Assessing the concurrent validity of days alive and at home metric

Background Most patients living with serious illness value spending time at home. Emerging data suggest that days alive and at home (DAH) may be a useful metric, however more research is needed. We aimed to assess the concurrent validity of DAH with respect to clinically significant changes in patient‐ and caregiver‐reported outcomes (PROs). Methods We drew data from a study that compared two models of home‐based palliative care among seriously ill patients and their caregivers in two Kaiser Permanente regions (Southern California and Northwest). We included participants aged 18 years or older (n = 3533) and corresponding caregivers (n = 463). We categorized patients and caregivers into three groups based on whether symptom burden (Edmonton Symptom Assessment System, ESAS) or caregiving preparedness (Preparedness for Caregiving Scale, CPS) showed improvements, deterioration, or no change from baseline to 1 month later. We measured DAH across four time windows: 30, 60, 90, and 180 days, after admission to home palliative care. We used two‐way ANOVA to compare DAH across the PRO groups. Results Adjusted pairwise comparisons showed that DAH was highest for patients whose ESAS scores improved or did not change compared with those with worsening symptoms. Although the mean differences ranged from less than a day to about 3 weeks, none exceeded 0.3 standard deviations. ESAS change scores had weak negative correlations (r = −0.11 to −0.21) with DAH measures. CPS change scores also showed weak, positive correlations (r = 0.23–0.24) with DAH measures. Conclusion DAH measures are associated, albeit weakly, with clinically important improvement or maintenance of patient symptom burden in a diverse, seriously ill population. See related article by Freed et al. and editorial by Ankuda et al. in this issue.