In the US, cancer is among the most common causes of death for children. [1] While environmental factors contribute to childhood cancer etiology, their contribution is not well understood. [2]

Because childhood cancer is rare, [1] environmental epidemiology studies on childhood cancer often employ case-control designs that estimate exposure from birth registry addresses. [2] Because the child’s residence may change over the exposure period, exposure misclassification is a major limitation in this study design. [3] To address this limitation, researchers have used commercially available databases containing residential addresses in attempts to better capture residential location over the exposure period. [4], [5], [6]

A previous study reported LexisNexis® Accurint® for Government data systems (LexisNexis) as the most complete and accurate of three residential data vendors evaluated. [7] However, several studies found completeness of LexisNexis records varies by race/ethnicity and participant’s age, [8], [9], [10], [11], [12], [13], [14] which could lead to selection and information bias in studies relying on LexisNexis records. Exclusion of participants because of non-random availability of information based on age, race, and ethnicity compromises the generalizability of study results: the study population may not accurately represent the population at risk. [3]

There is also potential for bias by case status in case control studies relying on residential databases. Information bias by case status could result in differential exposure misclassification, as well as differences in the case and control populations. [3] LexisNexis records for cancer patients (or parents of children with cancer) may be more likely complete and accurate than those for people without cancer because LexisNexis complies information through public records (e g court documents, government agencies, financial institutions, insurance carriers, and health care providers). Financial costs for adult cancer patients can lead to bankruptcy filings [15] and it is reasonable to suspect that the same may be true for families of children with cancer. Because of costly pediatric cancer treatment, families experience financial strain including reduction or loss of income [15] which may increase the likelihood of health care bills and bankruptcy, as well as application for public assistance and loans.

It is important to understand if reliance on residential databases introduces selection and information biases regarding case status in case-control studies, because of the potential for systematically biased and unreliable findings. Our objective is to evaluate the potential systematic bias regarding case status when using LexisNexis.