For cases and all controls, we examined the agreement between exposures during the first year of life assigned to birth residence and those assigned to LexisNexis addresses. Restricting to cases diagnosed at or after 1 year of age, we assessed the level of agreement between exposures assigned to birth residence and those assigned to diagnosis residence, for movers and non-movers combined, and movers by distance moved . These included: 1) continuous cumulative exposure estimates using Spearman correlation coefficients, since the exposure estimates were not normally distributed and, 2) dichotomous exposure indicators using Cohen’s kappa statistics, a robust agreement measure by taking into account the possibility of the agreement occurring by chance.Table 3.1 shows the child and maternal characteristics by residential mobility. Of 6,478 childhood cancer cases born in 1998-2011, 3,548 had a residential location at diagnosis at least 100 m away from address at birth. Cases diagnosed at an older age compared to those diagnosed within the first year of life were more likely to move between birth and diagnosis. Mothers of younger age, with lower education, a non-Hispanic ethnic background, those who used public health insurance,plastic pot or resided in non-metropolitan areas at delivery were more likely to move. Using 200m distance moved as an alternative cutoff, we classified 3,297 cases as movers accordingly; despite that, the predicting factors for residential mobility remained the same.
The information from LexisNexis suggested more than 20% of case and control mothers moved in child’s first year of life, with less than 4% moving multiple times . Using LexisNexis addresses, we identified similar predictors for residential mobility among cases and controls in their first year of life, and these patterns did not differ from those in comparing cases’ residential mobility from birth to diagnosis . Noticeably, for 42% of case mothers and 45% of control mothers LexisNexis provided only partial residential information at best during the child’s first year of life and these children were more likely to be born in earlier years , born to mothers of younger age, with lower education, having a Hispanic background or from Mexico, using public insurance type, or with a lower neighborhood SES. Based on LexisNexis, we calculated the cumulative proportions of mothers who moved over time for cases and controls . About 12% of the case mothers and 23% of the control mothers moved in the child’s first year of life. The proportions who moved increased with each year of the child’s age and the proportion for control mothers was slightly higher than that for case mothers at any time point. Within the first year of life, both control and case mothers seemed to move smoothly over time, i.e., no striking peaks appeared .With respect to distance moved, the median was 5.05 km among the cases who moved between birth and diagnosis. Overall, 30.1% of all cases moved less than 2 km, while more than half moved more than 4 km .
Cases with an older age at diagnosis, born to younger or Black mothers, or born in metropolitan areas were more likely to move further. Based on LexisNexis addresses, the median distance moved in child’s first year of life was 6.97 km for all cases , and 9.81 km for all controls, though the longest distance could be over 5,000 km, indicating an address out of the continental U.S. at one point. Under some circumstances, the distance moved was longer than 900 km even for some case mothers, because the case families resided outside of California after birth but came back to California before the cancer diagnosis. Exposures to specific carcinogens during the first years of life calculated for birth residence were compared with those calculated for diagnosis residence or LexisNexis addresses among cases diagnosed at or after age of 1, and birth residence was compared with LexisNexis addresses among all controls. Overall, the correlations between continuous exposure estimates within a 2km buffer of birth vs diagnosis residence were moderate to strong , and those between birth residence vs LexisNexis addresses were higher and did not differ by outcome status. When stratified by distance between birth and diagnosis residences, the correlations were very strong for those moved less than 2 km, moderate to strong for those moved between 2 km and 4 km, and relatively weak for those moved more than 4 km . The kappa statistics between dichotomous exposures were similar, also suggesting overall good agreement and decreasing agreement as the distance moved increased . This study aims to examine the extent to which using birth residence vs diagnosis address or reconstructed residential histories to estimate exposures to nearby agricultural pesticides in early childhood might lead to misclassification.
