1. Global Appraisal of Individual Needs (GAIN): How it acts as a key piece of infrastructure for supporting the move towards evidenced based practice Michael Dennis, Ph.D. Chestnut Health Systems, Normal, IL Presentation at the “Adolescent Treatment Summit”, Saratoga Springs, New York October 19-20, 2009. The meeting and presentation are sponsored by St. Peter’s Addiction Recovery Center (SPARC) and New York State Office of Alcoholism and Substance Abuse Services (OASAS). This presentation reports on treatment & research funded by the Center for Substance Abuse Treatment (CSAT), Substance Abuse and Mental Health Services Administration (SAMHSA) under contracts 270-2003-00006 and 270-07-0191, as well as several individual CSAT, NIAAA, NIDA and private foundation grants. The opinions are those of the author and do not reflect official positions of the consortium or government. Available on line at www.chestnut.org/LI/Posters or by contacting Joan Unsicker at 448 Wylie Drive, Normal, IL 61761, phone: (309) 451-7801, Fax: (309) 451-7763, e-mail: junsicker@Chestnut.Org

2. 1.Provide an overview of the role of the GAIN as a piece of infrastructure in support the move toward both evidence based practice and practice based evidence 2.Describe each of the measures, the reports that they use to help the assessment guide clinical decision making and illustrate how they provide a successively more detailed picture of client needs 3.Highlight our current work to using actuarial estimates of outcomes to improve placement decisions 4.Summarize the status of efforts to make the data available for secondary analysis and translate the software, measures and reports from English into Spanish, French, Portuguese and other languages Goals of this Presentation are to

3. Part 1. Provide an overview of the role of the GAIN as a piece of infrastructure in support the move toward both evidence based practice and practice based evidence

4. The Global Appraisal of Individual Needs (GAIN) is.. A family of instruments ranging from screening, to quick assessment to a full Biopsychosocial and monitoring tools Designed to integrate clinical and research assessment Designed to support clinical decision making at the individual client level Designed to support evaluation and planning at program level Designed to support secondary analyses and comparisons across individuals and programs

5. As of June 30, 2009, there were 1127 administrative units (agencies, grantees, counties, states) collaborating to use the GAIN in the U.S., State or County System GAIN-Short Screener GAIN-Quick GAIN-Full

6. Canada and other countries* * 1-10 Sites Other Countries: Brazil, China, Mexico, Japan State or County System GAIN-Short Screener GAIN-Quick GAIN-Full

7. So what does it mean to move the field towards Evidence Based Practice (EBP)? Introducing explicit intervention protocols that are – Targeted at specific problems/subgroups and outcomes – Having explicit quality assurance procedures to cause adherence at the individual level and implementation at the program level Introducing reliable and valid assessment that can be used – At the individual level to immediately guide clinical judgments about diagnosis/severity, placement, treatment planning, and the response to treatment – At the program level to drive program evaluation, needs assessment, performance monitoring and long term program planning Having the ability to evaluate client and program outcomes – For the same person or program over time, – Relative to other people or interventions

8. Key Issues that we try to address with the GAIN Instruments and Coordinating Center 1.High turnover workforce with variable education background related to diagnosis, placement, treatment planning and referral to other services 2.Heterogeneous needs and severity characterized by multiple problems, chronic relapse, and multiple episodes of care over several years 3.Lack of access to or use of data at the program level to guide immediate clinical decisions, billing and program planning 4.Missing, bad or misrepresented data that needs to be minimized and incorporated into interpretations 5.Lack of Infrastructure that is needed to support implementation and fidelity

9. 1. High Turnover Workforce with Variable Education Questions spelled out and simple question format Lay wording mapped onto expert standards for given area Built in definitions, transition statements, prompts, and checks for inconsistent and missing information. Standardized approach to asking questions across domains Range checks and skip logic built into electronic applications Formal training and certification protocols on administration, clinical interpretation, data management, coordination, local, regional, and national “trainers” Above focuses on consistency across populations, level of care, staff and time On-going quality assurance and data monitoring for the reoccurrence or problems at the staff (site or item) level Availability of training resources, responses to frequently asked questions, and technical assistance Outcome: Improved Reliability and Efficiency

