Next-Generation GAIN Software

1
Next-Generation GAIN Software

• David Smith, B.G.S. and Michael L. Dennis, Ph.D.
• Chestnut Health Systems, Bloomington, IL
• Workshop Presentation for the Joint Meeting on
  Adolescent Treatment Effectiveness
• Baltimore, Maryland
• March 29, 2006, Maryland A Room
• Preparation of this presentation was supported by
  funding from the Center for Substance Abuse
  Treatment (CSAT Contract no. 270-2003-00006). The
  contents of this presentation are the opinions of
  the authors and do not reflect the views or
  policies of the government. Available on line at
  www.chestnut.org/LI/Posters or by contacting Joan
  Unsicker at 720 West Chestnut, Bloomington, IL
  61701, phone [phone], fax (309)
  829-4661, e-Mail junsicker_at_chestnut.org

2
This think tank will

• Emphasize our commitment to making the GAIN
  Software widely available, adaptable and
  compatible with existing and new systems
• Provide a brief overview of the GAIN with some of
  the implications for the GAIN Software
• Review the history of GAIN Software
• Discuss features and capabilities that we would
  like to add to the next-generation of GAIN
  Software
• Solicit your input on what would make the GAIN
  Software more useful to clinicians, clinical
  researchers, and software developers.

3
GAIN Overview

• The Global Assessment of Individual Needs (GAIN)
  is a family of assessment instruments that is
  widely used in research and clinical settings
  throughout the United States and Canada.
• The GAIN has played a significant part in the
  renaissance of adolescent treatment research and
  is on the leading edge of the innovative use of
  assessment data in both research and clinical
  practice for adolescents and adults.

4
The GAIN is a Family of Instruments

• There are seven primary instruments
• GAIN-I a 100-page comprehensive biopsychosocial
  instrument
• GAIN-M90 the follow-up version of the GAIN-I
• GAIN-SS a 2-page screener for general
  populations
• GAIN-Q a 10-page quick assessment
• GAIN-QM the follow-up version of the GAIN-Q
• GAIN-CI a 58-page collateral instrument
• GAIN-CM the follow-up version of the GAIN-CI.

5
The GAIN-I is Comprehensive

• The current GAIN-I has
• A total of 1936 possible questions
• Hundreds of related instructions, transition
  statements and other text items
• 156 skips or conditional branches
• 314 internal consistency checks
• Hundreds of calculated variables per case to
  support clinical diagnosis and placement
  decision-making

6
The GAIN Instruments are Customizable

• Most GAIN instruments are customized
• Each is available as a Core with a set of
  required questions, and a Full with optional
  questions added.
• In addition, the makeup of the Core can vary
  by
• Individual Studies
• Regional Systems
• Individual Agencies or sites
• Populations within sites
• And Special Study questions can be added to the
  end of most instruments.

7
The GAIN is Constantly Evolving

• 1993 GAIN 1.x created for NIDA Training and
  Employment Program (TEP) as an integrated
  clinical and research instrument based on ASI,
  IAP, DATOS, several existing scales.
• 1996 GAIN 2.x revised for Drug Outcome Monitoring
  Study (DOMS) to focus more specifically on DSM,
  ASAM, JACHO/CARF and map onto epidemiological
  data based.
• 1998 GAIN 3.x revised for CYT and ATM to address
  problems in DOMS and incorporate GPRA versions 1
  2.
• 2000 GAIN 4.x revised to include several new
  modules to address specific NIDA and NIAAA
  research studies (not widely used).
• 2002 GAIN 5.x revised for SCY and other CSAT
  adolescent treatment studies to incorporate
  reasons for quitting, treatment history process
  measures, GPRA versions 3 4, several state
  reporting requirements. Currently on its fourth
  major revision (version 5.4.0).

8
The GAIN is Widely Used in Research

• 1995-1997 Drug Abuse Treatment Outcome Study
  (DOMS)
• 1997-2000 CSATs Cannabis Youth Treatment (CYT)
  experiments
• 1998-2003 CSATs 10 Adolescent Treatment Models
  (ATM)
• 2000-2003 CSATs Persistent Effects of Treatment
  Study (PETS-A)
• 2002-2007 CSATs 12 Strengthening Communities for
  Youth (SCY)
• 2002-2007 CSATs 12 Targeted Capacity Expansion
  TCE/HIV
• 2003-2009 NIDAs 14 individual research grants
  and CTN studies
• 2003-2006 CSATs 17 Adolescent Residential
  Treatment (ART)
• 2003-2007 CSATs 38 Effective Adolescent
  Treatment (EAT)
• 2004-2007 NIAAA/CSATs study of diffusion of
  innovation
• 2004-2009 CSAT 22 Young Offender Re-entry
  Programs (YORP)
• 2005-2008 CSAT 20 Juvenile Drug Court (JDC)
• 2005-2008 CSAT 16 State Adolescent Coordinator
  (SAC) grants

9
The GAIN is Widely Implemented

• The GAIN has played a role in
• Most of the studies that have supported the
  current Renaissance of Adolescent Treatment
  Research
• The development of clinical expert systems and
  statistical models to improve diagnosis,
  placement, treatment planning, program
  evaluation, and economic evaluations and
• Creating the infrastructure supporting the move
  toward evidence based practice.

