Using Multiple Methods to Identify Effective Clinician Incentives to Improve Depression Care

1
Using Multiple Methods to Identify Effective
Clinician Incentives to Improve Depression Care

• Lisa Meredith, Ph.D., RAND
• CCH/HSRC/EXPORT Methods Seminars
• March 7, 2006

2
Collaborators and Funding

• United Behavioral Health (UBH)
• Francisca Azocar, Ph.D.
• Joyce McCulloch, M.S.
• Robert Branstrom, Ph.D.
• RAND Corporation (RAND)
• Lisa S. Meredith, Ph.D
• Michael Schoenbaum, Ph.D.
• University of California, Los Angeles (UCLA)
• Susan Ettner, Ph.D.
• Mindy Morefield, M.A.
• Yuting Wong
• Supported by NIMH grant P30 MH 068639 to Dr.
  Kenneth B. Wells

3
Outline

• Overview of study background and objectives
• Methods used
• Community-based participatory research
• Multiple stakeholders
• Semi-structured interviews
• Modified Delphi panel
• Administrative data analysis
• Preliminary findings integrated throughout each
  of the different methods

4
Can incentives influence evidence-based care for
depression in a MBHO?

• Gap between evidence-based practice and health
  care as typically practiced in the real world
  (Quality Chasm report IOM, 2001)
• Changing the behaviors/practices of clinicians
  can increase evidence-based care
• Little is known about efficient and generalizable
  strategies for motivating clinicians to adopt
  evidence-based care
• No work on use of incentives with mental health
  specialists only primary care (RWJF)
• This project builds on and complements primary
  care incentive demonstrations

5
Study complements and builds on prior work in
area of clinician incentives

• Elicit and identify promising and feasible
  clinician incentive strategies (financial and
  non-financial) that will improve care for
  depression within a large MBHO
• Select the most promising option among the
  feasible alternatives
• Conduct a small pilot study to assess whether the
  most promising option is indeed feasible
• Develop and submit a follow-on R01 proposal that
  would implement and evaluate the most feasible
  clinician incentive packages

6
Clinician Incentives Pilot Development
Timing
Goals
Task

• Identify potential (modifiable) clinician
  practices.
• Understand how each practice will improve
  carefor depression (potential impact of change,
  pros and cons, sustainability for behavior
  change).

Round 1 Interviews
Dec 04 to March 05
(Research team picks 5 practices with most
potential)

• Identify incentives to clinician behavior change
  (pros and cons, impact, and sustainability).
• Query experiences with the incentive structures
  including consequences to implementation.

Round 2 Interviews
May and June 2005
(Research team generates list of
practice/incentive packages)

• Assess real-time feasibility of packages.
• Explore ability of databases to support
  intervention.
• Determine if reliable quality indicators can be
  constructed from existing claims and drug data.

Administrator Interviews
June and Nov 2005
(Research team conducts analysis to assess
tracking capability)

• Present top 5 options and elicit feedback.
• Engage group in discussion of practice/incentivep
  airs to pick most viable option.

Consensus Panel
November 2005
Pilot Study to determine feasibility
7
Community-Based Participatory Research (CBPR)
8
Community-Based Participatory Research(CBPR)

• Collaborative research approach that involves all
  partners in the research through an equitable
  process
• Recognizes the unique strengths that each brings
  to the project
• Process begins with
• a research topic that is important to the
  community
• combines knowledge with action to achieve social
  change
• To improve health outcomes/eliminate disparities

Lasker Weiss, 2003 Minkler Wallerstein, 2003
9
Community Partnership

• United Behavioral Health (UBH)
• One of the largest MBHOs in the country
• Serving 1,700 customers 24 million members
  nationwide
• UBH customers are large and small employers from
  different industries, health plans and public
  sector entities
• Their Employer and Health Plan Divisions and
  Public Sector are served by 7 care management
  centers in different regions of the country
• UBH service products include EAP, behavioral
  health, and disability services
• A national provider network with over 62,000
  clinicians
• Multiple specialties
• 7,500 psychiatrists 13,400 psychologists24,400
  masters level counselors 530 psychiatric nurses
  
• 2,500 facilities with locations in every state

10
What did the partnership do?

• Community partners (UBH administrators and
  clinicians) identified potential practices to
  adoptor change to improve care for depression
• Research team reviewed the list of practices and
  ranked them for feasibility to select the Top 5
• Community partners identified potential
  incentives to pair with the Top 5 practices
• Partners ranked the Top 5 practices in
  conjunction with their preferred
• Financial incentive
• Non-financial incentive

11
Semi-Structured Interviews
12
Round 1 Interviews Identify Clinician
Behaviors to Improve Depression Care

• 30-minute semi-structured telephone interviews
  with 14 stakeholders from MBHO or who contract
  with MBHO
• Initial contact letter co-signed by Ettner,
  Meredith, and Azocar with letterhead from all 3
  organizations
• Follow-up call to schedule interview 2 weeks
  after receipt of letter
• Open-ended questions about professional
  characteristics, potential clinician practice
  patterns, advantages/disadvantages, impact etc.

