We utilized nine years of data from the NSDUH (2003-2011); each year of the NSDUH survey employs a multistage probability sampling design to recruit a nationally-representative sample of the United States civilian, non-institutionalized population aged 12 and older. Since 1999, most questions in the NSDUH interview are administered as an audio computer-assisted self-interview to provide a private mode for responding to sensitive questions; for other less sensitive items computer-assisted personal interviewing is utilized. The design is state-based with a within-state independent, multistage area probability sampling, which allows estimating generalizable state-level prevalence rates. For each of the eight most populous states (e.g., California) approximately 3,600 respondents are interviewed annually; for each of the remaining 42 states and the District of Columbia, approximately 900 respondents are interviewed per state, per year. The design oversamples youths and young adults such that each state's sample is approximately equally divided into those 12-17 years, 18-25 years and 26 years and older. For Colorado considering year pairings of 2003-4, 2005-6, 2007-8, 2009-10 and 2010-11, sample sizes for 12-17 year olds, 18-25 year olds and those 26
