Estimating the Prevalence of Substance Abuse with Social Indicators
McRae, James A., Jr., Beebe, Timothy J., Harrison, Patricia A., Journal of Drug Issues
Governments are increasingly interested in estimating the prevalence of substance abuse with social indicators, largely because of the high cost of estimating prevalence with surveys of random samples of the population. With both the individual and county as the unit, we regress measures of the use of alcohol, marijuana, and other drugs on social indicators that fall into three categories: demographics, measures of social disorganization, and measures more directly related to the use of substances. The measures of explained variance are fairly low, but even more troubling is that the effects of several social indicators are in the "Wrong" direction. Reliance on social indicator data to supplant survey estimates of the prevalence of substance abuse requires further validation, attention to sources of bias in the indicator data, and replication of the models over time.
Collection of information on substance use is vitally important for planning prevention and treatment, allocation of resources, and evaluation of programs. Surveys of random samples of the population provide excellent means of estimating the prevalence of use of substances except for those which tend to be very rare, negatively sanctioned, or used by those who are difficult to sample. Estimating the prevalence of alcohol use with sample surveys is quite successful; estimating the prevalence of heroin use with sample surveys is less successful (Simeone, Frank, & Aryan, 1993). But even for alcohol, sample surveys have the disadvantage of being fairly expensive, especially if the sample must be large enough to allow estimates for small areas such as counties. Concern over costs has led policy makers to emphasize the development of methods of estimating the prevalence of substance abuse which would be substantially cheaper than surveys (Substance Abuse and Mental Health Services Administration, 1995, 1996).
One such effort, the social indicators approach, involves using data collected by various governmental agencies for diverse purposes to estimate indirectly the prevalence of substance abuse. The variables employed might be demographic factors (e.g., sex and age distributions), indicators of social disorganization (e.g., poverty and school dropout rates), or more direct indicators of substance use (e.g., drug and DWI arrest rates). These data are inexpensive to compile since they are routinely collected. The use of social indicators avoids the problem of basing rates of prevalence on small survey samples, since the indicators are based on the population within a specific geographic area; the denominator for each rate is the number of people at risk of experiencing the event, so the problem of inferring from small samples to the entire population is moot.
While there are several salutary aspects of the social indicator approach, it does have its potential limitations. In comparing social indicators across specific geographic areas, comparability of measures and procedures is crucial in order to ensure that differences reflect actual variation rather than methodological artifacts. Because the measures comprising social indicators are typically collected for purposes other than estimating levels of substance abuse, the comparability of measures in different geographic areas is unknown. In addition, local conditions such as population shifts, natural disasters, tourism patterns, and law enforcement resources and priorities can significantly affect the indicators within a given geographic area. Because there may be numerous and competing explanations for differences between geographic areas and even within a geographic area over time, more direct and systematic methods, such as surveys, may be needed. Joseph and Hollett (1993, p. 813) note that without "knowledge of local social geography, applications of indicators methodologies run the risk of being `black boxes."'
Systematic investigations assessing the concordance of survey and social indicator data in estimating the prevalence of substance abuse are warranted. The literature investigating this issue provides contradictory findings, however. While some researchers have found at least a reasonable correspondence between social indicator data and survey-based estimates of substance abuse treatment utilization (Ford & Schmittdiel, 1983; Joseph & Hollett, 1993) and substance abuse behavior (Kim, Wurster, Williams, & Helper, 1998a, 1998b, 1998c; Mammo & French, 1998), others have found little correspondence (Warheit, Holzer, & Robbins, 1979; Folsum & Judkins, 1997; McAuliffe et al., 1999).
The federal government is increasingly interested in estimating the prevalence of substance abuse with social indicators, largely because of the high cost of estimating prevalence with surveys of random samples of the population. This national movement should be informed by investigations of the effects of reliance on social indicators on the accuracy of estimates of the prevalence of substance abuse. This article contributes to the discussion about the utility of social indicators in estimating substance abuse. Our strategy is to use a model-based social indicator approach that incorporates survey and social indicator data. With both the individual and county as the unit, we regress survey-based measures of the use of alcohol, marijuana, and other drugs on social indicators that fall into three categories: demographics, measures of social disorganization, and measures more directly related to the use of substances.
