Public Health Institute, Alcohol Research Group, 2000 Hearst Avenue, Berkeley, CA 94709, USA
Received 21 August 1998; in revised form 16 November 1998; accepted 20 December 1998
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Alcohol's role in health problems is well documented (Popham et al., 1984; Rush, 1989
), but the role of drinking patterns and individual factors in use of ER and primary care services is less well understood. Over-representation of alcohol problems in primary care settings may be due to an increased rate of health problems in general among those who abuse alcohol, as well as to conditions related specifically to alcohol use. Alcohol mis-use has been identified as a risk factor for a number of diseases (Rush, 1989
), and patients who report alcohol-related problems have been found more likely than non-alcohol-problem patients to have other conditions which primary care physicians see and treat on a regular basis (Cleary et al., 1988
). One study found a probability sample of primary care clinic patients four times more likely to report physical health problems related to drinking during the last year, than those in the general population from which they came (Cherpitel, 1991
). In the comparative study cited above, 14% of the county clinic sample reported physical health problems related to drinking, compared to 5% in the HMO sample, and 3% in the general population (Cherpitel, 1994
). Additionally, in both the Epidemiologic Catchment Area (ECA) Study and the National Co-morbidity Survey, while most of those reporting an alcohol or other substance-use disorder during the previous year received no health services for their problem, among those who did, a large proportion obtained such services in the general medical sector (Regier et al., 1993
; Kessler et al., 1994
).
Given these prior studies and the variation in study findings within various segments of the population, little is known of the relationship of alcohol or other drug use and the use of ER or other primary care settings, and no general population studies have been reported which examined this relationship. To begin filling this gap in our knowledge, we report here on data from a representative sample of the US adult population from the 1995 National Alcohol Survey regarding the association of alcohol, drug use, and demographic characteristics with use of ERs and primary care services during the preceding year. This is the first year in which the National Alcohol Survey obtained data on use of health care services for illness in addition to injuries, making these analyses possible. Additionally, black and Hispanic respondents were oversampled, allowing an analysis of substance use and health services use by ethnicity. The likelihood of reporting an ER or primary care visit is reported by ethnicity, for males and females, separately for each demographic and substance-use characteristic, and multiple logistic regression is then used to analyse the association of substance-use variables, controlling for demographic characteristics, with reporting either an ER and/or primary care visit during the last year.
Findings reported here from a nationally representative sample on the association of alcohol, drug use, and demographic characteristics with primary care services utilization are important for a better understanding of selection factors which result in use of these services, and for identifying those settings which may provide the best potential for screening for alcohol-related problems. ER and primary care settings may provide one of the best opportunities for screening and subsequent intervention with problem drinkers, since these patients may be more motivated at a time when they are seeking care for a health problem.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Data collection
Face-to-face interviews, using a structured interview schedule of about an hour's length, were conducted by trained interviewers in the respondent's own home after obtaining informed consent. Hispanic respondents were given a choice of being interviewed in English or in Spanish, with a bilingual interviewer. The Spanish version of the questionnaire underwent a process of translation and independent back-translation.
Instruments
Among other items, respondents were asked questions related to quantity and frequency of usual drinking, frequency of drunkenness, consequences related to drinking, alcohol dependence, and frequency of drug use all in the last year and alcohol and drug-abuse treatment on a lifetime basis, as well as demographic characteristics. Respondents were also asked whether, during the last 12 months, they had an injury, and separately an illness, for which they received treatment, and if so, whether they went to an ER, a private doctor or clinic, or some other place for treatment. Those having sought treatment for either an injury or illness in the ER were considered to have made an ER visit, whereas those who went to either a private doctor, clinic, or some other source of traditional medical care were considered to have made a primary care visit.
A quantityfrequency (QF) typology was developed, based on questions regarding the quantity and frequency of usual drinking. Frequency of drinking was defined as: low (once per month); moderate (at least once per month but <3 times per week); high (
3 times per week). Quantity was defined as: low (never 5 drinks at any one time); moderate (
5 but <12 drinks at any one time); high (
12 drinks at any one time). A time was defined by each patient, but generally referred to a drinking occasion separated by at least 2 h from the next occasion. A drink was defined as a bottle or can of beer, a glass of wine or a mixed drink, each of which contains approximately 0.5 oz (15 g) of absolute alcohol. Five drinking categories were developed, based on this QF typology: (1) abstainer (no drinking during the last year); (2) infrequent (low frequency/any quantity); (3) light (moderate or high frequency/low quantity); (4) moderate (moderate or high frequency/ moderate quantity or moderate frequency/high quantity); (5) heavy (high frequency/high quantity).
Consequences of drinking included problems with friendships and social life, home life or marriage, outlook on life, financial position, work and employment opportunities or health, and has been used in prior general population surveys (Clark and Hilton, 1991; Midanik and Clark, 1995
).
