TABLE 3.
Summary of the β estimates (regression coefficients) from the meta-analysis of the association between BMI and telomere length as the outcome and relative telomere length (T/S ratio) as the independent variable1
Pooled sample3 | |||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
All (total pooled sample)2 | “Young” (18–60 y old) | “Middle”-aged (61–75 y old) | “Old” (>75 y old) | ||||||||||||||
Line | Ethnicity | n | Estimate | 95% CI | I 2, % | n | Estimate | 95% CI | I 2, % | n | Estimate | 95% CI | I 2, % | n | Estimate | 95% CI | I 2, % |
1 | Overall | 58 | −1.58* | −2.14, −1.01* | 32.7 | 55 | −2.58* | −3.92, −1.25* | 80.0* | 50 | −1.08* | −1.76, −0.39* | 0.0 | 29 | 0.20 | −1.40, 1.80 | 0.0 |
2 | White | 43 | −1.87* | −2.44, −1.31* | 8.1 | 40 | −2.80* | −4.77, −0.82* | 84.1* | 37 | −1.65* | −2.45, −0.86* | 0.0 | 21 | −0.28 | −2.29, 1.73 | 0.0 |
3 | African American | 2 | 5.66 | −6.60, 17.92 | 80.0* | 2 | 5.21 | −5.67, 16.08 | 68.7 | 2 | 0.08 | −6.20, 6.36 | 0.0 | 1 | −0.74 | −12.62, 11.14 | |
4 | Hispanic | 3 | 2.53 | −5.18, 10.25 | 17.7 | 3 | −0.42 | −4.19, 3.34 | 0.0 | 3 | 2.31 | −2.35, 6.97 | 0.0 | 2 | 27.29 | −40.32, 94.61 | 77.7* |
5 | Asian | 3 | −1.11 | −4.23, 2.02 | 60.4 | 3 | −4.50* | −5.75, −3.25* | 0.0 | 2 | 2.18 | −2.90, 7.27 | 0.0 | 0 | |||
6 | Native American | 1 | −2.64* | −3.60, −1.68* | 1 | −4.14* | −5.28, −3.00* | 1 | 2.23 | −1.00, 5.46 | 1 | 4.68 | −2.35, 11.71 |
1The unit of the estimates and 95% CI of the T/S ratio is 10−3. A random-effects model was used. Statistical heterogeneity was estimated by I2 statistics for each of the meta-analyses. *P < 0.05 or I2 >50%.“n” indicates number of studies. T/S ratio, telomere to single-copy gene ratio.
2Adjusted for age and sex.
3Adjusted for sex.