By specialty, percentage of physicians who reported that they usually or always recommend a 12-step program, such as Narcotics Anonymous, offer referral to a mental health professional or social worker, offer referral to a chemical dependency treatment program, or counsel the patient without other consultation or referral for diagnosed illicit drug abuse. Error bars indicate SE.
Friedmann PD, McCullough D, Saitz R. Screening and Intervention for Illicit Drug AbuseA National Survey of Primary Care Physicians and Psychiatrists. Arch Intern Med. 2001;161(2):248–251. doi:10.1001/archinte.161.2.248
Illicit drug abuse causes much morbidity and mortality, yet little is known about physicians' screening and intervention practices regarding illicit drug abuse.
We mailed a survey to a national sample of 2000 practicing general internists, family physicians, obstetricians and gynecologists, and psychiatrists to assess their screening and intervention practices for illicit drug abuse.
Of 1082 respondents (adjusted response rate, 57%), 68% reported that they regularly ask new outpatients about drug use. For diagnosed illicit drug abuse, 55% reported that they routinely offer formal treatment referral, but 15% reported that they do not intervene. In multivariate logistic regression models, more optimal screening and intervention practices were associated with psychiatry specialty, confidence in obtaining the history of drug use, optimism about the effectiveness of therapy, less concern that patients will object, and fewer perceived time constraints.
Most physicians reported that they ask patients about illicit drug use, but a substantial minority inadequately intervene in diagnosed drug abuse. Initiatives to promote physician involvement in illicit drug abuse should include strategies to increase physicians' confidence in managing drug problems, engender optimism about the benefits of treatment, dispel concerns about patients' sensitivity regarding substance use, and address perceived time limitations.
ILLICIT DRUG ABUSE causes much morbidity and mortality,1 and primary care physicians and psychiatrists are well positioned to identify and treat it.2 Few studies have examined physician screening for illicit drug abuse, but much research supports health benefits from substance abuse screening and intervention for alcohol or tobacco abuse3; these benefits are likely to extrapolate to illicit drug abuse. In addition, the Substance Abuse Coverage Study4 suggested that decreased criminal justice involvement would have the individual and societal benefit of greater recognition of and treatment referral for illicit drug abuse. Some professional and governmental organizations5- 8 recommend screening for drug abuse, while others9 equivocate, in large part because of the low prevalence of drug abuse in primary care settings and insufficient direct evidence of the benefits of screening. Although physicians adopt some screening and intervention practices despite a lack of consensus, little is known about such practices regarding illicit drug abuse. Thus, we conducted a national survey of primary care physicians and psychiatrists to examine screening and intervention practices for illicit drug abuse and the barriers to their implementation.10
As previously reported,11 we mailed a survey to a national systematic sample of family physicians, general internists, obstetricians and gynecologists, and psychiatrists, from September 15, 1997, to March 15, 1998. The institutional review board of the University of Chicago, Chicago, Ill, approved this research.
Using 5-point Likert scales, with responses from "never" to "always," the survey assessed how often the physician asked new adult outpatients about illicit drug use and how often they used 4 interventions in cases of diagnosed illicit drug abuse. The 4 interventions were recommending a 12-step program, offering referral to a mental health professional, offering referral to a chemical dependency treatment program, and counseling the patient without other consultation or referral. Dichotomous variables created from these questions indicated whether the physician usually or always offered any of the interventions for illicit drug abuse.
On 5-point Likert scales, the survey assessed physicians' confidence in obtaining patients' history of drug use, their interest in caring for drug-using patients, and the importance of 8 barriers to screening for substance abuse. The brief Substance Abuse Attitude Survey12,13 evaluated physician attitudes toward substance-abusing patients. Five subscales (permissiveness, nonmoralism, nonstereotyping, optimism in the effectiveness of treatment, and treatment intervention) are standardized along a 10-point continuum, with 10 representing the most positive attitudes. The questionnaire also inquired about the outpatient practice setting, including the percentage of outpatients who are women, Hispanic, black, uninsured, Medicaid insured, or aged older than 50 years or have a history of substance abuse.
We drew a self-weighted, national, systematic sample of 500 family physicians, 500 general internists, 500 obstetricians and gynecologists, and 500 psychiatrists with greater than 50% clinical practice time from the AMA (American Medical Association) Physician Masterfile,14 a listing of all licensed physicians. Of the 2000 mailings, 107 could not be delivered. The remaining 1893 included 856 eligible and 226 ineligible (retired or nonpracticing) respondents and 811 nonrespondents, yielding an adjusted response rate of 57%.15 Sex, geographic location, and age of respondents were similar to those listed in the AMA Physician Masterfile. The 3 mailing waves revealed similar rates of illicit drug screening, making response bias unlikely.16
We entered the explanatory variables (P≤.25) into stepwise multivariate logistic regression models to determine correlates (P≤.05, 2-tailed) of whether the physician reported that they usually or always ask about drug use or offer any intervention for diagnosed illicit drug abuse. Models controlled for the physicians' specialty, sex, age, and board certification. Statistical analysis was performed using SAS statistical software (version 6.12; SAS Institute, Cary, NC).
Of the 856 eligible respondents, 243 (28%) were family physicians; 195 (23%), internists; 222 (26%), obstetricians and gynecologists; and 196 (23%), psychiatrists. The mean ± SD age and length of practice were 46 ± 12 years and 19 ± 13 years, respectively; 29% were women, and 62% were board certified. There were 23% from the Northeast, 24% from the Midwest, 30% from the South, and 22% from the West. Twenty-six percent practiced in a group; 29%, hospital; 22%, solo; 4%, academic setting; 0.1%, health maintenance organization; and 19%, unclassified or other.
