Schools Are Better than Hospitals for Improving Health

Education, social support and early childhood development play a more important role in determining the overall health of a country’s population than the condition of their hospitals and other health services, according to researchers who look at the social determinants of health.

Oh what a difference a choice of graphic makes!

If you think about it for just five minutes, you know that it’s not just doctors, nurses, hospitals and pills that determine how healthy we are. Those critical health services are part of an entire constellation that includes our genetic makeup, personal choices, the physical environment we live in (to overstate the obvious, it’s hard to stay healthy in a war zone), our social support, educational level and socioeconomic status.

Here is the way that Canadian public health officials visualize all these factors on an official government website—as spokes coming off a central hub.




The hub-and-spoke diagram makes it seem as if all these factors are playing equal roles in determining how healthy a given population is.

But what if you ranked those factors in order of importance?

That’s what Michael Hayes of Simon Fraser University in Vancouver, Canada and Stephen Bezruchka of the University of Washington did in a panel discussion at the annual meeting of the Association of Health Care Journalists in Seattle on Saturday, April 18, 2009. (Search twitter for #ahcj09 to get some of the raw tweets published during the conference.)

Hayes and Bezruchka placed health services (the doctors-nurses-hospitals-clinics-medications part of the equation) at the absolute bottom of the list in importance. At the top of the list are social determinants, like education (particularly of mothers) and socioeconomic standing.

The whole list goes something like this, in order of importance:

Most Important
Social determinants of health (education, socioeconomic status, social support, etc)
Early childhood
Personal behavior
Physical Environment
Biology
Health Care Services
Least Important

This ordering of the list, which is based on 25 years worth of data, Hayes and Bezruchka said, leads to some startling conclusions. First off, we in the press focus far too much attention on health services (the docs, nurses, new treatments and medications) as opposed to education, crime rates, poverty when talking about health. Second, dollar for dollar, you would probably actually get a better health benefit investing disproportionately more in schools than in hospitals. And perhaps, third, Canadian public health officials need a new graphics team.

To develop that last point a little further. Don’t use circles when describing the factors that lead to health disparities. It makes everything seem equal. Use ranked lists that tell you what the most important factors are—provided, as always, you have the data to back it up.

What is still unclear to me is just how widespread the consensus is that social determinants of health are more important than health services in determining the overall health of any country’s citizens. (Chime in on comments below if you can enlighten us.)

I asked Hayes about consensus in a quick follow-up at the panel table prior to the start of the next session. He says there is consensus that there are social gradients to health and even what the major influences are but not on how they play out. One camp favors the psychosocial view: people who are optimists tend to be healthier and live longer than pessimists. The other camp is more materialist: poverty is key.

The point, Hayes said, is to link biology to social being. Then he gave a quick example. We tend to focus on the fact that smoking cigarettes is clearly a matter of personal choice, he said. But we ignore the role that advertising plays.

In my mind, I was thinking well, we haven’t really ignored advertising—I mean smoking was banned from television ads. But I guess there are still plenty of billboards, sports sponsorships, etc.

Earlier in the session, either Hayes or Bezruchka threw out the statement that the relative risk of dying from smoking in Japan is much lower than in the USA. And then strongly implied that social determinants of health (a better social safety net in Japan, more family support, etc) explained the difference.

It was a memorable line but begs to be checked out further. What are the absolute risks in both cases? (Even though I knew about the importance of absolute risk before I knew journalism professor Gary Schwitzer, I can hardly think of one without the other anymore.) What ways do smokers compare/differ in the US and Japan? (Seems like you would have to compare Japanese-American smokers to Japanese national smokers to be truly thorough.)

Later, I was reminded of a conversation I had had with a woman, who was a physician, who had developed tuberculosis. She would not have survived, she said, had it not been for her own mother and sisters, who kept her on track. They made sure she took her medications over the course of many months, even after the pills made her violently sick.

She knew intellectually that she had to continue her treatment but that was not enough to keep her alive. She needed the pills but that was not enough to keep her alive. Her knowledge, the pills and her family’s willingness to make sure she took them every day—even when she begged them not to—is what allowed her to survive.

For further reading on social determinants of health:
"Toward a Healthy Future"
The 1999 Canadian report that studied and quantified social determinants of behavior.

Bezruchka's webpage of resources on social factors in health inequalities (hat tip: @HoltzReport)

"Spreading the News: Social Determinants of Health Reportage in Canadian Daily Newspapers"
Hayes' own research into how much Canadian journalists are aware of the research on social determinants of health.

NB: This post is based on my writeup of my notes from AHCJ09.

Panel: "The healthy environment: it's not just medicine"
Panelist: Michael Hayes, Ph.D., associate dean, Faculty of Health Sciences, Simon Fraser University
Panelist: Stephen A. Bezruchka, M.D., senior lecturer, Department of Global Health, University of Washington
Moderator: Andrew Holtz, M.P.H., independent journalist, Portland, Ore.

These notes are as accurate as I can make them but if you use any of this for a published piece, you should be sure to do your own reporting to double check it--as I will if I write it up for formal publication.

I'll post other notes from the AHCJ conference as I write them up.

Update (4/21/09): Here are a few more thoughts on how to think about the non-medical factors related to health and how they relate to health services and each other.