Prevalence and detection of illicit drug disorders among hospitalized patients. Michael D. Stein; Joanne Wilkinson; Nancy Berglas; Patricia O'Sullivan.

Author's Abstract: COPYRIGHT 1996 Marcel Dekker, Inc.

The objective of this study was to determine the prevalence of a lifetime history of illicit drug dependence-abuse among hospitalized patients, and to determine the rate of identification of these patients by physicians. This cross-sectional study included patient interview and chart review in an acute-care teaching hospital. The participants were 235 randomly selected inpatients with medical, neurologic, or surgical diagnoses. The prevalence of lifetime history of dependence-abuse of at least one class of illicit drug was 11.9%. Across all classes of drugs, hospitalized patients had higher prevalences than community estimates. Only 18% of patients had documentation of having been asked about illicit drug use compared to 49% having been asked about alcohol use. There was no significant difference in physicians' asking across clinical services: Medicine 21 %, Neurology 15%, Surgery 14%. Smokers, unmarried persons, and patients without a regular physician were most likely to have been asked. Screening for drug abuse is not routinely performed and documented among hospitalized patients. Wider identification of persons at risk for drug use may allow for specific physician interventions.

Full Text: COPYRIGHT 1996 Marcel Dekker, Inc.

INTRODUCTION

Illicit drug abuse is a major public health problem in the United States. The Epidemiologic Catchment Area (ECA) Study and the National Comorbidity Survey sampled communities for the lifetime prevalence of mental disorders including substance abuse and dependence and found that drug disorders other than alcohol were present in approximately 6% of respondents (1, 2). An estimated 3 million Americans had drug abuse or dependence during the preceding 6 months (1). The long-term health consequences of drug abuse or dependence are unfavorable, with specific manifestations dependent on the type of drug used (3, 4). Individuals abusing drugs have higher health-care utilization rates than nondrug-abusing individuals (5). With total social and economic costs (treatment, productivity loss) annually attributed to drug abuse at $47 billion, the incentives for recognition and treatment are great (6).

In the past decade research has focused on the prevalence of alcoholism in a variety of medical settings. High prevalence of serious drinking problems has been documented in both ambulatory and inpatient populations, far exceeding community estimates (7-10). This overrepresentation of alcoholics in medical settings has led to the recommendation for formal screening of all patients using instruments such as CAGE or SMAST as a first step toward intervention (11).

As with alcohol abuse, the prevalence of drug abuse is higher among hospitalized patients than in the general population. Physicians caring for inpatients have a heightened opportunity to identify drug users and may be in a position to intervene. While the prevalence of illicit drug abuse in the community has been estimated, the prevalence among patients admitted to the hospital has not been documented.

The aims of this study were first to determine the proportion of hospitalized patients who have a lifetime history of drug disorders across the clinical services, and second, to determine the rate of identification of these patients by physicians.

METHODS

This study was conducted at Rhode Island Hospital, a 719-bed teaching hospital affiliated with Brown University Medical School that serves southeastern New England. Rhode Island Hospital does not provide acute services in Obstetrics.

Study patients were persons at least 18 years of age randomly selected (every tenth name) from daily admission logs during July and August 1993. Within 48 hours of admission, patients were approached by the study interviewer, a medical student experienced with psychiatric interviewing. If patients were not immediately available, repeated visits to their rooms were made. Patients were excluded if they were admitted for short stay (24-48 hour) elective procedures, or if they were too ill to be interviewed. The interview included questions about diet, smoking, and exercise as well as drug use. After obtaining informed consent, the average time of the interview was 15 minutes.

The Quick Diagnostic Interview Schedule-R (Q-DIS) was used as the gold standard for lifetime history of drug abuse or dependence. The Q-DIS is a version of a highly structured Diagnostic Interview Schedule (DIS-R), an instrument that has been widely used for assessing alcohol, drug, and mental disorders on the basis of its sensitivity, specificity, and predictive power for diagnosing drug abuse or dependence as defined by DSM-III-R criteria (11, 12). The Q-DIS, adapted by the creators of DIS-R, has been compared to the DIS-R and found to have high kappas for all drug classes ([kappa] > 0.75 personal communication, K. Bucholz), and allows for a shorter interview. The Q-DIS has the identical questions and the identical probing pattern as DIS-R, but cuts short full symptom lists once a person meets diagnostic criteria. Questions regarding lifetime use of individual classes of drugs (marijuana, amphetamines, barbiturates, cocaine, opiates, hallucinogens) are asked consecutively during the Q-DIS interview; these are computer scored according to set algorithms. At the end of the Q-DIS, patients were asked whether they used any of these drugs within the past month. In addition, we collected data concerning the patient's age, gender, race, marital status, education, residence (homeless or not), cigarette smoking, type of admission (emergency, elective), and whether they have a regular physician. Interviews were conducted in English and Spanish.

