PROMISing development for standardising the assessment of self-reported health worldwide? An exploration of the potential application of U.S. developed patient-reported outcome measures to Germany
Public Health, Healthcare Research, Social and Occupational Medicine
Final Report Abstract
In the course of growing focus on patient centeredness, with patient-reported outcome (PRO) data becoming an important source for medical decision making, it is increasingly important that PRO instruments meet minimum quality criteria, such as validity, reliability and responsiveness. In addition, it is crucial to provide evidence for measurement equivalence of a PRO instrument in order to compare PRO data across different interventions, patient groups, diseases, cultures and countries. To allow for such comparison, several initiatives have been started that are aimed at standardising PRO assessment, one of which is the Patient-Reported Outcomes Measurement Information System (PROMIS®) initiative established in the U.S. in 2004. Using the PROMIS v1.2 Physical Function (PF) item bank as an example, the present project was aimed at undertaking a thorough investigation of the psychometric performance of the item bank in a different cultural context (i.e., US-English versus German) and across different disease groups (i.e., psychosomatic medicine, rheumatology and cardiology) as well as investigating the concordance between self-reported data and results from a physical performance tests. The key findings of this project are that there was a high level of concordance between patients’ self-report and their scores based on a physical performance test. However, despite support for the psychometrics of the PF item bank, we also found evidence for differential item functioning across three patient populations, especially between rheumatology and cardiology. That is, PRO scores are not comparable between these groups if no further adjustments are undertaken. In addition, using different item subsets of the bank resulted in different PF scores, a finding that was particularly pronounced in rheumatology patients. Therefore, both the unidimensionality assumption as well as the assumption of measurement equivalence across disease groups were refuted as part of this project. In conclusion, this project provided evidence for the overall psychometric performance of the PROMIS PF item bank in German. However, issues were found regarding its dimensionality and the performance of specific PF items, especially those related to upper extremity, between disease groups. Our main recommendations resulting from this project are the development of disease-specific item parameters for items showing DIF and the application of subdomain- related content balancing for scoring of the generic PF construct. Such approach would ensure that items from all four PF subdomains are administered to a respondent; hence, the influence of DIF would be substantially reduced when calculating a patient’s total physical function score.
Publications
- (2017). Der Einfluss von Schmerz, Krankheitswahrnehmung und Depressivität auf die Erfassung der körperlichen Funktionsfähigkeit mittels unterschiedlicher Erhebungsmethoden. Deutscher Kongress für Psychosomatische Medizin und Psychotherapie, Berlin, Germany
Liegl G, Stengel A, Obbarius A, Knebel F, Buttgereit F, Rose M, Nolte S
- (2017). Generic self-reported and performance-based measures of physical function are highly correlated but differentially affected by pain and illness perception. Quality of Life Research 25(1 Supplement):1-196; ISOQOL 23rd Annual Conference, Copenhagen, Denmark
Liegl G, Kanlidere S, Stengel A, Obbarius A, Knebel F, Buttgereit F, Rose M, Nolte S
(See online at https://doi.org/10.1007/s11136-016-1390-7) - (2017). Prior execution of physical tasks influences patients’ self-assessment of physical function. Quality of Life Research 25(1 Supplement):1-196. ISOQOL 23rd Annual Conference, Copenhagen, Denmark
Liegl G, Kanlidere S, Knebel F, Obbarius A, Stengel A, Buttgereit F, Rose M, Nolte S
(See online at https://doi.org/10.1007/s11136-016-1390-7) - (2017). The impact of disease-related differential item functioning on PROMIS® Physical Function scores in cardiology and rheumatology patients. Quality of Life Research 26(Suppl 1):1; ISOQOL 24th Annual Conference, Philadelphia, USA
Liegl G, Rose M, Fischer HF, Buttgereit F, Knebel F, Stengel A, Nolte S
- (2017). The Value of PROMIS® Physical Function for Predicting Physical Performance in Patients with Various Conditions. PHO Conference 2017, Philadelphia, USA
Liegl G, Obbarius A, Rose M, Nolte S
- (2018). An initial psychometric evaluation of the German PROMIS v1.2 Physical Function item bank in patients with a wide range of health conditions. Clin Rehabil, 32(1):84–93
Liegl G, Rose M, Correia H, Fischer HF, Kanlidere S, Mierke A, Obbarius A, Nolte S
(See online at https://doi.org/10.1177%2F0269215517714297) - (2018). Möglichkeiten IRT-basierter Methoden zur konstruktbasierten Erfassung körperlicher Funktionsfähigkeit am Beispiel der PROMIS Physical Function Itembank. 27. Reha-Kolloquium | Deutscher Kongress für Rehabilitationsforschung, Munich, Germany
Liegl G, Fischer HF, Nolte S, Rose M
- (2020). Using subdomain-specific item sets affected PROMIS Physical Function scores differently in cardiology and rheumatology patients. J Clin Epidemiol.
Liegl G, Rose M, Knebel F, Stengel A, Buttgereit F, Obbarius A, Fischer HF, Nolte S
(See online at https://doi.org/10.1016/j.jclinepi.2020.08.003)