13
Stakeholder Characteristics (N14)
Includes 1 administrator, 2 quality/clinical
education specialists, and 1 medical
director Includes 2 benefits consultants, and 1
benefits manager
14
Top 5 Behaviors for Targeting Change
15
Round 2 Interviews Identify Clinician
Incentives to Pair with Top 5 Behaviors

• 15-minute telephone interviews with 9 of initial
  14 stakeholders from MBHO
• Follow-up letter co-signed by Ettner, Meredith,
  and Azocar with letterhead from all 3
  organizations
• Follow-up call to schedule interview 2 weeks
  after receipt of letter
• Questions about personal experience with
  clinician-based interventions, incentives to pair
  with top 5 behaviors, barriers associated with
  each

16
Results Types of Incentives
Money Resources Opportunity
Provide opportunities for clinicians to talk
with each other. Give the clinician more
empowerment CEUsWhen people feel more
informed, then they feel more competent.
more money, to attract more psychiatrists Ev
eryone wants to be paid more. You could also
give clinicians a bonus if
There is a resource issue because of the
national shortage of psychiatrists Provide
educational materials to therapists free of
charge
17
Classification of Incentives (Round 2
Interviews)
18
Modified Delphi Panel
19
Instead of relying on only one idea
20
Group Decision Process Delphi Method

• Structured process for collecting and distilling
  knowledge from a group of experts using
  questionnaires combined with controlled opinion
  feedback
• Makes discussion between experts possible
  without permitting a certain social interactive
  behavior as happens during a normal group
  discussion and hampers opinion forming (Wissema,
  1982)
• Developed by RAND in 1969 for technological
  forecasting
• Adopted for application in fields that have not
  yet developed to the point of having scientific
  laws e.g., education and health
• Recognizes human judgment as legitimate and
  useful input in generating forecasts

21
10 Steps for the Delphi Method

• Form a team to undertake and monitor a Delphi on
  a given subject
• Select one or more panels to participate in the
  exercise (panelists are usually content experts
  in the area to be investigated)
• Develop the first round Delphi questionnaire
• Test the questionnaire for proper wording (e.g.,
  ambiguities, vagueness)
• Administer the first questionnaires to the
  panelists
• Analyze the first round responses
• Prepare a second round of questionnaires (and
  possible testing)
• Administer the second round questionnaires to the
  panelists
• Analyze the second round responses (Steps 7 to 9
  are reiterated as long as desired or necessary to
  achieve stability in the results)
• Prepare a report and present the conclusions of
  the exercise

22
(No Transcript)
23
Top 5 Practices and Selected Incentives
Practices (columns) are ordered by rank (highest
on the left) CTCombination Therapy,
ATAntidepressant Treatment, RARapid Access to
Care, PEPatient Education, CEClinician
Education. Entries are the number of pre-panel
survey respondents selecting each type of
incentive.
24
Top 5 Practice-Incentive Pairs

• Combination Therapy
• Pay more to increase access to psychiatrists
  (tie)
• Bonus for incorporating patient education into
  treatment (tie)
• Provide feedback on patient or clinician outcomes
• Antidepressant Treatment
• Increase amount of reimbursement (tie)
• Pay more to increase access to psychiatrists
  (tie)
• Bonus for incorporating patient education into
  treatment (tie)
• Provide feedback on patient or clinician outcomes
  (tie)
• Access to preferred physicians (tie)
• Rapid Access to Care
• Pay more to increase access to psychiatrists
• Access to preferred physicians
• Patient Education
• Bonus for incorporating patient education into
  treatment
• Provide readily available information (free)
  through plan
• Clinician Education
• Bonus for incorporating patient education into
  treatment
• CEU-based education/training

25
Get more patients into psychotherapy AND
antidepressant therapy
26
(No Transcript)
27
Top 5 Practice-Incentive ProgramsFinal Panelist
Rankings (1-4, 1highest/first choice)
28
Results From Panel Meeting

• Top Practice Chosen Antidepressant Treatment
• Definition Patients on antidepressants receive
  adequate dosage and duration
• Financial Incentive Pay bonus for each patient
  meeting standard OR pay overall bonus if gtX of
  psychiatrists patients meet the standard
• Non-financial Incentive Provide ongoing
  feedback to psychiatrist about patient
  prescription filling
• Runner-Up Practice Chosen Combination Therapy
• Definition Patients receive antidepressant
  medication psychotherapy
• Financial Incentive Pay therapist per patient
  bonus for increasing access to psychiatrist OR
  for an additional session (with a new billing
  code) to work with the care manager to get the
  patient to psychiatrist
• Non-financial Incentive Incorporate education
  into treatment by giving clinicians on-line tools
  and mailed information

29
Administrative Data Analysis
30
UBH Sources of Information

• Lots of data but limited clinical information
• Claims
• Behavioral
• Medical pharmacy (only available for 40)
• Outpatient Treatment Review forms
• Audits
• Surveys
• Wait-period for claims
• Behavioral 40 of claims arrive in 2 weeks 98
  arrive within 90 days
• Pharmacy 7 days if through UHC

31
Select Databases

• UBH Enhanced Outpatient Model Implemented 1/1/06
  includes
• Open certification no authorization required
• Member Wellness Survey (RR pilot27) at baseline
  6 months measures outcomes
• Algorithms flag 2 of most severe cases at
  baseline
• Per patient claims report provides UBH Care
  Manager with diagnosis, type amount of service
  use
• Care Manager calls either clinician and/or
  patient to follow up with treatment plan

32
Implementation Questions

• Can an incentive enhance the proportion of MDD
  cases that get combination treatment?
• Preliminary process and system analyses
• What proportion of flagged patients get combo
  therapy currently with this care management
  system?
• If close to 100, nothing to incentivize
• What proportion of cases are already receiving
  antidepressants from their PCP (data not
  available to UBH in real-time)?
• How long for real-time behavioral and pharmacy
  claims data to answer this question?
• Can we talk with some care managers about how new
  system is working?
• Wait 3-4 months for PBH merger dust to settle?