From Comparative Effectiveness to Optimal Health Implementation Research

From Comparative Effectiveness to Optimal Health: Implementation Research Robert Wood Johnson Foundation Synergy Working Group June 18, 2009

Develop IT in Data… USA Switzerland Synthesize Interpret Canada UK Learn Use Source: Anderson and Frogner, Health Affairs 2009

Variation in medical expenses Mc. Allen, Texas Source: Fisher, et al. NEJM 2009

The Problem • US spends more than any other country • Only 60% of recommended care is utilized in practice – 61% of post-MI candidates received aspirin, a nearly-free treatment that reduces MI by 30% • Large practice variation exists • Many major guidelines and trials have no effect on clinical care How can we use CER to encourage good choices in health care? Mc. Glynn NEJM 2003

Evidence from Development to Practice Data Development Data Generation and Synthesis Help Stakeholders Learn the Data Interpretation Encouraging Data Use

Rationale CER without implementation is of no value Help Stakeholders Learn the Data X Encouraging Data Use

Develop Synthesize Interpret Learn Use Why data are not used? • • • Error Data reliability (Internal validity) Generalizability (External validity) Knowledge Resource limitation - time Patient preferences (perceived or real)

Public Press on Clinical Action – The Katie Couric Effect Cram et al. Archives of Internal Medicine, 2003

Develop Synthesize Interpret Learn Use Why data are not used? • • Error Data reliability (Internal validity) Generalizability (External validity) Knowledge Resource limitation - time Patient preferences (perceived or real) External Pressures

Culture and Social Environment: Mc. Allen vs Mayo Gawande New Yorker 2009

Develop Synthesize Interpret Learn Use Why data are not used? • • Error Data reliability (Internal validity) Generalizability (External validity) Knowledge Resource limitation - time Patient preferences (perceived or real) External Pressures – – – Social Environment Consumer pressure Fear of Litigation Perverse Incentives: reimbursement, approval Heuristics

Develop Synthesize Interpret Learn Use How to make CER work? • Data quality varies • Data translation can be a stopping point

Develop Encouraging Appropriate Data Use Synthesize Interpret Learn Use • Education-based • Payment reform-based • Social Environment

Develop Encouraging Appropriate Data Use Synthesize • Education-based Interpret Learn Use – Make research simple and usable – Evidence-based decision-support – Clinical Registries – Strengthen certification

Develop Synthesize Interpret Learn Use Encouraging Appropriate Data Use • Payment reform-based – Pay-for-performance – Value-Based Insurance Design: Florida’s Medicaid Enhanced Benefits, cheaper data-based drugs – Accountable Healthcare System – Refusal of reimbursement – Tort Reform

Develop Encouraging Appropriate Data Use Synthesize • Social Environment-based Interpret Learn Use – VA, Mayo, Geisinger – Safeway – Smoking access

Develop Synthesize Interpret Learn Use Unintended consequences • A dollar saved is a dollar lost • Overreliance on available data could make practitioners lose the art of medicine • Mutually-incompatible recommendations • Better care sometimes costs more • Perception of rationing • Effects on liability unclear • Punishes those who need help • Cherry-Picking

Develop Synthesize Interpret Learn Use Bridging Research and Practice: Policy Statement Any CER policy must include an implementation plan

Develop Synthesize Interpret Learn Use Stakeholders for implementation • • • Consumers/Patients Providers Researchers Foundations/Nonprofits/Faith-based Organizations Professional Associations 3 rd Party Payers Health Plans Hospitals and Care Organizations Device Manufacturers/Pharma

Develop Plan of Action • Federal Regulatory Approach Synthesize Interpret Learn Use • Involve professional societies for support and implementation • Convene IOM panel on implementation research, financial incentives and tort reform • Establish research emphasis on implementation • Allow Medicaid waivers for comparative implementation research • Encourage small-scale Medicare Demonstration projects