1 / 30

Evidence

Evidence. TERESA BERNARD GIBSON, PHD SENIOR DIRECTOR, HEALTH OUTCOMES NOVEMBER 16, 2011. OVERVIEW. Background Evaluating Effects Evidence 2 Studies Evidence in Progress Desirable Evidence. EVIDENCE.

maina
Download Presentation

Evidence

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Evidence TERESA BERNARD GIBSON, PHD SENIOR DIRECTOR, HEALTH OUTCOMES NOVEMBER 16, 2011

  2. OVERVIEW Background Evaluating Effects Evidence 2 Studies Evidence in Progress Desirable Evidence

  3. EVIDENCE • Probability of a commercial airline flight with a physician-passenger: 0.7 (range: 0.5-0.85) • Groeneveld et al. JAMA. 2001. 286(12): 1482-1489.

  4. RAND HEALTH INSURANCE EXPERIMENT Manning et al. Health Insurance and the Demand for Medical Care. 1988. Santa Monica, CA : The Rand Corporation. • Design • 6 Cities: Seattle, WA; Dayton, OH; Charleston, SC: Fitchburg-Leominster, MA; Franklin County, MA; Georgetown County, SC • Fee-for-Service Plans: • Free (0% coinsurance) • 25% coinsurance • 50% coinsurance • 95% coinsurance • Outpatient individual deductible • Maximum Dollar Expenditure per year • 5, 10 or 15 percent of income • No more than $1000 (average income was around $10K)

  5. RAND HEALTH INSURANCE EXPERIMENT • No selection of health plan • Assigned to health plan • Balanced characteristics of groups assigned to health plan options • Random sample from cities • Findings • “The data from the HIE clearly show that the use of medical services responds to changes in the amount paid out of pocket.” p. ix.

  6. EVALUATING EFFECTS • How do we know that an intervention works? • What are the effects of an intervention? • How does an intervention perform in a real-world setting?

  7. THE EVIDENCE • Strength/Intensity of the Intervention • Include brand and/or generic medications • Amount of reduction in cost-sharing • Percentage reduction in cost-sharing • Qualification for reduction in cost-sharing • Disease-specific, medication-specific, incentive for engagement? • Comparison group • Internal/external • Assignment to intervention/comparison group • Randomization, matching • Length of time studied

  8. THE EVIDENCE • Size of groups • Outcomes/Measures • Methods of Comparison • Falsification tests • Effects of the program on outcomes not related to the program

  9. THE EVIDENCE • Design elements • Context • VBID programs in place • Other programs in effect • An individual’s previous program • From  To

  10. INTERVENTIONS (Chronological order)

  11. INTERVENTIONS

  12. INTERVENTIONS

  13. INTERVENTIONS

  14. ADHERENCE EFFECTS

  15. ADHERENCE EFFECTS

  16. Gibson and Colleagues Health Affairs 2011; 30(1): 100-108. • Strength/Intensity of the Intervention • 3 Tier 10%/20%/35% 1 Tier 10% • Diabetes • Comparison group • Internal (same firm) • Selected sites received the VBID, other sites did not • One-to-one propensity score matching • VBID + Disease Management Disease Management • VBID/No Disease Management No Disease Management • Also matched Disease Management to similar employers in the MarketScan Database

  17. Gibson and Colleagues Health Affairs 2011; 30(1): 100-108. • Length of time studied • 3 years post-implementation • Size of groups • VBID + Disease Management: 1,876 • VBID, no Disease Management: 328 • Outcomes/Measures • Adherence/Medication Utilization • Receipt of Guideline-related Utilization • Spending • Methods of Comparison • Multivariate models, panel data framework

  18. RESULTS: PRESCRIPTION DRUG UTILIZATION

  19. RESULTS • The combination of VBID and disease management is more powerful than disease management alone: all effects higher • Prescription drug utilization • All antidiabetic medications • Oral antidiabetic medications • Insulin • Brand/generic antidiabetic medications • Adherence to recommended medical service guidelines • PCP Visits, HbA1c tests, Urinalysis, Lipid tests • Retinal eye exams unaffected • Patients may have been using vision care benefits (not medical benefits) to receive eye exams

  20. RESULTS: VBID WITHOUT DISEASE MANAGEMENT • The effects on those who opted out of a diabetes disease management program (VBID with no disease management compared to no disease management) resulted in higher levels of adherence to a 0.8 MPR threshold or higher, but no other lasting effects among other measures. • Among those who opted out, the sample size was smaller (n=328), which may have reduced statistical power • The majority were enrolled in an HMO (tighter medical management)

