• Background
  
• Organization
  
• Versions and Administration Time
  
• Summary of Psychometrics.
  
• Copyright and Availability
  
• References
  
• Table 1 Areas of the Assessment in the GAIN
  
• Table 2. Time to Administer the GAIN by Population, Mode, and Level of Care
  
• Table 3. Reliability of Scales

Background. Dennis and colleagues (1993; 1995; 1996; 1998) have been developing the Global Appraisal of Individual Needs (GAIN) to implement an integrated bio-psycho-social model of treatment assessment, planning and outcome monitoring that can be used for evaluation, clinical practice and administrative purposes. The current version was adapted for the Substance Abuse and Mental Health Services Administration’s (SAMHSA’s) Center for Substance Abuse Treatment’s (CSAT’s) Cannabis Youth Treatment (CYT) cooperative agreement study (Dennis, Diamond, Donaldson, Godley, Kaminer, Tims and others, 1998; CSAT TI 11320), but has actually evolved over a half dozen grants from CSAT, the National Institute on Drug Abuse, The National Institute on Alcohol Abuse and Alcoholism, and the Interventions Foundation. The GAIN embeds questions for documenting substance use disorder, attention deficit/hyperactivity disorder, oppositional defiant disorder, conduct disorder, and pathological gambling, (based on APA, 1994), dimensional patient placement criteria for intoxication/withdrawal, health distress, mental distress, and environment distress to guide movement among and between levels of care (based on ASAM, 1996), treatment planning (based on JCAHO, 1995), reporting requirements related to the state client data system (OAS, 1995), and measures of a core set of clinical status and service utilization outcomes used in the Drug Outcome Monitoring Study (DOMS; Dennis et al., 1995). The GAIN focuses on face valid measures of recency, past year symptoms and behavioral counts for the past 90 days because (in addition to use in scales) the individual item answers need to be communicated/understood by a wide range of clients, clinicians and policy makers (Sechrest, McKnight, & McKnight, 1996; JCAHO, 1995). Several of these measures have also been designed to compare with the National Household Survey on Drug Abuse (NHSDA; OAS, 1995), French and Martin’s (1997) estimates of unit cost of drug-related outcomes, and/or other major epidemiological data sets. The instrument has also been set up to be flexible and allow either oral or a proctored (i.e., staff observed) form of self-administration to adults and/or adolescents in both outpatient and inpatient settings.

Organization. The content of the GAIN is divided into eight areas: background and treatment arrangements, substance use, physical health, risk behaviors, mental health, environment, legal, and vocational. In each area, the questions check for major problem areas and the recency of any problems. If a given problem occurred in the past year, additional symptom-based questions (e.g., criteria for alcohol dependence) are asked for the past year to clarify the problem. If it occurred in the past 90 days, detailed behavioral counts are collected (e.g., days of alcohol use, days of drinking 5+ drinks per day, etc.). The GAIN also asks detailed questions about lifetime and current (past 90 days) service utilization, as well as changes in the client’s cognitive state (e.g. self efficacy to resist alcohol use, resistance to treatment, motivation to be in treatment, and what services the client currently wants from treatment). Table 1 provides a summary of its core measures related to both clinical status and service utilization. It can be administered orally or initially done as a self-administered assessment by a client (but then needs to be reviewed). It is designed to facilitate clinical reviews by keeping scales in modules on the same page and numbering them together.

