Beyond individuals: How and why neighborhoods matter for health
by Ana Diez Roux, MD, PhD, MPH | Apr 22, 2010
Ana Diez Roux is Professor of Epidemiology and Director of the Center for Integrative approaches to Health Disparities at the University of Michigan. Dr. Diez Roux received her medical degree from the School of Medicine of the Universidad de Buenos Aires and subsequently completed clinical training in pediatrics at the National Children's Hospital in Buenos Aires. She obtained an MPH and a PhD in Health Policy from Johns Hopkins School of Hygiene and Public Health. In addition to her faculty position, she is Director of the Robert Wood Johnson Health and Society Scholar's program. Dr. Diez Roux serves on the Blue Cross Foundation's National Advisory Panel on Social Determinants of Health.
Common explanations for why people differ in health include differences in biology, behaviors, individual knowledge/choice and early detection and treatment in medical care. While all are important, we also have to remember that in fact health can really be thought of as a patterned response of social groups to their environments. While workplaces and schools are also part of the picture, neighborhoods are one important environment in people’s lives.
Neighborhoods are contexts for a variety of physical exposures — not just the classic things like air pollution but also the built environment — as well as social exposures. The latter include connections between people and may affect health through a variety of mechanisms related to stress and support. Since the United States is very strongly segregated by race, ethnicity and social class, neighborhood may be an important contributor to social inequities in health. Policies implemented at the neighborhood level could have health consequences — including policies we don’t think of as health policies.
Health patterning by place. At least two factors play out in health patterning by space and place:
The paradigm for prevention has traditionally been individual based — focusing on things like genetic factors and the choices people make. Our studies show how chronic diseases may be influenced by neighborhood. Indeed, based on the research, the probability of developing coronary artery disease over a 10-year period is clearly patterned by neighborhood. These findings hold constant even after controlling and adjusting for income, education and occupation.
What might explain these patterns? Features of both the physical and social environment are important, including accessibility of recreational resources, design of public spaces, land use, aesthetic quality, availability and relative cost of “healthy” foods, availability of tobacco, noise, air pollution, safety and violence, social support/cohesion and social norms.
There are specific mechanisms through which features of the physical and social environment might operate to affect health. The physical and social environments are linked, with features of each affecting the other. Although we try to isolate these factors, these things go together in a very holistic way.
Food environment and dietary behavior. The type of stores available in a neighborhood affects people’s dietary behaviors. Supermarkets tend to offer a variety of healthy foods. Our research has shown that lower-income neighborhoods are 50 percent less likely to have supermarkets and over 50 percent more likely to have liquor stores. Lower-income neighborhoods also tend to have many more small grocery stores that do not offer much in the way of healthy food. Through our research, we discovered that those who lived in areas with the worst ranked local food environments were 27 to 40 percent less likely to have a good diet. The conclusion: Food environments vary by neighborhood and correlate with what people are eating. Even a highly motivated person will find it hard to change diet in an area without access to healthy food. This can become a vicious cycle. We have not created an environment that facilitates the adoption and maintenance of healthy behaviors. Changing the environment is amenable to policy. Focusing on the environment, broadly understood, should be an important piece of public health efforts.
The recreational environment and physical activity. In another example of health patterning by place, we discovered that neighborhoods with predominantly poor people and people of color are less likely to have a recreational facility. Not surprisingly, people with high access to a recreational facility are more likely to be physically active than those in areas with low access. Additionally, people who live in areas with higher density and mixed land use are substantially more likely to be walking. The way to increase physical activity is to build it into people’s daily lives.
Longitudinal research covering a five-year period found that people in better environments —those with better access to resources for physical activity and healthy food — were substantially less likely to develop diabetes over time and also less likely to increase in weight over time.
Role of research. Research can provide evidence of the link between health and policies not usually thought of as “health policies.” And researchers can partner with communities to determine the most promising strategies for intervention and action.
Current research has several limitations:
Research remains observational, which limits the ability to draw causal conclusions.
A large portion of the research remains cross-sectional, though studies are beginning to study how neighborhoods (and changes in neighborhoods) affect changes in health over time.
While we are very good at measuring characteristics of individuals, we are not yet good at measuring the environment.
Going forward, we need evidence including:
Rigorous observational studies. We have to measure changes in the environment, which is quite complicated.
Natural experiments. We have to learn a lot more from changes unrelated to health going on in neighborhoods, evaluating health impacts to provide evidence for what works. This requires researchers to partner with a variety of community groups.
Qualitative studies. These studies help us better understand how people relate to place and can inform both the design of quantitative studies and the research questions we may want to ask.
Simulation and systems approaches. As we try to think in a much more sophisticated way about how things are related to each other, simulation and systems approaches could be helpful. Many of the relationships we need to explore are very dynamic. What people buy, for example, is affected by what is in food stores — and vice versa. We’ll need to assess the impact of policies on a variety of outcomes and note unexpected impacts, such as the impact on air quality of a supermarket with a large parking lot.
Action based on “best available evidence” and systematic evaluation of this action. We need to remember that not taking action is also taking action.
Why focus on places/neighborhoods?
Place-based and individual inequalities are mutually reinforcing.
Neighborhood differences are not “naturally” determined; they result from specific policies and are amenable to intervention.
It’s not just about neighborhoods causing poor health; it’s also about neighborhood environments facilitating the ability of people to make the right choice.
Health is affected by many things that have nothing to do with health care — housing, social policies, community development policies.
Positive changes in neighborhood environments are likely to have multiple health and non-health benefits.
