Introduction: Risk-adjusted stroke mortality is a publicly-reported CMS quality measure. Patients (pts) admitted with stroke who are discharged to hospice prior to the first midnight are excluded in reported metrics. It is difficult to prognosticate outcomes early in ischemic stroke; additionally, the sudden nature of the disease makes it difficult for families to make rapid decisions. We hypothesized that the decision to proceed to comfort care in ischemic stroke pts takes longer than 24 hours.
Methods: This was an observational study of all ischemic stroke pts seen at Intermountain Medical Center in 2016 (n=581) who were discharged to hospice (n=27) or died in hospital on comfort care (n=27). A Wilcoxon sign rank test was used to compare time to decision regarding comfort care to the CMS standard of 24 hours. Differences in the time to decision or death for the pt characteristics were also compared using Wilcoxon-Mann-Whitney test.
Results: The average age was 80 years old ± 13 (range 44 - 98), 59% were female, all were Caucasian Non-Hispanic, and 65% reported a religious affiliation. NIH stroke score was reported in 42 subjects with an average of 15.9±7.9 (range 2-31). The mean time from admission to initiation of comfort care orders was 3.12 days (±3.6) with a median of 1.8 days and a range from 12.5 hours to 14.1 days. The time was significantly greater than 24 hours (p<0.0001) with only a third (n=18) having the time less than 24 hours and with only a fifth (n=11) having the time before the first midnight. The table below contains the differences in the times based on pt characteristics and none of these were significantly associated with time differences.
Conclusions: The decision to transition to comfort care or hospice often takes greater than 24 hours. If the stroke mortality metric is meant to exclude hospice pts or those that prefer comfort measures only, the time window for initiation of those orders should be extended beyond the first midnight.