Background People with rheumatic or neurologic conditions are commonly referred for occupational therapy because of decreased ability to perform ADL tasks. Upon referral, occupational therapists use a client-centred, occupationfocused approach when evaluating a person's perceived and observed quality of ADL task performance to clarify the nature and extent of the person’s problems, plan interventions and determine effectiveness of interventions. Evaluation of the extent of problems and change following intervention require the use of linear measures of ADL. The aim of this doctoral thesis was to develop and validate linear measures of perceived and observed quality of ADL task performance for occupational therapy clinical praxis and research. Methods To develop linear measures of observed quality of ADL task performance based on Rasch measurement methods, clients with ABI (n=70) were evaluated using a 3-category rating scale and the ADL Taxonomy. Similarly, to develop linear measures of perceived quality of ADL task performance women with rheumatic diseases (n=118) reported their ability on a questionnaire and in an interview based on the ADL Taxonomy and a 4- category rating scale. To further validate the ADL ability measures of the Assessment of Motor and Process Skills (AMPS) data were collected in two samples. In a sample of women with chronic widespread pain (CWP) or fibromyalgia (FM) (n=50) the ADL ability measures were validated with regard to discrimination, stability and sensitivity to changes. Discrimination was examined by comparing AMPS data from women with CWP/FM to those of healthy women. Stability and sensitivity to change were examined based on repeated AMPS observations pre and post intervention. In addition, sensitivity to change was examined in a sample of clients with ABI (n=36) using a pre- and post-test design. Finally, the validity of the AMPS ADL ability measures in relation to perceived ADL ability was examined in women with rheumatic diseases (n=118) and women with CWP/FM (n=50). Results Based on the results of the four studies in this thesis, we were able to conclude that it was possible to obtain linear measures of observed quality of ADL task performance based on a revised version of the ADL Taxonomy (ADL-O). It was also possible to obtain linear measures of perceived quality of ADL task performance based on revised versions of the ADL Taxonomy using the formats of questionnaires (ADL-Q) and interviews (ADL-I). Moreover, it was possible to establish further evidence of validity of the ADL ability measures of the AMPS. Thus, the AMPS ADL ability measures could discriminate between women with CWP/FM and healthy women. The ADL ability measures of remained stable when no intervention was provided and the ADL motor ability measures were sensitive to change among women with CWP/FM. Moreover, the ADL ability measures were sensitive to change among clients with low ADL ability following ABI. Finally, only low to moderate correlations were found between measures of perceived and observed ADL ability. Conclusions Rasch analyses of revised versions of the ADL Taxonomy suggest that the majority of the ADL tasks and rating scales focused on quality of ADL task performance can be used to obtain linear measures of quality of ADL task performance based on methods of questionnaire, interview or observation. Furthermore, the studies provided evidence of validity of the ADL ability measures of the AMPS in relation to discrimination, stability and sensitivity to change and relation to perceived ADL ability. The results may, therefore, enhance the use of ADL instruments that provide linear measures of perceived and observed quality of ADL task performance in occupational therapy clinical praxis and research.

Diss. (sammanfattning) Umeå : Umeå universitet, 2010.