THE ASSOCIATION BETWEEN A LONELINESS RESOURCE GUIDE AND HEALTH-RELATED QUALITY OF LIFE AMONG A MEDICARE ADVANTAGE POPULATION

Abstract

Introduction Loneliness and social isolation have been associated with worsening health-related quality of life (HRQOL), increased mortality, and other poor physical and mental health outcomes among older adults. It is believed early identification of loneliness and related patient characteristics can guide more targeted and effective interventions. The objective of this study was to measure the association between the dissemination of a loneliness resource guide and HRQOL among a sample of a Medicare Advantage population with a high propensity for loneliness. Methods A random sample of 50,000 individuals enrolled in a Medicare Advantage plan, by a national health and wellbeing company, was scored using a model to predict propensity for loneliness. The top 20% (N=10,000) of the sample, predicted to be most lonely and not eligible for Medicare before age 65 because of disability, was randomly assigned to an intervention (n=5,000) and control group (n=5,000). In October 2017 the intervention group was mailed a loneliness resource guide, which provided information, worksheets, and resources for loneliness. The content was designed to describe loneliness, educate about related risk factors, and direct individuals to appropriate resources. Health-related quality of life was measured by Healthy Days, a valid set of HRQOL measures developed by the Centers of Disease Control and Prevention (CDC), which measured the physical and mental Unhealthy Days in the past 30 days. Total Unhealthy Days (UHD) was reported as the sum of physical and mental Unhealthy Days. In September and October of 2017 the intervention and control groups were administered the Healthy Days measures, telephonically. Healthy Days data was collected again in January and February of 2018, after the intervention. A linear mixed model using repeated measures assessed the longitudinal association between the intervention group and UHD. Stratified post-hoc analyses were used to identify characteristics association with the greatest change in UHD. Results Average total UHD for the intervention group (n= 782) was 14.93 UHD (pre) and 14.17 UHD (post). Average total UHD for the control group (n= 813) was 15.35 UHD (pre) and 15.18 UHD (post). Dissemination of the loneliness resource guide was associated with a reduction of 0.58 UHD (95% CI: -2.15, +0.98) when compared to the control group. Stratified analyses identified the loneliness resource guide was associated with a reduction of 2.31 UHD (95% CI: -4.96, +0.35) compared to the control group among those without evidence of depression or disability (in the prior 2 years); Average total UHD for the intervention group (n= 262) was 11.77 UHD (pre) and 10.81 UHD (post). Average total UHD for the control group (n= 244) was 12.18 UHD (pre) and 13.53 UHD (post). Conclusions Our analyses identified the largest reduction in HRQOL among individuals with a high propensity for loneliness, who were non-disabled and without a history of depression. Future loneliness research should continue to explore this population. This research was funded by Humana Inc. was the study sponsor; no external funding was involved in this study.