|
Selevan et al. 348 (16): 1527, Figure 3 April 17, 2003 |
| Return to Article | Add to Personal Archive | PowerPoint Help |

Figure 3. Association between Age at Menarche and Blood Lead Concentrations.
Panel A shows the reported age at menarche for all girls. Delays in menarche were estimated with the use of PROC LIFEREG software (SAS). In Panel B, the age at menarche was modeled with the use of the same potential risk factors and confounding variables in the SUDAAN PROC SURVIVAL analysis. In Panel C, model-based estimates of delays in menarche were obtained by evaluating each model at blood lead concentrations ranging from 1 to 10 µg per deciliter, with the use of mean values for each group (shown in Panel B). Then delays were calculated by subtracting the median age for a blood lead concentration of 1 µg per deciliter from the median age for lead concentrations of 2 to 10 µg per deciliter. Since the PROC LIFEREG software cannot accommodate the complex sampling weights used in the third National Health and Nutrition Examination Survey and this analysis was not available in SUDAAN software, only the ages at menarche, and not the confidence intervals or significance levels, were estimated. To convert values for lead to micromoles per liter, multiply by 0.048.
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |