Communication Quality Predicts Psychological Well-Being and Satisfaction in Family Surrogates of Hospitalized Older Adults: An Observational Study

Alexia M. Torke, Christopher M. Callahan, Greg A. Sachs, Lucia D. Wocial, Paul R. Helft, Patrick O. Monahan, James E. Slaven, Kianna Montz, Emily S. Burke, Lev Inger

Many hospitalized older adults require family surrogates to make decisions, but surrogates may perceive that the quality of medical decisions is low and may have poor psychological outcomes after the patient's hospitalization. To determine the relationship between communication quality and high-quality medical decisions, psychological well-being, and satisfaction for surrogates of hospitalized older adults. Observational study at three hospitals in a Midwest metropolitan area. Hospitalized older adults (65+ years) admitted to medicine and medical intensive care units who were unable to make medical decisions, and their family surrogates. Among 799 eligible dyads, 364 (45.6%) completed the study. Communication was assessed during hospitalization using the information and emotional support subscales of the Family Inpatient Communication Survey. Decision quality was assessed with the Decisional Conflict Scale. Outcomes assessed at baseline and 4-6 weeks post-discharge included anxiety (Generalized Anxiety Disorder-7), depression (Patient Health Questionnaire-9), post-traumatic stress (Impact of Event Scale-Revised), and satisfaction (Hospital Consumer Assessment of Healthcare Providers and Systems). The mean patient age was 81.9 years (SD 8.32); 62% were women, and 28% African American. Among surrogates, 67% were adult children. Six to eight weeks post-discharge, 22.6% of surrogates reported anxiety (11.3% moderate-severe anxiety); 29% reported depression, (14.0% moderate-severe), and 14.6% had high levels of post-traumatic stress. Emotional support was associated with lower odds of anxiety (adjusted odds ratio [AOR] = 0.65, 95% CI 0.50, 0.85) and depression (AOR = 0.80, 95% CI 0.65, 0.99) at follow-up. In multivariable linear regression, emotional support was associated with lower post-traumatic stress (β = -0.30, p = 0.003) and higher decision quality (β = -0.44, p < 0.0001). Information was associated with higher post-traumatic stress (β = 0.23, p = 0.022) but also higher satisfaction (β = 0.61, p < 0.001). Emotional support of hospital surrogates is consistently associated with better psychological outcomes and decision quality, suggesting an opportunity to improve decision making and well-being.