Statistical analyses were conducted using SPSS (Version 24, IBM Corporation). Prevalence was reported as an exact point estimate and 95% adjusted Wald interval (Agresti and Coull 1998). The valid percent was reported if data were missing, determined by excluding cases with either no response or with responses reported as “unsure.” Differences between means for continuous variables were analyzed with a two-sided student's t-test or Welch's t-test (Welch 1947) as indicated by the results of Levene's test for equal variances (Levene 1960). One-sample z-tests with binomial p-values were used to compare prevalence estimates with the maximum population estimate based on literature review. Given low frequencies for some conditions, differences among dichotomous categorical variables were analyzed with a two-sided Fisher's exact test. Pearson's chi-squared test was used to compare frequencies for nominal variables with two or more categories. Correction for multiple comparisons was applied as indicated using the Bonferonni correction with α = 0.05 for the indicated number (n) of tests (Benjamini and Hochberg 1995). Continuous variables are reported as mean ± standard deviation with 95% confidence intervals in square brackets.
