Analyses of Substance Abuse and Treatment Need Issues

Part IV Estimating Substance Abuse Treatment Need From a National Household Survey

• Introduction
• History
• Recently Developed Methods
• Table 1:Twelve Month Estimates of Dependence From the NHSDA and the NCS for Persons 18-44 Years of Age
• Table 2:Comparison of Two Definitions of Drug Dependence from the NCS
• Table 3:Spearman Rank Order Correlations Between the NCS and the NHSDA for Criterion Level Prevalence
• Table 4:Population Size and the Percent Dependent For Selected Drug Abusing Populations Based on the 1991 NHSDA
• Future Research
• Conclusions
• References
• Appendix 1:DSM-III-R criteria for substance dependence and corresponding questions in the 1991-1993 NHSDA that represent them.

7. Estimating Substance Abuse Treatment Need From a National Household Survey

By Joan F. Epstein and Joseph C. Gfroerer

Paper presented at 37th International Congress on Alcohol and Drug Dependence, August 20-25, 1995, UCSD Campus, La Jolla, California

Introduction

For purposes of planning future demand on the health care system in general and the substance abuse treatment system in particular, it is important to be able to develop estimates of the number of people needing treatment for substance abuse (i.e., treatment need) on a regular basis. In order to do this, measurement of treatment need must distinguish low intensity substance use from drug use that requires intervention. The overall prevalence of substance use is a poor absolute measure of problem substance use.

Estimating treatment need is a difficult problem. Drug and alcohol consumption patterns and their consequences are complicated and dynamic. The modalities and philosophies of treatment are diverse. The applicability of even well ensconced and tested diagnostic criteria must be reestablished as new drugs and ways of administering them appear. Measuring treatment need involves both a scientific and clinical understanding of the substance abuse problem.

Estimating how many people in the general population need treatment is a different problem from diagnosing the need for treatment for an individual based on history taking, physical examination and information in previous records. In most household surveys it would be impractical to perform physical examinations or take a detailed history. In addition, household surveys may not fully cover key populations that may have a significant number of people needing treatment such as incarcerated and homeless individuals.

The main focus of this paper will be to describe recent attempts to estimate treatment need based on the National Household Survey on Drug Abuse (NHSDA). The NHSDA is potentially a valuable source for estimating treatment need since it is the only ongoing large national survey of substance use covering most of the population. Thus, if a reasonable method for estimating treatment need from the NHSDA could be developed, it would provide more timely estimates than other sources and could be used to measure changes in treatment need over time.

History

In recent years, several methods have been developed to estimate treatment need using the NHSDA. These methods were generally developed to estimate drug abuse treatment need, not alcohol treatment need.

The National Institute on Drug Abuse (NIDA) developed an illicit drug index in 1989 that defined heavy drug users who need treatment as those who had used illicit drugs at least 200 times in the past year. An estimated four million Americans met this criterion in 1988. By assuming that 25 percent of these drug users would stop using drugs without formal treatment and another 25 percent were hard core addicts that were too difficult to reach, the Office of National Drug Control Policy determined that there were 2 million people with drug problems for which treatment would be beneficial (1).

A shortcoming of NIDA's illicit drug index was that it was based solely on the frequency of drug use, without considering problems associated with use. To overcome this, another model based on reported problems and symptoms of abuse or dependence on illicit drugs was developed by NIDA. This method was an approximation of criteria in the Diagnostic and Statistical Manual of Mental Disorders, Third Edition (DSM-III) published by the American Psychiatric Association (2). This method produced an estimate of 1.6 million people with either cocaine abuse or dependence in 1985 (3). This method also produced an estimate of 3.5 million people with any illicit drug abuse or dependence in 1991.

A study sponsored by the Institute of Medicine (IOM) in 1990 developed a methodology for estimating drug abuse treatment need using a combination of frequency of use and problems/symptoms indicators from the NHSDA. This method for estimating treatment need is based on combining data from three separate variables: 1) frequency of drug use (highest frequency of use in the past month out of cocaine, marijuana and inhalants) 2) symptoms of dependence (based on all five of the six questions in the Drugs section of the 1991 NHSDA and 3) problems/consequences (based on all 11 questions in the Drug Problems section of the 1991 NHSDA). The IOM method also supplemented these NHSDA estimates with data from prison and homeless populations to produce a more comprehensive estimate of need. This method resulted in an estimated 4.6 million people needing treatment in 1988. (4)

Recently Developed Methods

A method for estimating treatment need from the NHSDA considered by the Office of Applied Studies (OAS) of the Substance Abuse and Mental Health Services Administration (SAMHSA) uses items in the NHSDA questionnaire that are similar to several but not all of the items defined in the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III-R) for substance dependence (5). In this method we are using dependence as a definition for treatment need. We considered using abuse or dependence as a definition of treatment need but didn’t because the NHSDA does not contain enough questions to adequately approximate abuse. The DSM-III-R defines a person as dependent for a substance if they meet 3 out of 9 criteria for that substance. This method combines items on symptoms and problems included in the NHSDA questionnaire to approximate 5 of the 9 DSM-III-R criteria for substance dependence and defines a person as dependent if they meet 2 out of these 5 criteria. These five criteria are 2, 5, 6, 7 and 8. Appendix 1 indicatesthe DSM-III-R criteria and the corresponding questions from the NHSDA that represent them. It also shows the criteria not covered by the NHSDA.