Alcohol dependence during the last year was measured by 24 items similar to those in the Alcohol Section of the Composite International Diagnostic Interview (CIDI; Wittchen et al., 1991), which operationalized both ICD-10 (International Classification of Diseases, 10th Revision) (World Health Organization, 1990) and DSM-IV (Diagnostic and Statistical Manual of Mental Disorders) (American Psychiatric Association, 1994
) criteria (Caetano and Room, 1994
). Items which operationalized ICD-10 included questions related to the six domains of craving, impaired capacity to control, withdrawal, tolerance, neglect of interests, and continued use despite problems; whereas DSM-IV included questions related to the seven domains of tolerance, withdrawal, drinking more than intended, unsuccessful efforts to control, giving up pleasures or interests to drink, spending a great deal of time in drinking activities, and continued use despite problems. A respondent was considered alcohol-dependent if positive on three or more domains in either of the diagnostic schemes considered separately.
Drug use during the previous year included any use in the following categories: speed or amphetamines, cocaine or crack, tranquillizers, heroin or opium, methadone, marijuana or tetrahydrocannabinol, and hallucinogens.
Data analysis
Respondents who self-identified as either white of Hispanic origin or black of Hispanic origin (Latino, Mexican, Central or South American, or any other Hispanic origin) were classified as Hispanic (n = 1585). Respondents who self-identified as black not of Hispanic origin were categorized as black (n = 1582). Of the remaining respondents, 1636 were classified as white and 122 other respondents were excluded from analyses here (Asian, Native American, and those of mixed race).
To adjust for the design effects inherent in multi-cluster sampling, SUDAAN (Software for Survey Data Analysis) (Research Triangle Institute, 1994) was used to correct for estimates of standard errors. Data were weighted to reflect the number of adults living in a sampled household which results in the probability of selection into the sample, non-response rates, and for analyses by ethnicity. Weights were also applied to account for the oversampling of black and Hispanic respondents.
Univariate logistic regressions were used to analyse the association of demographic characteristics and drinking and drug-use variables with reporting a visit to the ER and reporting a primary care visit during the preceding year. The univariate effect of drinking and drug variables was also analysed separately for males and females by ethnicity. Finally, multiple logistic regression was used, with simultaneous entry of demographic, drinking, and drug variables, separately for reporting an ER visit and for reporting a primary care visit. Odds ratios (OR) are reported for all variables in the equation and 95% confidence intervals (CI) for those variables found to be significant predictors.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
Table 4 shows the ORs and CIs from multivariate regression, with simultaneous entry of variables, for demographic, drinking, and drug-use characteristics on reporting an ER visit and reporting a primary care visit during the preceding year. Moderate alcohol use, reporting consequences related to drinking and having ever been in alcohol treatment were significantly associated with an ER visit, whereas none of the drinking or drug-use variables was associated with having made a primary care visit. Similar regressions were carried out by gender within each ethnic group (data not shown). Among white males, drinking consequences (OR = 5.42) and alcohol treatment (OR = 2.95) were significantly associated with an ER visit (with frequency of drunkenness showing a negative association), and among white females heavy drinking was found to be significantly associated with ER use (OR = 26.72). None of the variables was associated with a primary care visit among whites. For Hispanics, multivariate analyses were similar to univariate analyses, with frequency of drunkenness for males and drug treatment for females negatively associated with an ER visit, whereas alcohol treatment was positively associated with an ER visit for females (OR = 7.38). Among blacks, variables which were found to be associated with an ER visit in univariate analysis were no longer significant, although alcohol dependence among black males was associated with a primary care visit (OR = 6.29).
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Univariate analysis found heavy drinking among white males and frequent drunkenness among white females to be negatively associated with ER use, as was frequent drunkenness among Hispanic males. These negative findings may be related to the fact that QF categories were compared to abstainers as the reference category, whereas frequency of drunkenness was compared to no drunkenness as the reference group. It is well known that those who report current abstention may have health problems for which drinking is contra-indicated and for which treatment is obtained from ER and primary care providers on an on-going basis; thus abstention, itself, would be predictive of health services use. However, heavy drinking was found to be associated with ER use among white females using multivariate analysis.
Among black females, alcohol consumption was found to be positively associated with ER use (univariate analysis), with those reporting heavy drinking, drinking-related consequences, or alcohol dependence being from three to over five times more likely to have used the ER. National survey data has found high rates of abstention among black females, compared to others (Caetano and Kaskutas, 1995), and abstention in this group may be more a part of a cultural lifestyle and less related to health problems, than among other groups as suggested above. In multivariate analysis, alcohol consumption variables lost significance, however, possibly because of the large standard errors of ORs for both blacks and Hispanics, due to the heavy clustering of respondents in drawing the oversample for these two ethnic groups.
The data reported here suggest an association between treatment for alcohol or drug problems and service use among some subgroups, which may be indicative of general medical sector use for treatment of substance-related problems. Among Hispanics, drug treatment among males and alcohol treatment among females were associated with ER use, whereas drug treatment was also associated with ER use among black males, and alcohol treatment was associated with primary care use among black females. Among whites, substance-abuse treatment was only associated with ER use in multivariate analysis among males, and may suggest that blacks and Hispanics are more likely to use general medical services for substance-abuse treatment than speciality services, compared to whites.
