Burping, bloating, rumblings, and tummy pains. Patients with dyspepsia have crowded my GP days of late. I have turned my computer screen around for patients to deliver my familiar online show and tell of NICE guidelines, patient information leaflets, and google images of where your gullet, stomach, and guts are. Then I’ve printed off their blood test forms, and forms for the h.pylori stool test—no one is keen on collecting that in a pot.

A run of one symptom provides fertile hunting ground for some reading, and of course a dreaded e-portfolio entry. By chance a pair of articles on bmj.com this week could be just the thing: an uncertainties page on who we should “test and treat” for h.pylori, and a research paper by the similar authors.

The research paper is a systematic review and meta-analysis of randomised controlled trials, and asks whether searching and treating h.pylori infection in asymptomatic people reduces the incidence of gastric cancer? They found it might (relative risk 0.66, 95% confidence interval 0.46 to 0.95) if the benefit of eradication therapy was maintained. But the data are drawn from Asian populations and might not be generalisable to others. If they were, the number needed to treat might range from 15 for Chinese men to 245 for American women.

While a research paper is handy for a journal club, a bigger picture is painted in the uncertainties article in the education section of The BMJ. How good is “test and treat” for different types of dyspepsia such as ulceration, GORD, uninvestigated symptoms, functional dyspepsia, and the wider population?

Here are some useful factoids that I learned. H.pylori is the culprit in approximately 1 in 20 people with uninvestigated dyspepsia. The number needed to treat for epigastric pain and positive h.pylori is 13. A significant (but unquantifiable) concern is the role that h.pylori eradication might play in antibiotic resistance.

So, what will I do differently in future? Sadly, this e-portfolio box remains a little bald. Because it seems that despite my reading I’ll be doing and saying the same thing, although perhaps with a little more conviction.

Helen Macdonald is the analysis editor, The BMJ.