I’ve just spent five days—yes, five days—talking about health literacy. Before my five day conversation I’d never thought much about health literacy, but now I see myself as an expert. Pick a small enough subject and you can be a world expert in about 20 minutes. But health literacy is actually a big subject and increasing it could potentially make a huge difference to beleaguered patients and health care systems. Our marathon conversation took place in Frankfurt and included lots of grand and smart people, all of them grander and smarter than me. As in many conversations, we spent a lot of time talking across each other because we had different ideas of what we were talking about.

The Institute of Medicine has defined health literacy as “The degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.”

The “individuals” are patients and citizens, but in our conversation we talked as well about health care professionals, including doctors. If we think of health literacy as knowing the treatment for atrial fibrillation or whether to go to the doctor with a sore throat, then doctors are a lot more literate than patients.

If, however, we consider, as many do, health literacy to be the ability to derive and understand the statistical risk of various treatments versus no treatment of a condition from randomized trials and systematic reviews, then doctors are not much more literate than patients. Indeed, when it comes to sorting out sensitivity, specificity, positive predictive value, and the like then almost everybody is illiterate apart from those trained in epidemiology and statistics and using the concepts regularly.

Another source of confusion was whether to think about health literacy as simply a function of individuals. The Institute of Medicine’s definition suggests that it is, but a few lines after offering the definition the institute’s report recognises that it may be too narrow and says that literacy is “a shared function of cultural, social, and individual factors.” It is, in other words, a property of a system not just individuals.

So if we want to raise health literacy should we concentrate on individuals or systems? The two approaches are not mutually exclusive, but when resources are short, as they always are, what should be our priority? Some would emphasise teaching statistics and uncertainty to the school children, especially as statistical literacy can be useful not just in working out whether to have a mammogram but also whether to bet on Ben’s Red Nag in the 4.40 at Wincanton.

But others, including me, were skeptical that much would be gained unless we also thought about systems. We discussed, for example, the strong evidence from Jack Wennberg and others in the US that the amount of health care that people receive is strongly determined by the supply of health care institutions—what is called “supplier induced demand.” For example, people in Los Angeles receive twice as much health care as people in Minneapolis—simply because there are twice as many health care providers in Los Angeles. Far from producing benefit this leads to poorer outcomes among the people in Los Angeles, which may be explained by them being more likely to suffer from medical errors because they have more contact with the health system.

Improving health literacy among the people of Los Angeles seems unlikely to reduce the amount of care they receive. Yet providing this compelling data to those who organize and find health care could at least potentially reduce health care supply in Los Angeles and encourage more care based on evidence. So where did we get to after five days? We did agree that standard ways of presenting information on medical risk in journals, the media, patient information leaflets, and, indeed, everywhere would be a good step to raising literacy, and our recommended form will be published. But we never quite agreed on whether to concentrate on individuals or systems. Maybe we needed to talk for another five days.

Competing interest: Richard Smith is on the board of the Public Library of Science and has been an enthusiast for open access publishing for 15 years.