—Any system of quality monitoring should use the measures that have the greatest utility for the intended purposes. As Drs Wynia and Hasnaim-Wynia point out, outcomes measures used to make comparisons among health plans have methodologic problems, many of which are unlikely to be solved in the near future.Consider implementing breast cancer performance measures in the median-sized managed care plan (90 000 enrollees).1 The "screening" process measure (ie, the annual mammogram rate) would apply to 10% of the population, or 9000 women. Another process measure might examine the timeliness of follow-up for women with abnormal mammograms (540 or 6% of those screened). Among women with abnormal mammograms, 27 will have an incident case of breast cancer, and about 12 (45%) of those women will be diagnosed with late-stage cancer. Finally, we would expect 7 deaths per year from breast cancer. Regardless of the level of quality, the number
McGlynn EA, Brook RH. Assessing Quality of Care: Process Measures vs Outcomes Measures-Reply. JAMA. 1996;276(19):1551–1552. doi:10.1001/jama.1996.03540190023017
Customize your JAMA Network experience by selecting one or more topics from the list below.