The aim of the study was to evaluate whether poor compliance can be considered as the main cause of the low Helicobacter pylori ( H. pylori ) eradication rate observed in an ambulatory population.

Seventy-eight patients with non-ulcer dyspepsia or gastroduodenal ulcer in whom H. pylori infection was confirmed by urease Clo-test and histology or bacterial culture, received a 1-week triple therapy comprising lansoprazole 30 mg b.d., amoxicillin 1000 mg b.d. and clarithromycin 500 mg b.d. Compliance was assessed using MEMS^® containers (Medication Event Monitoring System) which recorded time of medicines consumption.

The overall H. pylori eradication rate was 65.4% (95% CI: 54.8-76.0%) (intention to treat). Sixty-nine subjects (88.5%) consumed greater than 85% of doses and were considered as “good compliers”. The major reason listed by the nine remaining patients for stopping treatment prematurely was side effects. In the population categorised as “good compliers”, H. pylori eradication rate was 69.6% (95% CI: 58.7-80.5%) (per protocol) indicating that compliance could not be considered as the sole reason for treatment failure. Bacterial culture in a subset of 30 patients further showed a H. pylori eradication rate of 73.9% (95% CI: 55.7-92.1%) in “good compliers” with a clarithromycin-sensitive H. pylori strain. On multivariate analysis, H. pylori eradication was inversely associated with poor compliance (P = 0.029). Presence of a gastroduodenal ulcer, age, gender and smoking habit did not differ significantly between the eradicated and noneradicated groups.

Although poor compliance and bacterial resistance were important factors in determining treatment success in our population, they could only explain 40% of failures suggesting that other factors must be involved.