1 The Science of Recovery Management Michael L. Dennis, Ph.D. Chestnut Health Systems 720 W. Chestnut, Bloomington, IL 61701, USA Presentation at “2007 National Association of Addiction Treatment Providers (NAATP) Conference”, May 20-23, 2007, San Diego, CA. The opinions are those of the authors and do not reflect official positions of the association or government. Available on line at or by contacting Joan Unsicker at 720 West Chestnut, Bloomington, IL 61701, phone: (309) , fax: (309) , This presentation was supported by funds from NIDA grant no. R37-DA11323, and R01 DA15523 and SAMHSA/CSAT contract no The opinions are those of the authors do not reflect official positions of the government or ATTCs. Please address comments or questions to the author at or A copy of these slides will be posted at and the conference

2 Problem and Purpose Over the past several decades there has been a growing recognition that a subset of substance users suffers from a chronic condition that requires multiple episodes of care over several years. This presentation will present 1.Epidemiological data to quantifying the chronic nature of substance disorders and how it relates to a broader understanding of recovery 2.The results of two experiments designed to improve the ways in which recovery is managed across time and multiple episodes of care.

3 Severity of Past Year Substance Use/Disorders (2002 U.S. Household Population age 12+= 235,143,246) Dependence 5% Abuse 4% Regular AOD Use 8% Any Infrequent Drug Use 4% Light Alcohol Use Only 47% No Alcohol or Drug Use 32% Source: 2002 NSDUH and Dennis & Scott under review

4 Problems Vary by Age Source: 2002 NSDUH and Dennis & Scott under review No Alcohol or Drug Use Light Alcohol Use Only Any Infrequent Drug Use Regular AOD Use Abuse Dependence NSDUH Age Groups Severity Category Adolescent Onset Remission Increasing rate of non- users

5 Higher Severity is Associated with Higher Annual Cost to Society Per Person Source: 2002 NSDUH and Dennis & Scott under review $0 $231 $725 $406 $0 $500 $1,000 $1,500 $2,000 $2,500 $3,000 $3,500 $4,000 No Alcohol or Drug Use Light Alcohol Use Only Any Infrequent Drug Use Regular AOD Use Abuse Dependence Median (50 th percentile) $948 $1,613 $1,078 $1,309 $1,528 $3,058 Mean (95% CI) This includes people who are in recovery, elderly, or do not use because of health problems Higher Costs

6 The Majority Stay in Tx Less than 90 days Source: Data received through August 4, 2004 from 23 States (CA, CO, GA, HI, IA, IL, KS, MA, MD, ME, MI, MN, MO, MT, NE, NJ, OH, OK, RI, SC, TX, UT, WY) as reported in Office of Applied Studies (OAS; 2005). Treatment Episode Data Set (TEDS): Discharges from Substance Abuse Treatment Services, DASIS Series: S-25, DHHS Publication No. (SMA) , Rockville, MD: Substance Abuse and Mental Health Services Administration. Retrieved from OutpatientIntensive Outpatient Short Term Residential Long Term Residential Level of Care Median Length of Stay in Days

7 Less Than Half Are Positively Discharged Source: Data received through August 4, 2004 from 23 States (CA, CO, GA, HI, IA, IL, KS, MA, MD, ME, MI, MN, MO, MT, NE, NJ, OH, OK, RI, SC, TX, UT, WY) as reported in Office of Applied Studies (OAS; 2005). Treatment Episode Data Set (TEDS): Discharges from Substance Abuse Treatment Services, DASIS Series: S-25, DHHS Publication No. (SMA) , Rockville, MD: Substance Abuse and Mental Health Services Administration. Retrieved from 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% OutpatientIntensive Outpatient Short Term Residential Long Term Residential Level of Care Discharge Status Other Terminated Dropped out Completed Transferred Less than 10% are transferred

