OBJECTIVE: To demonstrate the potential of the Health Plan Employer Data and Information Set (HEDIS) for the calculation of a performance measure for eye exams in the diabetic population using Veterans Health Administration (VA) administrative data. DESIGN: We calculated a 1-year HEDIS-defined patient denominator and three alternative denominators that considered coding factors in identifying a VA patient as diabetic. We calculated the HEDIS-defined numerator, along with alternative specifications that captured other types of eye exams. Finally, we supplemented national data with VA pharmacy and Medicare claims data to identify all VA diabetic patients at 14 selected VA facilities and to establish a more accurate picture of non-VA health care utilization. RESULTS: The national average annual HEDIS-defined eye exam rate in the VA was 26% in fiscal 1997 compared with 39% for managed care organizations. Medicare utilization raised this by 15 percentage points at 14 northeastern VA hospitals. Over 2 years, at least two-thirds of diabetic VA patients had some type of eye exam through VA or Medicare. CONCLUSION: A HEDIS measure of eye exams for VA patients with diabetes can be calculated using VA administrative data only. However, the question remains to what extent the denominator and numerator accurately and completely identify all diabetic patients using VA services and all appropriate eye exams. We recommend caution in interpreting the results of performance measurement across different health care sectors based on what we currently know are data system limitations.
OBJECTIVE: To demonstrate the potential of the Health Plan Employer Data and Information Set (HEDIS) for the calculation of a performance measure for eye exams in the diabetic population using Veterans Health Administration (VA) administrative data. DESIGN: We calculated a 1-year HEDIS-defined patient denominator and three alternative denominators that considered coding factors in identifying a VA patient as diabetic. We calculated the HEDIS-defined numerator, along with alternative specifications that captured other types of eye exams. Finally, we supplemented national data with VA pharmacy and Medicare claims data to identify all VA diabeticpatients at 14 selected VA facilities and to establish a more accurate picture of non-VA health care utilization. RESULTS: The national average annual HEDIS-defined eye exam rate in the VA was 26% in fiscal 1997 compared with 39% for managed care organizations. Medicare utilization raised this by 15 percentage points at 14 northeastern VA hospitals. Over 2 years, at least two-thirds of diabetic VApatients had some type of eye exam through VA or Medicare. CONCLUSION: A HEDIS measure of eye exams for VA patients with diabetes can be calculated using VA administrative data only. However, the question remains to what extent the denominator and numerator accurately and completely identify all diabeticpatients using VA services and all appropriate eye exams. We recommend caution in interpreting the results of performance measurement across different health care sectors based on what we currently know are data system limitations.
Authors: Richard R Owen; Carol R Thrush; Dale Cannon; Kevin L Sloan; Geoff Curran; Teresa Hudson; Mark Austen; Mona Ritchie Journal: J Am Med Inform Assoc Date: 2004-06-07 Impact factor: 4.497