258: EVALUATING PHYSICIAN VARIABILITY IN USE OF TIME-LIMITED TRIALS IN MECHANICALLY VENTILATED PATIENTS

Le résumé fourni par la source

Introduction: A time-limited trial (TLT) is a collaborative plan among patients, surrogates, and clinicians to try life-sustaining therapy for a defined duration for critically ill patients whose prognosis is uncertain. Prior work demonstrates variability in physicians’ use of TLTs, but little is known about the drivers of this variability. We sought to test the hypothesis that physician characteristics are associated with TLT use. Methods: This is a secondary analysis of an observational study in medical ICUs within 5 hospitals, focused on identifying TLTs in patients with respiratory failure receiving mechanical ventilation. Attending ICU physicians for all ventilated patients received 1) a query asking whether a TLT was being used with 3 options (yes, considering, or no) and 2) a voluntary demographic survey. We performed t-tests and Chi squared tests to evaluate the relationships between physician characteristics and TLT use. Results: Of 117 unique physicians, 27 (23.1%) never conducted or considered a TLT (non-adopters), 69 (59.0%) conducted at least 1 TLT, and 21 (17.9%) considered at least 1 TLT but never used a TLT. Excluding non-adopters, mean frequency of physician TLT use was 11.9% (SD 17.0%) of ventilated patients. Among 79 responders to the demographic survey (67.5% response rate), the proportion of non-adopters was 10/ 44 (22.7%) in physicians with 10 years or less in practice, 4/ 23 (17.4%) in those with 11 to 20 years, and 4/ 12 (33.3%) in those with > 20 years (p = 0.58). Among 29 female physicians, 6 (20.7%) were non-adopters compared to 12 (24.5%) of 49 male physicians (p = 0.70). When stratified by patient volume (# of ventilated patients cared for during study period), 43.3% of physicians in the lowest-volume quartile were non-adopters compared to 3.5% in the highest-volume quartile (p < 0.001). Conclusions: We found that ICU physicians with higher patient volume during our study were more likely to use or consider TLTs, but we did not find a relationship between physicians’ years in practice or gender. This finding could represent ascertainment bias or reflect that physicians who spend more time in the ICU are more likely to utilize TLTs. Future research should seek to uncover and characterize physicians’ reasoning about why and whether they use TLTs.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.