A Call to Establish Common Metrics for Consumer-reported Health Status Measurement

SF-36v2™ and SF-12v2™ Health Surveys Offer Substantial Improvements

New versions of the SF-36^® (SF-36v2™) and SF-12^® (SF-12v2™) Health Surveys, developed by scientists at QualityMetric Incorporated and collaborators, have been shown to produce substantial improvements over the originals. Improvements in item wording and format and a 6-fold increase in the ranges of scores covered were achieved for both surveys without increasing respondent burden. Survey developers, with over 10 years of experience in health outcomes measurement, recommend adoption of the SF-36v2™ or SF-12v2™ for clinical trials, disease management, risk prediction, population monitoring, and other studies where scientific validity and precise measurement are required. New up-to-date norms and guidelines are available for maintaining backward comparability with studies published to date, providing complete standardization between the surveys and allowing for comparison of data sets for trend analyses.

Used successfully in more than 600 randomized clinical trials reported in over 240 scientific and medical journals, the SF-36^®, SF-36v2™, SF-12^®, SF-12v2™, and SF-8™ are proven responsive in 44 disease conditions and are accepted by the FDA as proof of benefit for improved functioning and other patient-reported outcomes. Additionally, the SF-36v2 and the SF-12v2 have been adopted as the standard of measurement by key government agencies, including the Agency for Healthcare Research and Quality (AHRQ), which has adopted use of the SF-12v2 for the nationally significant Medical Expenditure Panel Study (MEPS).

Improvements Include:
After more than 10 years of testing, new versions of the SF-36^® (SF-36v2™) and SF-12^® (SF-12v2™) Health Surveys were published to correct deficiencies that had been identified in the original versions. Improvements, which are documented in the SF-36v2 user’s manual (Ware, Kosinski and Dewey, 2000) and in the SF-12v2 user’s manual (Ware, Kosinski, Turner-Bowker and Gandek, 2002) were implemented after careful study using both qualitative and quantitative methods in the US and other countries.

Briefly, the SF-36v2 and SF-12v2 improvements include:

• better instructions and questionnaire items to shorten and simplify the wording and make them more familiar and less ambiguous;
• an improved layout for questions and answers in the self-administered forms that makes it easier to read and complete, and that reduces missing responses;
• greater comparability with translations and cultural adaptations widely-used in the U.S. and in other countries;
• five-level responses in place of dichotomous response choices to greatly increase the range and precision of scores.

As shown in the figure, a 6-fold increase in the range covered by the SF-12v2 Role-physical functioning scale was achieved. As documented in the new SF-12v2 and SF-36v2 user’s manuals, comparable improvements were achieved for both role scales in both v2 forms. Specifically, the adoption of 5-choice (over dichotomous) response categories for items measuring both physical and emotional role functioning led to substantial increases in precision as indicated by the number of levels measured as well as the internal-consistency reliability of those scales for both SF-36 and SF-12 forms. Among the practical implications are virtual eliminations of the “floor” effects and substantial reductions in the ceiling effects for both SF-36v2 and SF-12v2 role functioning scales, in comparison with the original v1.

In the figure, norm-based scoring (NBS) algorithms are used to achieve a mean and SD that are 50 and 10, respectively, for both v2 measures to be comparable with SF-36v1 and SF-12v1. As explained in the user’s manuals, item response theory (IRT) thresholds are used in the figure to show the differences in ranges covered by v1 and v2 items in a large representative sample of the general US population in 1999 (see the SF-12 user’s manual for more information).

QualityMetric’s SF™ Generic Health Surveys (SF-36^®, SF-36v2™, SF-12^®, SF-12v2™, and SF-8™ Health Surveys) can be scored online accurately, securely, and reliably now, with exclusive new desktop scoring software set for launch in early 2004.