A Test of Recovery Management: Or was it?

JAMA Blog Photo.jpgAn important study was just published in the Journal of the American Medical Association.  Dr. Richard Saitz and colleagues conducted a trial comparing the results of 563 persons meeting criteria for alcohol or drug dependence randomly assigned to a chronic care management (CCM) protocol or usual primary care (UPC).  The CCM approach included sustained contact with a clinical care team that provided clinical assessment, primary care coordination, referral for addiction treatment (including pharmacotherapy and mutual aid referral), referral for mental health services, motivational enhancement therapy, relapse prevention counseling, proactive contacts following any missed service appointment, follow-up telephone contact, and emergency access over the 1-year study period.  The UPC approach included an appointment with a primary care physician, a referral list of addiction treatment providers, and a phone number to access motivational interviewing sessions.  More than 40% of the subjects in both arms of the trial exhibited improvements in health status at 12-month follow-up, but there were no significant differences in outcomes between the higher intensity CCM and lower intensity UPC groups based on rates of past-30 day abstinence (44% vs. 42%), addiction severity, health-related quality of life, or acute healthcare utilization.     

Findings from the Saitz et al. study results will be disappointing for those who have advocated a major shift in the design of addiction treatment from models of acute intervention to models of sustained recovery management (RM), and the results will be welcomed by those who have resisted these calls to transform the prevailing design of addiction treatment.  The CCM/UPC study deserves close scrutiny by both advocates and critics of RM due to the rigorous design of this study, e.g., large sample, randomization, use of ingredients with proven efficacy in the management of other chronic health conditions. But before this study is used to discount the move toward RM and to imbed models of RM within larger recovery-oriented systems of care (ROSC), several points should be considered related to the differences in the CCM model that was evaluated and recommended approaches to RM. 

There are four critical differences between addiction-focused disease management protocols used in primary medicine to treat such conditions as diabetes, heart disease, and asthma and RM models being advocated within the addictions field.  The neutral outcomes of the CCM/UPC comparison study leave unanswered the question of whether RM models might exhibit far greater efficacy than both CCM and UPC approaches to the management of severe alcohol and other drug problems.  Consider these differences between CCM and RM. 

CCM/RM Settings:  The CCM model tested in the CCM/UPC study was nested within a primary care clinic–a setting in which it would be very difficult to create a fully developed culture of recovery capable of engaging, retaining, and supporting individuals and families in long-term recovery.  In contrast, RM models are being tested within specialized addiction treatment facilities and recovery community organizations.  The extent to which long-term recovery outcomes differ across such service settings remains unclear and deserves rigorous evaluation.

CCM/RM Personnel:  The CCM multidisciplinary team in the just-published study consisted primarily of a nurse care manager and a social worker, with consultations available from internists and an addiction psychiatrist. Conspicuously absent from this list and key members of RM support teams are peer-recovery support specialists (e.g., recovery coaches), volunteers and alumni who are in recovery, and culturally indigenous healers.  These roles perform critical functions within the RM approach, including modeling the reality of long-term addiction recovery, providing stage appropriate recovery education and assertive linkage to indigenous recovery community institutions, e.g., recovery mutual aid, recovery community centers, recovery homes, recovery schools, recovery ministries, recovery cafes, and other recovery support institutions.  Whether such differences in service team composition (and in particular the inclusion of recovering people as staff and volunteers within these teams) affect long-term recovery outcomes deserve rigorous evaluation.         

Core CCM/RM Ingredients:  CCM and RM share many characteristics (e.g., assertive engagement and motivational strategies; tri-directional integration of primary care, addiction treatment, and mental health services; assertive follow-up), but the CCM protocol in the Saitz et al. study lacked key elements of the RM model, including diverse recovery representation in the design, conduct, and evaluation of the service model; strengths-based assessment (focus on recovery capital); patient-developed recovery plans; assertive engagement of family and extended family; peer-based recovery coaching; assurance of evidence-based addiction treatment options; minimal service thresholds (at least 90 days of support across levels of care); nesting extended service delivery within the natural environment of the patient/family;  mobilization of indigenous recovery support resources; assertive versus passive linkage to recovery mutual aid groups and other recovery community support institutions; and commitment for prolonged monitoring, support (specifically targeting critical windows of vulnerability) and, if and when needed, early re-engagement and recovery re-stabilization.  Also missing from the CCM and other disease management models that is a prominent element of RM proposals are adaptations of service designs for adolescents, communities of color and other special populations.

Duration of CCM/RM Interventions:  The CCM/UPC study measured outcome over a one-year time period.  The lack of differences between the CCM and UPC groups could have been influenced by this short window of time.  The presence and severity of addiction ebb and flow over what all too often are prolonged addiction careers.  While brief interventions (the UPC intervention) and the assessment protocol that were part of the CCM/UPC study may exert short-term periods of stabilization, more sustained and intensive interventions may be required to sustain and extend those gains to the goal of self-managed, long-term recovery. RM models are based theoretically on the proposition that the durability set point for addiction recovery (the point at which the risk of future lifetime addiction recurrence drops below 15%) is 5 years of stable recovery and suggest extending recovery checkups and other recovery supports as needed for that critical window of time.  All “chronic care” models of intervention and support are best evaluated within the lens of this longer time perspective. 

Two models of sustained support are emerging for the management of severe and complex substance use disorders–a disease management model adapted from primary medicine protocols used to manage a broad spectrum of chronic illnesses and a recovery management model drawn more specifically from the collective experience of addiction treatment and recovery community organizations. (Elements of the latter have already been rigorously evaluated and have demonstrated clinical and cost-effectiveness–see references).  Dr. Saitz and his colleagues have done the addictions field a great service by testing the adaptation of a CCM from primary medicine to addiction.  Now it is time to test the CCM model against models of sustained RM that have risen specifically within the addictions field.  Such a trial should include the broad spectrum of outcome measures used in the CCM/IPC study.  It is my prediction that the distinctive ingredients noted above will contribute to superior outcomes of RM models in such a trial. But such a conclusion awaits results of a randomized trial as rigorous as the CCM/UPC comparison study.     

Further Reading

Kelly, J. & White, W.  (Eds., 2011) Addiction recovery management: Theory, science and practice.  New York:  Springer Science.

Dennis, M. L., & Scott, C.K. (2007). Managing addiction as a chronic condition. Addiction Science & Clinical Practice, 4(1), 45-55. 

Dennis. M., Scott, C. K., & Funk, R. (2003).  An experimental evaluation of recovery management checkups (RMC) for people with chronic substance use disorders.  Evaluation and Program Planning, 26(3), 339-352.

Dennis M. L., & Scott, C. K. (2012).  Four-year outcomes from the Early Re-Intervention (ERI) experiment using Recovery Management Checkups (RMCs). Drug and Alcohol Dependence, 121, 10-17.

McCollister, K.E., French, M.T., Freitas, D.M., Dennis, M.L., Scott, C.K., & Funk, R.R. (in press).  Cost-Effectiveness analysis of Recovery Management Checkups (RMC) for adults with chronic substance use disorders: evidence from a four-year randomized trial, Addiction.

Scott, C. K., Foss, M. A., & Dennis, M. L. (2005).  Pathways in the relapse–treatment–recovery cycle over 3 years.  Journal of Substance Abuse Treatment, 28(Supplement 1), S63-S72.

White, W. (2008).  Recovery management and recovery-oriented systems of care:  Scientific rationale and promising practices.  Pittsburgh, PA:  Northeast Addiction Technology Transfer Center, Great Lakes Addiction Technology Transfer Center, Philadelphia Department of Behavioral Health & Mental Retardation Services