If you paid much attention to the health-care debate in Washington over the past few years, you may recall this zinger: The United States – often reputed by its own politicians to have the best health care system in the world – in fact was ranked just 37th among 191 countries by the World Health Organization, sitting awkwardly between Costa Rica and Slovenia.
We’re used to comparing our health records at the country level like this, looking at how we stack up against other nations on life expectancy or infant mortality or AIDS rates. And somehow it makes sense that these metrics would vary across national boundaries. We've all got different health care systems, different public health policies, even different genes.
When it comes to meaningful variation in the spread of specific illnesses, though, turns out cities, not countries, are where it's at.
This picture emerged earlier this week with the publication of the largest study ever undertaken in Europe comparing the health not of different countries, but of individual cities across the European Union. The project has published online health profiles of 26 cities, containing some particularly mysterious variations.
The mortality rate among women in Amsterdam from diseases of the respiratory system is substantially higher than the average in the 25 other cities. Cardiff, in the United Kingdom, has a relatively high proportion of overweight adults; people in Manchester suffer at a higher rate from depression and anxiety; while Maribor, Solvenia, contends with heavy episodic drinking among its young people (its adults, on the other hand, binge drink less often than the study-wide average).
"National-level data is absolutely vital," says James Higgerson, one of the researchers on the project from the University of Manchester. "But when you do data at the national level, the differences within a country can be lost. As we’ve been able to demonstrate, even within countries, there were differences in terms of rates of cancer, rates of obesity. And policies aren’t just made at the national level."
This raises a lot of questions about how decisions made at the local level – around smoking ordinances, bike lanes, green space, traffic management, drinking age, land use, air quality – might impact even cancer rates.
"High levels of lung cancer, even asthma, you can look at it and you can think, 'what is it that’s causing this?'" Higgerson says. "The work we’ve done can’t provide definitive answers, but it can provide information to start looking into it."
The study included pre-existing data on measures like age, sex, education level, the cause of death for certain diseases, health insurance access and vaccination prevalence in each location. But the researchers also conducted an extensive mail survey of 1,600 adults in each city (400 men and 400 women aged 18-65, and the same over 65), as well as in-person surveys of a similar number of 14-to-16 year-olds. A fourth piece of the project – interviews with policymakers in each city – is still underway.
The results suggest that cities have a kind of unique health fingerprint in much the same way countries – or even individuals – do. And it’s a little overwhelming to begin to think about why this might be. Higgerson focused his research in particular on the indicators around alcohol use and abuse. And he suggests that rates of use may vary according to details as locally specific as the presence of dense bar strips (which also encourage cheaper alcohol prices).
Each of the 45 health indicators included in the study comes attached to some implied policy responses, although it won’t be easy to figure out what they are, and where and when they’re needed.
"Perhaps this is just where epidemiology is getting to now," Higgerson says of this more fine-grained assessment of metro-level health reports. "There’s always been this belief that there were differences between areas, but perhaps there had not been the resources or the organization to do this."