OBJECTIVES: The International Classification of Functioning, Disability & Health is commonly used to measure outcomes in rehabilitation. As other instruments it assigns equal weight to each item. The objective is to examine the extent to which preference-weighted core elements of the ICF differ from unweighted assignments currently used in treatment decisions.

METHODS: A best–worst scaling experiment Case II is used to value body function. ICF dimensions relevant in terms of function (12 attributes, 3 levels each) are extracted from ICFs chapter on neuromusculoskeletal & movement-related functions. Stroke patients as well as members of the public are recruited. A fractional, efficient design is applied for the online survey (randomization, forced choice). Conditional and multinominal logit analyses are used as the main analysis method.

RESULTS: N=373 participants were recruited until August 2022. The study is currently collecting data of the patient population. In BWS II, ensuring “Control of voluntary movement functions” (SQRT: 1.396) and “gait pattern functions”(SQRT: 1.265) are most important, with “muscle endurance” (SQRT: 0.668) and Muscle tone functions (SQRT: 0.729) being least relevant. The level “No problem” always has lowest sqrt and highest worst counts, while a “complete problem” is always rated more important irregardless of the respective attribute/category. This means that respondents clearly expressed what should be the short-term focus in the rehabilitation. The more severe a problem, the more focus should be placed on this in the therapy.

CONCLUSIONS: If improvements of functions have effects on activities and these have effects in terms of health-related quality of life raises the question how the value of functions can be measured? The preliminary results show that unlike in the ICF body functions are not equally weighted by those affected. Results reveal differences between patient/public judgements and current clinical practice. This enhances the need for preference-based outcome evaluations.