What Does a PROMIS T-score Mean for Physical Function? JEFF HOUCK, PT, PHD 1, ZANE WISE, DPT 1, AMANDA TAMANAHA, DPT 1, LUKE SKERJANEC, DPT 1, ALEXANDRA WEGNER, DPT 1, CHRIS DASILVA 2, MICHAEL BASS 3, JUDITH BAUMHAUER, MD, MPH 2 1 GEORGE FOX UNIVERSITY 2 UNIVERSITY OF ROCHESTER SCHOOL OF MEDICINE 3 NORTHWESTERN UNIVERSITY POSTER #459
No Conflicts to Disclosure The authors have no potential conflicts of interest to disclose. Jeff Houck, PT, PhD 1, Zane Wise, DPT 1, Amanda Tamanaha, DPT 1, Luke Skerjanec, DPT 1, Alexandra Wegner, DPT 1, Chris Dasilva 2, Michael Bass 3, Judith Baumhauer, MD, MPH 2 The authors disclosures are listed in the FINAL AOFAS mobile APP
Patient Reported Outcomes Instrumentation System(PROMIS) Physical Function Scale (v1.0 = 124 items, v2.0>124 items) Item Response Theory Used to model responses to questions Responses on ALL items are determined in 4-8 questions 2010 US Census for Physical Function T-Score normalized to the US population 50 = average for the US population 10 points = 1 standard deviation of the US population Advantages Takes less than a minute to complete Model estimates score on ALL items Accuracy for individual patients is +/- 1-2 SD units Advantages Minimal floor and ceiling effects Disadvantage No tools exist to convert T-Scores into meaningful values for patients and clinical decisions
Purpose The purpose of this research was to link PROMIS PF T-scores with physical function activities and provide a visual map of this linkage to aid in treatment assessment and address concrete patient education. Empower clinicians to use PROMIS PF t-scores to answer concrete questions like: When will I be able to run again?
Methods Link each item response to a specific PROMIS PF t-score Obtained model parameters used generate PROMIS t-scores based on patient responses from developers (Micheal Bass/Chris Dasilva). e.g. Version 1.0 = 124 items with a total of 4 possible model parameters per item yielding 5 Likert scale estimates Model parameters were used to assign an estimated Likert scale response for each items the patient actually did not answer for any t- score
Methods (cont) Items were clustered into categories that are meaningful to clinicians The World Health Organization International Classification of Function(ICF) is widely used in models of disability. 36 PROMIS PF items were grouped into 9 distinct ICF categories including: Self care Hand arm use Transferring Walking Lift and Carry House Work Climbing Stairs Fitness, Running. Display the visual map of items in ways that are useful to clinicians
Results Simple Visual Map (Snapshot) Includes only 3 ICF categories Colors match patients expected ability Tasks ordered based on difficulty T-SCORE PROMIS PF ITEMS
Results Simple Visual Map (Snapshot) Includes 9 ICF categories Colors match patients expected ability Tasks ordered based on difficulty 36 PROMIS PF items T-SCORE PROMIS PF ITEMS
Conclusions PROs provide real time assessments and a road map to follow patients throughout a treatment course. Understanding the translation of the outcome score (T-score) to patient physical activity allows the patient and physician to have realistic expectations of recovery. Applying this PF and activity linkage data to cohorts of patients with common surgeries will allow patients to gain a better understanding of the recovery duration and return to activity timing. Providing this patient friendly knowledge will help enhance patient engagement and patient satisfaction.
References Ader DN. Developing the Patient-Reported Outcomes Measurement Information System (PROMIS). Med Care. 2007;45:S1-2. Amtmann D, Cook KF, Johnson KL, Cella D. The PROMIS initiative: involvement of rehabilitation stakeholders in development and examples of applications in rehabilitation research. Arch Phys Med Rehabil. 2011;92:S12-19. Cella D, Yount S, Rothrock N et al. The Patient-Reported Outcomes Measurement Information System (PROMIS): progress of an NIH roadmap coorperative group during its first two years. Med Care. 2007;45:S3-11. Fries JF, et al. Item response theory, computerized adaptive testing, and PROMIS: assessment of physical function. J Rheumatol. 2014;41:153-158. Habibi A, Sarafrazi A, Izadyar S. Delphi technique theoretical framework in qualitative research. Int J Sci Eng. 2014;4:8-13. HealthMeasures (Northwestern Univerisity). Available at: http://www.healthmeasures.net/explore-measurement-systems/promis. Accessed on April 13, 2017. Hsu C-C, Sandford BA. The Delphi technique: making sense of consensus. Prac Assess, Res & Eval. 2007;12(10). Lai JS, et al. How item banks and their application can influence measurement practice in rehabilitation medicine: a PROMIS fatigue item example. Arch Phys Med Rehabil. 2011;92:S20-27. Penfield RD. An NCME instructional module on polytomous item response theory models. J Educ Meas. 2014;33:36-48. Rose M, et al. The PROMIS physical function item bank was calibrated to a standardized metric and shown to improve measurement efficiency. J Clin Epidemiol. 2014;67:516 526. World Health Organization - ICF Browser. Available at: http://apps.who.int/classifications/icfbrowser/. Accessed on April 13, 2017.