A Method to Estimate Off-Schedule Observations in a Longitudinal Study

详细信息    查看全文

• 作者：Katherine W. Reeves PhD ; MPH^a ; ^b ; ^{kwreeves@schoolph.umass.edu} ; Roslyn A. Stone PhD^c ; Francesmary Modugno PhD ; MPH^b ; Roberta B. Ness MD ; MPH^b ; Victor G. Vogel PhD ; MHS^d ; Joel L. Weissfeld MD ; MPH^b ; ^e ; Laurel A. Habel PhD^f ; Marike Vuga PhD^b ; Jane A. Cauley DrPH ; MPH^b
• 关键词：Epidemiologic Methods ; Linear Interpolation ; Mammography ; Missing Data ; Multiple Imputation
• 刊名：Annals of Epidemiology
• 出版年：2011
• 出版时间：April 2011
• 年：2011
• 卷：21
• 期：4
• 页码：297-303
• 全文大小：299 K

文摘

Purpose

Data in epidemiological studies sometimes are collected off-schedule from planned study visits. In an ancillary study to the Study of Women’s Health Across the Nation (SWAN), longitudinal breast density data were collected retrospectively from mammograms that were not acquired at the study visits. We propose a method to estimate the off-schedule breast density measurements at the time of study visits.

Methods

This method uses local linear interpolation, with multiply imputed error terms drawn from assumed subject-specific normal distributions based on the within-subject standard deviations of mammographic density measurements. We evaluate the validity and implications of this approach.

Results

Coefficients of random intercept models used to assess the association between annual changes in body mass index and dense breast area estimated with this approach (β = −0.17, p = .46) differed from those obtained when each mammogram was matched to the nearest study visit (β = −0.30, p = .04). The proposed estimation approach had a small average prediction error (0.11 cm²).

Conclusions

Because matching does not incorporate breast density changes over time, our local linear interpolation with multiple imputation approach may provide more accurate results. The proposed approach is applicable to other epidemiologic studies with off-schedule data in which the missing variable changes linearly over relatively short periods of time.