After completion of the interview, the interviewer reviewed the patient's medical record to determine the clinical service the patient was admitted to (Medicine, Surgery, Neurology). The chart was also reviewed for any documentation of physician screening for drug use and for alcohol use. Any notation in the emergency room record, admission history, or daily progress notes of questions pertaining to illicit drug history, alcohol history, results of a toxicologic screen, or a blood alcohol level qualified as asking about drug use.

Overall, 240 persons were approached to become study subjects and 235 consented to study participation.

Data analysis was performed using chi-square statistics to determine associations on a P.C. version of SPSS.

RESULTS

As shown in Table 1, the sample included 45% females and 86% whites with a mean age of 40.8 years. This sample mirrors the characteristics of all admissions to Rhode Island Hospital during the study period (47% female, 83% white, mean age 44.7 years). One-third of the study participants smoked cigarettes and 35 % stated that they had a regular physician.

Table 1. Sample Characteristics (n = 235)

Variable                                 %
Female                                  45
White                                   86
Married                                 51
Employed                                61
Current smoker                          33
Homeless                                 5
Admission through emergency room        68
Has regular physician                   35
Years Education                      13.1 (2.3)
Mean Age                            40.8 (13.1)

Overall, only 18% of reviewed medical records contained information relating to patients' drug utilization (Table 2). Smokers, unmarried persons, and patients without a regular physician were most likely to have been asked about drug use, although only a minority of each group had documentation of being asked. Forty-nine percent of respondents had been asked about alcohol use.

Table 2. Factors Associated with Being Asked About Drug Use

                            n        %     p-value
Men                      24/128     19      .83
Women                    18/107     17

African-American           6/19     32      .19
Caucasian                36/216     17

Smoker                    20/78     26      .04
Nonsmoker                22/157     14

Has regular physician      7/83      8     .009
No regular physician     35/152     23

Employed                 23/142     16      .51
Unemployed                19/93     20

Married                  13/119     11      .02
Unmarried                29/126     21

Age < 45 years           31/143     22
Age >45 years             11/82     13      .08

Table 3 compares prevalence of a lifetime history of drug dependence or abuse in the Rhode Island Hospital population versus community prevalence estimates from the Epidemiologic Catchment Area Study. Across all categories of drugs, patients at Rhode Island Hospital had a higher prevalence. A history of cocaine dependence-abuse was the drug most commonly reported by respondents.

Table 3. Lifetime History of Drug Dependence-Abuse

                                  RIH(a)         ECA Study(b)
                                   (%)               (%)
Marijuana dependence-abuse         5.5               4.3
Amphetamine dependence-abuse       3.8               1.7
Barbiturate dependence-abuse       2.5               1.2
Cocaine dependence-abuse           7.6               0.2
Opiate dependence-abuse            3.4               0.7
Hallucinogen dependence-abuse      2.1               0.3

(a) Rhode Island Hospital.
(b)Epidemiologic Catchment Area Study data (1).

Table 4 shows the relationship of a lifetime history of drug dependence-abuse and recent drug use. Twenty-eight persons (11.9%) had a lifetime history of dependence-abuse for at least one drug. For all drug classes except opiates, only a minority of persons with a lifetime history of dependence-abuse had used drugs in the past month. The prevalence of illicit drug use in the past month was 7.6%.

Table 4. Relationship of Lifetime Drug Dependence-Abuse and Recent
Use

          Lifetime abuse(a)
          or dependence      Recent users(b)    Abusers still users
          n (prevalence)         n                      n
Marijuana        13              13                    .4
Amphetamines      9               0                     0
Barbiturates      6               1                     0
Cocaine          18               3                     2
Opiates           8               4                     4
Hallucinogens     5               0                     0

(a) DIS positive.
(b) Any use in last month.

