Taking Action on Environmental Health Could Impact U.S. Health Care Cost Effectiveness: A Q&A with Bobby Milstein

May 10, 2011, 7:20 PM, Posted by NewPublicHealth

The May 2011 issue of Health Affairs, which focuses entirely on environmental health (see the NewPublicHealth interview with Health Affairs Editor-in-Chief Susan Dentzer) includes an article by Bobby Milstein, PhD, MPH, and colleagues on the critical role taking action on environmental health would play in improving cost effectiveness of the U.S. health care system. Milstein leads the Syndemics Prevention Network at the Centers for Disease Control and Prevention.

NewPublicHealth spoke with Milstein about the Health Affairs article.

NPH: Tell us about the new study.

Bobby Milstein: This is an attempt to examine some of the major proposals for health system change in the United States. We look specifically at scenarios for expanding health insurance coverage, improving the delivery of preventive and chronic care in clinical settings, as well as wider efforts outside the doctor's office to protect people’s health by enabling healthier behaviors and building safer environments. One of the unique things about our approach in this study is its broad scope: it is relatively rare for a single study to look at interventions that deal with financing, clinical interventions, and wider population-level health protection efforts. To consider all of those dimensions in one single study--comparing the relative advantages and disadvantages of each strategy alone as well as what can happen if they were to be combined--gives us a pretty rich space to examine how the U.S. health system functions and how it can be improved over time.

NPH: And what were your key findings?

Bobby Milstein: What we discovered, in focusing on just these three types of interventions, is that each strategy alone could save hundreds of thousands--if not millions--of lives and would be a good economic value. That said, there’s nothing mutually exclusive about these options and if they were to be done together, they could be more effective in combination. Part of the story behind what happens when they’re combined is that the element of health protection--helping people to have healthier behaviors and live in safer environments--is a very critical ingredient in the mix because it is the one element among the things we studied that has a chance of lowering the sheer number of disease and injury events in this country. It can also ease demand on an already over-burdened health care delivery system. So, it not only keeps people healthy, but makes our health care delivery system work better for the people who need it. Those are two key reasons why investing in healthier behaviors and safer environments is likely to contribute to so many more lives saved and ultimately save money in the process.

Our paper also highlights the importance of choosing an appropriate time horizon when studying scenarios for health system change. This study shows that one might draw different conclusions about the impacts of these interventions in the short-term vs. the long-term. If we limit our scope of concern to policy scenarios that extend out over ten years, we’re going to miss some of the fuller effects of what these kinds of national commitments might accomplish. When we carry the analysis out through twenty-five years, we see some very powerful and potentially surprising results regarding the potential impact on health and cost. This is an instance where a long-term view may be needed to transform the functioning of the U.S. health system to be more effective.

NPH: Can you explain the new mathematical tool used to analyze the combinations?

Bobby Milstein: Most of us would like to know the likely impacts over time on health and cost for a range of different policy options. That sort of prospective information could help shape a better overall strategy. What we did, with several years of work at the CDC, is to combine information from many data sources into a single analytic framework. This tool, in turn, has an explicit set of pathways that link together more than a dozen policies that could alter health system dynamics. For example, we can identify how insurance coverage affects not only the receipt of care and the access to it, but also how the demand for that care could change over time based on the success of wider interventions to enable healthier behaviors and build safer environments. That is just one quick example of the work that we did to carefully fit together the moving pieces of our health system into a single tool-and we call that the “HealthBound Policy Simulation Model.” It is this HealthBound model that is the basis for the analysis that was published in Health Affairs.

NPH: Is this the first time the tool has been used?

Bobby Milstein: It is not. The model began as a prototype back in 2008 and one of the early steps that we took was to convene a stakeholder review by twenty-five leading policy analysts and scholars of health system dynamics in the U.S. Those colleagues gave us an early indication that such a tool was plausible, that it had the right policy ingredients, as well as the most essential inputs and outputs they wanted to see.

We then proceeded to develop the tool for a year and a half, primarily through interactions with about five hundred to a thousand colleagues in various venues. We have also been refining it by adding more research as an empirical basis and clarifying how it works. There was a prior publication about a year ago in May of 2010 in The American Journal of Public Health, which laid out the general design and technical features of the HealthBound model. It is also available online at the CDC website through an interface that people can use as an interactive game in their efforts to strengthen leadership and vision for health system change. Hundreds of people have used the HealthBound game in structured ways to think about health policy in this country.

NPH: Are there any immediate steps from the paper’s findings that policy-makers could take?

Bobby Milstein: Absolutely. As I said, there are three broad classes of policies that we considered: insurance, clinical care, and health protection.

There are many practical ways to fulfill the potential in each of these categories, and those specific steps might be different in different contexts. For instance, in the category of clinical interventions, we can think of efforts to improve the adherence to recommended guidelines for blood-pressure control, cholesterol control, smoking cessation and even getting people on aspirin who are at risk for cardiovascular disease. These are all important clinical measures that are well-proven-and just not used as widely as possible.

Turning to the wider community interventions, when we think about healthier behaviors and safer environments, that conversation must begin with efforts to prevent smoking, which is still the single leading cause of preventable death in this country. Comprehensive tobacco control could make tremendous gains both in health and costs. In addition, efforts to deal with obesity, physical inactivity, dietary improvements, and alcohol are just the top of the list of interventions for which guidelines and clear recommendations have already been made. The question is, do we have the will to implement them in this country?

In the environmental realm, we might think about things like reducing air pollution, improving the safety of motor vehicles, and other forms of injury prevention, any of which could be a practical step that people could think of as fitting within the results of the study that we’ve conducted here.

NPH: How else might the tool be used in the future?

Milstein: We’d like to see the HealthBound model used as a way to make sure that when diverse stakeholders get together and think about health system change that they do so with the benefit of a tool that can help them play out the likely consequences of different scenarios. That is just one among many uses. This study, as I mentioned, compares and combines just three broad classes of policies. The model itself, however, offers a much wider range of interventions-such as things having to do with the coordination of care in the clinical realm or social policies to address health equity more broadly. Users can explore a very diverse range of options within this same framework. So the HealthBound model stands as a rich resource for pursuing other policy studies.

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