The technology for calibrating transparent rating scale-based measures of health and functional status is over 50 years old. It is well researched and documented. It has been used to construct scientifically and legally defensible admissions and professional certification tests for over 30 years, in healthcare and other industries.[2,3]
So how is it that this technology is virtually unknown in health services research? I say virtually unknown, because there are a growing number of applications in healthcare of the mathematical models at issue.
What healthcare does not have, however, is an overarching plan for decentralized networks of different brands and configurations of instruments that measure the same things in the same units.
Such networks of measures traceable to reference standards are essential to the common product definitions, economic performance, and quality improvement records of every one of today's commercially successful industries, from housing to automobiles to electronics.
Despite the indisputable productivity of other industries' metrological infrastructures, healthcare persists in ignoring the proven measurement technologies it needs, even as they are being put to work in other fields, such as education.
Lacking common product definitions, healthcare consumers cannot make purchasing decisions that reward quality, and providers cannot systematically effect quality improvements.
Healthcare needs an institute for standards and technology that will create and maintain reference standard metrics for its patient-centered rating scale measures. It accordingly also needs to require traceability to these reference standards for all new rating scale instrument development and improvement research.
Quality improvement efforts in healthcare cannot be systematically coordinated without a metrological infrastructure. Quality improvement measurement systems should be designed from the ground up with the aim of calibrating transparent instruments traceable to reference standards.
That's my opinion. I'm William P. Fisher, Jr., Chief Science Officer at Avatar International, Inc.
- Rasch G. Foreword, afterword, by Benjamin D. Wright. In: Probabilistic Models for Some Intelligence and Attainment Tests. Chicago: University of Chicago Press; 1980:ix-xix, 185-199. Available at: http://www.rasch.org/memo63.htm Accessed November 7, 2006. [Reprint; original work published in 1960 by the Danish Institute for Educational Research].
- Smith RM, Julian E, Lunz M, Stahl J, Schulz M, Wright BD. Applications of conjoint measurement in admission and professional certification programs. Int J Educ Res. 1994;21:653-664.
- Kelley PR, Schumacher CF. The Rasch model: its use by the National Board of Medical Examiners. Eval Health Professions. 1984;7:443-454.
- Bezruczko N, ed. Rasch Measurement in Health Sciences. Maple Grove, Minn: JAM Press; 2005.
- National Institute for Standards and Technology. Appendix C: assessment examples. Economic impacts of research in metrology. In: Subcommittee on Research, ed. Assessing Fundamental Science: A Report From the Subcommittee on Research, Committee on Fundamental Science. Washington, DC: National Standards and Technology Council; 1996. Available at: http://www.nsf.gov/statistics/ostp/assess/nstcafsk.htm#Topic%207 Accessed November 7, 2006.
- Stenner AJ, Burdick H, Sanford EE, Burdick DS. How accurate are Lexile text measures? J Appl Meas. 2006;7:307-322.
- Kindig DA. Purchasing population health: aligning financial incentives to improve health outcomes. Health Serv Res. 199833:223-242.
- Berwick DM, James B, Coye MJ. Connections between quality measurement and improvement. Med Care. 2003;41(suppl):I30-38.
source - Medscape