An Evaluation of a Value-Based Health Plan Design at Group Health

1. An Evaluation of a Value-Based Health Plan Design at Group Health David Grossman, MD, MPH Group Health Research Institute Seattle, Washington

2. Disclosures and Funding • PI is employee and shareholder, Group Health Permanente medical group • Funding from AHRQ (R01 HS018913-01) and Group Health Cooperative

3. Improving Value of US Healthcare Expenditures • Increased purchaser focus on: • Improving value of expenditures • Reducing waste • Improving health outcomes for beneficiaries • Preventing chronic illness and complications • Two main levers • Health plan design • Delivery system design

4. Cost-Sharing and Health • Impact of larger cost-shares on chronic disease self-management • Chronic disease the major driver in health care costs • Re-consideration of indiscriminate cost-sharing • Consumer holds the early short term risk • Purchaser/health plan holds the longer term risk

5. Value-based Cost-Sharing • First iterations: • Preventive service coverage • Tiered pharmacy benefits • Generics • Brand-name • Non-preferred and non-formulary • Most recent efforts focused on pharmacy cost-sharing: reducing cost-shares • Pitney Bowes • University of Michigan employees

6. Science of Value-Based Design • Large body of evidence on impact of increased cost-shares • Tends to be focused on discrete services • Much smaller literature on impact of reducing cost shares • Even smaller literature on impact of cost-sharing on health outcomes and productivity • Tiny literature using control group with multiple outcomes

7. Worksite Wellness • Another approach to reducing costs and improving health • Focus on lifestyle change • Incenting health behavior • Healthy work environments • Change of work culture • Outcomes of interest • Health status and utilization • Absenteeism and presenteeism • Productivity • 77% of large employers offer these services • Health risk assessments are entry portal for engagement

8. Group Health’s Total Health Plan for Employees • Employer Aims • Improve productivity through • Better health of staff • Decreased absences • Improved on-the-job productivity • Decrease health expenditure trend rate • Mechanism • Incent healthy behaviors and improved chronic disease control through monetary incentives and value-based health benefit pricing • Reinforce culture of self-awareness, accountability and reporting of health and health behaviors through monetary incentives and culture change

9. Specific Aims • To assess the impact of the new value-based insurance design on: • PRIMARY: changes over time in employee self-reported: • health status • absenteeism due to illness and disability • presenteeism (i.e. lost productivity time at the workplace) • SECONDARY: • clinical quality scores for chronic illness care and preventive screenings, • lifestyle behavioral risk factors, • employee satisfaction with health benefits, • health services utilization by employees, and • employer-paid health costs for the employee population.

11. Total Health Design Overview • Value-based copayments • Preventive services (already 1st dollar): no change • Chronic disease cost-sharing decreased for • Selected Visits • Pharmacy • Worksite wellness and health promotion activities • Engagement tied to premium stabilization for 3 years • Health risk assessment annually, AND • Achievement of point threshold • Points aimed at both healthy and chronically ill staff

12. Visit Cost-Sharing • Waiver of co-pay for 2 visits/year for chronic care • Coronary Artery Disease • Diabetes • Hypertension • Congestive Heart Failure • Asthma • Mental Health (first ten visits) • Waiver of copay for chemical dependency visits and lactation service visits

13. Pharmacy Co-payments • Copayments reduced to zero for: • generic, mail dispensed meds for same diseases plus depression • Copayment reduced for brand name drugs for same diseases

14. Devices • Wavier of cost-sharing for: • Home BP monitors • Diabetic glucose monitors • Spaces for inhaled asthma meds

15. Obesity Management Programs • 50% discount for enrollment • 100% coverage (50% rebate) for diabetics that lose five percent of body weight

16. Cost-Shares Increased • Outpatient surgery • High cost imaging procedures • CT, MRI, PET

17. Total Health Website

18. Total Health Evaluation Design • Study Design • Quasi-experimental 2 group before/after design • Repeated measures • Control group: Kaiser Permanente Colorado employees

19. Survey Tool • Survey invitation to employees • Web survey tool • Paper survey on request • Domains: Instrument • Functional Status: (SF-12) • Workplace productivity: Work Health Interview

20. Outcomes • Primary • Health status change: Survey • Absenteeism due to illness: Survey +HR data • Productivity at work: Survey • Secondary • Care Quality scores • Chronic illness: HEDIS scores • Preventive services HEDIS scores • Lifestyle behavioral risk factors Survey • E.g. smoking, activity • Employee satisfaction Survey • Costs and service utilization Claims data • Health Risk Behaviors BRFSS, other • Tobacco • Alcohol • BMI • Physical Activity • Satisfaction w/ plan

21. Administrative Data • Health utilization/cost/quality • Group Health Research Institute data warehouse • Claims • Pharmacy • EMR data • Employee characteristics • Human Resources administrative data

22. Data Collection • Sample of 5000 employees invited to take e-survey tool • Active opt-out • Implied consent with survey completion • Separate permissions to link claims and HR data • 3 follow-up emails • No telephone follow-up

23. Statistical Power

25. Total Health – participation • >80% of all staff and spouses/domestic partners on the TH medical plan have taken the HRA • 73% are earning points on the wellness website

26. Progress to Date • Baseline survey completed early 2010 • Group Health: 70% response rate • KPCO: 60% response rate Permissions to link survey data • Approximately 60-64% agree to linkage with HR and/or medical data

27. Challenges and Strengths • Privacy issues/concerns • Employer is also provider of care • Key engagement of organized labor units • Validity of self-reported data • Study design and potential for confounding • Use of highly similar control group external to Group Health

28. Research Team • GHRI/UW KPCO • Paul Fishman Arne Beck • Nora Henrikson Debra Ritzwoller • Rebecca Hubbard Nancy Brace • Diane Martin • Rob Reid • Ellen Schartz • Aaron Scrol • Kay Theis