THE CHALLENGE OF BENIGN INCIDENTALOMAS: IDENTIFYING AND MAPPING THE NATURAL COURSE OF HIATAL HERNIAS

Background: To address the persistent ethnoracial and socioeconomic disparities in access to quality surgical cancer care, it is imperative to rigorously understand the role of clinician-level factors including clinician-to-clinician connectedness. For patients with gastric cancer, the pathway from primary care (PC) clinicians to gastroenterologists (GI) to cancer specialists (medical oncologists or surgeons) is referral-dependent and requires significant care coordination. However, the impact of clinician-to-clinician connectedness on access to quality gastric cancer surgical care, such as at National Cancer Institute-Designated Cancer Centers (NCI-CC), remains underexplored. This study evaluates how the connectedness between PC clinician or GI with cancer specialists at NCI-CC can influence receipt of gastrectomy for gastric cancer at NCI-CC.

Methods: Maryland’s All-Payer Claims Database was used to evaluate 667 patients who underwent gastrectomy for cancer from 2013-2018. Clinician-to-clinician connectedness was measured via referral linkages between clinicians. Two separate referral linkages, defined as ≥9 shared patients between two clinicians, were examined from: 1) PC clinicians to GI at NCI-CC and 2) GI to cancer specialists at NCI-CC. Multiple logistic regression models were used to determine associations between referral linkages and adjusted odds of undergoing gastrectomy at NCI-CC.

Results: Only 15% of gastric cancer surgeries were performed at NCI-CC. Patients treated by GI with stronger referral links to cancer specialists at NCI-CC were more likely to be <65 years of age, male, white, and privately insured (for all, p< 0.05). Every additional referral link between PC clinician and GI at NCI-CC and between GI and cancer specialists at NCI-CC increased the odds of receiving gastric cancer surgery at NCI-CC by 71% and 26%, respectively (Table). Black patients had half the odds as white patients in receiving gastrectomy at NCI-CC (OR: 0.53, CI:[0.30, 0.93]). However, adjusting for covariates including clinician-to-clinician connectedness weakened the observed negative effects of black race on receipt of gastrectomy at NCI-CC (OR: 0.63, CI:[0.10, 3.83]).

Conclusion: Patients of clinicians with low clinician-to-clinician connectedness and black patients are less likely to receive gastrectomy at NCI-CC. Clinician connectedness appears to be an actionable area of intervention to overcome existing disparities in access to quality surgical cancer care. These results are relevant to policy makers, healthcare systems, clinicians, and patient advocates seeking to achieve equitable access to quality cancer care.

Objective
Computed tomography (CT) imaging routinely detects incidental findings, including hiatal hernias. We utilized a natural language processing algorithm to identify incidental hiatal hernias, characterize their natural progression, and evaluate clinical follow-up.

Methods
Imaging of adult trauma patients from 2010-2020 who underwent CT chest and/or abdomen/pelvis was evaluated using an open-source natural language processor query for hiatal hernias. Patients who underwent subsequent imaging, endoscopy, fluoroscopy or operation were retrospectively reviewed.

Results
1,087 of 10,299 patients (10.6%) had incidental hiatal hernias: 812 small (74.7%) and 275 moderate/large (25.3%). 224 patients (20.7%) had subsequent imaging or endoscopic evaluation, with a mean follow up of 2.78 ± 2.79 years. Patients with moderate/large hernias were older (small vs moderate/large: 66.3±19.4 vs 79.6±12.6 years,p<0.001) and predominantly female (small vs moderate/large: 403[49.6%] vs 199[72.4%],p<0.001). Hernia size was not associated with hernia growth (small vs moderate/large: 13[7.6%] vs 8[15.1%],p=0.102) or symptomology (small vs moderate/large 55.6% vs 67.9%,p=0.110) (Table.1). Though patients with moderate/large hernia were more likely to have an intervention/referral (small vs moderate/large 6[3.5%] vs 7[13.2%], p=0.008), more than 80% of symptomatic patients (n=28) with moderate/large hernias had no follow-up. No patients underwent elective or emergent hernia repair. Three patients had surgical referral; however, only one was seen by a surgeon. One patient death was associated with a large hiatal hernia.

Conclusions
We demonstrate a novel utilization of an NLP to identify patients with incidental hiatal hernia in a large population, and found a 10.6% incidence with only 1.2% of these receiving a referral for follow-up. While most incidental hiatal hernias are small, moderate/large hernias have very high risk of loss-to-follow up and need referral pipelines to improve patient outcomes.