Validation of self-reported proximity to agricultural crops in a case–control study of neural tube defects

Abstract

Self-reported perinatal exposures to chemicals or pollutant sources in case-control studies of birth defects may be inaccurate due to misreporting among mothers. In a case-control study of neural tube defects delivered in California in 1987-1988, mothers of case and control infants were asked whether they lived within 0.25 mile (400 m) of agricultural crops. Responses were compared against a gold standard derived from historical agricultural land-use survey maps. The odds ratio for self-reported proximity to any crops (1.62, 95% confidence interval: 1.08, 2.43) appeared to be positively biased compared with the estimate for map-based proximity (1.17, 95% confidence interval: 0.79, 1.71). This pattern was also observed for proximity to specific crops such as nonpermanent and orchard crops. For vineyards, however, we observed an increased risk associated with map-based proximity (odds ratio ¼ 2.45, 95% confidence interval: 1.08, 5.58) but not with self-reported proximity (1.09, 95% confidence interval: 0.51, 2.34). The sensitivity of self-reported proximity to any crops was greater for case (65.7%) than control mothers (50.0%) while specificity was about the same for case and control mothers (87.5 vs. 89.3%), suggesting that control mothers under-reported proximity to crops. Differential reporting was also observed between geographic regions, urban and rural residents, and across levels of maternal employment and education. These results suggest differential reporting between case and control mothers as well as an influence from maternal demographic characteristics on reporting accuracy.

Maternal residential pesticide use and risk of childhood leukemia in Costa Rica

“…Nevertheless, non‐differential misclassification would have biased our effect estimates towards the null. Third, we cannot rule out recall bias, particularly for exposure to pesticide applications on farms or businesses near the home,47 as our analyses relied on maternally‐reported pesticide use. However, we have no evidence to believe that mothers’ report would have varied by timing or amount of exposure or by child sex.…”

Using satellite derived land cover information for a multi-temporal study of self-reported recall of proximity to farmland

“…In addition, in order to compare our results directly with the validation analysis of Rull et al (2006), the five categories were collapsed into two categories F r1/4 mile from farmland and 41/4 mile from farmland. Studies have shown pesticide drift from 300 m to over 1 km depending on the type of application and environmental conditions.…”

“…Studies have shown pesticide drift from 300 m to over 1 km depending on the type of application and environmental conditions. The 1/4 mile threshold was chosen because this was the closest intermediate distance presented to our participants for the range of drift from pesticides, and this approximate distance has also been used in other studies (Ward et al, 2000;Rull et al, 2006). Sensitivity, specificity, and percentage matching were calculated and compared between 1978 and 2001, and between cases and controls using Fisher's exact test and twotailed P-values.…”

“…Our findings of case-control differences in the two category analyses raise some concerns about the use of self-reported proximity to farmland in exposure assessments. Rull et al (2006) conducted a similar study in California (although it was atemporal) and observed some differences within strata of case and control mothers. Sensitivity was poor and tended to be higher for cases; conversely, specificity was high and similar in magnitude among cases and controls.…”

“…There has been only one validation study that compared recall of proximity to agricultural crops up to 5 years prior to interview. Land use survey maps served as the ''gold standard'' to which participant responses were compared (Rull et al, 2006). In that study of neural tube defects, differences in recall just after conception were identified between cases and controls, indicating potential recall bias.…”