Free
Poster Session
Issue Date: August 2016
Published Online: August 01, 2016
Updated: January 01, 2021
Metabolic Equivalent as an Underlying Component of ADL Measures
Author Affiliations
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
  • Medical University of South Carolina
Article Information
Neurologic Conditions / Stroke / Assessment/Measurement
Poster Session   |   August 01, 2016
Metabolic Equivalent as an Underlying Component of ADL Measures
American Journal of Occupational Therapy, August 2016, Vol. 70, 7011500028. https://doi.org/10.5014/ajot.2016.70S1-PO3086
American Journal of Occupational Therapy, August 2016, Vol. 70, 7011500028. https://doi.org/10.5014/ajot.2016.70S1-PO3086
Abstract

Date Presented 4/8/2016

This study identifies metabolic equivalent (MET) as a common underlying component in two activity of daily living measures. Regression models demonstrated that MET was a significant factor in explaining item difficulties for the FIM™ and the Stroke Impact Scale.

Primary Author and Speaker: Matthew Husband

Additional Authors and Speakers: Ickpyo Hong, Christine Harris, Emily Schoen, Clare Fitzmaurice, Danielle Kapustka, Craig Velozo, Erica Ingram

PURPOSE: The purpose of this study was to determine whether metabolic equivalent (MET) values contribute as an underlying measurement component in activity of daily living (ADL) measures. We hypothesized that MET values are significant components of two similar ADL measures.
RATIONALE: Presently, there are >85 published ADL measures, with the vast majority showing acceptable reliability and validity. However, it is difficult to use ADL scores in treatment planning without an underlying theory that supports the relative challenge of different ADL activities. Identifying an underlying measurement model would allow for communication between different measures and provide a rationale for treatment planning.
DESIGN: Cross-sectional design and multivariate regression models using the published Rasch-calibrated item difficulty levels of the Stroke Impact Scale–Physical Activity (SIS) and the FIM™ Motor scale
PARTICIPANTS: Six 2nd-yr occupational therapy student raters scored the number of muscle groups for SIS and FIM ADL items. The mean age of raters was 24 yr (standard deviation = 1); participants were 5 women and 1 man.
MEASURES: This study used published Rasch item calibrations from the 13 ADL items of the FIM Motor and 16 ADL items of the SIS Physical Activity domain. Muscle groups were defined as (1) shoulder and arm, (2) wrist and hand, and (3) lower extremity. Total number of muscle groups necessary for an activity was summed for each ADL item.
METHOD: The Rasch item difficulty values of the two measures and MET values were retrieved from the published papers (Ainsworth et al., 2000). The 6 raters assigned the number of muscle groups necessary to complete each of the ADL items on the FIM and SIS.
ANALYSIS: Regression models were developed with the two independent variables (MET values and the number of necessary muscle groups) and the dependent variable as the published item difficulty values of the FIM and SIS ADL items using SPSS Version 23.
RESULTS: The 6 raters demonstrated reliable ratings on the number of muscle groups for the test items of the FIM, intraclass correlation (ICC) (2, 6) = .95, and the SIS, ICC(2, 6) = .97. Our hypothesis was supported by MET values being significant factors in the two regression models. For the FIM model, MET values explained 37% of the variance (p = .03). For the SIS model, MET values explained 50% of the variance (p = .003). Number of muscle groups explained another 20% of the variance for the SIS measurement model.
CONCLUSIONS: The significance of the MET in the regression models for the FIM and SIS supports our hypothesis that there is an underlying common measurement model for ADL measures. Based on our findings, clinicians may be able to grade their treatments by giving patients a progression of ADL activities based on choosing ADLs with increasing MET value. Future research is needed to determine whether METs are common to other ADL measures.
LIMITATION: The descriptions of the MET values were not available for several test items (i.e., bladder and bowel management). This may introduce error in the regression models. Other independent variables may describe additional variance in ADL measures.
IMPACT STATEMENT: Identifying an underlying factor that determines item difficulty on various ADL measures may help clinicians develop client-centered treatment plans. Clients’ physical ability can be matched to ADL activities of a particular MET value; individualized treatment plans can then be designed by having clients perform ADL tasks with increased MET values.
References
Ainsworth, B. E., Haskell, W. L., Whitt, M. C., Irwin, M. L., Swartz, A. M., Strath, S. J., . . . Leon, A. S. (2000). Compendium of physical activities: An update of activity codes and MET intensities. Medicine and Science in Sports and Exercise, 32(9, Suppl.), S498–S504. http://dx.doi.org/10.1097/00005768-200009001-00009