Selective nonresponse bias in population-based survey estimates of drug use behaviors in the United States
- 369 Downloads
There is a trend of decreasing response rates in population surveys, and selective nonresponse represents a major source of potential bias in population-based survey estimates of drug use behaviors, especially estimates based on longitudinal designs.
This study compared baseline substance use behaviors among initial respondents who did respond (n = 34,653) and did not respond (n = 8440) to a 3-year follow-up interview in a prospective study of the general U.S. adult population. Differences in nonresponse rates were assessed as a function of past-year drug use behaviors both before and after adjustment for socio-demographic differences potentially associated with these behaviors, and the effects of interactions of the socio-demographic characteristics with the drug use behaviors were assessed in multivariate logistic regression models for response at the 3-year follow-up.
Weighted and unweighted nonresponse rates varied between alcohol users and users of other drugs such as cocaine and marijuana, with rates of nonresponse being higher in the latter drug categories. There were also significant differences in nonresponse rates as a function of frequency of use and demographics. More specifically, being married tends to reduce the probability of non-response, while older age, being male, being Asian or Hispanic, and having lower education all substantially increase the probability of nonresponse at Wave 2, even after controlling for relevant covariates.
This study provides the substance abuse field with a methodology that users of longitudinal data can apply to test the sensitivity of their inferences to assumptions about attrition patterns.
KeywordsNonresponse bias Longitudinal Population-based Survey estimates Drug use
The development of this manuscript was supported by research grants R01DA036541 and R01DA031160 from the National Institute on Drug Abuse, National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors would like to thank the respondents for their participation in the study and the anonymous reviewers and editorial staff for their helpful comments to a previous version of this article.
Compliance with ethical standards
Conflict of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest.
- 1.Substance Abuse and Mental Health Services Administration (2006) Trends in substance abuse treatment admissions: 1993 and 2003. The DASIS report. Office of applied studies. http://oas.samhsa.gov/2k6/TXtrends/TXtrends.cfm. Accessed 26 January 2015
- 2.Substance Abuse and Mental Health Services Administration (2012) Results from the 2011 national survey on drug use and health: summary of national findings, NSDUH series H-44, HHS publication no. (SMA) 12-4713. Substance Abuse and Mental Health Services Administration, RockvilleGoogle Scholar
- 9.Bachman JG, O’Malley PM, Schulenberg JE, Johnston LD, Bryant AL, Merline AC (2002) The decline of substance use in young adulthood: changes in social activities, roles, and beliefs. Lawrence Erlbaum Associates, MahwahGoogle Scholar
- 10.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE, Miech RA (2014) Monitoring the future national survey results on drug use, 1975–2013. Volume II: college students and adults ages 19–55. University of Michigan Institute for Social Research, Ann ArborGoogle Scholar
- 11.Substance Abuse and Mental Health Services Administration (2014) Results from the 2013 national survey on drug use and health: summary of national findings, NSDUH series H-48, HHS publication no. (SMA) 14–4863. Substance Abuse and Mental Health Services Administration, RockvilleGoogle Scholar
- 12.Grant BF, Kaplan K, Shepard K, Moore T (2003) Source and accuracy statement for wave 1 of the national epidemiologic survey on alcohol and related conditions (NESARC). National Institute on Alcohol Abuse and Alcoholism, BethesdaGoogle Scholar
- 13.Grant BF, Kaplan KD (2005) Source and accuracy statement for the wave 2 national epidemiologic survey on alcohol and related conditions (NESARC). National Institute on Alcohol Abuse and Alcoholism, RockvilleGoogle Scholar
- 15.Grant BF, Dawson DA, Stinson FS, Chou PS, Kay W, Pickering R (2003) The alcohol use disorder and associated disabilities interview schedule-IV (AUDADIS-IV): reliability of alcohol consumption, tobacco use, family history of depression and psychiatric diagnostic modules in a general population sample. Drug Alcohol Depend 71:7–16CrossRefPubMedGoogle Scholar
- 26.Little RJA, Vartivarian S (2005) Does weighting for nonresponse increase the variance of survey means? Surv Methodol 31:161–168Google Scholar
- 29.Lepkowski JM, Couper MP (2002) Nonresponse in the second wave of longitudinal household surveys. In: Groves et al. (ed) From survey nonresponse, pp 259–271. Wiley, New YorkGoogle Scholar
- 30.Kott PS (2012) Why one should incorporate the design weights when adjusting for unit nonresponse using response homogeneity groups. Surv Methodol 38:95–99Google Scholar
- 31.West BT (2009) A simulation study of alternative weighting class adjustments for nonresponse when estimating a population mean from complex sample survey data. In: Proceedings of survey research methods section, 2009 joint statistical meetings, pp 4920–4933Google Scholar
- 32.Johnston LD, O’Malley PM, Bachman JG, Schulenberg JE, Miech RA (2014) Monitoring the future national survey results on drug use, 1975–2013. Volume I: secondary school students. University of Michigan Institute for Social Research, Ann ArborGoogle Scholar