Generalist Physicians and Addiction Care: Title and subTitle BreakFrom Turfing to Sharing the Turf

For many years physicians shunned addiction care—traditionally the turf of chemical dependency counselors and mental health professionals. But over the past quarter century, physicians have become more involved in research and clinical care of patients with substance use disorders. Generalist physicians in particular have been drawn to the addiction field for 4 reasons. First, there is growing awareness of the epidemiology of substance use disorders. Epidemiologic studies indicate high lifetime prevalence of substance use disorders and, particularly in regard to alcohol, that persons with lower-severity problems greatly outnumber those who meet formal criteria for abuse or dependence.¹ Generalist physicians have long been aware of the episodic infectious and traumatic complications that bring such patients to emergency departments and of the difficulties in engaging such patients in primary care.² Nonetheless, more than two thirds of individuals with substance use disorders have seen a primary care physician in the previous 6 months, offering the possibility of beginning a therapeutic relationship.³

Second, overwhelming evidence has emerged that drug addiction can be treated successfully, and a number of studies have revealed little difference in outcomes as a function of treatment intensity.⁴ At the same time, brief interventions appear to be efficacious and feasible within the time limits of primary care practice.⁵ Third, generalists, who with the advent of selective serotonin reuptake inhibitors became more comfortable providing the initial pharmacological treatment for some psychiatric disorders, soon discovered underlying substance problems. Finally, during the 2 decades of the HIV/AIDS epidemic, generalist physicians have directly confronted the sequelae of addiction in the management of HIV-infected patients.⁶

Increasingly, addiction is recognized as a chronic relapsing disease that requires diverse service components for successful treatment over the long term.⁷ In addition, increasing evidence supports the treatment of substance use at empirically defined levels as a "risk factor" worthy of intervention—much like clinically silent hypertension warrants treatment to prevent adverse consequences.⁸ Although in theory the model of comprehensive, coordinated, and continuous primary care suits the management of the continuum of substance use disorders,^{9 - 10} the current system delivers fragmented care.¹¹ Little evidence exists to guide how primary care might fit into a better system of care for patients with addictive disorders. What is the role of the primary care physician at the various stages of substance use severity, and should this role be different for different substances? Is a wider role for primary care physicians feasible or desirable, given their burden of responsibility for an increasing number of disorders? Is a primary care focus more or less desirable than a disease management focus? Two studies in this issue of THE JOURNAL begin to answer some of these questions.

In one study, Fiellin et al¹² recruited and trained 6 primary care physicians to provide office-based methadone maintenance and primary care to a select group of opiate-dependent patients. Patients stable in methadone maintenance were randomly assigned to receive methadone in their current treatment program (n = 24), or through 1 of these 6 physicians (n = 22). Patient eligibility criteria were strict and applied to only 12% to 14% of patients in the methadone treatment program. These criteria included having at least 12 drug-free months, as well as stable transportation, income, and living situations. At follow-up, nearly half of patients in both study groups returned to drug use, indicating the chronic relapsing nature of addiction. However, patients assigned to primary care practices were significantly more satisfied with their care than were those in the methadone program, a finding that likely has great meaning given the small sample size. These differences in satisfaction reveal as much about the burdensome nature of methadone clinics as they do about the nonstigmatizing care delivered in office practices. Satisfaction has been shown to be associated with treatment adherence in other diseases, and duration of treatment (a likely effect of satisfaction) is known to be the key to long-term addiction outcomes.

As in other diseases, some patients with substance use disorders may need prolonged care from a specialist, while others can be cared for by generalists with specialist consultation as needed.⁹ This model of collaboration is well developed for other chronic conditions. For example, most generalists can manage patients with stable angina and refer patients with unstable coronary syndromes to a cardiologist. Such collaboration is rare in the management of addiction. However, the findings of Fiellin et al¹² suggest that some stable opioid-dependent patients can be managed as well in the primary care setting (with as-needed consultation) as in a specialty methadone clinic, thereby freeing scarce clinic slots for less stable patients.¹³

Whereas Fiellin et al integrated care at the level of the individual practitioner, Weisner et al, in another article in this issue,¹⁴ studied 2 organizational models of substance abuse treatment and primary care in a large health maintenance organization (HMO). In a randomized design, 592 adult patients seeking addiction treatment received either integrated care, where addiction care and primary care were co-located, or usual treatment for this HMO, where addiction care and primary care were separated. Although nearly two thirds of patients reached and maintained 30 days of abstinence, no difference in abstinence rates could be detected between these groups. The integrated care patients had significantly greater reductions in emergency department and hospital use, but because medical service use was low and addiction treatment utilization high, the savings did not offset their higher overall cost per member per month. As a service delivery model for a heterogeneous group of addicted patients, integrated care was no more effective yet was more costly than usual care.

However, a preplanned subgroup analysis suggested that patients with substance abuse–related medical conditions in the integrated care group had significantly greater abstinence rates and longer periods of abstinence than patients receiving usual care. These improvements came at a price, estimated as $1581 over 6 months per abstinent, medically comorbid patient. The usual cautions apply regarding subgroup analyses.¹⁵ Yet this cost, for a period of time when an individual can reorganize his or her newly substance abuse-free life, does not seem unreasonable. Integrated care might prove even more cost-effective in uninsured or publicly insured groups that, unlike the Kaiser Permanente cohort studied by Weisner et al, lack primary care and have high rates of potentially preventable emergency department and hospital use.¹⁶ Although the integration of medical and substance abuse treatment may be cost-effective for the medically ill substance-abusing subpopulation,¹⁷ at a practical level, identifying these individuals at treatment entry and matching them to programs with co-located primary care may prove difficult to implement.¹⁸

The authors of these 2 studies provide careful empirical data on the question of the impact of primary care—a question that has circulated in the addiction field for a decade. The benefits and costs of involving primary care physicians in the care of patients with other chronic diseases rarely have been tested as rigorously. The controversy surrounding the role of generalists in addiction treatment seems similar to the debate about models of care for persons with HIV/AIDS.¹⁹ Certain skills related to providing care for HIV-infected patients are reasonable to expect of all generalists, including risk screening, testing and counseling, staging, and perhaps beginning antiviral treatment. But no single model of care will work for all patients across all clinical settings. In the HIV/AIDS arena, a relatively small number of physicians—willing to overlook financial and administrative disincentives—provide the great majority of care to patients with more complicated clinical pictures. The system of care for substance-using patients will likely take shape in the same way as new therapies become available.

Continuity of care from a well-meaning physician does not necessarily result in better patient health habits; think of the difficulties in changing smoking behaviors, diet, or exercise. Because patterns of drug and alcohol use fluctuate, a major effect of many substance abuse interventions is motivational. Physicians—whether generalists or addiction specialists—who provide an ongoing, trusting relationship, if they have learned the requisite skills, may be able to evoke from patients reasons for changing and commitment to change that endure. It is rewarding but challenging work. Time will tell how many generalist physicians really want a share of this turf.