Versions and Administration Time. There are currently two main versions: the GAIN-Initial (Version GAIN-I 12/99: 90 min) and the GAIN-Monitoring 90 days (Version GAIN-M90 12/99: 20 min.). The GAIN-I is a detailed bio-psycho-social that has many "skip outs" so administration time naturally varies depending on how much has been going on in the individual’s life. It also varies by how it is delivered. Table 2 provides data on the time to complete a GAIN-I for 168 adolescents and 73 adults. There is considerable range in the expected duration of the GAIN depending on population, mode and level of care/severity. Adults in general, adults entering outpatient treatment and adolescents entering short-term residential treatment generally took longer. Oral administration was always faster - often by 20% or more. Other administration refers to doing a proctored self-administration (i.e., with staff present), doing this and oral on some parts, and/or doing split assessments. The latter involves asking a select number of items to decide whether to admit and where to place someone (about 20 minutes), then going back over the whole assessment to complete the information with only those who were admitted after they were in treatment (time estimates are only for those who did get admitted and completed the whole assessment). Because it contains fewer skip-able questions and is orally administered, administration time on the GAIN-M90 is almost always completed within 15-30 minutes with a median time of approximately 22 minutes.

Summary of Psychometrics. In both adolescent and adult outpatient and inpatient samples, the GAIN has repeatedly demonstrated excellent internal consistency on core scales related to the frequency of substance use (alpha=.8+), alcohol use (alpha=.9+), drug use disorders (Cronbach’s alpha =.9+), and alcohol use disorders (alpha=.9+). The drug frequency index (days of use, days of heavy use, days of problem use, days of marijuana use, days of cocaine use, days of heroin use) is more correlated with lifetime drug problems (r=.36, p <.0001) than days of any use alone is (r=.29, p<.0001). The alcohol frequency index (days of use, days of heavy use and days of use causing problems) is more correlated with our measure of lifetime alcohol problems (r=.51, p<.0001) than simple days of alcohol use (r=.41, p<.0001). Excellent internal consistency has also been established for measures of general mental distress, PTSD, ADHD, conduct disorder, traumatic victimization, social support, environmental risk, illegal activity, and school problems. Table 3 provides a summary of the internal consistency for an initial sample of 136 adolescents and 211 adults (Dennis, Funk, et al., 1998). For both adults and adolescents, the GAIN is also able to reliably predict initial level of care placements (Kappa .4-.6 depending on model) and clinical subgroups that are used for making case mix adjustments between treatment units or levels of care (Kappa=.85 to .89 depending on model). The former is more reliable than most modern psychiatric diagnostic tests (which typically range from .4 to .6), and the latter is considered close to perfect (Kraemer, 1992).

Copyright and Availability. The GAIN and its products are tools that are proprietary products owned by Chestnut Health Systems either exclusively or jointly and protected under U.S. copyright laws. The current work is in beta test form, but can be used for evaluation and research under a non-exclusive, non-transferable, limited license at the cost of $1 plus any materials/assistance requested. Copies of the instruments, general terms of the license agreement and other products can be downloaded for inspection-only from www.chestnut.org/li/cyt/gain. For hard copies or further assistance please contact Joan Unsikcer by e-mail at junsicker@chestnut.org or directly at the Lighthouse Institute, Chestnut Health Systems, 720 West Chestnut, Bloomington IL, 61701, Telephone: 309-827-6026, Fax : 309-829-4661.

References

Table 1 Areas of the Assessment in the GAIN

Substance Use

Core Clinical Outcomes: Recency and days of alcohol & drug use (overall & by 11 drug types ‡), days of heavy use ‡, days use interfered with life, peak use of alcohol, marijuana, and other drugs; Clinical Symptom Counts: Alcohol and drug problems/abuse/dependence overall and by drug*‡, withdrawal symptoms*, treatment acceptance/resistance, problem orientation, self-efficacy to resist drugs*; Core Utilization Outcomes: Recency and number of days in detoxification ‡, self help ‡, and 11 types of substance abuse treatment levels of care; current use of medication for substance abuse problems; Other Utilization Data: Lifetime number detoxification, self-help, 11 types of substance abuse treatment, and identification of prior providers; urgency and specific list of things client wants.

