This Article  • Full text  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Related articles  • Similar articles in JAMA  Topic Collections  • Diabetes Mellitus  • Quality of Care, Other  • Statistics and Research Methods  • Neurology  • Alert me on articles by topic

	Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings.

To the Editor: In their Commentary, Drs Nolan and Berwick¹ advocate the widespread adoption of all-or-none measures in the assessment of health care quality. The all-or-none measure works in a multiplicative sense, compared with composite measures that are additive. For example, for diabetes care they note 5 activities that could be measured by a present/absent indicator and recorded as 1 (present) and 0 (absent). The all-or-none score for 3 possible patients would be calculated as: patient A: 1 x 1 x 1 x 1 x 1 = 1; patient B: 1 x 0 x 1 x 1 x 1 = 0; and patient C: 0 x 0 x 0 x 0 x 0 = 0.

The composite scores, which are summed, would yield scores of 5 (patient A), 4 (patient B), and 0 (patient C).

The all-or-none measurement hides the differences in care received by patients B and C, while the composite measurement reveals these differences. . . . [Full Text of this Article]

Joseph G. Van Matre, PhD
jgv@uab.edu
School of Business
University of Alabama
Birmingham

RELATED ARTICLES

All-or-None Measurement of Health Care Quality
Mark K. Chelmowski
JAMA. 2006;296(4):392-393.
EXTRACT | FULL TEXT  

All-or-None Measurement of Health Care Quality—Reply
Thomas W. Nolan and Donald M. Berwick
JAMA. 2006;296(4):393.
EXTRACT | FULL TEXT  

All-or-None Measurement Raises the Bar on Performance
Thomas Nolan and Donald M. Berwick
JAMA. 2006;295(10):1168-1170.
EXTRACT | FULL TEXT