Two of the big stories for 2016 have been income inequality as a key issue in the presidential election debates and the increasing focus on population health in the U.S. health care system. These two are strongly connected. Indeed, if you believe the classic definitions of population health, income inequality is the primary driver of disparities in health status. But, in the contemporary United States context, “population health” tends to mean doing a better job at managing the care of patients along the continuum, keeping them healthy by focusing more on prevention and by targeting social and medical interventions for those folk that we are at highest financial risk for.
In contrast, scholars outside the United States view population health as much closer to the classic notions of social determinants of health. International (non-U.S.) researchers and policymakers attempt to isolate the contribution that socioeconomic factors make to health status and intervene accordingly, politics willing. In the United States, however, population health is more about the health and health care of those people we are at financial risk for, rather than factors affecting the health status of the entire population such as housing, education and income.
This distinction matters. Are we talking everyone in the population or just the crowd we are responsible for? And are we really going to pull policy and program levers like income redistribution, education, early childhood development, housing, transportation and social policy to improve health and health care, or do we limit ourselves to factors we are more comfortable with, like giving wellness the old college try and doing a better job with diabetic registries?
Defining Population Health
My cynical definition of population health is it’s the phrase CEOs use during a lull in the conversation at a hospital board meeting when they really don’t have a strategy. As Monty Python said in the famous Piranha Brothers sketch in reference to a gangster: “Cruel but fair.”
Part of the confusion about population health is mixing terms between classical views of population health (mainly from outside the United States) and the emerging contemporary trends in population health management, here in the United States. Both are valid and indeed are connected.
The View from Outside the United States
Wikipedia, whence all true knowledge comes (actually this definition is from scholars David Kindig and Greg Soddardt) defines population health as “the health outcomes of a group of individuals, including the distribution of such outcomes within the group.” It is an approach to health that aims to improve the health of an entire human population.
In my view, the seminal works of classic population health are the Whitehall studies in the 1960s and 1970s and the work of the Canadian Institute of Advanced Research Determinants of Health Group led by my old colleague and mentor, Canadian health economist Bob Evans, in the 1990s.
The Whitehall studies of health status and death rates in the British Civil Service showed that despite the fact that all civil servants in England have the same access to health care, there was a strong correlation between social status and health even after adjusting and controlling for health-affecting factors such as obesity, exercise, smoking and drinking.
In the 1990s Evans brought together multiple researchers from disciplines as diverse as medicine, economics, political science, animal biology and sociology to distill the knowledge about health disparities. The group produced the classic work “Why are some people healthy and others not?”
Key connections were made in that research between socioeconomic status and health. It produced important insights linking the brain chemistry of individuals and populations and found that psychosocial stress was a key variable of health and longevity. As Robert Sapolsky of Stanford has shown in the field and the laboratory: If you’re going to be a baboon it’s better to be a leader than a follower. It’s healthier to be at the top of the hierarchy than at the bottom.
I recently reconnected with all of the members of the group at a reunion to celebrate Bob Evans’ retirement from the University of British Columbia in Vancouver. Over a couple of days in conference we heard from many of the key thinkers in that group and other notable researchers who have picked up the mantle of population health research. What struck me was the enduring resilience of the conclusions that Evans and his group identified, namely that there was a strong, if not dominant, role that income inequality and socioeconomic status played in determining the health of individuals and populations. And that role persisted even after accounting for risky personal behaviors such as overeating, indolence, smoking and imbibing.
One particularly poignant presentation for me was by Canadian physician John Frank, policy scholar and epidemiologist who now leads a health policy research unit at the University of Edinburgh (also my alma mater). John’s presentation highlighted the uncomfortable truth of how residents of public housing estates on the fringe of my native Glasgow had life expectancy 15 years less than the populations in the leafy suburbs where I grew up. John described how despite Scotland’s progressive health policies, the sad truth of disparity persisted. Growing up in an awful place with drunken parents who have no job and no hope is not the best way to live a long, healthy life. Being poor will kill you.
Population Health in the United States
In practice, most hospitals and health systems in the United States have a more “hopey-changey” definition of population health, to misquote Sarah Palin. In the contemporary context, population health leaders such as my pal Dr. David Nash, dean of the nation’s first college of population health at Thomas Jefferson University in Philadelphia, is leading us to focus the field at a more practical level on a set of initiatives that can be pursued to deliver Don Berwick’s noble Triple Aim: better health of the population, better patient care and lower per capita costs.