10. 2. Heterogeneous Needs and Severity Multiple domains Focus on most common problems Participant self description of characteristics, problems, needs, personal strengths and resources Behavior problem recency, breadth, and frequency Utilization lifetime, recency and frequency Dimensional measures to measure change with interpretative cut points to facilitate decisions Items and cut points mapped onto DSM for diagnosis, ASAM for placement, and to multiple standards and evidence- based practices for treatment planning Computer generated scoring and reports to guide decisions Treatment planning recommendations and links to evidence-based practice Basic and advanced clinical interpretation training and certification Outcome: Comprehensive Assessment

11. 3. Lack of Access to or use of Data at the Program Level Data immediately available to support clinical decision making for a case Data can be transferred to other clinical information system to support billing, progress reports, treatment planning and on-going monitoring Data can be exported and cleaned to support further analyses Data can be pooled with other sites to facilitate comparison and evaluation PC and web based software applications and support Formal training and certification on using data at the individual level and data management at the program level Data routinely pooled to support comparisons across programs and secondary analysis Over three dozen scientists already working with data to link to evidence-based practice Outcome: Improved Program Planning and Outcomes

12. 4. Missing, Bad or Misrepresented Data Assurances, time anchoring, definitions, transition, and question order to reduce confusion and increase valid responses Cognitive impairment check Validity checks on missing, bad, inconsistency and unlikely responses Validity checks for atypical and overly random symptom presentations Validity ratings by staff Training on optimizing clinical rapport Training on time anchoring Training answering questions, resolving vague or inconsistent responses, following assessment protocol and accurate documentation. Utilization and documentation of other sources of information Post hoc checks for on-going site, staff or item problems Outcome: Improved Validity

13. 5. Lack of Infrastructure Direct Services Training and quality assurance on administration, clinical interpretation, data management, follow-up and project coordination Data management Evaluation and data available for secondary analysis Software support Technical assistance and back up to local trainer/expert Development Clinical Product Development Software Development Collaboration with IT vendors (e.g., WITS) Over 36 internal & external scientists and students Workgroups focused on specific subgroup, problem, or treatment approach Labor supply (e.g., consultant pool, college courses) Outcome: Implementation with Fidelity

14. Across measures, the GAIN has a Common Factor Structure of Psychopathology Source: Dennis, Chan, and Funk (2006) CFI=.92, RMSEA=.06 allowing for age

15. Alcohol and Other Drug Abuse, Dependence and Problem Use are Age Related Source: 2002 NSDUH and Dennis & Scott, 2007 0 10 20 30 40 50 60 70 80 90 100 12-1314-1516-1718-2021-2930-3435-4950-64 65+ Other drug or heavy alcohol use in the past year Alcohol or Drug Use (AOD) Abuse or Dependence in the past year Age Severity Category Over 90% of use and problems start between the ages of 12-20 It takes decades before most recover or die Percentage

16. Co-occurring Mental Health Problems are Common, but the Type of Problems also Changes with Age Source: Chan, YF; Dennis, M L.; Funk, RR. (2008). Prevalence and comorbidity of major internalizing and externalizing problems among adolescents and adults presenting to substance abuse treatment. Journal of Substance Abuse Treatment, 34(1) 14-24. Internalizing Disorders go up with age Externalizing Disorders go down with age (but do NOT go away)

17. Progressive Continuum of Measurement (Common Measures) Screening to Identify Who Needs to be “Assessed” (5-10 min) – Focus on brevity, simplicity for administration & scoring – Needs to be adequate for triage and referral – GAIN Short Screener for SUD, MH & Crime – ASSIST, AUDIT, CAGE, CRAFT, DAST, MAST for SUD – SCL, HSCL, BSI, CANS for Mental Health – LSI, MAYSI, YLS for Crime Quick Assessment for Targeted Referral (20-30 min) – Assessment of who needs a feedback, brief intervention or referral for more specialized assessment or treatment – Needs to be adequate for brief intervention – GAIN Quick – ADI, ASI, SASSI, T-ASI, MINI Comprehensive Biopsychosocial (1-2 hours) – Used to identify common problems and how they are interrelated – Needs to be adequate for diagnosis, treatment planning and placement of common problems – GAIN Initial (Clinical Core and Full) – CASI, A-CASI, MATE Specialized Assessment (additional time per area) – Additional assessment by a specialist (e.g., psychiatrist, MD, nurse, spec ed) may be needed to rule out a diagnosis or develop a treatment plan or individual education plan – CIDI, DISC, KSADS, PDI, SCAN Screener Quick Comprehensive Special More Extensive / Longer/ Expensive

