Alyssa M Bamer1, Kara McMullen1, Anne Deutsch2,3, Mitch Sevigny4, Tracy Mroz5, Shelly A Wiechman6, Jeffrey C Schneider7, Dagmar Amtmann8. 1. Department of Rehabilitation Medicine, University of Washington, University of Washington Center on Outcomes Research in Rehabilitation, 12360 Lake City Way, Suite 502, Seattle, WA, 98125, USA. 2. Shirley Ryan AbilityLab and Northwestern University, 355 E Erie St, Chicago, IL, 60611, USA. 3. RTI International, 230 W Monroe St, Chicago, IL, 60606, USA. 4. Spinal Cord Injury Model System, Craig Hospital, 3425 S. Clarkson Street, Englewood, CO, 80113, USA. 5. Department of Rehabilitation Medicine, University of Washington, 1959 NE Pacific Street, Box 356490, Seattle, WA, 98195, USA. 6. Department of Rehabilitation Medicine, University of Washington, 325 Ninth Ave, Box 359612, Seattle, WA, 98104, USA. 7. Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Research Institute, Spaulding Rehabilitation Hospital, 300 1st Avenue, Boston, MA, 02129, USA. 8. Department of Rehabilitation Medicine, University of Washington, University of Washington Center on Outcomes Research in Rehabilitation, 12360 Lake City Way, Suite 502, Seattle, WA, 98125, USA. dagmara@uw.edu.
Abstract
BACKGROUND: The Satisfaction with Life Scale (SWLS) is a widely used measure of subjective well-being. Recent evidence indicates the fifth item of the scale reduces the reliability of the scale and is inappropriate for use in traumatic injury populations. The purpose of this study was to develop a linking procedure between the five-item version of the SWLS and a modified four-item version, which removes the problematic item, for use in Spinal Cord (SCI), Traumatic Brain (TBI), and Burn Injury populations. METHODS: Proration (i.e. adding the mean of the four items to their total) was identified as a potential linking solution that could be easily implemented in clinical or research settings. The validity of the proration approach was evaluated by examining mean differences, cross group classification by SWLS category, score correlations, the intraclass correlation coefficient, and visual inspection of Bland-Altman plots in a large sample of SCI, TBI, and Burn Injury survivors who were participants in the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR) Model Systems' National Databases. RESULTS: A total of 17,897 (SCI n = 8566, TBI n = 7941, and Burn n = 1390) participants were included in this study. SWLS scores ranged from 5 to 35, and the average score difference between directly derived and prorated scores was 0.39 points. A large majority of the sample (93%) had score differences of < 4 points (i.e. approximately 0.5 SD). The correlation between the prorated and directly derived scores was very high (r = 0.97) and the ICC value indicated excellent reliability (ICC = 0.97). CONCLUSIONS: This study provides a valid scoring approach for researchers or clinicians who don't want to lose continuity with previously collected data but prefer to switch to the modified four-item version of the SWLS. Clear guidance is provided for traumatic injury researchers or clinicians on how to implement the proration scoring approach.
BACKGROUND: The Satisfaction with Life Scale (SWLS) is a widely used measure of subjective well-being. Recent evidence indicates the fifth item of the scale reduces the reliability of the scale and is inappropriate for use in traumatic injury populations. The purpose of this study was to develop a linking procedure between the five-item version of the SWLS and a modified four-item version, which removes the problematic item, for use in Spinal Cord (SCI), Traumatic Brain (TBI), and Burn Injury populations. METHODS: Proration (i.e. adding the mean of the four items to their total) was identified as a potential linking solution that could be easily implemented in clinical or research settings. The validity of the proration approach was evaluated by examining mean differences, cross group classification by SWLS category, score correlations, the intraclass correlation coefficient, and visual inspection of Bland-Altman plots in a large sample of SCI, TBI, and Burn Injury survivors who were participants in the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR) Model Systems' National Databases. RESULTS: A total of 17,897 (SCI n = 8566, TBI n = 7941, and Burn n = 1390) participants were included in this study. SWLS scores ranged from 5 to 35, and the average score difference between directly derived and prorated scores was 0.39 points. A large majority of the sample (93%) had score differences of < 4 points (i.e. approximately 0.5 SD). The correlation between the prorated and directly derived scores was very high (r = 0.97) and the ICC value indicated excellent reliability (ICC = 0.97). CONCLUSIONS: This study provides a valid scoring approach for researchers or clinicians who don't want to lose continuity with previously collected data but prefer to switch to the modified four-item version of the SWLS. Clear guidance is provided for traumatic injury researchers or clinicians on how to implement the proration scoring approach.
Authors: David S Tulsky; Pamela A Kisala; David Victorson; Denise Tate; Allen W Heinemann; Dagmar Amtmann; David Cella Journal: Arch Phys Med Rehabil Date: 2011-10 Impact factor: 3.966
Authors: Dagmar Amtmann; Fraser D Bocell; Alyssa Bamer; Allen W Heinemann; Jeanne M Hoffman; Shannon B Juengst; Marta Rosenberg; Jeffery C Schneider; Shelley Wiechman; Kara McMullen Journal: Assessment Date: 2017-02-01
Authors: Dagmar Amtmann; Kara McMullen; Alyssa Bamer; James A Fauerbach; Nicole S Gibran; David Herndon; Jeffrey C Schneider; Karen Kowalske; Radha Holavanahalli; A Cate Miller Journal: Arch Phys Med Rehabil Date: 2017-10-05 Impact factor: 4.060
Authors: Willem A Arrindell; Irene Checa; Begoña Espejo; I-Hua Chen; Danilo Carrozzino; Phuong Vu-Bich; Huong Dambach; Paula Vagos Journal: Int J Environ Res Public Health Date: 2022-03-15 Impact factor: 3.390