Patient Characteristics Associated With Occurrence of Preoperative Goals-of-Care Conversations

IMPORTANCE Communication about patients' goals and planned and potential treatment is central to advance care planning. Undertaking or confirming advance care plans is also essential to preoperative preparation, particularly among patients who are frail or will undergo high-risk surgery. OBJECTIVE To evaluate the association between patient risk of hospitalization or death and goals-of-care conversations documented with a completed Life-Sustaining Treatment (LST) Decisions Initiative note among veterans undergoing surgery. DESIGN, SETTING, AND PARTICIPANTS This retrospective cross-sectional study included 190 040 veterans who underwent operations between January 1, 2017, and February 28, 2020. Statistical analysis took place from November 1, 2021, to November 17, 2022. EXPOSURE Patient risk of hospitalization or death, evaluated with a Care Assessment Need (CAN) score (range, 0-99, with a higher score representing a greater risk of hospitalization or death), dichotomized as less than 80 or 80 or more. MAIN OUTCOMES AND MEASURES Preoperative LST note completion (30 days before or on the day of surgery) or no LST note completion within the 30-day preoperative period prior to or on the day of the index operation. RESULTS Of 190 040 veterans (90.8% men; mean [SD] age, 65.2 [11.9] years), 3.8% completed an LST note before surgery, and 96.2% did not complete an LST note. In the groups with and without LST note completion before surgery, most were aged between 65 and 84 years (62.1% vs 56.7%), male (94.3% vs 90.7%), and White (82.2% vs 78.3%). Compared with patients who completed an LST note before surgery, patients who did not complete an LST note before surgery tended to be female (9.3% vs 5.7%), Black (19.2% vs 15.7%), married (50.2% vs 46.5%), and in better health (Charlson Comorbidity Index score of 0, 25.9% vs 15.2%); to have a lower risk of hospitalization or death (CAN score <80, 98.3% vs 96.9%); or to undergo neurosurgical (9.8% vs 6.2%) or urologic surgical procedures (5.9% vs 2.0%). Over the 3-year interval, unadjusted rates of LST note completion before surgery increased from 0.1% to 9.6%. Covariate-adjusted estimates of LST note completion indicated that veterans at a relatively elevated risk of hospitalization or death (CAN score >= 80) had higher odds of completing an LST note before surgery (odds ratio [OR], 1.29; 95% CI, 1.09-1.53) compared with those with CAN scores less than 80. High-risk surgery was not associated with increased LST note completion before surgery (OR, 0.93; 95% CI, 0.86-1.01). Veterans who underwent cardiothoracic surgery had the highest likelihood of LST note completion before surgery (OR, 1.35; 95% CI, 1.24-1.47). CONCLUSIONS AND RELEVANCE Despite increasing LST note implementation, a minority of veterans completed an LST note preoperatively. Although doing so was more common among veterans with an elevated risk compared with those at lower risk, improving proactive communication and documentation of goals, particularly among higher-risk veterans, is needed. Doing somay promote goal-concordant surgical care and outcomes.