18. 18 Part 2. Describe each of the measures, the reports that they use to help the assessment guide clinical decision making and illustrate how they provide a successively more detailed picture of client needs

19. Next slides will Describe the difference in the breadth of information you get with different levels of assessment Summarize validation studies to date Illustrate the difference using data from a single sample (Reclaiming futures project) Demonstrate that multi-morbidity is the norm and varies by type of client and program

20. GAIN-Short Screener (GSS) Administration Time: A 3- to 5-minute screener Purpose: Used in general populations to – identify or rule-out clients who will be identified as having any behavioral health disorders on the 60-120 min versions of the GAIN – triage area of problem – serve as a simple measure of change – Easy for administration and interpretation by staff with minimal training or direct supervision Mode: Designed for self- or staff-administration, with paper and pen, computer, or on the web Scales: Four screeners for Internalizing Disorders, Externalizing Disorders, Substance Disorders, Crime/Violence, and a Total

21. Response Set: Recency of 20 problems rated past month (3), 2- 12 months ago (2), more than a year ago (1), never (0) Interpretation: Combined by cumulative time period as: – Past month count (3s) to measure of change – Past year count (2s or 3s) to predict diagnosis – Lifetime count (1s, 2s or 3s) as a measure of peak severity – Can be classified within time period low (0), moderate (1-2) or high (3) – Can also be used to classify remission as – Early (lifetime but not past month) – Sustained (lifetime but not past year) Reports: Narrative, tabular, and graphical reports built into web based GAIN ABS and/or ASP application for local hosting GAIN-Short Screener (GSS) (continued)

22. GAIN-Short Screener (GSS )

23. GAIN SS Psychometric Properties Total Disorder Screener (TDScr) 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 01234567891011121314151617181920 Prevalence (% 1+ disorder) Sensitivity (% w disorder above) Specificity (% w/o disorder below) (n=6194 adolescents) Low Mod. High At 3 or more symptoms we get 99% prevalence, 91% sensitivity, & 89% specificity Using a higher cut point increases prevalence and specificity, but decreases sensitivity Total score has alpha of.85 and is correlated.94 with full GAIN version Source: Dennis et al 2006

24. GSS Performance by Subscale and Disorders Prevalence Sensitivity Specificity Screener/Disorder 1+ 3+ 1+ 3+ 1+ 3+ Internal Disorder Screener (0-5) Any Internal Disorder 81% 99% 94% 55% 71% 99% Major Depression 56% 87% 98% 72% 54% 94% Generalized Anxiety 32% 56% 100% 83% 44% 83% SuicideIdeation 24% 43% 100% 84% 41% 79% Mod/High Traumatic Stress 60% 82% 94% 60% 55% 90% External Disorder Screener (0-5) Any External Disorder 88% 97% 98% 67% 75% 96% AD, HD or Both 65% 82% 99% 78% 51% 85% Conduct Disorder 78% 91% 98% 70% 62% 90% Substance Use Disorder Screener (0-5) Any Substance Disorder 96% 100% 96% 68% 73% 100% Dependence 65% 87% 100% 91% 30% 82% Abuse 30% 13% 89% 25% 14% 28% Crime Violence Screener (0-5) Any Crime/Violence 88% 99% 94% 49% 76% 99% High Physical Conflict 31% 46% 100% 70% 38% 77% Mod/High General Crime 85% 100% 94% 51% 71% 100% Total Disorder Screener (0-5) Any Disorder 97% 99% 91% 47% 89% Any Internal Disorder 58% 63% 100% 98% 8% 28% Any External Disorder 68% 75% 100% 99% 10% 37% Any Substance Disorder 89% 92% 99% 92% 20% 51% Any Crime/Violence 68% 73% 100% 96% 10% 32% Recommend Triage as 0=Not likely 1-2 Possible 3+=Likely Moderate (1+) gives best result for sensitivity High (3+) gives best result for specificity