There was no significant difference in physicians' documentation of asking about drug use across clinical services: Medicine 21%, Neurology 15%, Surgery 14%; p = .40. Admissions clearly related to illicit drug use (overdose, opiate withdrawal) occurred in 0.4% of total admissions. Of those persons with a history of lifetime dependence, 46% had documentation of being asked about drug use compared to 14% of persons with no history of drug dependence (p < 0001).

DISCUSSION

The prevalence of drug abuse-dependence in hospitalized patients has not been previously documented. Our results indicate that a history of drug abuse or dependence is present in 11.9% of patients admitted to our hospital. For each drug class, the prevalence of lifetime abuse-dependence for hospitalized patients exceeds the prevalence noted in previous community samples (1, 2).

The Diagnostic Interview Schedule identifies persons with a lifetime history of drug dependence or abuse (11). Screening questionnaires that frame questions in terms of lifetime use patterns may be less threatening than questions regarding current illicit drug use, although past use of illicit drugs may or may not reflect current use patterns. Yet recognizing lifetime diagnoses can be valuable to clinicians due to the risk of relapse among substance abusers. Thus, having information regarding lifetime problems may offer insights into current or future care.

When patients were asked a single question about illicit drug use in the past month, a difference was noted between estimated lifetime abuse and recent use. For all drugs, only a minority of past abusers were still using. It is impossible to know whether this reflects high recovery rates or whether patients find it easier to report having a problem in the past than acknowledging a current one. Our methods do not allow us to differentiate current drug use from current abuse or dependence. These findings suggest, however, that questions directed solely at lifetime use may not identify many persons who may have active drug problems.

The prevalence of cocaine lifetime abuse-dependence in our study is strikingly higher than that found in the ECA Study. In part, this may be due to the age of the Epidemiologic Catchment Area data. This survey, performed in the early 1980s, missed the rise in cocaine use that occurred in the late 1980s that is detected in the Rhode Island Hospital cohort (13). In part, the demographics and socioeconomic features of this hospitalized sample are not the same as those of the ECA community populations. Still, the lifetime prevalence of 7.6% is concerning, particularly given the limited long-term treatment options for cocaine dependence and high rates of return to drug use (14).

Because acute complications of drug use are not commonly the principal reason for hospital admission, persons with a history of drug abuse or dependence may be difficult to identify unless questions about drug use are asked explicitly (15). We found that only 18% of hospital admissions had any documentation of being asked about illicit drug use while 49% of patients had been asked about alcohol use. While it seems unlikely that physicians do not record answers to drug questions as thoroughly as responses to alcohol questions, there are many reasons physicians may not ask about drug use. First, they may judge prevalence of drug use to be far lower than alcohol use and therefore not time worthy. Second, there are no well-validated brief drug screens analogous to the CAGE that have been disseminated or publicized (16, 17). One difficulty in developing such a screen is differentiating between particular classes of illicit drugs. Third, perhaps physicians may not trust patient responses to even the simplest drug-related questions (Have you ever used illicit or illegal drugs?), and so may avoid asking altogether. Finally, because physicians are not trained in the diagnosis and treatment of drug disorders, and may not know what to do with a positive response to a drug screen, or alternatively are pessimistic about the rehabilitation prospects for a substance abuser, they do not ask at all (18).

While only 18% of all patients had documentation of being asked about drug use, physicians identified 46% of persons who were Q-DIS-positive and had a lifetime history of dependence-abuse. Therefore, physicians may be using information or clues not captured by this study that allow them to target their screening at a high-risk subgroup. One could argue that physicians should systematically screen all hospitalized patients for drug problems in a population where the disorder prevalence is 11.9%. For a drug screen to have widespread use, it would presumably be brief and have high predictive value.

As a cross-sectional survey, this study relies on retrospective reports to assess the prevalence of lifetime disorders. In addition, diagnostic assessments are based on a single structured interview with a nonclinician. While one limitation of research in drug abuse remains the determination of "cases," we used a shortened version of the well-validated Diagnostic Interview Schedule as a case finder. The use of this instrument is unlikely to result in false-positives; more likely, an underestimate of prevalence may occur (18). Nonetheless, improved diagnostic precision might have been accomplished using multiple or ancillary information from informants. The prevalences described here, therefore, should be viewed as estimates rather than definitive diagnoses.

The primary limitation in evaluating physicians' detection of drug use is that physicians may not document what they ask patients, particularly if a particular question results in a negative response. Incomplete charting is not uncommon in medical practice (19). However, physicians documented alcohol questions nearly three times as often as drug questions, suggesting that drug questions are simply not considered part of the routine interview.

