- Posts: 23
So part of what you have is correct, and part is not. To use your abbreviations, PFUE and PFLE are subsets of the generic PF item bank. PFLE uses the same item calibrations as the generic bank, and thus scores should be comparable regardless of which you administer (within the accuracy of the score). The PFUE items received their own calibrations as part of v2.0, when administered exclusively as the UE-CAT or UESF7a. Scores from PFUE are not comparable to scores from the generic PF item bank if using v2.0 (or later if there is a later).
Then regarding the generic PF-CAT and your questions:
1. Your synopsis is not correct. Both the PFUE and PFLE are better at measuring poorer physical functioning than the generic bank. The generic bank has a much wider range of accurate measurement. When administering the PF-CAT, most items that are administered are not part of either item subset. And even when individuals have poor physical function, the UE items are rarely chosen.
2. There is a manuscript under review at the time of writing this (March 2019) that describes the UE item bank development. It does state that UE items are rarely chosen in the generic PF-CAT, but does not expand on this assertion. Additionally Kaat et al. (2017; DOI: 10.1097/BOT.0000000000000924) documents the extreme score inconsistency between scores derived from the generic item bank (using an 8-item short form) and the original UE-CAT (v1.2). That was an impetus for the development and re-calibration of the UE v2.0 item bank.
3. I think there is a lot of information in this response. Please ask additional questions if this raises more than it answers.
I am looking for information regarding the relationships between the computer adaptive test (CAT) versions of the PROMIS Physical Function, PROMIS Physical Function Upper Extremity (PFUE), and PROMIS Physical Function Lower Extremity (PFLE; aka Mobility). It is my understanding that the PFUE and PFLE item banks are subsets of Physical Function item bank, and the items are limited to upper extremity or lower extremity specific functions, respectively. There are a number of studies comparing Physical Function, PFUE, and PFLE to each other and to legacy/static PRO measures (e.g., QuickDASH). I am not looking for this type of information.
I am specifically interested in the Physical Function CAT as I have been anecdotally informed that as patients respond that they have higher levels of physical function the Physical Function CAT provides them with more items that are typically found in the PFLE item bank. And, conversely, as patients respond that they have lower levels of physical function the Physical Functional CAT provides them with more items that are typically found in the PFUE item bank.
1. Is the above synopsis true/accurate?
2. If it is, is this documented anywhere (peer reviewed publication or otherwise)?
3. Any additional information regarding this topic and the functional relationships between the PROMIS Physical Function assessments would be greatly appreciated.