25. GAIN SS Total Score is Correlated With Level Of Care Placement: Adolescents

26. GAIN SS Total Score is Correlated With Level Of Care Placement: Adults

27. GAIN SS Can Also be Used for Monitoring 10 9 11 9 10 8 3 22 0 4 8 12 16 20 Intake3 Mon 6 9 12 Mon 15 Mon 18 Mon 21 Mon 24 Mon Total Disorder Screener (TDScr) 12+ Mon.s ago (#1s) 2-12 Mon.s ago (#2s) Past Month (#3s) Lifetime (#1,2,or 3) Track Gap Between Prior and current Lifetime Problems to identify “under reporting” Track progress in reducing current (past month) symptoms) Monitor for Relapse

28. GAIN Short Screener Profile: Reclaiming Futures (Range based on 0/1-2/3+ Symptoms) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

29. GAIN Short Screener Number of Problems Mod/Hi in Reclaiming Futures Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192) 93% endorsed one or more problems (40% 4 or more)

30. Construct Validity of GSS Internalizing Disorder Screener Source: Education Service District 113 (n=979) and King County (n=1002)

31. Construct Validity of GSS Externalizing Disorder Screener Source: Education Service District 113 (n=979) and King County (n=1002)

32. Construct Validity of GSS Substance Disorder Screener Source: Education Service District 113 (n=979) and King County (n=1002)

33. Construct Validity of GSS Crime/Violence Screener Source: Education Service District 113 (n=979) and King County (n=1002)

34. Adolescent Rates of High (2+) Scores on Mental Health (MH) or Substance Abuse (SA) Screener by Setting in Washington State Source: Lucenko et al (2009). Report to the Legislature: Co-Occurring Disorders Among DSHS Clients. Olympia, WA: Department of Social and Health Services. Retrieved from http://publications.rda.dshs.wa.gov/1392/http://publications.rda.dshs.wa.gov/1392/ Problems could be easily identified Comorbidity is common

35. Adult Rates of High (2+) Scores on Mental Health (MH) or Substance Abuse (SA) Screener by Setting in Washington State Source: Lucenko et al (2009). Report to the Legislature: Co-Occurring Disorders Among DSHS Clients. Olympia, WA: Department of Social and Health Services. Retrieved from http://publications.rda.dshs.wa.gov/1392/http://publications.rda.dshs.wa.gov/1392/ Lower than expected rates of SA in Mental Health & Children’s Admin

36. Source: Lucenko et al (2009). Report to the Legislature: Co-Occurring Disorders Among DSHS Clients. Olympia, WA: Department of Social and Health Services. Retrieved from http://publications.rda.dshs.wa.gov/1392/http://publications.rda.dshs.wa.gov/1392/ Adolescent Client Validation of Hi Co-occurring from GAIN Short Screener vs Clinical Records by Setting in Washington State Two page measure closely approximated all found in the clinical record after the next two years

37. Adult Client Validation of Hi Co-occurring from GAIN Short Screener vs Clinical Records by Setting in Washington State Source: Lucenko et al (2009). Report to the Legislature: Co-Occurring Disorders Among DSHS Clients. Olympia, WA: Department of Social and Health Services. Retrieved from http://publications.rda.dshs.wa.gov/1392/http://publications.rda.dshs.wa.gov/1392/ Higher rate in clinical record in Mental Health and Children’s Administration (But that was past on “any use” vs. “abuse/dependence” and 2 years vs. past year