There are problems with this method. One is that it classifies dependence based on the presence of 2 out of 5 criteria rather than the full 3 out of 9 criteria specified in the

DSM-III-R. Another is that the questionnaire wordings in the NHSDA do not precisely reproduce the DSM-III-R criteria and in some cases these wordings only partially cover a criteria.

In order to evaluate this method, OAS performed an in depth analysis that compared the consistency of estimates for alcohol, marijuana, cocaine, and psychotherapeutics from the NHSDA to the National Comorbidity Survey (NCS). The NCS, conducted by the University of Michigan’s Institute for Social Research, was designed to provide nationally representative estimates of 14 psychiatric disorders, including substance abuse and dependence. It employed a stratified, multistage area probability sample of 8,098 respondents in the household population. It contained a modified version of the Composite International Diagnostic Interview (CIDI). The CIDI is a structured diagnostic interview designed to provide estimates of DSM-III-R disorders and includes questions to cover all nine dependence criteria specified in the DSM-III-R (6).

Table 1 compares the 12-month dependence estimates from the NHSDA and the NCS. Generally, NCS and NHSDA estimates for marijuana and cocaine are similar. Summary estimates that included marijuana, cocaine, psychotherapeutics and/or alcohol were also computed for the NHSDA and the NCS and found to be similar. For 18-44 year olds, the prevalence estimates were 2.0 percent and 1.8 percent from the NCS and NHSDA, respectively. Prevalences for any substance (alcohol, marijuana, cocaine or psychotherapeutics) were 6.4 percent (NCS) and 6.9 percent (NHSDA).

Table 1:Twelve Month Estimates of Dependence From the NHSDA and the NCS for Persons 18-44 Years of Age

To assess using 2 out of 5 criteria instead of the full 3 out of 9 criteria in the NHSDA, we constructed dependence estimates from the NCS using five of the nine criteria that are approximated in the NHSDA and compared them to the corresponding NCS estimates generated from the full set of nine criteria (Table 2). Generally the smaller set of criteria represented the full set very well for cocaine dependence, reasonably well for marijuana dependence and psychotherapeutics, but not as well for alcohol.

Table 2:Comparison of Two Definitions of Drug Dependence from the NCS

To assess reporting differences resulting from differences in the way questions are worded, we compared the overall prevalence rates for each of the criteria, and also for combinations of them with each other and with other variables. Differences were difficult to detect because prevalences and sample sizes were small. However with few exceptions, NHSDA and NCS estimates are similar for marijuana, cocaine and alcohol. For psychotherapeutics, the NHSDA and NCS were slightly more discrepant. To further evaluate differences in questionnaire wordings between surveys, we compared the rank orders of criterion level prevalences between surveys using Spearman Rank Order Correlation Coefficients (Table 3). NHSDA and NCS rankings of criteria were similar for marijuana, cocaine and alcohol. However, the Spearman rank correlation for psychotherapeutics was 0, indicating that the criterion level prevalances were not close in rank order between the NHSDA and the NCS for psychotherapeutics.

Table 3:Spearman Rank Order Correlations Between the NCS and the NHSDA for Criterion Level Prevalence

The research described above indicates that the NHSDA approximation of the DSM-III-R definition of drug dependence provides prevalence estimates which are comparable to the NCS. With few exceptions, the rates of individual criteria and of overall 12-month dependence found in the NHSDA were similar to those in the NCS. This method, which resulted in an estimated 2.6 million people dependent on illicit drugs in 1991, has been used to approximate substance related dependence in several studies (7-11). In a study of AFDC it was used to estimate substance related impairment (7) and in a study to evaluate the substance abuse and mental health services block grant allotment formula it was used to estimate drug and alcohol dependence as a measure of treatment need (8).

While the above method seems to approximate DSM-III-R dependence, its applicability to treatment need estimation is a separate question. To address this (for illicit drug use) we evaluated this method and compared it to the IOM method (9). This research compared these methods in terms of their inclusion of various categories of drug abusers that may or may not need treatment. For example it classified 40 percent of weekly cocaine users and 29 percent of past year needle users as dependent (Table 4).