10
All Adolescent and Adult, Clinical and Research
Sites Using the GAIN since 1993
New Hampshire
Washington
Maine
Minnesota
Montana
Vermont
North Dakota
Oregon
Massachusetts
Wisconsin
Idaho
New York
South Dakota
Michigan
Rhode Island
Wyoming
Iowa
Pennsylvania
Connecticut
Nebraska
Nevada
Indiana
Ohio
New Jersey
Illinois
West Virginia
Utah
Colorado
Delaware
Virginia
Kansas
Missouri
District Of Columbia
Kentucky
California
North Carolina
Maryland
Tennessee
Oklahoma
Arkansas
South Carolina
Arizona
New Mexico
Mississippi
Alabama
Georgia
Number of Sites
Texas
Louisiana
None (yet)
1 to 14
Alaska
15 to 30
Florida
31 to 88
Hawaii
3/06
Puerto Rico
11
Where is the GAIN Going?

• Growth
• The number of sites has doubled every year for
  five years and is projected to continue to do so
  for the next five years.
• Better Integration
• Across records from multiple sources (e.g.,
  participant, collateral, urine) and/or over time.
• Into existing and new information systems
  supporting diagnosis, placement, treatment
  planning, monitoring, and billing.
• Decision Support
• Clinical, including how to better use assessment
  information in diagnosis, placement, and
  treatment planning.
• Supervisory, including monitoring of and
  technical assistance to staff, grantee or clinic
  sites to support supervisors, administrators, and
  funders.

12
Where is the GAIN Going?

• Flexibility
• More specialized versions, different languages,
  self administration, and better
  modularization/set up for local customization
  (subsets, new items).
• Technical Innovation
• Incorporation of computer adaptive testing (CAT)
  to shorten the administration time and other
  complex statistical modules to improve validity
  and provide clinical guidance.

13
The Evolution of GAIN Software

• GAIN Software has been evolving for over a
  decade
• 1993 Version 1 (FICS) was written in Fortran on
  DOS for the PC-AT, math-coprocessor emulators,
  24-page narrative report that nobody read.
• 1997 Version 2 (DOMS) written in Microsoft
  Access95 with direct synchronization and a
  statistical summary.
• 1998 Version 3 (ABSLite) written in Access97
  for data-entry only with direct synchronization
  and data exports, limited reports.
• 2000 Version 4 (ABS) written in Visual Basic
  with Jet database engine with data export and
  statistical summaries (ICP).
• 2003 Version 5 (ABS) Update of software to
  address HIPAA requirements, allow interactive
  interviews and add clinical narrative report
  (GRRS), GPRA tool and GRL.

14
Current GAIN Software

• Working in hundreds of agencies around the
  country.
• Easy to use user training takes less than half
  a day.
• Clinical reports available immediately after an
  online interview or after data has been keyed.
• Data can be pooled over server/network/internet,
  uploaded from a remote/laptop on demand, or
  exported and sent via e-mail, FTP, or HTP (in a
  password-protected file).
• Features for interactive administration, data
  entry, editing, note making, rekeying and
  resolution, read-only, report generation.
• Privacy/security features to aid in complying
  with HIPAA, 42 CFR and other privacy and security
  policies.
• Deployable over LAN/WAN and Internet using
  Terminal Server or Citrix.

15
Moving to the Next-Generation GAIN Software

• Up to now weve talked about the context in which
  were working on the new GAIN Software.
• Now wed like to talk about how we are
  envisioning that new software and begin what we
  hope will be an ongoing conversation about how it
  will look and work.

16
Initial Decisions

• The GAIN is too complex for the GAIN Coordinating
  Center to be able to define and test for multiple
  developers of GAIN software.
• The GCCs core competency is the GAIN content
  we dont want to become a big software developer.
• We want to create new GAIN software, not a new
  case management system.

17
What We Want From a New System

• Anyone designing a new system these days wants
  many of the same things
• Open
• Scalable
• Secure
• Flexible
• Maintainable
• Internet-capable

18
What We Want From a New System

• In our environment there are several other things
  that are important to us
• Maximum Clinical utility
• Maximum Research utility
• Easy interaction with statistical/software/CAT
  modules
• Easy interface with other systems

19
Next-Generation

• We want to build on our existing software
  strengths
• End-user ease of use
• Support for Data Submission process
• While we add
• Ease of Startup/Implementation
• Ease of Local Support
• Ease of Tailoring Instruments
• Ease of integration with other systems
• Robust, modern, standard platforms
• Enhanced remote access
• Enhanced research and clinical utility
• Flexibility and extensibility

20
Next-Generation

• Overview of typical system
• Implementation Options (CHS Hosted, Other Hosted
  (Datacenter), Local Installation, Laptop)
• Focus on features
• Customization (Templates)
• Clinical Reporting (GRRS)
• Integration
• Discussion

21
Next-Generation GAIN Software Overview

• Based on
• Web browser interface
• SQL database
• .NET codebase

22
GAIN Software System Architecture
23
Implementation Options

• Internet Hosted Cross-system
• Internet Hosted System-based
• Locally Hosted
• Stand-alone

24
Hosting Options
25
Publicly Hosted
26
Agency Hosted
27
3rd Party Interaction
28
GRRS Process