38. Other Validations Confirmatory Factor Analysis Dennis, Chan & Funk (2006) found that the 20 item GSS and its four subscales were highly correlated (.84 to.94) with the full scale, had 90% sensitivity and over 90% area under the curve relative to the full GAIN; Confirmatory factors analysis also found it to be consistent with the overall model of psychopathology after allowing for age (CFI=.92; RMSEA=.06). Substance Disorders: McDonnell and colleagues (2009) found that the 5-item GAIN SS Substance Disorder Screener had 92% sensitivity and 85% correct classification relative to the Diagnostic Inventory Scale for Children (DISC) Predictive Scales (DPS; Lucas et al 2001) and 88% sensitivity and 88% correct classification relative to the CRAFFT (Knight et al 2001) Internalizing Disorders: McDonnell and colleagues (2009) found that the 5-item GAIN SS Internalizing Disorder Screener had 100% sensitivity and 75% correct classification relative to the Youth Self Report (YSR; Achenbach et al, 2001) and that the 5-item GAIN SS Externalizing Disorder Screener had 89% sensitivity and 65% correct classification to the YSR. Riley and colleagues (2009) found that the 5-item GAIN SS’s Internalizing Disorder Screener had 92% sensitivity and 80% area under the curve relative to the Structured Clinical Interview for DSM (SCID) and was more efficient relative to 11 item Addiction Severity Index (ASI) psychiatric composite score (McLellan et al., 1992), 10 item K10 (Kessler et al., 2002) and the 87 item Psychiatric Diagnostic Screening Questionnaire (PDSQ; Zimmerman and Mattia, 2001)

39. GAIN Quick (GQ) Administration Time: 20-30 minutes (depending on severity and whether reasons for quitting module used) Training Requirements: 1 day (train the trainer) plus certification within 1-2 months Mode: Generally Staff Administered on Computer (can be done on paper or self administered with proctor) Purpose: Designed for use in targeted populations to support brief intervention or referral for further assessment or behavioral intervention. Not originally designed for follow-up. Scales: The GQ has total scale (99-symptoms) and 15 subscales (including more detailed versions of the GSS scales and subscales plus scales for service utilization, sources of psychosocial stress, and health problems).

40. GAIN Quick (GQ) (Continued) Response Set: Breadth (past year symptom counts for behavior and lifetime for utilization) and Prevalence (past 90 days) Interpretation: – Items can be used individually or to create specific diagnostic or treatment planning statements – Items can be summed into scales or indices for each behavior problem or and for recent service utilization overall – All scales, indices and selected individual items have interpretative cut-points to facilitate clinical interpretation and decision making – Some people repeat just days items for follow-up. Reports: Narrative, tabular, graphical, validity and “motivational interviewing” reports built into web based GAIN ABS; Program level reports available in SPSS/Excel

41. GAIN Quick Profile of Reclaiming Futures Sites (Range based on 0-24% / 25-74% / 75-100% of Symptoms) Source: Reclaiming Futures Chicago, IL, Dayton, OH, Portland, OR and Santa Cruz, CA sites (n=475). * Summary Measure Risk Stress Health More detail within each area

42. GAIN Quick Number of Problems Mod/Hi 97% endorsed one or more problems (69% 4 or more problems) Source: Reclaiming Futures Chicago, IL, Dayton, OH, Portland, OR and Santa Cruz, CA sites (n=475).

43. Validations Titus and colleagues (2008) found that the internal consistency of the full scales ranged from.82 to.90 among adults and adolescents with alpha above.7 for 7 of 8 subscale scores for adults and 7 of 8 subscale scores for the adolescents. Titus and colleagues (2008) found that the mental health scales from the GAIN quick have good internal consistency (.86 to.90), are correlated with the full GAIN dimensional measures (.92 to.97) and.99 to 100% sensitivity relative to the full GAIN.

