The American Journal of Surgery 216 (2018) 1127e1128

Contents lists available at ScienceDirect

The American Journal of Surgery

journal homepage: www.americanjournalofsurgery.com

Disparities in peptic ulcer disease: A nationwide study

Joaquim M. Havens*, Manuel Castillo-Angeles, Stephanie L. Nitzschke, Ali Salim
Division of Trauma, Burns, and Surgical Critical Care, Brigham and Women's Hospital, Boston, MA, USA

a r t i c l e i n f o a b s t r a c t

Article history: Background: While advances in diagnosis and treatment of peptic ulcer disease have led to a decrease in
Received 17 May 2018 hospital admissions the socioeconomic distribution of these benefits is unknown.
Received in revised form Methods: We designed a retrospective cohort study using the National Inpatient Sample from 2012 to
29 August 2018
2013 including all patients that were admitted for peptic ulcer disease. We compared the types of ulcer
Accepted 30 August 2018
related complications, the rates of intervention and the outcomes based on race and insurance status.
Meeting Presentation: This study was pre- Results: Of 42,046 patients admitted for peptic ulcer disease 80.25% had an ulcer related complication.
sented at the 2018 Pacific Coast Surgical Black patients had the lowest rates of bleeding and highest rates of perforation and were less likely to
Association Annual Meeting; February 20, undergo surgery for their complication but mortality was not different from white patients. Uninsured
2018; Napa, California. patients also had lower rates of bleeding and higher rates of perforation and they were at increased risk
for death.
Keywords: Conclusions: Unlike other surgical conditions insurance status, not race, predicts mortality in peptic ulcer
Peptic ulcer disease disease.
Disparities
© 2018 Elsevier Inc. All rights reserved.
Race
Current word count: 600

Hospitalizations for peptic ulcer disease (PUD) have decreased
over the past 20 years due to significant advances in diagnosis and
treatment.1 However, PUD requiring intervention still represents a
critical source of morbidity and mortality.2e4 Healthcare in-
equalities have been previously described in this population,
particularly patients of lower socioeconomic standing having
increased rates of complicated PUD.5,6 Nevertheless, a nationwide
study assessing the role of race and insurance status in complica-
tion rates and treatment choice has not been performed in this
setting. We sought to evaluate for disparities in the management
and outcomes for PUD in the United States.
We queried the Nationwide Inpatient Sample (NIS, calendar
years 2012e2013) to identify all inpatient admissions with a pri-
mary diagnosis of PUD. The NIS is maintained by the Healthcare
Cost and Utilization Project (HCUP) and captures a 20% stratified
sample of discharges from all participating hospitals. Cases were
identified using International Classification of Diseases, 9th Revi-
sion, Clinical Modification (ICD-9-CM) principal/admitting diag-
nostic (531e533) and procedure codes (44.40e44.42, 44.43, 44.44).
PUD admissions were classified by ulcer site: gastric, duodenal and
* Corresponding author. Division of Trauma, Burns, and Surgical Critical Care,
Brigham and Women's Hospital, Harvard Medical School, 75 Francis St, Boston, MA,
02115, USA.
E-mail address: jhavens@bwh.harvard.edu (J.M. Havens).
unspecified; by complication: hemorrhage and perforation; and by
procedure: esophagogastroduodenoscopy (EGD), surgery and
catheter embolization. Patient-level characteristics including age,
sex, race, payment source, median household income, weekend
admission, and disease severity were obtained. Charlson comor-
bidity index (CCI) was determined from ICD-9-CM codes using the
ICDPIC module in Stata version 13.0 (StataCorp, Inc.). Hospital-level
characteristics including geographic region, teaching status, and
hospital bed size were also collected.
The primary endpoint was in-hospital mortality. The secondary
endpoint was undergoing any of the defined procedures. Multi-
variable logistic regression models were used to determine if race
and insurance status were independent predictors of undergoing a
procedure and of mortality. A P value < 0.05 was considered sta-
tistically significant. This study was approved by the Partners
Healthcare Institutional Review Board with a waiver of consent.
There were 42,046 patients identified representing a weighted
total of 210,000 PUD admissions during the study period. Mean age
was 64.8 years (SD 17.2) and 48% were female. Patients were pre-
dominantly white (68%). An ulcer related complication occurred in
80.25%, and hemorrhage was the most frequent complication
(86.67%). A surgical or endoscopic procedure was performed in
75.65% of patients. The overall mortality was 2.6%. Perforation had
the highest mortality (5.96%). Complications differed by race and
insurance status. Black patients had the lowest frequency of

https://doi.org/10.1016/j.amjsurg.2018.08.025
0002-9610/© 2018 Elsevier Inc. All rights reserved.
1128 J.M. Havens et al. / The American Journal of Surgery 216 (2018) 1127e1128

Table 1
Proportion of complications by race.

