The State of U.S. Health: Q&A with Christopher Murray, Institute for Health Metrics and Evaluation

Jul 15, 2013, 1:23 PM

 Christopher Murray, MD, DPHIL, Director of the Institute for Health Metrics and Evaluation at the University of Washington Christopher Murray, MD, DPHIL, Director of the Institute for Health Metrics and Evaluation at the University of Washington

The final plenary session at this year’s NACCHO Annual included a talk by Christopher Murray, MD, DPhil, director of the Institute for Health Metrics and Evaluation at the University of Washington on how data is used to measure health, evaluate interventions and find ways to maximize health system impact. Dr. Murray was a lead author on three pivotal studies published last week that used data to assess the state of health in the United States compared with 34 other countries and county level data on diet and exercise. One of the key findings is that Americans are living longer, but not necessarily better—half of healthy life years are now lost to disability instead of mortality; and dietary risks are the leading cause of U.S. disease burden.

NewPublicHealth spoke with Dr. Murray about the study findings, their impact and upcoming research that can add to the data public health needs to improve the health of all Americans.

NewPublicHealth: Tell us about the three studies that were published this week using the Institute’s research.  

Dr. Murray: The study in JAMA [Journal of the American Medical Association] is an analysis of a comprehensive look at the health of the United States in comparison to the 34 OECD [Organisation for Economic Co-operation and Development] countries. The study looks at both causes of death and premature mortality through over 290 different diseases and puts them all together in a comprehensive analysis of what the contributors are to lost healthy life. That study also looks at the contribution to patterns of health in the U.S., from major environmental, behavioral, and metabolic risk factors. In each of those categories, there are important findings:

  • The U.S. spends the most on healthcare but has pretty mediocre outcomes and ranks about 27th for life expectancy among its peer countries.
  • For many large, important causes of premature death, the U.S. does pretty poorly. And we also see a big shift towards more and more individuals having major disability—from mental disorders, substance abuse, and bone and joint disease.
  • On the risk factor front, the big surprise is that diet is the leading risk factor in the U.S. It is bigger than tobacco, which is second and then followed by obesity, high blood pressure, high blood sugar, and physical inactivity. Diet in this study is made up of 14 subcomponents, each analyzed separately and then put together.
Data visualization of global causes of disease and injury Data visualization of global causes of disease and injury

The two papers in the journal Population Health Metrics explore county findings. One study looked at the trends by county, year by year, from 1985 to 2010, and that study found:

  • The gap between the best and the worst counties is continuing to widen. There are a whole slew of counties with low life expectancies where there’s essentially been no change in a generation in life expectancy. And that is particularly true for women.
  • In about 40 percent of counties there’s no statistically significant improvement for women from 1985 to 2010. But at the top, there have been really huge improvements in life expectancy, so the net effect is the gap between best and worst just keeps widening every single year.
  • A very significant finding is that even in some quite poor counties, we’ve seen very substantial improvements in life expectancy and I think the reason that’s important is it suggests that it is possible to overcome the barriers that are obviously related to socioeconomic status and health and see improvements in some disadvantaged places.

NPH: And what did the second study published in Population Health Metrics find?

Dr. Murray: That study is a county-level assessment of obesity and physical activity from 2001 to 2011, and there, the good news on physical activity is we’ve seen a number of communities, most notably in Kentucky, Georgia, and Florida, with very large improvements in the number of individuals getting 150 minutes per week of moderate activity or 75 minutes of vigorous activity. In fact, the biggest improvements are 18-percentage point improvements, so that’s pretty interesting. The national trend isn’t very much, maybe a 5 percentage point improvement, but there are communities with spectacular improvements.

And, the other part of that study is about obesity. We don’t see any communities where obesity is decreasing but there are some communities where there’s been essentially no increase. On the flip side, in general, obesity has gone up and the worst increases are 16-percentage point ones.

NPH: Did the studies show an impact of the increase in physical exercise on obesity?

Dr. Murray: Yes, for every percentage point improvement of the population achieving 150 minutes of exercise per week, there’s a 0.1 percentage point lower obesity rate. The impact is there, but it’s small. I think the way to think about that is it could be a time lag or it could be that the energy and balance that’s been driving the obesity epidemic is big enough that even with the improvements in physical activity at the population level, it’s not enough to counteract the caloric intake. Or there’s still a net energy imbalance, and so obesity is still rising.

NPH: Is it possible that the levels of obesity are such that the current recommendations on physical activity are insufficient? Do we need more than 150 minutes a week?

Dr. Murray: Those are the questions we should be asking because I think the 150 minutes is driven by the epidemiological evidence about the benefits of physical activity in their own right. So all these places where we see these big improvements should be getting a very distinct health benefit from physical activity itself, independent of its effect on obesity. But if you add up the caloric value of 150 minutes of exercise weekly, it isn’t that much. And so, if we want to think of physical activity as a mechanism for reducing obesity, yes, we might need to think of much more, and of course, about other strategies that reduce energy intake.

NPH: What should a county that hasn’t had success yet be learning from a county that has?

Dr. Murray: That’s what we’re hoping this sort of data will trigger because I think the next step is to say what happened in Kentucky on physical activity? What happened in some of these communities in Georgia? What happened in Beaufort, South Carolina with big improvements in life expectancy while many of the neighbors had no improvement?

Part of the challenge there is there’s no obvious central location where people record county by county all the different policy initiatives and programs that are implemented. And, that’s a real gap in being able to make the how-we-did-it type story that’s so critical here, which is to figure out both from success and less successful places what’s different about them. So, I think that’s the next wave of work for the public health community. Even if we don’t ever have a registry of all the policy efforts that are made by different public health authorities and physical activity programs at least you can do some peer-wise analysis and say let’s take a successful county and one that looks pretty similar, perhaps based on socioeconomic status, that didn’t seen progress and see if we can do some compared analysis to help share strategies that work.

NPH: What are other areas that your research center in particular is looking at in terms of greater life expectancies? What’s the next data set?

Dr. Murry: We want to continue documenting some of the other key outcomes that are measurable, like tobacco use and access to insurance. The other one is a bigger task, but I think is one that would be enormously powerful and that is to try to measure how much is spent on public health programs compared with medical care across each county because if you had that, you could then start to test the question: if we spent more money on public health, would there be a causal link?

And, if you start to have detailed outcomes at the county level of, say, diabetes, death rates or disease burdens, can we look at how much is spent on diabetes, and is there a relationship? Can we demonstrate just how inefficient spending on medical care for some conditions may be? Or, the contrary, that there are communities that do a great job of giving access to care and serving their needs but at a pretty low cost. So, I think the next big agenda there is to pair measurement of outcomes that matters over time to where the money goes. We’re flying blind on this very expensive medical care system, and if you could actually show where resources have a big effect, I think that would be very powerful.

People should have that information at the local level so they can prioritize their biggest health problems and eventually use this as a template for evaluation of what works and what doesn’t by linking where the money and the policies are to the outcomes that are achieved.

Bonus Links:

This commentary originally appeared on the RWJF New Public Health blog.