44. GAIN Quick (GQ): In Transition Strengths: Length (20-30min) in desired range, range of topics, efficiently categorizes, narrative reports to support screening, brief intervention, and referral to treatment Problems: – Lacks scales or recency to support analyses or outcomes related to “change over time” – Item response choices do not provide information about lifetime problems – Current Personal Feedback Report focuses only on substance use and does not address the other content areas of the GAIN-Q – Only about 60% of the items can be directly imported into the GAIN-I

45. GAIN Quick (GQ): In Transition Plans for Version 3: – Keep focus on screening, brief intervention and referral to treatment – Break out sections for Crime/Violence, HIV risk, Work and School problems – Subsume GSS and add similar screeners in other GAIN Q areas with recency response to address change and lifetime issues – Change measures for each symptom count and days items – Create reasons for change items in each area to support brief intervention, reducing number of items in substance use – Make all questions importable into full GAIN – Expand narrative report to have more treatment planning statements and to allow motivational interviewing within each area Plans for Version 4: Add computer adaptive testing (CAT) component to get at more detailed diagnosis where needed

46. GAIN Initial (GI) Administration Time: Core version 60-90 minutes; Full version 110- 140 minutes (depending on severity) Training Requirements: 3.5 days (train the trainer) plus recommend formal certification program (administration certification within 3 months of training; local trainer certification within 6 months of training); Advanced clinical interpretation recommended for clinical supervisors and lead clinicians Mode: Generally Staff Administered on Computer (can be done on paper or self administered with proctor) Purpose: Designed to provide a standardized biopsychosocial for people presenting to a substance abuse treatment using DSM-IV for diagnosis, ASAM for placement, and needing to meet common (CARF, COA, JCAHO, insurance, CDS/TEDS, Medicaid, CSAT, NIDA) requirements for assessment, diagnosis, placement, treatment planning, accreditation, performance/outcome monitoring, economic analysis, program planning and to support referral/communications with other systems

47. Scales: The GI has 9 sections (access to care, substance use, physical health, risk and protective behaviors, mental health, recovery environment, legal, vocational, and staff ratings) that include 103 long (alpha over.9) and short (alpha over.7) scales, summative indices, and over 3000 created variables to support clinical decision making and evaluation. It is also modularized to support customization Response Set: Breadth (past year symptom counts for behavior and lifetime for utilization), Recency (48 hours, 3-7 days, 1-4 weeks, 2-3 months, 4-12 months, 1+ years, never) and Prevalence (past 90 days), patient and staff ratings Interpretation: – Items can be used individually or to create specific diagnostic or treatment planning statements – Items can be summed into scales or indices for each behavior problem or type of service utilization – All scales, indices and selected individual items have interpretative cut-points to facilitate clinical interpretation and decision making GAIN Initial (GI) (continued)

48. Reports: – Validity Report (VR): identifying missing/bad data and potentially problematic areas of assessment – Individual Clinical Profile (ICP): lab report with graphical and tabular summary with links back to the items – GAIN Recommendation and Referral Summary (GRRS): Draft of biopsychosocial narrative for clinician to use for initial assessment summary, diagnosis, placement and treatment planning – Personal Feedback Report (PFR): used to support Motivational Interviewing (MI) / Motivational Enhancement Therapy (MET) – Program Profile: program level report that allows comparison of client characteristics, services received and outcomes between programs, cohorts or types of clients. GAIN Initial (GI) (continued)

49. GAIN Initial Profile: Substance Problems Past Year (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

50. GAIN Initial Profile: Substance Problems by Time (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

51. GAIN Initial Profile: Motivation and Readiness (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

52. GAIN Initial Profile: Crime/Violence (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

53. GAIN Initial Profile: Environmental Risk (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

54. GAIN Initial Profile: Internalizing Disorders (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

55. GAIN Initial Profile: Externalizing Disorders (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

56. GAIN Initial Profile: Personality Disorders (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

57. GAIN Initial Profile: General Factors / Stress (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

58. GAIN Initial Profile: Other Problem Scales (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

59. GAIN Initial Profile: Measures of Behavior Change (Range based range of clinical/logical/statistical rules) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

60. GAIN Initial Number of Problems Mod/Hi 99% endorsed one or more problems (98% 4 or more) Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

61. GAIN Treatment Planning/Placement Grid Problem Recency/Severity NonePast Current (past 90 days)* Low-Mod | High Severity Treatment History None Past Current. 1. No Problem 2. Past problem Consider monitoring and relapse prevention. 3. Low/Moderate problems; Not in treatment Consider initial or low invasive treatment. 4. Severe problems; Not in treatment Consider a more intensive treatment or intervention strategies. 0. Not Logical Check under- standing of problem or lying and recode. 5. No current problems; Currently in treatment Review for step down or discharge. 6. Low/Moderate problems; Currently in treatment Review need to continue or step up. 7. Severe problems; Currently in treatment Review need for more intensive or assertive levels. * Current for Intoxication & Withdrawal = Past 7 days

