Differences in the prevalence rates of past-year DSM-IV and DSM-5 AUD diagnoses were tested in the full sample, using the SAS McNemar test statistic for differences of proportions in paired data, (http://support.sas.com/documentation/cdl/en/procstat/63104/HTML/default/viewer.htm#procstat_freq_sect008.htm). This statistic accounts for the inherent positive correlation of each individual’s DSM-IV and corresponding DSM-5 diagnoses resulting from the many common symptom item indicators shared by the two sets of criteria. All other statistical analyses employed SUDAAN software (Research Triangle Institute, 2008) to adjust variance estimates for the complex, multi-stage sample design of the NESARC.