Physical Health

Core Clinical Outcomes: Physical health distress ‡, days of medical problems, days interfered with life; recency of pregnancy ‡ and baby birth weights ‡§; recency of infectious diseases ‡¥; Clinical Symptom Counts: Health distress*‡, specified medical conditions (by ICD-9 areas)§¥; Core Utilization Outcomes: Recency and number of health-related emergency room visits ‡§, nights in hospital ‡§, outpatient surgical procedures, and outpatient doctor visits; current use of medication for physical health problems §; Other Utilization Data: Lifetime number emergency room admissions, hospital stays, outpatient surgery, and identification of prior providers related to physical health problems; urgency and specific list of things client wants.

Risk and Prevention

Core Clinical Outcomes: Days used ‡ and shared needles; times had sex and unprotected sex ¥; number of sexual partners; days without food; days of exercise; Clinical Symptom Counts: Needle risk behaviors*‡, sexual risk behaviors ¥*, risk reduction (attempts to change); Core Utilization Outcomes: Number of diet/nutrition, exercise, HIV, other health prevention, and relaxation classes; Other Utilization Data: Urgency and specific list of things client wants.

Mental and Emotional Health

Core Clinical Outcomes: Recency and days of mental problems ¥, days interfered with life, recency and days bothered by bad memories, recency and days of attention or behavior problems; Clinical Symptom Counts: Somatic/depressive/anxiety symptoms ¥*, traumatic stress symptoms ¥*, attention deficit/hyperactivity/conduct disorder symptoms*, personality disorder symptoms (by DSM-IV cluster); Core Utilization Outcomes: Recency and number of mental-health-related emergency room visits ‡¥, nights in the hospital‡§, and outpatient doctor visits ‡§; current use of medication for mental health problems; Other Utilization Data: Lifetime number emergency room admissions, hospital stays, prior diagnoses and identification of prior providers related to mental health problems; Urgency and specific list of things client wants.

Environment and Interpersonal Relationships

Core Clinical Outcomes: Recency of homelessness ¥, days lived in housing that had alcohol and/or drug use ¥; days lived in a controlled environment; recency and frequency of violent behaviors ¥; recency and days physically, sexually or emotionally abused; Clinical Symptom Counts: Interactions with children, child functioning §, living/vocational/social environment risk, verbal/physical violence*¥, victimization/traumatization §, psycho-social stressors*, social support*, spirituality and religious support*; Core Utilization Outcomes: Days in public and emergency housing ¥; child-days of foster care §; child-days of institutional care §; Other Utilization Data: Urgency and specific list of things client wants.

Legal

Core Clinical Outcomes: Recency and days of illegal activity ¥; days of primary support from illegal activity ¥; Clinical Symptom Counts: Property/interpersonal/substance- related illegal activities ‡§*; Core Utilization Outcomes: Times arrested overall and by offense ‡§; days in probation ‡§, parole/jail/prison ‡§, and juvenile detention; Other Utilization Data: Urgency and specific list of things client wants.

Vocational

Core Clinical Outcomes: Recency and days of school/training and work attended, sick while there and missed ‡§; recency and days of financial problems; and recency and days of gambling; Clinical Symptom Counts: Training problems §*, work problems ‡*, financial problems*, gambling problems§*; Core Utilization Outcomes: Days in training ‡; forms of public assistance received § and poverty level; Other Utilization Data: Urgency and specific list of things client wants.

Table 2. Time to Administer the GAIN by Population, Mode, and Level of Care

		Minutes
Population/Mode/Level of Care	N	Min	Median	90th Percentile	Max
Adolescents (All)	168	15	69	120	795
Oral Administration (Across Levels)	73	27	60	113	780
Outpatient	27	27	53	88	105
Short-term Residential	25	30	65	107	780
Long-term Residential	21	45	72	135	735
Other Administration (Across Levels)	95	15	75	120	795
Outpatient	22	25	62	105	146
Short-term Residential	28	30	100	150	795
Long-term Residential	45	15	67	120	145
Adults (All)	73	30	86	127	360
Oral Administration (Across Levels)	2	30	45	60	60
Methadone	1	60	60	60	60
Residential	1	30	30	30	30
Other Administration (Across Levels)	71	30	90	129	360
Outpatient	16	50	100	270	360
Methadone	36	30	95	130	232
Residential	19	30	60	90	115
Combined	241	15	75	120	795
Oral Administration	75	27	60	112	780
Oral Administration	166	15	80	125	795

