Kruskal-Wallis was used to test percentage of plants in flower using R statistical software because these data were not normally distributed.
We used remote sensing derived land cover data to estimate changes in forest area in this study because these data were available consistently and contiguously over the conterminous US over the time period of study.
Because these data were missing due to systematic causes, and thus missing not at random (MNAR) [45], no methods for imputation of missing data at the visit level were applied in the primary analyses.
Nonparametric statistics were used with both behavioral and electrophysiological data, because these data were not normally distributed.
Because these data were modeled, we did not impose an LOD on the estimated levels.
Urine output criteria were not used in diagnosis or staging because these data were not available.
Hemiplegia and peripheral vascular disease were omitted from the index because these data were not collected.
Because these data were not available for the year 2008, we relied on data for 2007.
We were unable to consider infertility history, because these data were not publicly available.
†Denominator is 47 because these data were unavailable for 4 patients.
Because these data were non-normally distributed, we plotted medians and 95% confidence intervals.
