Table I.
Sperm concentration and total sperm count in first and last years of meta-regression analysis with percentage change and slope per year, for all men and by fertility and geographic groupsa.
| Category | N (estimates) | First year | First year SC (million/ml) | Last year | Last year SC (million/ml) | Percentage change/year | Slope (95% CI), million/ml/year |
|---|---|---|---|---|---|---|---|
| All men | 244 | 1973 | 92.8 | 2011 | 66.4 | −0.75 | −0.70 (−0.72 to −0.69) |
| Unselected Western | 110 | 1973 | 99.0 | 2011 | 47.1 | −1.40 | −1.38 (−2.02 to −0.74) |
| Fertile Western | 65 | 1977 | 83.8 | 2009 | 62.0 | −0.81 | −0.68 (−1.31 to −0.05) |
| Unselected Other | 30 | 1986 | 72.7 | 2010 | 62.6 | −0.58 | −0.42 (−1.24 to 0.40) |
| Fertile Other | 39 | 1978 | 66.4 | 2011 | 75.7 | 0.42 | 0.28 (−0.44 to 1.00) |
| Category | N (estimates) | First Year | First year TSC (million) | Last year | Last year TSC (million) | Percentage change/year | Slope (95% CI), million/year |
| All men | 244 | 1973 | 295.7 | 2011 | 212.0 | −0.75 | −2.23 (−2.31 to −2.16) |
| Unselected Western | 110 | 1973 | 337.5 | 2011 | 137.5 | −1.58 | −5.33 (−7.56 to −3.11) |
| Fertile Western | 65 | 1977 | 277.4 | 2009 | 209.5 | −0.76 | −2.12 (−4.31 to 0.07) |
| Unselected Other | 30 | 1986 | 212.4 | 2010 | 167.3 | −0.88 | −1.88 (−4.77 to 1.01) |
| Fertile Other | 39 | 1978 | 189.2 | 2011 | 233.2 | 0.70 | 1.33 (−1.20 to 3.86) |
aFor all men: simple linear regression weighted by sample size. For all other categories: Meta-regression model weighted by sperm concentration (SC) SE, adjusted for fertility group, time x fertility group interaction, geographic group, time × geographic group interaction, age, abstinence time, semen collection method reported, counting method reported, having more than one sample per men, indicators for study selection of population and exclusion criteria (some vasectomy candidates, some semen donor candidates, exclusion of men with chronic diseases, exclusion by other reasons not related to fertility, selection by occupation not related to fertility), whether year of collection was estimated, whether arithmetic mean of SC was estimated, whether SE of SC was estimated and indicator variable to denote studies with more than one estimate. Total sperm count (TSC) meta-regression models weighted by TSC SE, adjusted for similar covariates and method used to assess semen volume.