8 Multiple Co-occurring Problems are Correlated with Severity and Contribute to Chronicity 0% 20%40%60%80% 100% Health Distress Internal Disorders External Disorders Crime/Violence Criminal Justice System Involvement Dependent (n=1221) Abuse/Other (n=385) 0% 20% 40%60%80% 100% Dependent (n=3135) Abuse/Other (n=2617) Adolescents Adults Source: GAIN Coordinating Center Data Set Exception Adolescents More likely to have externalizing disorders Adults more likely to have internalizing disorders[

9 Pathways to Recovery Study (Scott & Dennis) Recruitment: 1995 to 1997 Sample: 1,326 participants from sequential admissions to a stratified sample of 22 treatment units in 12 facilities, administered by 10 agencies on Chicago's west side. Substance:Cocaine (33%), heroin (31%), alcohol (27%), marijuana (7%). Levels of Care: Adult OP, IOP, MTP, HH, STR, LTR Instrument:Augmented version of the Addiction Severity Index (A-ASI) Follow-up:Of those alive and due, follow-up interviews were completed with 94 to 98% in annual interviews out to 8 years (going to 10 years); over 80% completed within +/- 1 week of target date. Funding: CSAT grant # T100664, contract # NIDA grant 1R01 DA15523 (Scott & Dennis)

10 Pathways to Recovery Sample Characteristics 0% 20%40%60%80% 100% African American Age Female Current CJ Involved Past Year Dependence Prior Treatment Residential Treatment Other Mental Disorders Homeless Physical Health Problems

11 Substance Use Careers Last for Decades Percent in Recovery % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Median duration of 27 years (IQR: 18 to 30+) Source: Dennis et al 2005 (n=1,271) Years from first use to 1+ years abstinence

12 Substance Use Careers are Longer, the Younger the Age of First Use Percent in Recovery Years from first use to 1+ years abstinence % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Source: Dennis et al 2005 (n=1,271) under 15* * Age of 1 st Use Groups * p<.05 (different from 21+)

13 Substance Use Careers are Shorter the Sooner People get to Treatment Percent in Recovery Years from first use to 1+ years abstinence % 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Source: Dennis et al 2005 (n=1,271) * 10-19* Years to 1 st Tx Groups * p<.05 (different from 20+)

14 It Takes Decades and Multiple Episodes of Treatment Years from first Tx to 1+ years abstinence Median duration of 9 years (IQR: 3 to 23) and 3 to 4 episodes of care Percent in Recovery 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Source: Dennis et al 2005 (n=1,271)

15 The Cyclical Course of Relapse, Incarceration, Treatment and Recovery: Adults In the Community Using (53% stable) In Treatment (21% stable) In Recovery (58% stable) Incarcerated (37% stable) 6% 13% 28% 30% 8% 25% 31% 4% 44% 7% 29% 7% Treatment is the most likely path to recovery P not the same in both directions Source: Scott et al 2005 Over half change status annually

16 Source: Scott et al 2005 Predictors of Change Also Vary by Direction In the Community Using (53% stable) In Recovery (58% stable) 13% 29% Probability of Relapsing from Abstinence + times in treatment (1.21) - Female (0.58) + homelessness (1.64)- ASI legal composite (0.84) + number of arrests (1.12) - # of sober friend (0.82) - per 77 self help sessions (0.55) Probability of Transitioning from Using to Abstinence - mental distress (0.88)+ older at first use (1.12) -ASI legal composite (0.84) + homelessness (1.27) + # of sober friend (1.23) + per 8 weeks in treatment (1.14)

17 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Using (N=661) 1 to 12 ms (N=232) 1 to 3 yrs (N=127) 3 to 5 yrs (N=65) 5 to 8 yrs (N=77) % Days of Psych Prob (of 30 days) % Above Poverty Line % Days Worked For Pay (of 22) % of Clean and Sober Friens % Days of Illegal Activity (of 30 days ) Other Aspects of Recovery by Duration of Abstinence of 8 Years 1-12 Months: Immediate increase in clean and sober friend 1-3 Years: Decrease in Illegal Activity; Increase in Psych Problems 3-5 Years: Improved Vocational and Financial Status 5-8 Years: Improved Psychological Status Source: Dennis, Foss & Scott (under review)

