | 229 | 442 | |
| Number of males | 89 | 180 | |
| Sperm concentration (%) | — | — | 0.24 |
| Model 1 | −13.06 (−27.12 to 1.01) | −2.30 (−5.79 to 1.19) | |
| Model 2 | −24.76 (−41.20 to −8.33) | −4.16 (−8.06 to −0.26) | |
| Model 3 | −22.01 (−40.18 to -3.85) | −3.54 (−7.93 to 0.85) | |
| Total sperm count (%) | — | — | 0.16 |
| Model 1 | −11.29 (−23.84 to 1.27) | −1.65 (−5.96 to 2.66) | |
| Model 2 | −23.12 (−38.77 to -7.47) | −2.45 (−7.33 to 2.43) | |
| Model 3 | −22.05 (−37.04 to -7.06) | −0.65 (−5.69 to 4.38) | |
| Sperm motility (% motile) | 0.47 | ||
| Model 1 | −2.41 (−5.48 to 0.67) | −0.43 (−1.43 to 0.57) | |
| Model 2 | −4.39 (−8.12 to −0.66) | −0.72 (−1.96 to 0.51) | |
| Model 3 | −5.15 (−8.96 to −1.34) | −0.58 (−2.02 to 0.86) | |
| Progressive motility (%)2 | — | — | 0.61 |
| Model 1 | −1.42 (−3.48 to 0.64) | −0.27 (−1.05 to 0.51) | |
| Model 2 | −2.21 (−4.77 to 0.35) | −0.48 (−1.36 to 0.40) | |
| Model 3 | −3.09 (−5.68 to −0.5) | −0.43 (−1.43 to 0.57) | |
| Sperm morphology (% normal)3 | — | — | 0.85 |
| Model 1 | −0.30 (−0.73 to 0.14) | −0.03 (−0.21 to 0.15) | |
| Model 2 | −0.62 (−1.15 to −0.09) | −0.05 (−0.24 to 0.15) | |
| Model 3 | −0.74 (−1.32 to −0.16) | −0.02 (−0.25 to 0.22) | |
Sperm concentration and total count data were log-transformed to achieve normality. Thus, estimates represent percentage differences for total count and concentration and absolute differences for all other semen parameters. All analyses were conducted using cluster-weighted generalized estimating equations. Model 1 was adjusted for age (continuous), race (White, non-White), education (high school or some college, college, graduate), smoking status (ever, never), history of varicocele (yes, no) and abstinence time (<2 days, 2–3 days, 3–4 days, and ≥4 days). Model 2 was adjusted for the variables in Model 1 plus BMI (continuous). Model 3 was adjusted for the variables in Model 2 plus height (continuous).
Contains data from 658 samples because an evaluation on progressive motility was not performed for all samples.
Contains data from 639 samples because a morphologic evaluation was not performed for all samples.