This revised application, first submitted on 107/01/04 (R01 HL81627-01A1), which proposes to investigate the contribution of adverse early life socioeconomic status (SES) to the development of chronic diseases in adulthood, is a current area of interest in epidemiologic research. Early life SES is often assessed at midlife using father's occupation but little has been done to evaluate the accuracy of these proxy approaches. Further, early life SES is sometimes ascertained after a cohort's inception, which creates the potential for biased findings since loss to follow-up and mortality tend to be greater among those with lower SES. We will address these issues in the context of the Life Course SES, Social Context, and Cardiovascular Disease (LC-SES) Study, which collected early life SES from survivors of a cohort of middle-aged and older participants about 15 years after baseline.
The first aim of the study is to acquire early life SES data from historical records (birth and death certificates, social security records and individual census records) for approximately 2800 decedents, many of whom died prior to participating in the LC-SES Study.
The second aim i s to assess the comparability of early life SES obtained from historical records to retrospectively recalled early life SES in the subset of decedents who died after participating in the LC-SES Study. This data will be then used in sensitivity analyses to assess the impact, if any, of recall error. The third study aim is to compare methods for addressing missing early life SES data (complete case analysis, multiple imputation analysis, and historical record analysis). The comparison of methodologies will be used to assess the impact of survivorship bias in studies of the association between SES across the life course on CVD-related health conditions in mid to later adulthood. In summary, the proposed study seeks to provide validation data on retrospectively obtained early life SES and to provide estimates of the potential magnitude of recall error and survivorship bias;both would be valuable contributions to an understudied area in life course research.
| Bang, Heejung; Chiu, Ya-Lin; Kaufman, Jay S et al. (2013) Bias Correction Methods for Misclassified Covariates in the Cox Model: comparison offive correction methods by simulation and data analysis. J Stat Theory Pract 7:381-400 |
| Patel, Mehul D; Rose, Kathryn M; Owens, Cindy R et al. (2012) Performance of automated and manual coding systems for occupational data: a case study of historical records. Am J Ind Med 55:228-31 |