Table 4:Population Size and the Percent Dependent For Selected Drug Abusing Populations Based on the 1991 NHSDA

The following conclusions were made based on this research: 1) Under both methods many populations of serious drug users are only partially assigned as needing treatment 2) A very small proportion of persons using prescription drugs are assigned as needing treatment under both methods 3) the two methods do not estimate the same people as needing treatment 4) the DSM-III-R dependence method defines a smaller more seriously impaired group of drug abusers as needing treatment compared to the method in the study sponsored by IOM, which defines a broader group of drug abusers as needing treatment.

This research demonstrated that the DSM-III-R dependence method was too restrictive. Substantial proportions of several important populations such as IV drug users, heroin users, and heavy cocaine users were classified as not needing treatment by the algorithm, and a high percentage of persons who actually reported receiving treatment in the past year were classified as not needing treatment at any time during the past year.

Based on this research the following recommendations were made for estimating drug abuse treatment need from the NHSDA:

1.Count some heavy drug users who report no problems due to their drug use as needing treatment, to account for respondents who deny problems.

2.Account for persons who use more than one drug and who may have an overall drug problem not concentrated in one drug.

3.Use independent sources to improve estimates of treatment need from populations that are not well represented in the NHSDA such as the criminal justice population and the treatment population.

4.Consider certain categories of drug users to need treatment regardless of whether they are dependent (e.g. IV drug users).

5.Keep the algorithm as simple as possible so it can be used by different studies.

Based on these recommendations an improved method for estimating drug abuse treatment need was developed by OAS.

In this new method the definition of treatment need refers to the population that had a serious drug problem at any time in the past year (i.e., period prevalence). The following is our revised definition for estimating treatment need:

Persons with at least one of the following criteria are counted as having needed treatment at some time during the year before they were interviewed:

1. Drug Dependence

Dependent on any illicit drug in the past year. This is based on the algorithm that approximates the DSM-III-R definition.

2. Heavy Drug Use

Any of the following in the past year:

a. Used heroin at least once.

b. Used marijuana daily.

c. Frequent use (52+ days/weekly) of some other illicit drug.

3. IV Drug Use

Used heroin, cocaine, or stimulants with a needle in the past year.

4. Treated for Drug Abuse

Had treatment for any illicit drug in the past year.

Of course, there is considerable overlap between these categories of drug abusers. The overlap is largest for dependence and heavy drug use. Approximately 40 to 50 percent of the persons meeting the criteria for dependence also meet the criteria for heavy drug use and approximately 30 to 40 percent of the persons meeting the criteria for heavy drug use also meet the criteria for dependence.

The new method also includes a ratio adjustment to outside counts of arrest and treatment data, to partially account for underestimation by the NHSDA due to underreporting and/or undercoverage of hard-core drug use. It is believed that estimates of drug abuse prevalence from the NHSDA are generally conservative because of the potential undercoverage of heavy drug using populations and underreporting of drug use among survey respondents. To attempt to correct for this underestimation of "hard core" drug use, OAS developed a ratio estimation procedure. This procedure links NHSDA data with outside sources of data (primarily the Uniform Crime Reports and the National Drug and Alcohol Treatment Unit Survey) to result in adjusted prevalence estimates (10).

Figure 1 compares estimates of treatment need from the new method with and without ratio estimation for years 1991 to 1993. It shows the contribution of each component of the treatment need definition hierarchically. That is, the darkest shading represents all persons meeting the criteria for drug dependence. The bars showing heavy drug use represent persons who meet the criteria for heavy drug use but not drug dependence. Similarly, the chart shows data for those who meet the criteria for IV drug use but not drug dependence or heavy drug use and those who meet the criteria for treatment but not drug dependence, heavy drug use or IV drug use.

This method resulted in an estimated 3.2 percent of the population age 12 and older needing treatment in 1993. This translates to an estimated 6.7 million people. Comparing the adjusted and non-adjusted estimates by components of the definition and by year shows the differential impact of the ratio adjustment. For 1993 the ratio adjustment increased the estimate of treatment need by about 30% while for 1991 and 1992 it increased the estimate of treatment need by about 20%. The ratio adjustment had the strongest effect on the "treated for drug use" component of the estimate (40% increase). This is not surprising since treatment is one of the variables used in the ratio estimation procedure. The percent of persons needing treatment appears to be decreasing from 1991 to 1993 but these differences are not statistically significant. This trend does not appear for the estimates of dependence. This new method has been used to estimate the number of people needing treatment in a study of substance abuse in state and metropolitan areas (11).

Future Research

Improving our methods of estimating treatment need using the NHSDA is a continual process. We are studying ways to improve the coverage and quality of questions in the NHSDA and the analysis used to estimate treatment need. More questions on drug dependence have been added to the 1994 NHSDA questionnaire to cover more criteria from the DSM-III-R. Questions on drug dependence have been added to the 1995 NHSDA questionnaire to address 6 of the 7 criteria defined in the DSM-IV (12).