The acute-care hospital may be only one of many places where screening for drug abuse problems could be worthwhile. As has been noted, simple physician warnings about alcohol use in the medical setting may have therapeutic value (20); this may be true of illicit drug use as well. Certainly identifying those persons who are currently using may allow for physician intervention and motivation toward appropriate drug treatment in any medical context.

REFERENCES

(1.) Regier, D. A., Farmer, M. E., Rae, D. S., et al., Comorbidity of mental disorders with alcohol and other drug abuse, JAMA 264:2511-2518 (1990). (2.) Kessler, R. C., McGonagle, K. A., Zhao, S, et al., Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States, Arch. Gen. Psychiatry 51:8-19 (1994). (3.) Warner, E. A., Cocaine abuse, Ann. Intern. Med. 119:226-235 (1993). (4.) Hser, Yi., Angoin, M. D., and Powers, K., A 24-year follow-up of California narcotic addicts, Arch. Gen. Psychiatry 50:577-584 (1993). (5.) Stein, M. D., O'Sullivan, P. S., Ellis, P., et al., Utilization of medical services by drug abusers in detoxification, J. Substance Abuse 5:187-193 (1993). (6.) Kamerow, D. B., Pincus, H. A., and Macdonald, D. 1., Alcohol abuse, other drug abuse, and mental disorders in medical practice, JAMA 255:2054-2057 (1986). (7.) Cleary, P. D., Miller, M., Bush, B. T., et al., Prevalence and recognition of alcohol abuse in a primary care population, Am. J. Med. 85:466-471 (1988). (8.) Buchsbaum, D. G., Buchanan, R. G., Lawton, M. J., and Schnoll, S. H., Alcohol consumption patterns in a primary care population, Alcohol and Alcoholism 26:215-220 (1991). (9.) Cyr, M. G., and Wartman, S., The effectiveness of routine screening questions in the detection of alcoholism, JAMA 259:51-54 (1988). (10.) Moore, R. D., Bone, L. R., and Geller, G., Prevalence, detection, and treatment of alcoholism in hospitalized patients, JAMA 261:403-407 (1989). (11.) Robins, L., Helzer, J., Croughan, J., et al., National Institute of Mental Health Diagnostic Interview Schedule: Its history characteristics, and validity, Arch. Gen. Psychiatry 38:381389 (1981). (12.) Griffin, M. L., Weiss, R. D., Mirin, S. M., et al., The use of the diagnostic interview schedule in drug-dependent patients, Am. J. Drug Alcohol Abuse 13:281-291 (1987). (13.) National Household Survey on Drug Abuse: Population Estimates 1991. National Institute on Drug Abuse, Department of Health and Human Services ADM 92-1887, Rockville, Maryland, 1991. (14.) Rawson, R. A., Obert, J. L., McCann, M. J., et al., Cocaine abuse treatment: A review of current strategies, J. Substance Abuse 3:457-491 (1991). (15.) Dans, P. E., Matricciani, R. M., Otter, S. E., and Reuland, D. S., Intravenous drug abuse and one academic medical center, JAMA 263:3173-3176 (1990). (16.) Rost, K., Burnham, M. A., and Smith, G. R., Development of screeners for depressive disorders and substance abuse disorder history, Med. Care 31:189-200 (1993). (17.) Gavin, D. R., Ross, H. E., and Skinner, H. A., Diagnostic validity of the drug abuse screening test in the assessment of DSM-III drug disorders, Br. J. Addict. 84:301-307 (1989). (18.) O'Connor, P. G., Bigby, J., and Gallagher, D., Substance abuse and AIDS: A faculty development program for primary care providers, J. Gen. Int. Med. 8:266-268 (1993). (19.) Jencks, S. F., Recognition of mental distress and diagnosis of mental disorder in primary care, JAMA 253:1903-1907 (1985). (20.) Walsh, D. C., Hingson, R. W., Merrigan, D. M., et al., The impact of a physician's warning on recovery after alcoholism treatment, JAMA 267:663-667 (1992).

Michael D. Stein, M.D., To whom correspondence and reprint requests should be addressed at the Division of General Internal Medicine, Rhode Island Hospital, 593 Eddy Street, Providence, Rhode Island 02903. Telephone: (401) 444-5248. FAX: (401)