Prevalence of DSM-III-R alcohol abuse and/or dependence among selected occupations: United States, 1988 - Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised

Alcohol Health & Research World,  Spring, 1992  by Frederick S. Stinson,  Samar Farha DeBakey,  Rebecca A. Steffens

• E-mail
• Print
• Link

United States, 1988

There has been much curiosity among researchers and the general public alike regarding the prevalence of alcohol problems among different occupations. The authors explored this issue using data from the 1988 National Health Interview Survey and found considerable variation in the prevalence of alcohol problems both across occupations and within occupational groups.

This Epidemiologic Bulletin examines the prevalence of alcohol abuse and/or alcohol dependence among specific occupations. The findings presented here are from the 1988 National Health Interview Survey (NHIS), the first U.S. national level survey to implement psychiatric definitions of alcohol abuse and alcohol dependence as articulated in the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III-R) (American Psychiatric Association 1987). The DSM-III-R definitions are the most current definitions now in use for the diagnosis of substance abuse and dependence.

BACKGROUND

The Alcohol Epidemiologic Data System (AEDS) has received many inquiries from researchers, journalists, and members of the general public about alcohol problems in different groups of the U.S. population. A frequently asked question involves the prevalence of alcohol problems among different occupational groups.

We used the 1988 NHIS to explore the prevalence of alcohol use disorders. The definition of alcohol use disorders used here includes alcohol abuse, in addition to alcohol dependence. The occupational classification used in the 1988 NHIS (U.S. Bureau of the Census 1980) defines more than 500 detailed occupations. A different examination of prevalence estimates of dependence among occupational groups using the NHIS data set was conducted by Parker and Harford (see the article by Parker and Harford, pp. 97-105). We will discuss this study in more detail later in this article.

DATA SOURCE

Data used in the analyses reported here are from the 1988 NHIS, an ongoing, cross-sectional household interview survey designed by the National Center for Health Statistics (NCHS) and conducted by interviewers from the U.S. Bureau of the Census. The survey collects information on the health and other characteristics of each member of the households sampled.

The 1988 NHIS included a set of supplementary questions designed and funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). These questions were asked of one randomly selected person aged 18 or older in each sample household; a total of 43,809 persons were interviewed, representing a response rate of 87 percent. Among the questions were 34 items that could be used to assess alcohol abuse or alcohol dependence; these items were based on criteria articulated in the DSM-III-R.

Although the 1988 NHIS used a very large sample, the fact that answers come from a sample of all possible respondents means that statistical procedures must be used in order to assess the significance of any apparent differences found in analyses of these data. This survey did not use a random sample, because a truly random sample would not have provided enough respondents representing certain small population groups of special interest (e.g., in 1988 the NHIS oversampled blacks). In order to increase the reliability of estimates for these relatively small groups in the U.S. population, the 1988 NHIS used a complex multistage sampling design, described elsewhere (Massey et al. 1989). Because of the complex sampling design, it was necessary to use special statistical procedures in calculating variance estimates for assessing statistical significance. This was accomplished through the use of SUDAAN (Research Triangle Institute 1991), a set of statistical computer programs that adjust for sample design characteristics.

ALCOHOL ABUSE/DEPENDENCE

The prevalence estimates reported here for alcohol disorders involve respondents who met DSM-III-R criteria for alcohol abuse and/or alcohol dependence in the 12 months prior to the interview. The assignment of a diagnosis of alcohol abuse or dependence is based on responses to 34 questionnaire items in the 1988 NHIS. A detailed explanation of the items and the criteria for the diagnosis of alcohol disorders are presented elsewhere by Grant and co-workers (1991).

In summary, a diagnosis of alcohol dependence requires self-reports of various symptoms that meet at least three of the following nine DSM-III-R dependence criteria: tolerance; withdrawal; drinking to relieve or avoid withdrawal symptoms; drinking larger amounts than intended; a great deal of time spent drinking; activities given up for drinking; continued drinking despite problems; neglected responsibilities or obligations; or impaired control. The DSM-III-R also includes a duration requirement (i.e., that some symptoms persisted for at least 1 month or occurred repeatedly over a longer period of time). In the 1988 NHIS, the duration requirement was satisfied if respondents reported two or more symptoms in the past year for each of two dependence criteria, with the following two exceptions. First, because withdrawal is a syndrome, or a cluster of symptoms, a respondent must report at least two symptoms of withdrawal to meet the duration criterion. It should be noted, however, that withdrawal is not a necessary criterion for alcohol dependence. Second, any symptom of tolerance was deemed to meet the duration requirement.