Table 3. Reliability of Scales

		Adolescents (n=136)		Adults (n=211)
Index	No. of Items	Mean Inter-Item r	Cronbach’s Alpha	Mean Inter-Item r	Cronbach’s Alpha
A1. Substance Abuse and Dependence
Alcohol Issues Index (AII)	5	0.42	0.78	0.42	0.78
Alcohol Abuse Index (AAI)	4	0.49	0.79	0.52	0.81
Alcohol Dependence Index (ADI)	7	0.41	0.83	0.62	0.92
Alcohol Problem Index (API)	16	0.43	0.92	0.50	0.94
Alcohol Dependence Index Lifetime (ADIL)	7	0.36	0.79	0.62	0.92
Drug Issues Index (DII)	5	0.34	0.72	0.47	0.81
Drug Abuse Index (DAI)	4	0.37	0.70	0.48	0.79
Drug Dependence Index (DDI)	7	0.38	0.81	0.60	0.91
Drug Problem Index (GDPI)	16	0.37	0.90	0.49	0.94
Drug Dependence Index Lifetime (DDIL)\1	7	0.23	0.69	0.49	0.86
B1. Current Intoxication and Withdrawal
Alcohol Frequency Index (AFI)	3	0.83	0.93	0.77	0.91
Drug Frequency Index (DFI)	3	0.63	0.83	0.68	0.86
Current Withdrawal Index (CWI) (past wk)	22	0.30	0.91	0.40	0.94
B2. Physical Health/Biomedical Conditions
Health Distress Index (HDI) hdi8 (vers. 1296+)\2	8	0.24	0.62	0.33	0.75
Health Problems Index (HPI) hPi3z(vers. 1296)	3	0.72	0.88	0.28	0.81
B3. Mental/Emotional/Behavioral Conditions
Somatic Symptom Index (SSI)	4	0.39	0.68	0.67	0.89
Depressive Symptom Index (DSI)	6	0.57	0.87	0.79	0.96
Anxiety Symptom Index (ASI)	11	0.28	0.78	0.58	0.94
General Mental Distress Index (GMDI)	21	0.32	0.90	0.60	0.97
B4. Treatment Acceptance and Resistance
Treatment Motivation Index (TMI)	5	0.33	0.72	0.18	0.52
Treatment Resistance Index (TRI)\3	3	0.39	0.64	0.34	0.60
B5. Relapse Potential
Self Efficacy Index (SEI)	5	0.43	0.80	0.21	0.54
Problem Orientation Index (POI)	5	0.52	0.85	0.13	0.38
B6. Recovery Environment and Social Support
Living Environment Risk Index(LERI) (1296)\4	6	0.10	0.33	0.35	0.72
Vocational Environment Risk Index (VERI) (1296)	6	0.21	0.61	0.49	0.80
Social Environment Risk Index (SERI) (1296)\4	6	0.23	0.69	0.50	0.85
Environmental Risk Index (ERI) (all ver)	15	0.16	0.75	0.27	0.85
Other
General Victimization Index (GVI)	15	0.31	0.88	0.30	0.87
Personal Sources of Stress Index (PSSI)\3	15	0.19	0.47	0.22	0.59
Other Sources of Stress Index (OSSI)\3	7	0.19	0.60	0.35	0.80
General Social Support Index (GSSI)	9	0.37	0.84	0.38	0.85
Illegal Activities Index (IAI3z)	3	0.49	0.74	0.57	0.80
Training Index (TI3z) (days in school)	3	0.70	0.87	0.58	0.79
Vocational Index (VI7z)	7	0.31	0.75	0.29	0.74