62. Reclaiming Futures ASAM Placement Cells Source: Reclaiming Futures Portland, OR and Santa Cruz, CA sites (n=192)

63. Other Common Treatment Planning Needs: Reclaiming Futures Source: Reclaiming Futures (n=192)

64. 64 Part 3. Highlight our current work to using actuarial estimates of outcomes to improve placement decisions

65. CSAT Adolescent Treatment GAIN Data from 203 level of care x site combinations Outpatient General Group Home Short-Term Residential Outpatient Continuing Care Intensive Outpatient Long-term Residential Moderate-Term Residential Early Intervention Other Corrections Levels of Care Source: Dennis, Funk & Hanes-Stevens, 2008

66. Ratings of Problem Severity (x-axis) by Treatment Utilization (y-axis) by Population Size (circle size) 12% 20% 14% 8% 14% 12% -0.20 0.00 0.20 0.40 0.60 0.80 1.00 -0.200.000.200.400.600.801.00 Average Current Problem Severity Average Current Treatment Utilization. A Low-Low B Low- Mod C Mod-Mod D Hi-Low E Hi- Mod F. Hi- Hi (CC) G. Hi-Mod (Env Sx/ PH Tx) 9% H. Hi-Hi (Intx Sx; PH/MH Tx) 12%

67. Variance Explained in 10 NOMS Outcomes \1 Past month \2 Past 90 days *All statistically Significant 26% 24% 11% 25% 15% 33% 26% 18% 14% 8% 24% 0%5%10%15%20%25%30%35% No AOD Use No AOD related Prob. No Health Problems No Mental Health Prob. No Illegal Activity No JJ System Involve. Living in Community No Family Prob. Vocationally Engaged Social Support Count of above Percent of Variance Explained Source: CSAT 2007 AT Outcome Data Set (n=11,013)

68. Predicted Count of Positive Outcomes by Level of Care: Cluster A Low - Low (n=1,025) Person “A” does better in Outpatient Person “B” does better in Higher Levels of Care

69. Best Level of Care*: Cluster A Low - Low (n=1,025) Source: CSAT 2007 AT Outcome Data Set (n=11,013)

70. Best Level of Care*: Cluster B Low - Mod (n=1,654)

71. Best Level of Care*: Cluster C Mod-Mod (n=1209) Source: CSAT 2007 AT Outcome Data Set (n=11,013)

72. Best Level of Care*: Cluster D Hi-Low (n=687)

73. Best Level of Care*: Cluster F Hi-Hi (CC) (n=968) Source: CSAT 2007 AT Outcome Data Set (n=11,013)

74. Best Level of Care*: Cluster Cluster H Hi-Hi (Intx/PH/MH) (n=1,017)

75. Source: CSAT 2007 AT Outcome Data Set (n=11,013) Best Level of Care*: Cluster E Hi-Mod (n=1,190)

76. Best Level of Care*: Cluster G Hi-Mod (Env/PH) (n=749) Source: CSAT 2007 AT Outcome Data Set (n=11,013)

77. Best (x) by Actual (y) Level of Care Placement 3132 797 1968 2339 0 500 1000 1500 2000 2500 3000 3500 Outpatient (n=3132) Intensive Outpatient (n=797) OP - Continuing Care (n=1968) Residential (n=2339) Higher Best Lower

78. 553/771=72% unmet need 218/224=97% to targeted 771/982=79% in need Exploring Need, Unmet Need, & Targeting of Mental Health Services in AAFT Size of the Problem Extent to which services are currently being targeted Extent to which services are not reaching those in most need At Intake. After 3 mon No/Low Need Mod/High Need Total Any Treatment6218224 No Treatment205553758 Total211771982

79. Mental Health Problem (at intake) vs. Any MH Treatment by 3 months *3+ on ASAM dimension B3 criteria Source: 2008 CSAT AAFT Summary Analytic Dataset

80. Why Do We Care About Unmet Need? If we subset to those in need, getting mental health services predicts reduced mental health problems Both psychosocial and medication interventions are associated with reduced problems If we subset to those NOT in need, getting mental health services does NOT predict change in mental health problems Conversely, we also care about services being poorly targeted to those in need.