18 Percent Sustaining Abstinence Through Year 8 by Duration of Abstinence at Year 7 36% 66% 86% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1 to 12 months (n=157; OR=1.0) 1 to 3 years (n=138; OR=3.4) 3 to 5 years (n=59; OR=11.2) 5+ years (n=96; OR=11.2) Duration of Abstinence at Year 7 % Sustaining Abstinent through Year 8. It takes a year of abstinence before less than half relapse Even after 3 to 7 years of abstinence about 14% relapse Source: Dennis, Foss & Scott (under review)

19 Post Script on the Pathways Study There is clearly a subset of people for whom substance use disorders are a chronic condition that last for many years Rather than a single transition, most people cycle through abstinence, relapse, incarceration and treatment 3 to 4 times before reaching a sustained recovery. It is possible to predict the likelihood risk of when people will transition Treatment predicts who transitions from use to recovery and self help group participation predicts who stays in recovery. “Recovery” is broader than abstinence and often takes several years after initial abstinence

20 The Cyclical Course of Relapse, Incarceration, Treatment and Recovery: Adolescents In the Community Using (75% stable) In Treatment (48% stable) In Recovery (62% stable) Incarcerated (46% stable) 5%5% 12% 7%7% 20% 24% 10% 26% 7 % 19% 7%7% 27% 3%3% Source: 2006 CSAT AT data set More likely to relapse (OR=1.4 to 1.8) More likely to be diverted to treatment (OR=4.0) Treatment is still the most likely path to recovery Avg of 39% change status each quarter

21 The Cyclical Course of Relapse, Incarceration, Treatment and Recovery: Adolescents In the Community Using (75% stable) In Treatment (48 v 35% stable) In Recovery (62% stable) 12% 7%7% Source: 2006 CSAT AT data set Probability of Transitioning to Tx - Age (0.7) +Weeks in Cont. Environ. (1.4) + Times urine tested (1.7) + Treatment Motivation (1.6) Probability of Transitioning to Recovery - Age (0.8) - Freq. Of Use (0.23) + Female (1.7), + Non-White (1.6) + Self efficacy to resist relapse (1.4) + Sub. Abuse Tx Index (1.96) 19%

22 The Early Re-Intervention (ERI) Experiments (Dennis & Scott) ERI 1ERI 2 RecruitmentRecruited 448 from Community Based Treatment in Chicago in 2000 (84% of eligible recruited) Recruited 446 from Community Based Treatment in Chicago in 2004 (93% of eligible recruited) DesignRandom assignment to Recovery Management Checkups (RMC) or control Follow-UpQuarterly for 2 years (95-97% per wave) Quarterly for 4 years (95 to 97% per wave) Data SourcesGAIN, CEST, Urine, Salvia Staff logs GAIN, CEST, CAI, Neo, CRI, Urine, Staff logs PublicationDennis, Scott & Funk 2003; Scott, Dennis & Foss, 2005 Dennis & Scott (in press); Scott & Dennis, (under review) Funding Source NIDA grant R37-DA11323

23 Sample Characteristics of ERI-1 & -2 Experiments 0% 20%40%60%80% 100% African American Age Female Current CJ Involved Past Year Dependence Prior Treatment Residential Treatment Other Mental Disorders Homeless Physical Health Problems ERI 1 (n=448) ERI 2 (n=446)

24 Recovery Management Checkups (RMC) in both ERI 1 & 2 included: Quarterly Screening to determining “Eligibility” and “Need” Linkage meeting/motivational interviewing to: –provide personalized feedback to participants about their substance use and related problems, –help the participant recognize the problem and consider returning to treatment, –address existing barriers to treatment, and –schedule an assessment. Linkage assistance –reminder calls and rescheduling –Transportation and being escorted as needed

