Physician-level variability in opportunistic salpingectomy uptake at the time of hysterectomy for ovarian cancer prevention.

38 Background: As there is no effective screening or characteristic symptoms, > 70% of ovarian cancers are diagnosed at advanced stage with poor prognosis, highlighting the importance of prevention. The American College of Obstetricians and Gynecologists and the Society of Gynecologic Oncology have both endorsed new evidence that most ovarian cancers originate from the fallopian tubes, rather than from the ovaries. This makes opportunistic salpingectomy (OS) (prophylactic removal of both fallopian tubes while conserving the ovaries) a novel method for ovarian cancer prevention. This study aimed to examine physician-level variability in their trajectory of adopting OS in hysterectomy practice. Methods: Using the Premier Healthcare Database, we identified 33,401 patients aged 18-49 years who underwent an inpatient hysterectomy in 2011-2021 (operated on by 1,297 physicians). We studied premenopausal patients as this is the group who would benefit the most from ovarian preservation. OS was identified using procedure codes. We focused on inpatient hysterectomy because only ICD procedure codes (routinely available in inpatient but not outpatient billing data) can definitively identify OS. Patients with any cancer or elevated risk for cancer were excluded. We estimated a risk-adjusted rate of OS for each physician-year using a logistic regression (adjusting for patients’ age, surgical indication/route, comorbidities, and year) and applied a finite mixture model to identify latent groups of physicians who exhibited distinct trajectories of adopting OS. Results: In the overall sample, OS use among inpatient hysterectomy increased from 6.2% in 2011 to 74.5% in 2021. Physicians varied widely in OS use (e.g., 10th-to-90th percentile range of risk-adjusted OS rate in 2021: 19.7-97.9%). We identified two latent groups of physicians: one exhibiting a low-adoption trajectory (risk-adjusted OS rate increased from 0.4% in 2011 to 49.1% in 2021) and the other following a high-adoption trajectory (13.9% in 2011 to 85.3% in 2021). They accounted for 39.4% and 60.6% of the physicians, respectively. Physicians at teaching hospitals (odds ratio [OR] = 1.60, 95% confidence interval [CI]: 1.12-2.28) or in the Northeast (OR = 3.97, 95% CI: 2.40-6.55) or West (OR = 2.31, 95% CI: 1.53-3.46) were more likely, whereas physicians with more Medicaid patients were less likely (OR = 0.90 for each 10-percentage-point increase in the proportion of patients with Medicaid insurance, 95% CI: 0.83-0.97), to follow a high-adoption trajectory. Conclusions: Despite an overall rapid increase in OS use, physicians varied in their trajectory of adopting OS at the time of hysterectomy. Differences in OS uptake by non-clinical factors suggest potential sociodemographic and geographical inequalities in OS use.