DATA SOURCES AND MEASURES
The data for this paper come from the 1996/1997 Minnesota Adult Household Survey (described in Beebe, Harrison, & McRae, 1999), the 1995 Minnesota Student Survey (described in Minnesota Department of Children, Families, and Learning, 1995), and various state agencies.
Adult Household Survey. The Adult Household Survey (AHS) is a telephone survey of a stratified random sample of 7,508 Minnesota residents aged 18 years or older conducted by the Gallup Organization in 1996 and 1997. The AHS includes items which measure the amount and frequency of use of alcohol and other substances.
We employ four measures of use within the past year. Drinks measures the number of alcoholic drinks consumed in the past year and is the product of responses to questions which ask about the frequency of drinking and the typical amount consumed. Drunks measures the number of times that the respondent consumed more than five alcoholic drinks in the past year. Marijuana measures the number of times that the respondent used marijuana or hashish in the last year, and other drugs measures the number of other drugs (hallucinogens, cocaine, opiates, sedatives, stimulants, or inhalants) that the respondent used during the past year. For marijuana and other drugs, interviewers asked respondents to include only use for non-medical reasons.
Minnesota Student Survey. The Minnesota Student Survey (MSS) is an anonymous, self-administered, paper-and-pencil questionnaire completed in the classroom by 126,922 public school students (72% of enrollment) in grades 6, 9, and 12 in 1995. The MSS contains a variety of questions on the frequency of use of alcohol and other drugs. We restrict attention to the 45,534 ninth graders, since they show substantial use but have not suffered the attrition which characterizes twelfth graders. We employ measures of drinks, drunks, marijuana, and other drugs which are similar to those defined for AHS.
Social indicators. We employ five demographic variables, six measures of social disorganization, and four more direct measures of substance use. The demographic variables are: the age of the respondent, the percentages of the population in the county between the ages of 20 and 29, 30 and 54, and 55 and 69, and the percentage of adults in the county between the ages of 30 and 54 who are male. The measures of social disorganization include the percentage of children in the county who were enrolled in the free school lunch program in the 1994-1995 school year, the percentage of children who were living in poverty, the rate of suspensions from school in the 1993-1994 school year, the percentage of students in the ninth through twelfth grades who dropped out of school in the 1994-1995 school year, the rate of removal of children from their homes because of abuse or neglect in 1993, and the rate of arrests for juveniles in 1992. The more direct measures of substance use are liquor sales per capita in 1993, the rate of arrests for driving while intoxicated in 1994, the rate of arrests for violation of drug laws in 1994, and the rate of admissions of county residents for treatment for chemical dependency in 1994.
This research was supported in part by contract #277-95-1035 from the Center for Substance Abuse Prevention of the Substance Abuse and Mental Health Services Administration. We thank Chow-hong Lin and Michael Luxenberg for assistance with computations.
Beebe, T.J., Harrison, P.A., & McRae, Jr., J.A.
1999 Rural/urban differences in chemical dependency treatment: Results from the Minnesota Adult Household Survey. Minnesota Medicine, 82(11), 4650.
Empey, L.T., & Erickson, M.L.
1966 Hidden delinquency and social status. Social Forces, 44, 546-554. Ford, W. E., & Schmittdiel, C. J.
1983 Predicting alcoholism service needs from a national treatment utilization survey. The International Journal of the Addictions, 18(8), 1073-1084. Folsom, R.E., & Judkins, D.R.
1997 Substance abuse in the states and metropolitan areas: Model based estimates from the 1991-1993 National Household Surveys on Drug Abuse-- methodology report. Rockville, MD: SAMSHA, Office of Applied Studies. Gold, M.
1970 Delinquent behavior in an American city. Belmont, CA: Brooks/Cole.
Joseph, A.E., & Hollett, R.G.
1993 On the use of socio-demographic indicators in local health planning: A Canadian non-metropolitan perspective. Social Science Medicine, 37(6), 813-822.