  21. RESULTS: SPENDING • VBID and disease management group • All-cause • Prescription drug spending • Medical spending: • Total (medical plus drug spending): • Diabetes-related • Prescription drug spending: • Medical spending: • Total spending: •  Spending effects for the VBID group without disease management:

  22. Choudhry and Colleagues NEJM; online first November 14, 2011. • MI FREEE Trial • Post Myocardial Infarction Free Rx Event and Economic Evaluation (MI FREEE) Trial • Strength/Intensity of the Intervention • Existing Plan $0 copayment for statins, beta blockers and ACEI/ARB • Patients post-Acute Myocardial Infarction hospitalization • Comparison group • Cluster randomized by plan sponsor (e.g., employer, union) within a large insurer randomized to full coverage or existing plan prior to intervention • Plan sponsors agreed to participate or not

  23. Choudhry and Colleagues NEJM; online first November 14, 2011. • Length of time studied • Varied, median 294 days (IQR: 201-663 days) • Size of groups • Full coverage (2,845 patients, 1,494 plan sponsors) • Usual coverage (3,010 patients, 1,486 plan sponsors) • Outcomes/Measures • Medication Utilization • Percent of days covered, Percent fully adherent • Clinical • First major vascular event or revascularization, total vascular events and revascularization, first major vascular event • Spending

  24. RESULTS (FULLY COVERED VS. USUAL COVERAGE) • Adherence/Medication Utilization • Percent of days covered for all classes (ACEI/ARB, beta blockers, statins, all classes) • Percent of patients that were fully adherent for all medication classes • Clinical Outcomes • First major vascular event or revascularization • Major vascular events (fatal or nonfatal MI, stable angina, CHF, stroke): • Total major vascular events or revascularization:

  25. RESULTS • Effects on Health Spending • Patient Spending (out of pocket): • Prescription drug: • Medical: • Total: • Total Spending (health plan, patient): • Prescription drug: • Medical: • Total:

  26. EVIDENCE IN PROGRESS • Incentive-based wellness programs • Biometric screenings, counseling to interpret results • Health Risk Appraisal • Incentive platform to improve health and manage health • VBID component for condition(s) • Prescription drugs • Supplies • Office visits

  27. DESIRABLE EVIDENCE • Role of office visit cost-sharing • Brand and generic medications • Substitutes and/or complements • Income and use of brand and generic medications • Cost-sharing and children • Cost-sharing and productivity • Cost-sharing and mortality • Pilote et al. 2002 (no effect), Choudhry et al. 2011 (death rates from cardiovascular causes lower in intervention group) • After Acute Myocardial Infarction

  28. DESIRABLE EVIDENCE • Optimal health insurance • Balance moral hazard with risk aversion • Randomization to intervention/control groups • Effects of combining incentives and programmatic elements • Incentive-based wellness programs

  29. REFERENCES Chang A, Liberman JA, Coulen C, Berger JE, Brennan TA. Value-based insurance design and antidiabetic medication adherence. Am J Pharm Benefits 2010; 2(1): 39-44. Chernew ME, Shah MR, Wegh A, et al. Impact of decreasing copayments on medication adherence within a disease management environment. Health Affairs 2008; 27(1): 103-112. Choudhry NK, Fischer MA, Avorn J, et al. At Pitney Bowes, value-based insurance design cut copayments and increased drug adherence. Health Affairs 2010; 29(11): 1995-2001. Choudhry NK, Avorn J, Glynn RJ, et al. Full coverage for preventive medications after myocardial infarction. NEJM 2011; online first. Gibson TB, Wang S, Kelly E, et al. A value-based insurance design program at a large company boosted medication adherence for employees with chronic illnesses. Health Affairs 2011a; 30(1): 109-117. Gibson TB, Mahoney J, Ranghell K, Cherney BJ, McElwee N. Value-based insurance plus disease management increased medication use and produced savings. Health Affairs 2011b; 30(1): 100-108.

  30. REFERENCES Kim YA, et al. Evaluation of Value-Based Insurance Design with a Large Retail Employer. AJMC 2011; 17(10): 682-690. Maciejewski ML, Farley JF, Parker J, Wansink D. Copayment reductions generate greater medication adherence in targeted patients. Health Affairs 2010; 29(11): 2002-2008. Zeng F, An JJ, Scully R, Barrington C, Patel BV, Nichol MB. The impact of value-based benefit design on adherence to diabetes medications: A propensity score-weighted difference in difference evaluation. Value in Health 2010; 13(6): 846-852.

More Related