81. Residential Treatment need (at intake) vs. 7+ Residential days at 3 months Opportunity to redirect existing funds through better targeting Source: 2008 CSAT AAFT Summary Analytic Dataset

82. 82 Part 4. Summarize the status of efforts to make the data available for secondary analysis and translate the software, measures and reports from English into Spanish, French, Portuguese and other languages

83. 83 We currently pool data from Center for Substance Abuse Treatment (CSAT) grantees annual and make it available for secondary analysis: –Requires abstract length proposal/ feasibility –Requires agreement to respect privacy and not attempt to re-identify –We will get permission from any active grantees –No cost to the end user Over 36 scientist and evaluators have already accessed the data and about 1-2 more come get approval each month We can also negotiate access to additional data from individual grantees and/or regional projects Secondary Analysis

84. 84 Status of Translations Language Short Screener Other InstrumentsSoftwareReports English Done Spanish Done In progress French In progress Planned Portuguese DoneIn progressNot yetNot Yet Hmong, Japanese, Russian, Pilipino, Punjabi, Vietnamese DoneNot yet Not Yet

85. 85 Acknowledgments and Contact Information This presentation was supported by analytic runs provided by Chestnut Health Systems for the Substance Abuse and Mental Health Services Administration's (SAMHSA's) Center for Substance Abuse Treatment (CSAT) under Contracts 207-98-7047, 277-00-6500, 270-2003-00006 and 270- 2007-00004C using data provided by the following 152 grantees: TI11317 TI11321 TI11323 TI11324 TI11422 TI11423 TI11424 TI11432 TI11433 TI11871 TI11874 TI11888 TI11892 TI11894 TI13190TI13305 TI13308 TI13313 TI13322 TI13323 TI13344 TI13345 TI13354 TI13356 TI13601 TI14090 TI14188 TI14189 TI14196 TI14252 TI14261 TI14267 TI14271 TI14272 TI14283 TI14311 TI14315 TI14376 TI15413 TI15415 TI15421 TI15433 TI15438 TI15446 TI15447 TI15458 TI15461 TI15466 TI15467 TI15469 TI15475 TI15478 TI15479 TI15481 TI15483 TI15485 TI15486 TI15489 TI15511 TI15514 TI15524 TI15524 TI15527 TI15545 TI15562 TI15577 TI15584 TI15586 TI15670 TI15671 TI15672 TI15674 TI15677 TI15678 TI15682 TI15686 TI16386 TI16400 TI16414 TI16904 TI16928 TI16939 TI16961 TI16984 TI16992 TI17046 TI17070 TI17071 TI17334 TI17433 TI17434 TI17446 TI17475 TI17476 TI17484 TI17486 TI17490 TI17517 TI17523 TI17535 TI17547 TI17589 TI17604 TI17605 TI17638 TI17646 TI17648 TI17673 TI17702 TI17719 TI17724 TI17728 TI17742 TI17744 TI17751 TI17755 TI17761 TI17763 TI17765 TI17769 TI17775 TI17779 TI17786 TI17788 TI17812 TI17817 TI17825 TI17830 TI17831 TI17864 TI18406 TI18587 TI18671 TI18723 TI19313 TI19323 TI655374. Any opinions about this data are those of the authors and do not reflect official positions of the government or individual grantees. It is available at www.chestnut.org/li/posters. Comments or questions can be addressed to Michael Dennis, Chestnut Health Systems, 448 Wylie Drive, Normal, IL 61761. Phone 1-309-451-7801; E-mail: mdennis@chestnut.org. www.chestnut.org/li/posters mdennis@chestnut.org More information on the GAIN is available at www.chestnut.org/li/gain or by e-mailing gaininfo@chestnut.org.www.chestnut.org/li/gain gaininfo@chestnut.org