WinsLosses and ErrorsTies: Quality of Care for Acute Myocardial Infarction in the VA Health Care Sys

1
Wins/Losses and Errors/TiesQuality of Care for
Acute Myocardial Infarction in the VA Health Care
System

• Laura A. Petersen, M.D., M.P.H.1
• Sharon-Lise T. Normand, PhD2
• Barbara J. McNeil, MD, PhD2
• 1 Houston Center for Quality of Care and
  Utilization Studies,
• an HSRD Center of Excellence
• 2Harvard Department of Health Care Policy
• Supported by VA HSRD IIR 94-054
• VA HSRD Career Development Award Program RCD
  95-306 RWJ Foundation Generalist Physician
  Faculty Scholar Award Program
• AHRQ RO1-HS08071

2
Background

• Acute myocardial infarction (AMI) is a common,
  costly, and clinically significant condition, and
  represents a signal event in the natural history
  of ischemic heart disease
• Appropriate quality of care for AMI improves
  survival
• Quality of care for AMI may indicate a hospitals
  or systems ability to deliver highly skilled,
  specialized acute medical care

3
Goals

• To compare process of care and outcomes after
  acute myocardial infarction in VA and non-VA
  hospitals, controlling for differing patient
  characteristics

4
Methods - Patient Samples

• Age gt 65
• Male
• Clinically confirmed discharge diagnosis of acute
  myocardial infarction
• 29,249 FFS Medicare patients from 1,530 non-VA
  acute care hospitals in CA, FL, MA, NY, OH, PA,
  and TX
• Random sample of 2,486 patients from 81 VA
  hospitals nationwide

5
Wins

• Petersen LA, Normand SL, Leape LL, McNeil BJ.
  Comparison of use medications after acute
  myocardial infarction in the Veterans Health
  Administration and Medicare. Circulation,
  2001104(24)2898-2904.

6
Use of Thrombolytic Therapy at Arrival or Aspirin
at Discharge in VA Relative to Medicare

All comparisons significant at plt0.05
7
Use of ACE Inhibitors or Beta-Blockers at
Discharge in VA Relative to Medicare

plt0.05 comparison NS
8
Losses and Errors

• Petersen LA, Normand SL, Leape LL, McNeil BJ.
  Regionalization and the underuse of angiography
  in the Veterans Affairs Health Care System as
  compared with a fee-for-service system. N Engl J
  Med 2003 3482209-17.

9
Age-Adjusted Rates of Cardiac Procedures Among
ACC Class I Patients
RR 0.72 95 CI (0.67-0.78)
RR 0.82 95 CI (0.78-0.87)


Diagnostic angiography within 90 days after
index admission. Procedures within 90 days among
patients who underwent angiography within 90 days
after index admission.
10
Summary

• Differences in underuse not eliminated by
  adjustment for patient characteristics and
  clustering
• Differences eliminated by adjustment for
  procedure availability in the regionalized system

11
Ties

• Petersen LA, Normand SL, Daley J, McNeil BJ.
  Outcome of myocardial infarction in Veterans
  Health Administration patients as compared with
  Medicare patients. N Engl J Med
  20003431934-41.

12
Adjusted Odds of Mortality Medicare Relative to
VA(Using full sample 29,249 Medicare and 2,486
VA)

• Logistic Regression OR 95CI
• 30-day 0.93 (0.81-1.07) c0.800
• 1-year 0.93 (0.83-1.05) c0.799
• Adjusting for 30 sociodemographic, clinical and
  hospital variables cArea under ROC curve
• Results confirmed with propensity score adjustment

13
Summary

• Wins VA patients were equally likely (in the
  case of beta-blockers) or more likely (in the
  case of thrombolytic therapy, ACE inhibitors, or
  aspirin) than Medicare patients to receive
  medications of known benefit after AMI
• Losses/Errors Rates of use of angiography and
  cardiac revascularization procedures were
  significantly lower in the VA than in Medicare,
  even among groups where angiography was deemed
  clinically needed

14
Summary and Conclusions (2)

• However, once admitting hospital procedure
  availability was controlled for, there was no
  difference in angiography use between patients
  cared for in VA and Medicare
• The findings suggest that underuse of angiography
  could be remedied by changes in policy regarding
  availability of angiography services and
  regionalization of cardiac technology

15
Summary and Conclusions (3)

• Ties VA patients were somewhat sicker than
  Medicare patients. Yet, we found no significant
  differences in 30-day or 1-year mortality between
  Medicare and VA AMI patients