S1642 Factors Associated With Failure to Complete Helicobacter pylori Eradication Testing in an Urban Tertiary Care Center

Abstract

Introduction: Global trends in antibiotic resistance have limited progress in Helicobacter pylori eradication rates. Current data reflecting local antibiotic resistance and cure rates obtained from eradication tests are critical, but these tests may be inconsistently completed. As the first phase of a quality improvement initiative, we studied 481 patients treated for H. pylori at our institution to identify barriers to eradication test completion. Methods: Patients > 18 years and treated for H. pylori during the 25-month study period were identified via ICD-10 code query. Demographic data were collected along with data pertinent to the H. pylori diagnosis communication, provider characteristics, antibiotic prescription, and eradication test. Chi-squared testing was used to compare eradication test completion and cure rates between different groups. Multivariable logistic regression models were used to identify factors associated with incompletion of eradication tests and eradication failure. Results: Overall, 301 patients (63%) completed eradication tests (minimum follow-up 8 months since antibiotic prescription); 71% of these patients had confirmed cure; 352 patients reported treatment completion; 83% of these (n= 292) completed eradication testing (Table). Univariate analysis identified multiple factors associated with failure to complete eradication tests, including male sex, African-American race, and letters to communicate the diagnosis. Multivariate analysis showed that communication via letter was associated with lower odds of eradication test completion compared to telephone call (OR = 0.29, CI: 0.10 - 0.84). Scheduled treatment encounters (i.e. in-office or telehealth encounters) for antibiotic prescription were also associated with lower odds of eradication test completion (OR = 0.20, CI: 0.06 - 0.68). Patients who had eradication testing ordered on the same day as antibiotic prescription were not more likely to complete eradication testing overall, but were more likely to complete it early (21.4% vs 9.2%, P = 0.004). Conclusion: Our patients and prescribers may benefit from a standardized protocol to guide diagnosis communication, follow-up, and eradication testing which will be evaluated in the second phase of this quality improvement project. The protocol will include i) 2 methods of communicating the H. pylori diagnosis to each patient and deferral of ii) follow-up visit and iii) eradication testing order until after antibiotic completion. Table 1. - Adjusted Odds Ratios for Eradication Test Completion Exposure Adjusted OR 95% CI P-value Age 1.009 (0.978 - 1.041) 0.575 BMI 0.984 (0.935 - 1.035) 0.529 Sex Male 0.616 (0.278 - 1.361) 0.231 Race/ethnicity White vs African-American 0.357 (0.105 - 1.213) 0.099 White vs Hispanic 0.794 (0.215 - 2.928) 0.729 White vs Asian/Other 42.890 (2.203 - 835.148)* 0.013 Primary language English 0.740 (0.239 - 2.287) 0.600 H. pylori history Prior treatment 1.284 (0.465 - 3.551) 0.630 Diagnosis communication Telephone vs letter 0.288 (0.099 - 0.836)* 0.022 Telephone vs MyChart 0.316 (0.067 - 1.483) 0.144 Treatment encounter None vs in-office/telehealth 0.203 (0.060 - 0.684)* 0.010 Treatment regimen Bismuth vs clarithromycin 1.061 (0.411 - 2.736) 0.903 Bismuth vs other 1.375 (0.463 - 4.089) 0.567 Prescriber department Gastroenterology 1.538 (0.452 - 5.226) 0.491 Prescriber level Physician 1.047 (0.294 - 3.730) 0.943 Prescribed by endoscopist 0.683 (0.275 - 1.699) 0.412 Test and abx ordered same day 0.990 (0.429 - 2.287) 0.981 Abx nonadherence 0.930 (0.251 - 3.448) 0.913