As leaders of the population health management movement, Nash and others are exhorting us (appropriately, in my view) to:
Segment high-risk populations. Identify those 5 percent of patients who account for 50 percent of costs and mange their care aggressively.
Harness advanced analytics. Use big data to identify heavy utilizers, monitor populations, and target and measure performance.
Use patient registries and medical homes. Effective registry strategies, coupled with high-performing teams in medical home models, can improve health and reduce spending.
“No outcome, no income.” Nash’s famous quote captures the essence of a change toward value-based payment in a population health world.
Go upstream. More and more we learn that the solutions look more like social work than medical care.
Eat your own cooking. Health systems employees are normally self-insured, so starting your population health journey with your own employees can be both a learning and a financial opportunity if you can improve performance.
Focus on the whole population. Medical care isn’t everything, and traditional public health and classic social determinants are important where policy and program levers exist.
Meet people in their lives. Socioeconomic context matters: What works for affluent suburbs may not work in the projects.
Emphasize wellness and prevention. There is a thirst for health reflected in popular culture —harness and help this.
Think outside the box. Some of the solutions may be weird, unfamiliar and counterintuitive, but try them anyway.
Partner, partner, partner. Whether it be health plans or skilled nursing facilities, schools or landlords, retail clinics or barber shops, we have to learn to partner with any and all who can advance the Triple Aim.
Nash and other leaders are fully cognizant of the role that the classic social determinants of health play in creating health disparities. But most American practitioners of population health management are focusing much more on these health care–related activities than on addressing the fundamental causal drivers such as socioeconomic status. And this is maybe as it should be in the U.S. context given how squeamish we Americans all are about anything that smells too much like socialism, right, Bernie?
Income Inequality
Mitt Romney may have started it with his comments about the 47 percent who don’t pay tax. He may have lost the 2012 election because he alienated hard-working people who actually do pay FICA taxes but don’t pay income taxes. (And by the way pay a higher share of their income to such taxes than did Romney, just sayin’.)
Angst about the gap between the high-rollin’ top 1 percent versus the growing discontent, dismay and disillusionment of the rapidly shrinking and sinking middle-class has been the backdrop of this presidential campaign season.
Bernie unabashedly wants to redistribute income, get money out of politics, make college free, make health care free and pay for it all through higher taxation on the very wealthy. Hillary is being dragged every day in that direction.
Trump uses the economic angst of displaced factory workers to whip up support for a xenophobic fantasy of good jobs at good wages being returned to the United States because the CEO will receive a threatening phone call from the new president. Or because deported Mexican factory workers will not be able to climb over a 20-foot-high wall to return to their families in the United States.
It is undeniable that income inequality is increasing. The causes are globalization and technology. Neither can be stopped easily. But as the social determinants of health literature tells us, their effects can and should be ameliorated by massive investments in human capital to lift all to achieve their human potential, starting particularly in early childhood. This may come too late, and likely will not help, many of the angriest who are unlikely to be retrained as Ruby on Rails programmers and who want to remain mining coal in narrow seams that are never going to be reopened.
Public support for Addressing Income Inequality
I have had a research partnership with Bob Blendon and his colleagues at the Harvard School of Public Health for 30 years, and I continue to learn so much from Blendon about public opinion, politics and policy.
Blendon and longtime collaborator John Benson recently published a seminal systematic review of all the available polls (national and international) on the public’s interest in and willingness to address income inequality. Let me paraphrase what they found:
Everyone sees the inequality. Overall, two-thirds of the public believe that the gap between the rich and poor is getting larger. This includes about three-fourths of Democrats as well as a majority (56 percent) of Republicans.
Democrats rate it more important. Seven in 10 Democrat registered voters described economic inequality as very important compared with just 42 percent of Republicans.
Countries vary in their concern. In terms of how concerned they are about the rich versus poor gap, overall just 46 percent of American adults described it as a very big problem in the United States or ninth among 13 industrialized nations. By contrast 84 percent of Greeks, 74 percent of Spaniards and 73 percent of Italians deemed it so in their countries.
Democrats look a lot like Europeans. The partisan difference in the United States is striking: Nearly 60 percent of Democrats consider the income gap to be a very big problem compared with 49 percent of independents and only 19 percent of Republicans. (Democrats are like the French, Republicans are from another planet — my words not Blendon’s.)
Self-reliance versus government to blame. The two reasons (out of six offered) American cited most often for income inequality were “people working harder than others” (24 percent) and “the government’s economic policies” (24 percent) with worker pay, the educational system and the tax system also receiving some credit for the inequality of the six variables asked about.