Most physicians (68%) reported that they usually or always ask new outpatients about illicit drug abuse. This reported inquiry was associated with psychiatry or obstetrics and gynecology compared with family medicine, younger physician age, lack of board certification, greater confidence in obtaining the patient's history of drug use, greater optimism in the effectiveness of therapy, less agreement with the statement, "patients don't want to be asked about substance use," fewer perceived time constraints, and a practice with fewer older but more black patients (Table 1). There was a trend toward an association between more screening and the percentage of patients who are current or past alcohol or drug abusers (P = .05).
For patients with diagnosed illicit drug abuse, 55% of physicians reported that they usually or always offer referral to a chemical dependency treatment program and 61% reported that they recommend a 12-step program. A similar proportion reported that they offer referral to a social worker or mental health professional, and few physicians reported that they counsel the patient without referral. Of the specialties, psychiatrists were most likely to recommend a 12-step program, while family physicians were slightly more likely to offer referral to a formal treatment program (Figure 1). Although most physicians reported that they offer more than 1 of these interventions, 15% reported that they do not routinely offer any intervention. Psychiatrists, female, and board-certified physicians were more likely to intervene, while obstetricians and gynecologists were least likely (Table 1). Greater confidence in obtaining the patient's history of drug use, fewer perceived time constraints, and fewer patients with a history of substance abuse were also associated with a greater propensity to intervene.
In this national survey, 32% of primary care physicians and psychiatrists reported that they do not inquire routinely about illicit drug use. This finding is not surprising given the lack of brief, well-validated screening tools for illicit drug abuse, the dearth of evidence for benefits of screening, and conflicting recommendations.5- 9 Only 55% of physicians reported that they routinely recommend formal addiction treatment to drug-abusing patients, and a substantial minority reported that they do not regularly intervene at all, despite a consensus favoring referral of drug-abusing patients to specialized treatment.9 Psychiatrists were most likely to intervene but, like many of their primary care colleagues, were more likely to refer drug-abusing patients to a 12-step program than to a formal treatment program. This practice pattern is contrary to the available evidence, which provides strong support for formal addiction treatment, especially the use of methadone maintenance, but few data regarding the effectiveness of 12-step programs for drug abuse.4
Psychiatrists and obstetricians and gynecologists were most likely to screen for drug use, but obstetricians and gynecologists were least likely to intervene. This cross-sectional study cannot discern whether differences in training or stigmatizing attitudes mediated the influence of specialty. In distinction to prior research, stigmatizing attitudes contributed little to inadequate screening and intervention practices toward substance-abusing patients.13,17 The exception was in physicians who are optimistic about the effectiveness of treatment, a trait that increased screening. Expectations about treatment have been previously found to be related to drug abuse screening and intervention.18- 21 This finding highlights the need for research demonstrating downstream benefits from such screening.
Younger age, which is highly correlated with more recent medical school graduation, was also associated with screening.20 Curiously, board certification was correlated with less propensity to screen for illicit drug abuse but more propensity to intervene. We can only speculate that board-certified physicians do not screen because of conflicting recommendations but are more likely to follow the solid consensus favoring intervention.9
A correct perception of the lower prevalence of illicit drug abuse in the older population might underlie decreased screening for illicit drug abuse among physicians who treat older patients. Greater screening among physicians who treat more black patients might reflect the negligible increased prevalence of drug use among the black population (7.5% of black individuals vs 6.4% of white individuals).22 Alternatively, it might indicate prejudice regarding susceptibility to drug abuse.
Physicians' confidence in obtaining the patient's history of drug use was correlated with both screening and intervention. Self-efficacy has been previously associated with substance abuse practices.11,20 Development of brief, accurate screening tools might further augment confidence in obtaining the history of drug use. Concern that patients do not want to be asked about drug use implies a lack of familiarity with reports that patients willingly disclose sensitive information to physicians.23,24 The association between screening and intervention and perceived time constraints suggests the need for prescreening questionnaires or chart prompts and the need for training physicians and other staff in brief therapeutic strategies, to facilitate screening and intervention in busy clinical settings. We cannot determine the direction of the trend association between more substance-abusing patients in the physician's practice and a greater propensity to screen. One can speculate that patients with easily recognized histories of substance abuse might have more severe addictive disorders and that prior treatment failures might have left the physician skeptical about intervention.
These findings require validation in other physician samples. A major limitation of this study is its reliance on reported, not actual, practices. Also, social desirability may have biased respondents' reports of their screening and intervention practices. This study's strengths are national representation and a response rate comparable to other physician surveys.25 The findings suggest that initiatives to promote physician involvement with illicit drug abuse should include strategies to increase physicians' confidence in managing these problems, to engender optimism about the benefits of addiction treatment, to dispel physician concerns about patients' sensitivity about substance issues, and to address perceived, rather than actual, time constraints.
Accepted for publication June 16, 2000.
This research was funded by a grant from the Home Health Care Research Program and the summer research program of the Pritzker School of Medicine (Ms McCullough), the University of Chicago, Chicago, Ill; Generalist Physician Faculty Scholar awards from the Robert Wood Johnson Foundation, Princeton, NJ (Drs Friedmann and Saitz); Mentored Clinical Scientist Career Development Award K08-DA00320 (Dr Friedmann), and grant R01-DA10019 (Dr Saitz) from the National Institute on Drug Abuse, and grant R01-AA10870 from the National Institute on Alcohol Abuse and Alcoholism (Dr Saitz), Bethesda, Md; and faculty development grant T26-SP08355 from the Center for Substance Abuse Prevention, Rockville, Md (Dr Saitz).
We thank the participating physicians.
Corresponding author: Peter D. Friedmann, MD, MPH, Division of General Internal Medicine, Rhode Island Hospital, 593 Eddy St, Providence, RI 02906 (e-mail: firstname.lastname@example.org).