Kim, S., Wurster, L., Williams, C., & Helper, N.
1998a Algorithms for resource allocation of substance abuse prevention funds based on the estimated need: A case study on state of Florida-part 1. Journal of Drug Education, 28(2), 87-106.
Kim, S., Wurster, L., Williams, C., & Helper, N.
1998b Algorithms for resource allocation of substance abuse prevention funds based on composite risk-factor index score: A case study on state of Floridapart 2. Journal ofDrug Education, 28(3), 169-184.
Kim, S., Wurster, L., Williams, C., & Helper, N.
1998c Algorithms for resource allocation of substance abuse prevention funds based on social indicators: A case study on state of Florida-part 3. Journal of Drug Education, 28(4), 283-306.
Mammo, A., & French, JR
1998 Using social indicators to predict addiction. Substance Use and Misuse, 33(12), 2499-2513.
McAuliffe, W.E., LaBrie, R.A., Pollock, N., Lomuto, N., Fournier, E.A., & Betjemann, R.
1999 Measuring interstate variation in drug abuse. Drug and Alcohol Dependence, 53(2), 125-145.
Minnesota Department of Children, Families, and Learning
1995 Perspectives on youth: Minnesota student survey 1989, 1992, 1995. St. Paul: Minnesota Department of Children, Families, and Learning. Sherman, R.E., Gillespie, S., & Diaz, J.A.
1996 Use of social indicators in assessment of local community alcohol and other drug dependence treatment needs within Chicago. Substance Use and Misuse, 31(6), 691-728.
Simeone, R. S., Frank, B., & Aryan, Z.
1993 Needs assessment in substance misuse: A comparison of approaches and case study." The International Journal of the Addictions, 28(8), 767-792. Substance Abuse and Mental Health Services Administration.
1995 Request for proposals (rfp) no. 277-95-1032, state prevention needs assessment: Alcohol and other drugs. Rockville: SAMHSA.
Substance Abuse and Mental Health Services Administration.
1996 Substance abuse in states and metropolitan areas: Model based estimates from the 1991-1993 National Household Surveys on Drug Abuse summary report. Rockville: SAMHSA.
Substance Abuse and Mental Health Services Administration
2000 Core Protocol Guide -- Social Indicators. Retrieved October 26, 2000 from the World Wide Web: http://www.drug-watch.com/download/ SocIndlO.pdf.
Warheit, G. J., Holzer, III, C. E., & Robbins, L.
1979 Social indicators and mental health planning: An empirical case study. Community Mental Health Journal, 15(2), 94-103.
JAMES A. MCRAE, JR., TIMOTHY J. BEEBE, PATRICIA A. HARRISON
James A. McRae, Jr., Ph.D. is a Senior Research Scientist in the Health Care Research Division of the Minnesota Department of Human Services. Research interests include substance use, health care, and quantitative methods. Timothy Beebe, Ph.D., is a Senior Research Associate in the Division of Health Services Research and Policy, School of Public Health, University of Minnesota. His research interests include assessing effects of managed health care on the health, well being, and satisfaction of Minnesota's Medicaid population. He is also interested in general survey research methods. Patricia A. Harrison, Ph.D., is Director of Research and Assessment at the Minneapolis Department of Health and Family Support and a licensed psychologist. Her areas of interest include mental health and substance abuse screening and treatment services.…
Questia, a part of Gale, Cengage Learning. www.questia.com
Publication information: Article title: Estimating the Prevalence of Substance Abuse with Social Indicators. Contributors: McRae, James A., Jr. - Author, Beebe, Timothy J. - Author, Harrison, Patricia A. - Author. Journal title: Journal of Drug Issues. Volume: 31. Issue: 4 Publication date: Fall 2001. Page number: 977+. © Florida State University for and on behalf of The Florida State University Board of Trustees Spring 2008. Provided by ProQuest LLC. All Rights Reserved.
This material is protected by copyright and, with the exception of fair use, may not be further copied, distributed or transmitted in any form or by any means.