Type of complication Race Total

White Black Hispanic Other

Hemorrhage, n (%) 24,952 (86.63) 4103 (84.95) 2644 (88.31) 2779 (88.31) 34,478 (86.70)
Perforation, n (%) 3433 (11.92) 673 (13.93) 311 (10.41) 333 (10.58) 4750 (11.94)
Both, n (%) 418 (1.45) 54 (1.12) 32 (1.07) 35 (1.11) 539 (1.36)
Total, n (%) 28,803 (100) 4830 (100) 2987 (100) 3147 (100) 39,767 (100)

Table 2 data analysis.

Multivariable analysis of the association between race, insurance status and
outcomes.
Outcome Race
White Black Hispanic
Procedure Odds ratio (95% CI) Ref. 0.90 (0.84e0.97) 0.59 (0.04e0.631)
Mortality Odds ratio (95% CI) Ref. 0.86 (0.70e1.06) 1.04 (0.83e1.33)
CI¼ Confidence Interval: Procedure ¼ surgical or endoscopic intervention.
hemorrhage, but the highest frequency of perforation when
compared with white and Hispanic patients (Table 1). Uninsured
patients had lower frequency of hemorrhage (76% vs. 85%), but
higher frequency of perforation (22% vs. 13%) when compared with
privately insured patients (p < 0.01). Race was highly predictive of
undergoing a procedure but not of mortality (Table 2). Being
uninsured (OR 1.52, (1.02e2.26)) was an independent predictor of
in-hospital mortality in patients with bleeding but not perforation.
This nationwide study shows that PUD represents a group of low
frequency but highly morbid emergency procedures. Consistent
with prior literature, we found racial differences in ulcer related
complication rates.5,6 Although we identified disparities in rates of
intervention, these disparities did not result in mortality differ-
ences. Unlike other surgical conditions insurance status, not race,
predicts mortality in PUD. We acknowledge this study has limita-
tions including retrospective design and the use of an administra-
tive database, which includes issues with coding accuracy and
makes it difficult to demonstrate a cause-effect relationship. In the
absence of clinical data, it is difficult to determine the factors that
influence the decision to perform an intervention, however, this
raises the question of overtreatment in the white population.
Considering these results future studies should address the balance
between intervention and outcomes in the treatment of PUD.
Author contributions
Dr. Havens had full access to the data in the study and takes
responsibility for the integrity of the data and the accuracy of the
Study concept and design: Havens, Castillo-Angeles, Nitzschke,
Salim.
Acquisition, analysis, or interpretation of data: Havens, Castillo-
Angeles, Salim.
Drafting of the manuscript: Havens, Castillo-Angeles.
Critical revision of the manuscript for important intellectual con-
tent: All authors.
Obtained funding: None.
Statistical analysis: Havens, Castillo-Angeles.
Administrative, technical, or material support: Havens, Salim.
Study supervision: Havens, Salim.
Conflict of interest disclosures
None.
Funding/support
None.
References
1. Wang YR, Richter JE, Dempsey DT. Trends and outcomes of hospitalizations for
peptic ulcer disease in the United States, 1993 to 2006. Ann Surg. 2010;251(1):
51e58.
2. Hasadia R, Kopelman Y, Olsha O, Alfici R, Ashkenazi I. Short- and long-term
outcomes of surgical management of peptic ulcer complications in the era of
proton pump inhibitors. Eur J Trauma Emerg Surg; 2018. https://doi.org/10.1007/
s00068-017-0898-z [Epub ahead of print].
3. Sarosi Jr GA, Jaiswal KR, Nwariaku FE, Asolati M, Fleming JB, Anthony T. Surgical
therapy of peptic ulcers in the 21st century: more common than you think. Am J
Surg. 2005;190(5):775e779.
4. Scott JW, Olufajo OA, Brat GA, et al. Use of national burden to define operative
emergency general surgery. JAMA Surg. 2016;151(6), e160480.
5. Crooks CJ, West J, Card TR. Upper gastrointestinal haemorrhage and deprivation:
a nationwide cohort study of health inequality in hospital admissions. Gut.
2012;61(4):514e520.
6. Smith JW, Mathis T, Benns MV, Franklin GA, Harbrecht BG, Larson G. Socioeco-
nomic disparities in the operative management of peptic ulcer disease. Surgery.
2013;154(4):672e678. discussion 678-679.