158 January 2014 HEALTH SERVICES RESEARCH SPINE Volume 39, Number 2, pp 158-163 ©2014, Lippincott Williams & Wilkins Psychometric Properties of the PROMIS Physical Function Item Bank in Patients With Spinal Disorders Man Hung, PhD ,* Shirley D. Hon, BS,† Jeremy D. Franklin, MA,‡ Richard W. Kendall, DO,* Brandon D. Lawrence, MD,* Ashley Neese, BS,* Christine Cheng, BS,* and Darrel S. Brodke, MD* DOI: 10.1097/BRS.0000000000000097 Study Design. Patient-reported outcomes provide vital information when assessing effectiveness of clinical care. Yet, most patient- reported outcome instruments are limited by lack of validation and reliability to measure PF adequately. As part of the Patient-Reported Outcomes Measurement Information System (PROMIS), a PF item bank consisting of 124 items has been developed. Objective. There is validation evidence for the PROMIS PF item bank in the general orthopedic patient population in general, but has yet to be validated in the patient with spinal disorders. This study aims to evaluate the psychometric properties of the PROMIS PF item bank specifically for patients presenting with spine-related complaints. Summary of Background Data. Data were collected from adult patients visiting a university spine clinic for back and neck problems. All patients older than 18 years were eligible to participate. A total of 438 patients (49% male) were enrolled in this prospective study. Patients were 18- to 89-year old and presented with back problems (n = 286) and neck problems (n = 152). All patients were administered a 131 item questionnaire. Methods. Conventional descriptive statistics such as means, standard deviations, and proportions were conducted to examine patient characteristics. A Rasch model was used to examine the psychometric properties of the instrument including dimensionality, floor/ceiling effects, reliabilities, and item bias. Results. Results showed that a single PF dimension was supported by the data ( i.e., unexplained variance was 2.9%). The instrument had 1.7% ceiling effect and 0.2% floor effect. Item reliability was From the *University of Utah School of Medicine, Salt Lake City, UT; †University of Utah College of Engineering, Salt Lake City, UT; and ‡University of Utah College of Education, Salt Lake City, UT. Acknowledgment date: June 27, 2013. Revision date: October 14, 2013. Acceptance date: October 21, 2013. The device(s)/drug(s) is/are FDA-approved or approved by corresponding national agency for this indication. No funds were received in support of this work. Relevant financial activities outside the submitted work: board membership, royalties, stock/stock options, and grants/grants pending. Address correspondence and reprint requests to Man Hung, PhD, Department of Orthopaedic Surgery Operations, University of Utah School of Medicine, 590 Wakara Way, Salt Lake City, UT 84108; E-mail: Man.Hung@hsc.utah.edu T reatment outcomes in the population of patients with spinal disorders are commonly measured on the basis of clinical data including fusion rates, range of motion, muscle strength, deformity correction, and neurological deficits. 1 Although these measurements can provide informa- tion on patient outcomes, the measurements alone do not always accurately represent the patient’s health state nor do they necessarily represent outcomes that are important to patients. Patient-reported outcome (PRO) instruments are widely used to capture the patients’ perception of general health, quality of life, daily function, and pain. 2 The most com- monly used PROs in a spine population include the Oswestry Disability Index, Roland-Morris Disability Questionnaire, Neck Disability Index, Medical Outcomes Study Short Form questionnaire (SF-6, SF-12, SF-36), Japanese Orthopaedic Association myelopathy questionnaire, the Myelopathy Dis- ability Index (MDI), visual analogue scale, and the EuroQol Five Dimension. 1 Gathering this outcomes information often overburdens the patient with time-consuming questionnaires that have inaccuracies and poor psychometrics, such as floor and ceiling effects, when testing patients with either severe functional disability or extreme functional ability. 3 In an effort to develop improved PROs for patients with a variety of chronic illnesses and conditions, the Patient- Reported Outcomes Measurement Information System (PROMIS) was developed and funded by the National Insti- tutes of Health. 3 During the past decade, the PROMIS group has developed assessment tools that provide precise and efficient measurement of patient-reported health status for 1.00 and person reliability was 0.99. We found evidence of item response bias associated with sex, age, and education in some items. Conclusion. The PROMIS PF item bank adequately addressed outcomes of patients with spinal disorders as reliabilities were excellent, minimal ceiling/floor effect existed, and item bias was limited. Future effort should be focused on eliminating, rescaling, or modifying those items that had item bias. Key words: physical function, spine, PROMIS, orthopedic, psychometrics, Rasch, patient-reported outcomes. Level of Evidence: 2 Spine 2014;39:158–163 Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.