Abstract
Background
Physical function is a key outcome for monitoring patient recovery. Performance-based outcome measures (PBOMs) provide an unbiased and reliable method for assessing lower limb physical function. Musculoskeletal clinical practice and research are significantly impacted by the lack of a PBOM of lower limb physical function that can be used in populations with diverse functional abilities, such as community cohorts with knee osteoarthritis.
Aim
The primary aim of this thesis was to develop a multi-activity comprehensive PBOM to assess lower limb physical function and assess its initial measurement properties in people with knee osteoarthritis. Secondary aims that emerged during the development of the thesis were to develop a self-reported outcome measure (SROM) to assess lower limb physical function and compare self-reported to externally-rated physical function.
Methods
A scoping review identified physical activities included in existing PBOMs and SROMs of common knee musculoskeletal conditions. A discrete choice experiment (DCE) explored the perceived difficulty and importance of the candidate physical activities and created rank orders. Preliminary versions of the Performance of Lower Limb Scale (POLLS) and a self-reported version (self-POLLS) were developed and tested using Rasch analysis on a sample with diverse levels of physical function, including people with and without osteoarthritis. Initial measurement properties assessed were structural and construct validity, internal consistency, and floor and ceiling effects. A secondary analysis of the tested sample assessed the concordance of self-reported physical function compared to observed physical function.
Results
A total of 143 articles were included in the scoping review. A wide range of physical activities was identified in 93 different outcome measures. Climbing stairs, arising from sitting, and walking short distances were the most commonly included physical activities. Outcome measures for people with knee osteoarthritis did not include unilateral or challenging activities. PBOMs assessed discrete physical activities. Key limitations identified in most SROMs were the broad description of the physical activities and the lack of challenging or unilateral activities.
A total of 613 responses were analysed in the DCE. The perceived difficulty of physical activities was negatively correlated with the perceived importance. The perceived difficulty was similar for people with osteoarthritis compared to people without osteoarthritis. Differences in the rating of the perceived importance of some activities were found between these groups, with people with osteoarthritis rating some easier activities as more important than those without osteoarthritis. Samples of 101 and 127 participants were recruited for testing the preliminary POLLS and self-POLLS, respectively. The final versions of both measures fitted the Rasch model, achieved unidimensionality, exhibited sufficient construct validity and excellent internal consistency. No floor or ceiling effect was found.
The comparison between the self-POLLS and POLLS results revealed that participants generally overestimated their abilities. Participants were more accurate when reporting limitations.
Conclusions
This thesis developed a PBOM (POLLS) and a SROM (self-POLLS) to assess lower limb physical function. These measures are limb-specific, cover diverse functional abilities, and were demonstrated to be valid and internally consistent. While the self-POLLS provides a feasible alternative in contexts where externally-rated assessment cannot be conducted, the POLLS is recommended for accurately assessing lower limb physical function.