Tasha Straszewski1, Colleen A Ross2, Carley Riley3,4, Brita Roy5, Matthew C Stiefel6. 1. Institute for Healthcare Improvement, Boston, USA. 2. Kaiser Permanente Colorado, Denver, USA. 3. Cincinnati Children's Hospital Medical Center, Cincinnati, USA. 4. University of Cincinnati College of Medicine, Cincinnati, USA. 5. Yale School of Medicine, New Haven, USA. 6. Social Health Practice, Kaiser Permanente, Oakland, USA. Matt.Stiefel3@gmail.com.
Abstract
PURPOSE: We investigated the relationship between measures of self-reported health and well-being and concurrent and prospective healthcare utilization and costs to assess the added value of these self-reported measures in understanding utilization and cost. METHODS: Kaiser Permanente members (N = 6752) completed a 9-item survey measuring life evaluation, financial situation, social support, meaning and purpose, physical health, and mental health. Responses were linked to medical record information during the period 12 months before and after the survey. RESULTS: Correlations between health and well-being measures and healthcare utilization and cost variables were generally weak, with stronger correlations for future life evaluation and selected health measures (ρ = .20-.33, ps < .001). Better overall life evaluation had a significant but weak association with lower total cost and hospital days in the following year after controlling for age, sex, and race/ethnicity (p < .001). Full multivariate models, adjusting for age, sex, race/ethnicity, prior utilization, and relative risk models, showed weak associations between health and well-being measures and following year total healthcare cost and utilization, though the associations were relatively stronger for the health variables than the well-being variables. CONCLUSION: Overall, the health and well-being variables added little to no predictive utility for future utilization and cost beyond prior utilization and cost and the inclusion of predictive models based on clinical information. Perceptions of well-being may be associated with factors beyond healthcare utilization. When information about prior use is unavailable, self-reported health items have some predictive utility.
PURPOSE: We investigated the relationship between measures of self-reported health and well-being and concurrent and prospective healthcare utilization and costs to assess the added value of these self-reported measures in understanding utilization and cost. METHODS: Kaiser Permanente members (N = 6752) completed a 9-item survey measuring life evaluation, financial situation, social support, meaning and purpose, physical health, and mental health. Responses were linked to medical record information during the period 12 months before and after the survey. RESULTS: Correlations between health and well-being measures and healthcare utilization and cost variables were generally weak, with stronger correlations for future life evaluation and selected health measures (ρ = .20-.33, ps < .001). Better overall life evaluation had a significant but weak association with lower total cost and hospital days in the following year after controlling for age, sex, and race/ethnicity (p < .001). Full multivariate models, adjusting for age, sex, race/ethnicity, prior utilization, and relative risk models, showed weak associations between health and well-being measures and following year total healthcare cost and utilization, though the associations were relatively stronger for the health variables than the well-being variables. CONCLUSION: Overall, the health and well-being variables added little to no predictive utility for future utilization and cost beyond prior utilization and cost and the inclusion of predictive models based on clinical information. Perceptions of well-being may be associated with factors beyond healthcare utilization. When information about prior use is unavailable, self-reported health items have some predictive utility.
Authors: Eric S Kim; Kaitlin A Hagan; Francine Grodstein; Dawn L DeMeo; Immaculata De Vivo; Laura D Kubzansky Journal: Am J Epidemiol Date: 2016-12-07 Impact factor: 4.897
Authors: Karen J Blumenthal; Yuchiao Chang; Timothy G Ferris; Jenna C Spirt; Christine Vogeli; Neil Wagle; Joshua P Metlay Journal: J Gen Intern Med Date: 2017-03-24 Impact factor: 5.128
Authors: Nancy A Perrin; Matt Stiefel; David M Mosen; Alan Bauck; Elizabeth Shuster; Erin M Dirks Journal: Am J Manag Care Date: 2011-12-01 Impact factor: 2.229