25 RMC Protocol Adherence Rate by Experiment 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Follow-up Interview (93 vs. 96%) d=0.18 Treatment Need (30 vs. 44%) d=0.31* Linkage Attendance (75 vs. 99%) d=1.45* Agreed to Assessment (44 vs. 45%) d=0.02 Showed to Assessment (30 vs. 42%) d=0.26* Showed to Treatment (25 vs. 30%) d=0.18* Treatment Engagement (39 vs. 58%) d=0.43* Range of rates by quarter * P(H: RMC1=RMC2)<.05 ERI-1 ERI-2 ERI 2 Generally averaged as well or better than ERI 1 Improved Screening Improved Tx Engagement Quality assurance and transportation assistance reduced the variance

26 ERI-1 Time to Treatment Re-Entry 0%0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Days to Re-Admission (from 3 month interview) Percent Readmitted 1+ Times 60% ERI-1 RMC* (n=221) 51% ERI-1 OM (n=224) *Cohen's d=+0.22 Wilcoxon-Gehen Statistic (df=1) =5.15, p < = -200 days Revisions to the protocol

27 ERI-2 Time to Treatment Re-Entry 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0% Days to Re-Admission (from 3 month interview) Percent Readmitted 1+ Times 55% ERI-2 RMC* (n=221) 37% ERI-2 OM (n=224) *Cohen's d=+0.41 Wilcoxon-Gehen Statistic (df=1) =16.56, p < = -384 days The size of the effect is growing every quarter

28 ERI-1: Impact on Outcomes 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% of 630 Days Abstinent (d=0.04) of 7 Subsequent Quarters in Need (d= -0.19) * of 90 Days Abstinent (d= -0.05) of 11 Sx of Abuse/Dependence (d=-0.02) Still in need of Tx (d= -0.21) * Percentage OM RMC * p<.05 79% 33% 80% 21% 44% 79% 27% 79% 21% 34% RMC Broke the Run Less Likely to be in Need of Treatment Months 4-24 Final Interview No effect on Abstinence/Symptoms

29 ERI-2: Impact on Outcomes 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% of 630 Days Abstinent (d=0.29)* of 7 Subsequent Quarters in Need (d= -0.32) * of 90 Days Abstinent (d= 0.23)* of 11 Sx of Abuse/Dependence (d= -0.23)* Still in need of Tx (d= -0.24) * Percentage OM RMC * p<.05 68% 49% 68% 27% 57% 76% 37% 76% 19% 46% Months 4-24 Final Interview Significant Increase in Abstinence RMC Broke the Run Less Likely to be in Need of Treatment Less Symptoms

30 Source: ERI experiments (Scott, Dennis, & Foss, 2005) Impact on Primary Pathways to Recovery (incarceration not shown) In the Communityy Using (71% stable) In Treatment (35% stable) In Recovery (76% stable) 27% 5% 8% 33% 18% 17% Transition to Tx - Freq. of Use (0.7) + Prob. Orient. (1.4) + Desire for Help (1.6) + RMC (3.22) Again the Probability of Entering Recovery is Higher from Treatment Transition to Recov. - Freq. of Use (0.7) - Dep/Abs Prob (0.7) - Recovery Env. (0.8) - Access Barriers (0.8) + Prob. Orient. (1.3) + Self Efficacy (1.2) + Self Help Hist (1.2) + per 10 wks Tx (1.2) 32% Changed Status in an Average Quarter

31 Post Script on ERI experiments Again, severity was inversely related to returning to treatment on your own and treatment was the key predictor of transitioning to recovery The ERI experiments demonstrate that the cycle of relapse, treatment re-entry and recovery can be shortened through more proactive intervention Working to ensure identification, showing to treatment, and engagement for at least 14 days upon readmission helped to improve outcomes ERI 2 also demonstrated the value of on-site proactive urine testing versus the traditional practice of sending off urine for post interview testing