We are also studying alternative definitions of treatment need. For example, different categories of substance abusers could be included. There are also alternatives to the definition of treatment need that can account for a number of different circumstances and population characteristics. The following are examples of some possible alternative criteria to use in the definition of treatment need:

•Define treatment need differently for different populations.

•Use different definitions for adolescents and adults (e.g., heavy drug use for adolescents might include daily marijuana users while heavy drug use for adults might not)

•Have a separate definition for pregnant women or women with children. (e.g., any drug use for pregnant women might be defined as needing treatment)

•Define heavy drug use differently for each drug.

•Do not include marijuana use in the definition of heavy drug use.

•Classify substance abusers into categories based on various criteria and assume varying proportions of each category needs treatment.

We are also working on a method for estimating treatment need for alcohol that is similar to the method developed for drug abuse. As with drug abuse it has been difficult to define heavy drug use for alcohol since some individuals can consume heavily or regularly with few problems while others have severe problems at much lower doses and frequencies.

Conclusions

Due to the complex nature of the alcohol and drug abuse problem careful methods and sophisticated knowledge are required to understand and quantify the extent of the need for treatment. This paper has presented some methods that have recently been considered by the OAS for estimating drug abuse treatment need using the NHSDA and research on these methods. It has shown that the NHSDA approximation to the DSM-III-R definition of drug dependence may be a reasonable method for estimating dependence based on comparing estimates from the NHSDA with the NCS. It has also shown that the NHSDA approximation of the DSM-III-R definition of drug dependence may not be a sufficient method for estimating treatment need. This may be due to the DSM-III-R definition of drug dependence not being an adequate model for defining treatment need.

References

1)Office of the National Drug Control Policy. (1989) National Drug Control Strategy.

2)American Psychiatric Association Committee on Nomenclature and Statistics. (1980) Diagnostic and Statistical Manual of Mental Disorders, ed 3. (Washington DC, APA).

3)Adams, E.H., Gfroerer, J.C.: Risk of Cocaine Abuse and Dependence. In: the Epidemiology of Cocaine Use and Abuse, edited by Susan Schober and Charles Schade. National Institute on Drug Abuse Research Monograph 110. DHHS Pub. No. (ADM) 91-1787, Alcohol, Drug Abuse, and Mental Health Administration, 1991.

4)Gerstein, D.R., Harwood H.J. (1990) Treating Drug Problems, Volume 1. National Academy Press.

5)American Psychiatric Association. (1987) Diagnostic and Statistical Manual of Mental Disorders 3rd edn rev. (Washington DC, APA).

6)Kessler, R. C., McGonagle, K.A., Zhao, S., Nelson C., Hughes, M., Eshleman, S., Wittch, H.U., Kendler, K.S., Lifetime and 12-Month Prevalence of DSM-III-R Psychiatric Disorders in the United States: Results from the National Comorbidity Survey (NCS). Arch Gen Psychiatry. 51:8-9, 1994.

7)Office of the Assistant Secretary of Planning and Evaluation and the National Institute on Drug Abuse. (1994) Patterns of Substance Use and Substance-Related Impairment Among Participants in the Aid to Families With Dependent Children Program (AFDC).

8)Burnam, A., Reuter P., Adams J., Palmer A., Model K., Rolph J., Heilbrunn J., Marshall G., McCaffrey D., Wenzel S., Kessler R. (1994) Review and Evaluation of the Substance Abuse and Mental Health Services Block Grant Allotment Formula. Prepared for the U.S. Department of Health and Human Services, Center for Substance Abuse Treatment, DRU-635-2-HHS.

9)Epstein, J.F., Gfroerer, J.C. : Estimating Substance Abuse Treatment Need From the NHSDA. American Statistical Association 1994 Proceedings of the Section on Survey Research Methods.

10)Wright, D., Gfroerer, J.C., Epstein, J.F. : Ratio Estimation of Hard-Core Drug Use. American Statistical Association 1994 Proceedings of the Section on Survey Research Methods.

11)Substance Abuse in States and Metropolitan Areas: Model Based Estimates from the 1991-1993 National Household Survey on Drug Abuse, In preparation for the U.S. Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, Research Triangle Institute 1995.

12)American Psychiatric Association.(1994) Diagnostic and Statistical Manual of Mental Disorders 4th edn. (Washington DC, APA).

Appendix 1:DSM-III-R criteria for substance dependence and corresponding questions in the 1991-1993 NHSDA that represent them.

This page was last updated on June 03, 2008.

SAMHSA, an agency in the Department of Health and Human Services, is the Federal Government's lead agency for improving the quality and availability of substance abuse prevention, addiction treatment, and mental health services in the United States.