Variables found to be associated with ER use were not found to be associated with primary care service use (and vice versa), and drinking and drug-use characteristics appeared to have a greater association with ER use than with primary care use. Factors found here to be associated with ER and primary care services use suggest that these settings may be useful for identifying those with alcohol- and drug-related problems among specific gender/ethnic groups.
One explanation for the limited association found in this study of drinking and drug-use variables with reporting an ER or primary care visit during the last year may be related to the fact that those who reported use of such services may not necessarily be heavy users of these services, as are many of those sampled in clinical settings who would have a greater probability than less frequent users of falling into a sample drawn from such settings. ER and primary care use was based on reporting even one such visit during the preceding year, and it is possible that drinking and drug-use variables would have a stronger association with health service use among multiple users of such services, than among less frequent users.
Findings from previous research in clinical settings, which have found alcohol-related problems to be over-represented, may be due to the socio-demographic characteristics of the patients seen in the facility studied. The present data support these findings; drinking variables were not uniformly associated with service use across type of facility or across ethnic and gender subgroups. Future research is necessary to determine whether drinking and drug-use characteristics vary by frequency of health services use, and how associations may vary by type of primary care service, gender, and ethnicity.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Caetano, R. and Kaskutas, L. A. (1995) Changes in drinking patterns among Whites, Blacks and Hispanics: 19841992. Journal of Studies on Alcohol 56, 558565.[ISI][Medline]
Caetano, R. and Room, R. (1994) Alcohol dependence in the 1990 U.S. national alcohol survey: Operationalizing and comparing two nosological systems. Drug and Alcohol Review 13, 257267.
Cherpitel, C. J. (1988) Drinking patterns and problems associated with injury status in emergency room admissions. Alcoholism: Clinical and Experimental Research 12, 105110.[ISI][Medline]
Cherpitel, C. J. (1991) Drinking patterns and problems among primary care patients: a comparison with the general population. Alcohol and Alcoholism 26, 627633.[ISI][Medline]
Cherpitel, C. J. (1992) Drinking patterns and problems: a comparison of ER patients in an HMO and in the general population. Alcoholism: Clinical and Experimental Research 16, 11041109.[ISI][Medline]
Cherpitel, C. J. (1993) Alcohol consumption among emergency room patients: comparison of county/community hospitals and an HMO. Journal of Studies on Alcohol 54, 432440.[ISI][Medline]
Cherpitel, C. J. (1994) Alcohol use among primary care patients: comparing an HMO with county clinics and the general population. Drug and Alcohol Dependence 36, 167173.[ISI][Medline]
Cherpitel, C. J. (1995) Alcohol use among HMO patients in the emergency room, primary care and the general population. Journal of Studies on Alcohol 56, 272276.[ISI][Medline]
Clark, W. B. and Hilton, M. (eds) (1991) Alcohol in America: Drinking Practices and Problems. State University of New York Press, Albany, NY.
Cleary, P., Miller, M., Bush, B., Warburg, M., Delbanco, T. and Aronson, M. (1988) Prevalence and recognition of alcohol abuse in a primary care population. American Journal of Medicine 85, 466471.[ISI][Medline]
Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., Wittchen, H.-U. and Kendler, K. S. (1994) Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Co-Morbidity Survey. Archives of General Psychiatry 51, 819.[Abstract]
Magruder-Habib, K., Durand, A. M. and Frey, K. A. (1991) Alcohol abuse and alcoholism in primary health care settings. Journal of Family Practice 32, 406413.[ISI][Medline]
Malla, A. and Merskey, M. (1987) Screening for alcoholism in family practice. Family Practice Research Journal 6, 138147.[Medline]
Midanik, L. T. and Clark, W. B. (1995) Drinking-related problems in the U.S.: description and trends (1984 1990). Journal of Studies on Alcohol 56, 395402.[ISI][Medline]
Popham, R. E., Schmidt, W. and Israelstam, S. (1984) Heavy alcohol consumption and physical health problems: a review of the epidemiologic evidence. In Research Advances in Alcohol and Drug Problems, Vol. 8, Smart, R. et al. eds, pp. 149182. Plenum Press.
Regier, D. A., Narrow, W. E., Rae, D. S., Manderscheid, R. W., Locke, B. Z. and Goodwin, F. K. (1993) The de facto U.S. mental and addictive disorders service system: epidemiologic catchment area prospective 1-year prevalence rates of disorders and services. Archives of General Psychiatry 50, 8594.[Abstract]
Research Triangle Institute (1994) Software for Survey Data Analysis (SUDAAN). Version 6.30, Research Triangle Institute, Research Triangle Park, North Carolina, USA.
Rush, B. (1989) The use of family medical practices by patients with drinking problems. Canadian Medical Association Journal 140, 3539.[Abstract]
Wittchen, H. U., Robins, L. N., Cottler, L. and Sartorius, N. (1991) Cross-cultural feasibility, reliability and sources of variance of the Composite International Diagnostic Interview (CIDI) results of the multicentre WHO/ADAMHA field trials (wave I). British Journal of Psychiatry 151, 645653.
World Health Organization (1990) Mental and behavioral disorders, diagnostic criteria for research. International Classification of Diseases, Injuries and Causes of Death. World Health Organization, Geneva.