What’s really driving high-cost use of health care

So-called “high-cost users” of health care – the 5% of the population who use nearly two-thirds of health care resources – have become a major policy focus, both in Canada and abroad. But in the rush to save money on the care of patients who are already high-cost users, too little attention has been paid to what truly drives high-cost use of the system, and what might be done to prevent patients from becoming high-cost users in the first place.

While research identifying a small segment of the population as intensive users of the health care system is over 30 years old, the recent wave of interest has been driven in large part by the popularizing of the idea of “hot spotting”: the theory that health care costs can be reduced by strategically improving the coordination and delivery of care for high-cost users. This wave of interest has driven a range of initiatives across Canada, from intensive focus on frequent emergency department users in the Eastwood region in Edmonton, to the creation of Health Links across Ontario.

The trouble is that while initiatives like Health Links incorporate evidence-based approaches to reducing emergency visits and hospital admissions, they don’t actually seem to save the system much money.

How could this be? The first reason is what’s been called the “rigid structure of health care costs.” While interventions like these may improve the quality of care (a worthy goal in its own right), improvements in quality tend to translate into increases in capacity (shorter wait times, more available beds in hospitals, etc.), not cost savings.

Second, all of these interventions target patients who are already high-cost users of the health care system. In other words, interventions like Health Links – and other versions of “wrap around care” – tend to target patients only after they have already become very sick, often with multiple chronic conditions, such as congestive heart failure, chronic obstructive pulmonary disorder, and cancer. Once a patient is older and has developed multiple chronic conditions, it may be that a substantial portion of the health care resources they consume are medically necessary, and simply can’t be significantly reduced through better coordinated care or fewer visits to the emergency department. If the central policy goal of concentrating on high-cost users is to save money, then the best approach may not be to focus so exclusively on people who are already high-cost users, but also strive to prevent people from becoming high-cost users in the first place.

So what, at a fundamental level, is driving high-cost use? In a word, poverty.

This should be no great revelation – research has shown for some time that there is an association between high-cost use and socio-economic status. Yet this knowledge has not been enough to shape the policy agenda with respect to high-cost users, perhaps in part because solving poverty seems a monolithic problem well beyond the means of a health care system to address.

But we have begun to learn a great deal more about how the different dimensions of socio-economic status contribute to high-cost use, knowledge that could form the basis for more comprehensive, more effective approaches to reducing high-cost use.

Research published today in the American Journal of Preventative Medicine (full disclosure: we are co-authors on this paper), examines the role of a multitude of social and economic factors in someone’s odds of becoming a high-cost user in the future. These factors include the neighborhood one lives in, ethnicity, home ownership and many others. What we found was the single strongest predictor – even stronger than income – of future high-cost use is someone’s access to sufficient, safe, nutritious food (what public health professionals call food security).

This research opens up the possibility that interventions which target specific dimensions of socioeconomic status – such as reinventing Canada’s broken food bank system – could be part of an inter-sectorial approach to reducing high-cost use. On a broader scale, it suggests the upfront costs of policy options such as a minimum guaranteed income, Housing First programs, and appropriate funding for Ontario’s poverty reduction strategy could be offset, at least to some degree, by savings in the health care system. This may be particularly true for children who grow up in poverty, where small investments early in life could pay out significantly later on.

Of course, public health and community health have long focused on these broader issues. But these efforts are largely disconnected from the work being done within the health care system, and continue to suffer from inadequate funding. Initiatives like Health Links have highlighted the benefit of improving integration within the health care system. But this research indicates that we must go farther, and integrate across the spectrum of population health.

Recognizing the role socio-economic factors play in driving high-cost use also reminds us that while the “problem” of high-cost users is typically cast in terms of cost savings and sustainability of the health care system, it is fundamentally a problem of health disparities and social inequity. Reducing high-cost use is the right thing to do, not just economically, but ethically.

There has been much excellent work to date across Canada to improve the coordination and quality of care for high-cost users of the health care system. This work should continue, but the role of socio-economic status in driving high-cost use needs to become a bigger part of the conversation, and we need to consider a wider range of interventions and policies if we truly want to reduce costs and improve equity in our health care system.