We investigated the maternal and child characteristics that may predict early life residential mobility and assessed patterns of mobility among childhood cancer cases and controls in California. We found that case’s age at diagnosis, maternal age at birth, maternal education, maternal race/ethnicity, payment type for prenatal care, and/or maternal urban/rural residence at birth were associated with both residential mobility and the distance moved between birth and diagnosis. The overall agreement between the first year of life exposures to agricultural pesticides 2-km surrounding the birth residence and diagnosis address or reconstructed LexisNexis residential history was moderate to strong , but level of misclassification might depend on the demographic characteristics and distance moved. Few studies have examined the patterns of residential mobility in early childhood to identify the potential for exposure misclassification . As with previous studies that focused on residential mobility during pregnancy, higher residential mobility in early childhood was associated with a range of maternal sociodemographic characteristics such as lower maternal age, lower maternal education, and lower individual or neighborhood level SES measures . We found cases born to Hispanic mothers were less likely to move; however, findings for maternal race/ethnicity vary across studies at different geographic locations, years, and population compositions . The likelihood of moving between birth and diagnosis of cancer increases with case’s age at diagnosis, suggesting that assessed at the birth residence may be less likely to represent the true exposures as children grow up. If demographic factors are potential confounders that affect outcome risks, exposure misclassification will likely be differential. Other factors proposed by previous studies that were linked to mobility during pregnancy but not available in the present study include marital status, smoking and alcohol use ; they may also be associated with residential mobility in early childhood. Using address information obtained from LexisNexis, we found similar predictors for mobility in cases and controls; however, we observed that those with complete address information in the first year of life from LexisNexis were quite different from those with missing or partially missing address information regarding maternal and child characteristics. Earlier validation studies suggested a match rate of 70-85% when comparing interview based residential history with reconstructed LexisNexis history, but subjects were mostly mid-aged or older and their residential mobility might be quite different from women at child-bearing age. For example, findings from the California Teachers Study suggested that women under the age of 40 years had a lower match rate . Regarding distances moved from birth to diagnosis, our study reported a median of 5.05 km among all cases,grow bag with the smallest median for cases diagnosed at less than 12 months and largest median for cases diagnosed at age 2+. The median distance moved between LexisNexis addresses in the first year of life was 6.97 km for cases age of 1+ years and 9.81 km for controls. Our findings are consistent with previous studies that focused on residential mobility during pregnancy that reported median distances <10 km. The degree of exposure misclassification is a function of the distance moved in relation to the spatial heterogeneity of the exposure. For given environmental exposures such as pesticides, spatial variation may depend on geographic areas or by the buffer size used. For example, estimated exposures to agricultural pesticides within 500 m of a residence will be more sensitive to the variation of exposures due to mobility compared to those using 2 km or even larger buffer radius. Similarly, other environmental epidemiological studies of child health outcomes that used a smaller buffer radius or a shorter distance to exposure sources such as traffic-related air pollution, electronic magnetic fields, or power lines might be more vulnerable to an inaccurate location.
The use of LexisNexis has a remarkable advantage in identifying the timing of moves, providing important information for future research that assesses time-sensitive exposures or involves outcomes with a susceptible period after a child’s birth. In our study mothers of case and control children were more likely to move in child’s first year of life, compared to the following years in early childhood. Control mothers seemed to have slightly higher mobility than the case mothers throughout the entire study period, likely because the case families were diagnosed within California and therefore their residences were California-based at least at birth and at diagnosis, while control families could completely move out of California but were still captured in the LexisNexis database with the nationwide search ability. Within the first year of life, case mothers moved more in the latter half of the first year but this could be random variation due to a relatively smaller sample size compared to the controls. In general, we had good agreement between exposures to agricultural pesticides within a 2-km buffer of birth residence and LexisNexis addresses for cases and controls in their first year oflife. The level of misclassification introduced by residential mobility may be acceptable if our study period of interest is the child’s first year of life, since only around 20% of cases and control mothers changed their residence during that period and it wasn’t differential with regard to disease status. For all cases who moved between birth and diagnosis, we have moderate to good agreement between exposures to agricultural pesticides within a 2-km buffer of birth residence and diagnosis residence. For those moved within a short distance , misclassification is minimal; yet a longer distance could be problematic with regards to exposure assessment, in particular when the buffer radius is smaller than the change in location. The misclassification will be differential if the factors associated with residential mobility are also risk factors for the disease. In assessing patterns of and examining factors that may predict residential mobility in early childhood, our strength is that we identified a large number of childhood cancer cases born and diagnosed in California through California birth certificates and the California cancer registry. Unlike smaller questionnaire-based studies, our results were unlikely to be influenced by selection bias introduced by participation or recall bias; though the analysis based on LexisNexis addresses could be subject to potential selection bias, the conclusions drawn from it were comparable to those derived by examining birth vs diagnosis residence in cases only. To our knowledge, only a couple of studies examined the post-partum residential mobility either in a statewide sample of California women or in enrolled childhood leukemia cases of a Northern California Study but both focused on factors associated with residential mobility and changes in neighborhood SES. Our study, on the other hand, for the first time calculated distance moved in early life of cases and examined the potential for exposure misclassification by distance moved. Though pregnancy period is the most critical period for most environmental exposures, first year of life exposures could also substantially influence young children’s health outcomes . Therefore, it is crucial to assess exposures not only during pregnancy but also in the early life of children. This study also has a few limitations. We were unable to obtain a secondary address from a registry for our population-based controls that is comparable to cases’ diagnosis residence, limiting our ability to examine the patterns of residential mobility parallel in cases and controls; however, the reconstructed LexisNexis residential histories allowed us to compare moving patterns in cases vs controls. Additionally, only the latitude and longitude of the diagnosis residence on the cancer reports were available, so we cannot directly compare the street number and names of birth and diagnosis residences to ascertain residential mobility.