32 These studies provide converging evidence demonstrating that substance use disorders are often chronic in the sense that they last for years and the risk of relapse is high the majority of people accessing publicly funded substance abuse treatment have been in treatment before, are likely to return, have a variety of co-occurring problems and may need several additional episodes of care before they reach a point of stable recovery. Yet over half do make it to recovery and the odds of getting to and staying in recovery can be improved with proactive management. Though we did not have time to go over them today, similar studies and findings are coming out with adolescents and young adults

33 We need to.. Educate policy makers, staff and clients to have more realistic expectations Redefine the continuum of care to include monitoring and other proactive interventions between primary episodes of care. Shift our focus from intake matching to on-going monitoring, matching over time, and strategies that take the cycle into account Identify other venues (e.g., jails, emergency rooms) where recovery management can be initiated Evaluate the costs and determine generalizability to other populations through replication Explore changes in funding, licensure and accreditation to accommodate and encourage above

34 Sources and Related Work American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (DSM-IV-TR) (4th - text revision ed.). Washington, DC: American Psychiatric Association. Chan, Y.-F., Dennis, M. L., & Funk, R. (in press). Prevalence and comorbidity of major internalizing and externalizing problems among adolescents and adults presenting to substance abuse treatment. Journal of Substance Abuse Treatment. Dennis, M.L., Chan, Y.-F., & Funk, R. (2006). Development and validation of the GAIN Short Screener (GSS) for psychopathology and crime/violence among adolescents and adults. American Journal on Addictions, 15, Dennis, M.L., Foss, M.A., & Scott, C.K (under review). Correlates of Long-Term Recovery After Treatment. Evaluation Review. Dennis, M. L., Scott, C. K. (in press). Managing substance use disorders (SUD) as a chronic condition. NIDA Science and Perspectives. Dennis, M. L., Scott, C. K., Funk, R., & Foss, M. A. (2005). The duration and correlates of addiction and treatment careers. Journal of Substance Abuse Treatment, 28, S51-S62. Dennis, M. L., 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), Epstein, J. F. (2002). Substance dependence, abuse and treatment: Findings from the 2000 National Household Survey on Drug Abuse (NHSDA Series A-16, DHHS Publication No. SMA ). Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Retrieved from GAIN Coordinating Center Data Set (2005). Bloomington, IL: Chestnut Health Systems. See Kessler, R. C., Nelson, G. B., McGonagle, K. A., Edlund, M. J., Frank, R. G., & Leaf, P. J. (1996). The epidemiology of co-occurring mental disorders and substance use disorders in the national comorbidity survey: Implications for prevention and services utilization. Journal of Orthopsychiatry, 66, Office Applied Studies (2002). Analysis of the 2002 National Survey on Drug Use and Health (NSDUH) on line at Office Applied Studies (2002). Analysis of the 2002 Treatment Episode Data Set (TEDS) on line data at Scott, C. K., & Dennis, M. L. (under review). Results from Two Randomized Clinical Trials evaluating the impact of Quarterly Recovery Management Checkups with Adult Chronic Substance Users. Addiction. Scott, C. K., Dennis, M. L., & Foss, M. A. (2005). Utilizing recovery management checkups to shorten the cycle of relapse, treatment re-entry, and recovery. Drug and Alcohol Dependence, 78, Scott, C. K., Foss, M. A., & Dennis, M. L. (2005). Pathways in the relapse, treatment, and recovery cycle over three years. Journal of Substance Abuse Treatment, 28, S61-S70. World Health Organization (WHO). (1999). The International Statistical Classification of Diseases and Related Health Problems, tenth revision (ICD-10). Geneva, Switzerland: World Health Organization. Retrieved from