Physicians-in-training and advanced practice providers perceptions in managing perioperative obstructive sleep apnea: a multi-institutional survey.

Study objectives: Physicians-in-training (residents, fellows) and Advanced Practice Providers (APPs) receive limited education on sleep disorders, including obstructive sleep apnea (OSA). They often assess patients first. We aimed to understand their views on OSA and screening for OSA in the perioperative period.

Methods: Electronic-mail surveys were sent to physicians-in-training and APPs, in five categories of practice (Anesthesiology, Internal Medicine, Family Medicine, Obstetrics/Gynecology/Gynecologic Oncology, and Surgery) at nine major institutions. Cochran-Mantel-Haenszel (CMH) tests evaluated associations between participant characteristics (clinical role, physician years of training, APP years of practice, categories of practice) and survey responses (perception of OSA, perioperative risk factor, screening/managing of perioperative OSA) stratified by institution. False discovery rate (FDR) adjustment accounted for multiple comparisons (FDR-adjusted-p-values) of associations between multiple characteristics and a given response. Breslow-Day tests evaluated the homogeneity of odds ratios from CMH tests.

Results: We received 2236/6724 (33.3%) responses. Almost all (97%) agreed OSA represents a risk factor for perioperative complications. Many (37.9%) were unaware which screening tool was used at their institution, with differences by clinical role (FDR-adjusted-p<0.001), with APPs reporting not knowing more than residents and fellows, and across category of practice (FDR-adjusted-p<0.001). While 66.5% routinely asked perioperative patients about signs/symptoms of OSA, 33.5% did not. There were differences by clinical role (FDR-adjusted-p<0.001), as APPs reported asking about OSA more frequently than residents/fellows; and, by category of practice (FDR-adjusted-p<0.001) as Anesthesia and medical specialties reported asking about OSA more than surgical services. Importantly, approximately half of the respondents in Surgery (48%) and OB/GYN (46%) reported not routinely asking. Differences also existed by physician PGY clinical training (FDR-adjusted-p=0.005) with those with higher PGY reporting they asked about OSA more often.

Conclusions: Significant differences exist in screening by clinical roles and categories of care. This highlights the importance of improving provider education on the role of OSA in perioperative risk assessment and patient care.