ASAM PPC

A standardized treatment matching tool that has been in use since the early 1990’s is the American Society of Addiction Medicine Patient Placement Criteria (ASAM-PPC). The ASAM criteria allow a “clinician to systematically evaluate the severity of a patient’s need for treatment along six dimensions, and then utilize a fixed combination rule to determine which of four levels of care a substance abusing patient will respond to with the greatest success.”(Turner et al, 1999). The four levels of care are: Outpatient Treatment (level 1), Intensive Outpatient/Partial Hospitalization (level 2), Medically Monitored Intensive Inpatient Treatment (level 3), and Medically Managed Intensive Inpatient Treatment (level 4). (Kosanke et al, 2002). The hope is that by reducing mismatches involving overtreatment, clinics are able to allocate resources more efficiently and by reducing mismatches involving undertreatment they reduce the risk of adverse treatment outcomes.

Questions have been raised as to the validity and effectiveness of the ASAM-PPC under typical clinic settings. McKay et al (1997) performed a study to determine the predictive validity of the ASAM-PPC for inpatient versus intensive outpatient care. Using ASAM-PPC criteria the study determined whether the level of care received by 159 cocaine-dependent and 133 alcohol-dependent male patients was correctly matched. Patients were followed up at 3, 6 and 12 months post-rehabilitation and matched and mismatched clients were compared. There were some positive results when looking at short-term outcomes. Cocaine patients correctly matched to inpatient care had lower drop-out rates then mismatched to day care. Also, cocaine patients correctly matched to either inpatient or outpatient care had better drinking outcomes at 3 months then those who received mismatched care. But the authors feel that the overwhelming evidence did not support the predictive validity of the ASAM-PPC with regards to differentiating level 2 or level 3 treatment. Both alcohol and cocaine patients correctly matched to the appropriate level of care did not have significantly better outcomes then those who were mismatched. Also, alcohol patients determined by the ASAM-PPC to need inpatient care had no better results than those mismatched to day care. For cocaine patients, those correctly matched to intensive treatment did have slightly better results than those mismatched but they were determined not to be of statistical significance. The authors do point out that the ASAM-PPC is an improvement over previous matching protocols (Cleveland Criteria) in promoting more cost-effective treatments. However, they feel that in regard to the specific population studied, the criteria for inpatient care may still be too broad.

Another concern was based on whether the complexity of the ASAM criteria would yield acceptable results when administered by the average clinician. Turner et al (1999) studied a computerized decision-making algorithm that automatically performs the complex information integration required by the ASAM-PPC criteria. One variable in determining the effectiveness of any assessment tool is how consistently each counselor performs the screening. If a computerized ASAM-PPC is shown to be as effective as one given by a trained intake counselor, hypothetically you would be able to remove the human element which might lead to inconsistent results when determining level of care. 593 substance abusing adults were assessed using the computerized ASAM-PPC and results were analyzed to determine whether the level of care assignments showed significant differences on a variety of clinical measures. According to the authors, “the algorithm showed acceptable discrimination between each of three ASAM Levels of Care across numerous clinical subscales.” The authors conclude that the computerized model succeeded in integrating diverse clinically relevant factors within the level of care designations.

Magura et al (2003) performed a study examining the predictive validity of the ASAM PPC for matching alcoholism patients to different levels of care. Recommendations for level of care were obtained using the standardized computer algorithm mentioned above and a clinical evaluation protocol. The hypotheses were that patients matched to the recommended level of care would have better outcomes than those mismatched to undertreatment and patients mismatched to overtreatment would have no better outcomes than those matched to the appropriate level of care. The data supported the original hypotheses (matching to level of care is optimal, undertreatment is clinically harmful and overtreatment is a waste of resources). Overall, the mean number of days of any alcohol use during the last 30 days did decrease for all groups, though those with matched care and overtreatment had better results. Also, the computer driven algorithm assigned patients to overtreatment more frequently than clinician-recommended levels of care. The authors conclude that in general the ASAM-PPC was effective in reducing both undertreatment and overtreatment.

Kosanke et al (2002) examined the effectiveness of the ASAM-PPC in a substance abuse treatment program which tried to minimize barriers to treatment. All patients were able to enter treatment in a timely manner (immediately for residential treatment, approximately three days for outpatient) and treatment slots were available in all levels of care offered (levels 1-3). The study took place at a treatment center that accepts both public and private insurance coverage. Of the 281 patients in the study, overall 88% showed up for their initial treatment appointment. The authors also assessed the amount of mismatches in treatment placement and whether patients received overtreatment or undertreatment. Clients were matched correctly 72% of the time. When mismatched, patients were more likely to receive overtreatment than undertreatment (59% to 41%). The reasons given for overtreatment were availability of Medicaid coverage for inpatient treatment, referral sources’ philosophy of gradually “stepping down” from inpatient detoxification, social pressures on patients and mandated treatment. The reasons for under treatment were work schedule conflicts, patient reluctance, insurance coverage and interference with family or personal responsibilities. One problem this study raises is that in the “real world” not all clinics will be able to offer immediate access to the appropriate level of care as determined by a counselor. Availability of treatment slots may impact where patients are placed. Also, having to wait for treatment may reduce initial engagement rates.

The initial results on the validity of the ASAM-PPC are rather mixed. There were definitely positive results reported in placing patients in the most appropriate care as well as cost-effective benefits that may be realized. And the ASAM-PPC was shown to be successful as a computerized model, potentially removing human error in treatment matching. The one common theme in the literature reviewed was the call for further research into the validity of this widely used assessment tool.