Government deemed responsible outside the United States. Other countries have a greater consensus that governments are to blame for income inequality.
Democrats say government should act; Republicans don’t. A full 81 percent of Democrats believe government should do more to reduce the gap between rich and poor compared with just 34 percent of Republicans. When asked how much the government should do to reduce the gap between the rich and everyone else, 62 percent of Democrats said “do a lot” compared with just 23 percent of Republicans.
Tax the rich. Finally, three-quarters of Democrats believe that the government should redistribute wealth by placing heavy taxes on the rich compared with only 29 percent of Republicans.
That’s the election right there, whether it is income inequality or Obamacare. The question is, Are rich people going to write a check for poor people?
Obamacare has actually been one of the principal vehicles of income redistribution in the last 12 years. Expanding Medicaid and providing highly subsidized coverage to low-income folk has on balance effectively increased the minimum wage by as much as $2 an hour for those who receive the benefits.
If Democrats are elected, this trend will likely continue in the form of greater subsidies for the low-income to purchase health insurance and limit co-payments and deductibles, further expansions of coverage under Medicaid (including the undocumented), increases in the minimum wage, and redistributive taxation from the rich toward lower- and middle-income household groups.
If Republicans prevail, it is less likely that the income distribution will become more equal and it is much more likely that it will become more unequal.
But with Trump, who knows? He still hasn’t released his tax returns at this time of writing so that probably means either he doesn’t make as much money or he hasn’t paid as much tax as he should have. Trump might actually take on the hedge fund managers and their carried interest. Especially if he doesn’t have as big a carried interest as he bragged about.
Elections matter for all kinds of reasons. But this time, people, it’s really serious. Good luck with that.
Income Inequality and Population Health
The income inequality debate and the population health debate collided in the last few months with the release of several blockbuster studies linking income to health.
The Princeton economist and Nobel laureate Angus Deagan and colleagues released a startling report that documented declining life expectancy of low-income white males in the United States in contrast with rising life expectancy for all other ethnic and income groups in the United States and in almost all developed countries. They attributed the alarming statistics to the rise of drug, alcohol and opioid use, and much higher rates of depression and suicide. These are the combined health effects of an economy that cannot provide decent jobs to sustain lower-income households.
More recently, a massive data study (not big data, massive data) by Stanford and Harvard researchers led by Raj Chetty documented similar trends in life expectancy and highlighted that the core variation across time and across the country could be correlated to a number of socioeconomic characteristics.
The Chetty study confirmed what we knew: Life expectancy correlates with income. It is a gradient; the richer you are, the longer you live. And the gap is widening over time.
But most interestingly, the most recent study documents that there is significant variation in life expectancy among low-income groups in particular cities. For example, life expectancy for the bottom quartile of income was as much as 4.5 years higher in San Francisco and New York than in Dallas and Detroit. The correlates of income distribution on mortality were surprising. For example, having a high percentage of immigrants had a very strong positive effect on life expectancy for low-income groups. The conclusion I draw is that liberals have coat tails and that if you’re going to be poor it’s better to be poor in a place where there is interest in social policies that are more progressive than in a place where social policies are more Darwinian.
Support for this hypothesis also comes from the work of Elizabeth Bradley and colleagues at Yale. Professor Bradley was the first person I know to pull together health spending and social spending by country to show that we spend way more on health than most developed nations but way less on social spending than most other countries, putting us in the middle of the pack on the combined measure of social and health spending. The French lead the way in such combined spending, quelle surprise.
Bradley’s most recent study with colleagues at Yale showed that in the United States context, states that spent more proportionately on social spending than on health had better outcomes in particular:
“We find that states with a higher ratio of social to health spending (calculated as the sum of social service spending on public health spending divided by the sum of Medicare and Medicaid spending) had significantly better subsequent health outcomes for the following seven measures: adult obesity; asthma; mentally unhealthy days; days or with activity limitations; and mortality rates for lung cancer, acute myocardial infarction, and type II diabetes. Our study suggests that broadening the debate beyond what should be spent on health care to include what should be invested in health — not only in health care but also in social services and public health — is warranted.”
Better Opportunities mean Better Health
The message from all of these studies is clear. It’s what I tell my kids: Be in the top 1 percent and you’ll do just fine.
But that doesn’t help typical American households who have seen incomes decline over the last 20 years in real terms and indeed, as I’ve hammered in this column over and over again, any increment in compensation that lower income folk may have received has gone to health care benefits.
They probably would have been better off (and much healthier) with the money, or jobs, or housing, or education, or social services. But, that is not what was on offer. We should all think about that.