Data Source and Sample. Data were from the second wave of the India Human Development Survey (IHDS-II), conducted in 2011-2012. The IHDS was designed to be a nationally representative survey, comprised of thirty-three states and union territories of India except for small populations living in island states of Andaman Nicobar and Lakshadweep (Xxxxx & Xxxxxxxx, 2015). It was carried out by the University of Maryland and National Council of Applied Economic Research. Adult respondents were interviewed regarding health, education, employment, economic status, marriage, fertility, social capital, and gender relations (Xxxxx & Xxxxxxxx, 2015) (Xxxxx et al., 2010). The first wave of the survey was conducted in 2004-2005 with a sample of 41,554 households from 1503 villages and 971 urban neighborhoods in 383 of the 602 districts in India (Xxxxx & Xxxxxxxx). Approximately 83% of those interviewed in IHDS-I were resurveyed for IHDS-II, which consisted of 42,152 households and 204,568 individuals. Up to two ever-married women ages 15-49 were eligible for interview; a total of 39,523 meeting these criteria were interviewed in IHDS II (Xxxxx & Xxxxxxxx, 2015). Among mothers in the sample of ever-married women ages 15-49 years, data for 19,489 children were recorded. Of these, 14,458 were 0-5 years old and were eligible to provide data on growth status, the primary outcome of this study. Therefore, we began by creating a dataset of children ages 0 to 5 years with anthropometric data. Data from the women’s questionnaire, household questionnaire, and child anthropometry module were compiled and linked to each child aged 0 to 5. Child, maternal, and household characteristics were merged with the child- level anthropometric data using a combination of household, child, and maternal identification variables to create an analytic dataset. We excluded from analysis children with missing data for covariates of interest (see Supplemental Figure 2). A total of 26% were excluded due to missing anthropometry, and another 10% were excluded due to missing demographic (n= 12) or parental variables (n= 1429). The final analytic sample size was 9202 children ages 0-5 years with complete covariates of interest. The children in the analytic sample came from 7,017 households (average of ~1.3 children per household).
Data Source and Sample. Data utilized for this study was acquired from The National Survey on Drug Use and Health (NSDUH), which is funded and administered by the Substance Abuse and Mental Health Services Administration, an agency of the United States Department of Health and Human Services (DHHS).[4] NSDUH provides a wide range of statistical information on the use of drugs, alcohol, and tobacco by the U.S. population aged 12 or older as well as data regarding the prevalence of SUDs and SUD treatment utilization.[4] NSDUH does not collect survey data from homeless persons who do not reside in a shelter, military personnel on active duty, and person who are institutionalized in facilities such as correctional facilities or hospitals.[4] Data from NSDUH has been collected annually since 1971 and is representative at both the national and state level.[4] For the purposes of this study, 5 years of NSDUH data (2008-2012) were pooled together for the analysis. I restricted my sample to only include adults over the age of 18 years who were classified with having either illicit drug or alcohol abuse or dependence in the past year (N= 26,822). After excluding persons with missing data for any of the variables included in my models, the final sample size was 26,287 adults.
