Socioeconomic disparities in the operative management of peptic ulcer disease

Abstract

Over the last 60 years, there has been a nationwide decrease in the number of operations performed for peptic ulcer disease (PUD). In contrast, the experience at our university-based safety net hospital (SNH) was that ulcer operations are still performed frequently. We hypothesized that differences in frequency of PUD operation may occur in hospitals that serve different patient populations. The purpose of this study was to evaluate our experience with PUD and compare it with national trends. We received institutional review board approval and performed this retrospective study of patients undergoing operation for PUD between January 2008 and December 2011. Patient records at 2 hospitals (a private community hospital and a university SNH) with similar admission numbers and geographic catchment were examined for PUD risk factors, Helicobacter pylori status, insurance/income status, type of operation, and surgical outcomes. A case-matched control group of medically treated patients were identified after primary diagnosis of PUD by endoscopy at the SNH. Univariate and multivariate analyses were performed. The total number of operations for PUD performed at the SNH was greater than those performed at the private hospital from 2008 to 2011 (142 vs 24; P < .001). The private hospital followed national trends over the same time period with a decrease in operations for PUD of approximately 93% between 1967 and 2008 (115 to 8 operations per year nationally and 119 to 6 at the private hospital). In contrast with the national and local private hospital experience, the number of operations for PUD at SNH increased from 27 per year in 1985 to 36 per year in 2008. Additionally, 43% of patients at the SNH had no insurance, and 61% resided in the poorest quartile of zip codes compared with the 3% uninsured patient rate at the private hospital for a similar group of patients. At both hospitals, most operations were emergent (range, 83-92%) and treated with omental patch (45%), gastric wedge resection (15%), vagotomy and antrectomy (19%), or vagotomy and pyloroplasty (14%). At the SNH, the H pylori infection rate was less (48% vs 83%; P < .001) and nonsteroidal anti-inflammatory drug (NSAID) use was greater (76% vs 63%; P < .01) in the 142 surgical patients compared with the 320 medical controls. Adjusted risk ratios demonstrated insurance status, NSAID use, and lower socioeconomic class were all equally predictive of operative ulcer disease when compared with medical controls. Our study provides 2 observations. First, patients of lower socioeconomic standing may have increased rates of complicated PUD owing to multiple medical factors and other factors related to healthcare. Second, surgical care for PUD retains a clinically important role within this patient population.