1 / 38

EPI235: Epi Methods in HSR

EPI235: Epi Methods in HSR. April 17, 2007 L5 Program Evaluation with Longitudinal Data 1: Applications (Dr. Schneeweiss) Various examples of applications in Health Services Research. Strengths and limitations of time series analysis . Background reading:

farren
Download Presentation

EPI235: Epi Methods in HSR

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. EPI235: Epi Methods in HSR • April 17, 2007 L5 • Program Evaluation with Longitudinal Data 1: Applications (Dr. Schneeweiss) • Various examples of applications in Health Services Research. Strengths and limitations of time series analysis . • Background reading: • Soumerai SB, Avorn J, Gortmaker S, Ross‑Degnan D. Payment restrictions for prescription drugs in Medicaid: Effects on therapy, cost, and equity. New Engl J Med 1987;317:550‑556. • Soumerai SB, Ross-Degnan D, Avorn J, McLaughlin TJ. Effects of Medicaid drug-payment limits on admission to hospitals and nursing homes. N Engl J Med 1991;325:1072-1077

  2. What is Program Evaluation?

  3. Program Evaluation

  4. Observational Design: Simple pre-post comparisons Intervention Outcome rate Time Assumptions for causal inference: 1. The pre experience represents the post experience had there been no intervention

  5. Observational Design: Simple pre-post comparisons Intervention Outcome rate Time Threat to causal inference: Single pre-post estimates are averages of an underlying trend independent of the Intervention

  6. 25% cost sharing in Quebec Tamblyn, JAMA 2000

  7. Intervention Time Observational Design: no concurrent controls Assumptions for causal inference: 1. Close temporal relation 2. Extrapolation of baseline trend is equal to the counterfactual experience Schneeweiss, Health Policy 2000

  8. Fixed ¢50 cost sharing in SC Nelson, Med Care 1984

  9. Intervention Control group Intervention group Time Observational Design: With concurrent controls Assumptions for causal inference: 1. … 2. Control trend is equal to the counterfactual experience of intervention group Schneeweiss, J Clin Epi 2002

  10. 100% 95% New Jersey % outside nursing home 90% New Hampshire 85% Baseline Cap After Cap 80% 3 prescription caps in NH: Soumerai NEJM 1991

  11. Intervention Control group R Intervention group Time … or randomization? Assumptions for causal inference: 1. Subjects comply with their assigned ‘treatment’ = policy

  12. Complex statistical analyses are less convincing for decision makers

  13. Formulary delisting in BC

  14. More Examples:

  15. More Examples: • Prescription drug use • Surgical site infections • Contraindicated drug use

  16. Medicaid prior authorization: Use Smalley NEJM 1995

  17. Medicaid prior authorization: $$ Smalley NEJM 1995

  18. Reducing surgical site infections after C-section: Hospital A Open squares = utilization indicator (cesarean sections receiving perioperative antibiotic prophylaxis) Open circles = timing indicator (antibiotic within 1 hour of delivery) Solid diamonds = surgical site infection rate after cesarean section Period 1 was a baseline period. Periods 2 and 3 were successive intervention periods. Weinberg et al.: Arch Intern Med 2001

  19. Reducing surgical site infections after C-section: Hospital B Open squares = utilization indicator (cesarean sections receiving perioperative antibiotic prophylaxis) Open circles = timing indicator (antibiotic within 1 hour of delivery) Solid diamonds = surgical site infection rate after cesarean section Period 1 was a baseline period. Periods 2 and 3 were successive intervention periods. Weinberg et al.: Arch Intern Med 2001

  20. FDA Risk Management: Effectiveness of a ‘Dear Doctor’ letter The Intervention: 'Dear Doctor' letters concerning interactions between cisapride and a series of drugs. A letter in 1995 described a risk of prolonged QT intervals and serious ventricular arrhythmia in patients who received macrolide antibiotics and imidazole antifungals in conjunction with cisapride. A June 1998 letter that expanded the list of contraindicated comedications had wider distribution than an earlier one, was accompanied by substantial Internet and media coverage, and was complemented by an effort to inform large pharmacy dispensing information organizations of the warnings against concurrent use of the named drugs.

  21. Weatherby LB et al. PharmEpi Drug Safety 2001

  22. Lecture on LPUs: Weatherby LB et al. Clin Pharm Ther 2002

  23. The Policy Model

  24. Policy Model: reference pricing

  25. The Clinical Model

  26. More on the Clinical Model:

  27. Clinical Model: reference pricing

  28. Policy Model vs. Clinical Model

More Related