Effective Clinical Practice
LETTER TO THE EDITOR
Effective Clinical Practice, May/June 2000.
I would like to make three comments about Sorokin's paper. (1)
First, if the erroneous and irritating "report card" obtained from flawed administrative databases had not been published, would this group have been motivated to review their practices?
Second, despite incorrect information in the report card, the author notes that chart reviews confirmed "genuine deficiencies." These results suggest that, despite its flaws, the report card had a beneficial effect.
Finally, the author questions whether report cards, such as the Health Plan Employer Data and Information Set (HEDIS) measures, will really improve care. The answer is a qualified "no," because most of the old HEDIS measures are largely process indicators. The report card Dr. Sorokin describes combined process indicators (measurement of hemoglobin A1c, cholesterol, blood pressure, and microalbumin in diabetics) with surrogate outcomes, which were still process indicators (anticoagulation in atrial fibrillation and ß-blocker use in coronary artery disease). Such measures require very selective use if the goal is to improve care and outcomes. Simply documenting a process does not, per se, improve care.
The focus of report cards must shift from simply process measures to more relevant endpoints. Surrogate endpoints are a step forward (e.g., the percentage of patients with hemoglobin A1c below 8%, the percentage of those with blood pressure at or below 135/85 mm Hg, or the percentage of patients achieving a low-density lipoprotein cholesterol level below 130 mg/dl). Such modifications to the quality indicator for diabetics have already been recommended by the Diabetes Quality Improvement Program and are being incorporated into newer HEDIS protocols. (2)
Ultimately, however, we need measurements of the real outcomes of interest: reduced progression to renal insufficiency and to diabetic retinopathy and visual impairment and decreased incidence of stroke, myocardial infarction, and end-stage renal disease.
Report cards are useful tools for patients and physicians when they look at clinically important and relevant outcomes, thereby enabling both to choose the best practice to achieve these ends. (3)
Robert Matz, MD
Mount Sinai Medical Center
New York, NY
1. Sorokin R. Alternative explanations for poor report card performance. Eff Clin Pract. 2000;3:25-30.
2. Peters A, Galier P. Quality assurance in diabetes, in Taylor SI, ed. Current Review of Diabetes. Philadelphia: Current Medicine;1999:115-22.
3. Matz R. Outcomes in diabetic patients. Diabetes Care. 1997;20:1497.
The Author Responds
Dr. Matz clearly delineates what my article apparently did not: The report card directly motivated the chart review, and the chart review itself was a valuable exercise. Thus, the beneficial effect of the flawed report card was not in its data but in its effect as a stimulus for self-evaluation. I do not know that this is what the producers of the report card intended or whether it should be the primary goal of report cards.
I am unwilling to dismiss surrogate outcomes. Thus, randomized, placebo controlled trials support ß-blocker use after myocardial infarction (1) and retrospective reviews describe underutilization of ß-blockers (2) and document poorer outcomes. (3) ß-Blocker use after acute myocardial infarction satisfies me as a clinically important indicator. I do not see the need to prove the mortality data in every measurement. Although I agree that, as Dr. Matz notes, merely documenting a process (or outcome) does not improve care, it does define the task at hand.
I share the belief that well-constructed report cards could become a powerful tool. There is, however, tension between the desire to accurately measure real outcomes and the pressure to start measuring something (or anything). This pressure produces the weaker process variables Dr. Matz decries and the methodologic problems described in my article. I agree with Dr. Matz that improving the indicators would greatly increase the value of report cards.
Rachel Sorokin, MD
1. Freemantle N, Cleland J, Young P, Mason J, Harrison J. Beta blockade after myocardial infarction: a systematic review and meta regressions analysis. BMJ. 1999; 318:1730-7.
2. Ellerbeck EF, Jencks SF, Radford MJ, et al. Quality of care for medicare patients with acute myocardial infarction. JAMA. 1995;273:1509-14.
3. Soumerai SB, Mc Laughlin TJ, Spiegelman D, Hertzmark E, Thibault G, Goldman L. Adverse outcomes of the underuse of beta blockers in elderly survivors of acute myocardial infarction. JAMA. 1997;277:115-21.