Back

Early Population-Level Impact of Helicobacter pylori Eradication on Gastric Cancer Deaths in Japan: A Counterfactual Analysis of Short-Term Divergence

Kowada, A.

2026-02-26 epidemiology
10.64898/2026.02.24.26346975 medRxiv
Show abstract

BackgroundHelicobacter pylori infection accounts for 98% of gastric cancer (GC) cases in Japan. Since 2013, the nationwide expansion of H. pylori eradication therapy to chronic gastritis patients has created a unique opportunity to evaluate its population-level impact on GC primary prevention. However, short-term reductions in GC deaths are difficult to interpret given the long natural history of gastric carcinogenesis. This study aimed to assess the early impact of population-level eradication on GC deaths. MethodsWe applied a two-layer analytic framework consisting of a counterfactual analysis comparing observed GC deaths during 2013-2021 with expected GC deaths had eradication uptake remained at pre-2013 levels. This was combined with a structured, time-dependent, multilayer state-transition model to estimate GC deaths prevented by eradication using GC incidence integrated with age-dependent H. pylori prevalence. ResultsObserved GC deaths declined from 48,632 in 2013 to 41,624 in 2021, whereas counterfactual GC deaths declined more modestly, from 49,794 to 45,654. The divergence between observed and counterfactual GC deaths widened steadily from 1,162 in 2013 to 4,030 in 2021. Model-based estimates indicated that eradication prevented 1,427 GC deaths during 2013-2021, with annual GC deaths prevented increasing from 17 in 2015 to 417 in 2021, particularly among adults aged 50-79. ConclusionsThis study demonstrates that H. pylori eradication has already contributed to a 10.4% reduction in GC deaths in Japan by 2021, with annual expansion of primary prevention effects. This framework supports evidence-based evaluation of short-term reductions in GC deaths attributable to H. pylori eradication in high-prevalence settings.

Published in Population Health Metrics · not in our set (fewer than 10 published preprints to learn from) · training set

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.