Allman, Ava-Ann

Abstract Rationale: The Deterioration Index (DTI) may serve as a crucial tool for predicting adverse patient outcomes; however, there is a paucity of literature attempting to externally validate this tool in specific clinical scenarios. In this study, we analyzed the relationship between DTI scores during rapid response team (RRT) activations and subsequent ICU transfer, initiation of goals of care (GOC) conversations, and 90-day mortality. Methods: This was a retrospective single-center review of adult patients who had an RRT activation while hospitalized between January 1, 2023 and December 31, 2023. This study was approved by our hospital institutional review board. Deterioration scores (0-100) were recorded every 15 minutes during the RRT event, and the highest score of each encounter was selected for analysis. The primary outcome was transfer to the ICU as a direct result of the RRT activation. Secondary outcomes included GOC discussions following RRT activation and 90-day mortality. GOC discussions, defined as physician-led conversations that specifically addressed code status, treatment priorities, or broader goals of care, met criteria only if initiated within 72 hours of the analyzed RRT activation. Logistic regression was used to assess the relationship between DTI scores during RRT activations and the aforementioned variables. Results: A total of 1643 patients were included in the analysis. For every 1-point increase in the deterioration score, the odds of ICU transfer increased by 2.7% (OR: 1.027, CI: 1.021-1.034, p<0.0001), the odds of a goals of care discussion within 72 hours increased by 3.3% (OR: 1.033, CI: 1.025-1.040, p<0.0001), and the likelihood of mortality within 90 days increased by 2.6% (OR: 1.026, CI: 1.019-1.033, p<0.0001). Conclusion: Our findings contribute to external validation of DTI predictive value during RRT events. DTI scores were associated with key clinical outcomes, including ICU transfer and 90-day mortality. There was also an association between DTI scores and the likelihood of physician-led GOC discussions, suggesting that DTI may demonstrate predictive value not only of patient outcomes, but also the likelihood of specific non-medical physician actions. Further validation of these scores in prospective studies, specifically in the context of acute decompensation events, could help facilitate the integration of DTI-based RRT interventions into routine hospital practice.

Abstract Rationale: Medical trainees are often involved in the care of hospitalized patients, yet there are minimal data about correlation of trainee involvement with outcomes for patients with rapid response team (RRT) activations. Methods: This was a retrospective single-center review of adult patients on a primary medicine service who had an RRT activation between January 1, 2023 and December 31, 2023. This study was approved by our hospital institutional review board. Patients were included if they were at least 18 years old and had a hospitalization with an RRT activation. Outcomes analyzed included transfer to the ICU, 90-day mortality, and goals of care discussions. A subset analysis was performed of patients who remained on the floor after RRT. Statistical analysis was performed with Pearson's chi-square and Kruskal-Wallis tests. Results: We analyzed 1686 adult patients who had an RRT activation within the study timeframe. Mean age was 67.3 years (SD 16.8), 792 (47.0%) were female, and 1489 (88.3%) identified as white. Trainees were involved with 1016 (60.3%) patients; 935 of these were on an internal medicine team and 81 were on a family medicine team. Patients were more likely to transfer to the ICU if they were on a primary floor team with trainee involvement (p = 0.006). There was no difference in goals of care discussion occurrence prior to RRT (p = 0.257) or 90-day mortality (p = 0.877). Subset analysis of patients who remained on the floor after RRT was performed. There was no difference in occurrence of goals of care discussions after RRT (p = 0.169), code status change (p = 0.237), or transition to comfort care (p = 0.381) for patients on teams with trainee involvement and those without. Of the 925 patients who remained on the floor after RRT, goals of care discussion occurred after RRT in 210 (22.7%). Conclusions: In this study, we demonstrate that trainee involvement does not correlate with occurrence of goals of care discussions either before or after RRT activations. Trainee involvement also does not correlate with 90-day mortality for patients with decompensation events. However, patients on teams with trainee involvement are more likely to transfer to the ICU as the result of an RRT activation. It is possible that this reflects the increased complexity of patients who are admitted to teaching services; however, further investigation is needed.