The Spine Blog

Friday, January 31, 2014

From the Desk of Darrel Brodke, MD

The future of value-driven, evidence-based decision-making in spine care depends on high quality and valid outcomes measures.  Ten years ago, the PROMIS Network was founded by the NIH to begin the process of building and organizing patient-reported outcome measures that more accurately reflect the patient’s health in a variety of domains. Banks of questions (“items”), validated for specific dimensions of the human condition (physical function, pain that interferes with function, depression, social isolation, anxiety, etc.) were developed using a psychometric paradigm for the design, analysis, and scoring of questionnaires, known as item response theory (IRT).


Among the PROMIS domains, physical function is of prime interest for our patients seeking treatment of spinal disorders.  The study by Hung et al, “The Psychometric Properties of the PROMIS Physical Function Item Bank in Spine Patients”, took the foundational step in understanding the value of the PROMIS Physical Function (PF) item bank (a bank of 124 questions with level of difficulty on a continuum of function from almost no function to extremely high-level activity).  The entire bank of questions was administered to 438 patients to prospectively investigate the psychometric properties of each item in the PROMIS Physical Function item bank in a heterogeneous spine patient population.  The PF item bank was found to have excellent coverage, with minimal floor and ceiling effects, minimal unexplained variance, and excellent reliability.


Validating the entire item bank of questions for assessing patient physical function also validates an important aspect of this tool, computerized adaptive testing (CAT), which allows for administration of a small subset of the total bank of questions and maintains the accuracy of the entire 124 question bank.  Future studies will help us understand if the CAT administered PROMIS Physical Function Item Bank reduces patient burden, improves floor and ceiling effects and maintains precision as compared to our current scales, as has been shown to be true in other areas of musculoskeletal medicine.1-4




1.       Hung MNickisch FBeals TCGreene TClegg DOSaltzman CL.  New paradigm for patient-reported outcomes assessment in foot & ankle research: computerized adaptive testing.  Foot Ankle Int. 2012 Aug; 33(8): 621-6.

2.       Hung MBaumhauer JFLatt LDSaltzman CLSooHoo NFHunt KJNational Orthopaedic Foot & Ankle Outcomes Research Network. Validation of PROMIS ® Physical Function computerized adaptive tests for orthopaedic foot and ankle outcome research. Clin Orthop Relat Res. 2013 Nov; 471(11): 3466-74.

3.       Hung MStuart ARHiggins TFSaltzman CLKubiak EN.  Computerized Adaptive Testing using the PROMIS Physical Function Item Bank Reduces Test Burden with Less Ceiling Effects Compared to the Short Musculoskeletal Function Assessment in Orthopaedic Trauma Patients.  J Orthop Trauma. 2013  Dec 27. [Epub ahead of print]

4.       Hung MFranklin JDHon SDCheng CConrad JSaltzman CL.  Time for a paradigm shift with computerized adaptive testing of general physical function outcomes measurements.  Foot Ankle Int. 2014  Jan; 35(1): 1-7.  Epub 2013 Oct 7.