Q&A: A closer look into a JAHA study on greenness and heart disease

By Amanda Torres

University of Miami Miller School of Medicine Public Health Sciences researchers find an environmental factor that may be associated with reducing heart disease.

The Journal of the American Heart Association has published a study that found that neighborhood greenness may be associated with reduced cardiovascular disease (CVD). The findings could lead to new preventative strategies to lower CVD risk.

In this Q&A, Scott Brown, Ph.D., principal investigator of the study and research associate professor of public health, Joanna Lombard, M.Arch., professor of architecture and public health, who also conducted the study, and José Szapocznik, Ph.D., professor of public health sciences, architecture, psychology, and educational & psychological studies, explain the process of starting and conducting the study, as well as continuing to research the association through a follow-up study.

What led you to focus on neighborhood greenness and heart disease?  

For the last 15 years, we have been looking at the built environment and health. More recently, we’ve become interested in greenness and health, and greenness is a vegetative presence, such as trees and green spaces.

Going back over 35, almost 40 years, there have been studies suggesting that greenness may be related to different health outcomes. Studies in the early 80s suggested that having a view of nature from hospital windows could be associated with faster recovery from gallbladder surgery and require less pain medication than somebody with a view of a brick wall. Other studies subsequently showed that having nature in a courtyard was associated with more opportunities for social interactions, as well as for opportunities for physical activity. Nature could also improve air quality and reduce urban heat island effects.

In the last 10 years or so, we have been looking at larger data sets and higher power studies with more study participants, where it was found that the same measure of greenness in a sample of a quarter-million Medicare beneficiaries was related to reducing the likelihood of obesity-related conditions, such as diabetes and hypertension.

Recently, we published a study that showed that there were reduced odds of depression and Alzheimer’s disease when surroundings had more greenness.

This area of literature has been taking off in the last few years and we focused on heart disease because it is the leading cause of death in the U.S., as well as worldwide, and secondly, we identified that through Medicare data, we could get four specific heart disease diagnoses, which had not been done in literature before.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine

What were the four types of cardiovascular disease?

We focused on acute myocardial infarction, which is a heart attack, ischemic heart disease, also known as coronary heart disease, which is one of the most common forms of heart disease, as well as heart failure, which is a heart condition where a heart is not pumping blood as efficiently as it should be. Lastly, we looked at atrial fibrillation, which is characterized by irregular heartbeat and can increase risk for a stroke.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine

What sample and locations did the study focus on? 

For this study, we focused on the entire population of Medicare beneficiaries in Miami-Dade County who lived in the same address in 2010 and 2011. This is a sample that closely mirrors the population of Miami-Dade that are 65 and older because this is the largest insurance program for people in that age range. The sample is approximately 67 percent Hispanic, 11.15 percent black, and 23.29 percent non-Hispanic or white.

We had previous funding from the U.S. Housing and Urban Development (HUD) a few years ago, where we had initially proposed a one-county test to look at the relationship between built environment and health and became aware that Centers for Medicare and Medicaid Services, or CMS, provide a zip-plus four. This is essentially the nine-digit zip code, which could be translated into a specific census block.

We were interested in looking at block-level determinants of health because a lot of research in this area historically had used larger geographical areas, such as a zip code, and we know that a zip code can be heterogeneous where there could be one block that is completely green, and another that is a vacant lot with no greenery.

Going back 10 years, our prior published work and data had also suggested that the block where an elder lives may be particularly important in determining their health, social interaction, and physical activity.

We also had discussions with Maria Nardi, the director of the Miami-Dade County Parks and Recreation and Open Spaces, which is the third largest county park system in the country. We asked her what geographic unit of analysis would be the most helpful or impactful for her in her role as a park director, and she said the block-level of analysis because one could be strategic on said level. For example, if it is identified that a block has a particular risk for poor health as a result of low greenery, that is the easiest geographical level to do an intervention because you can plant trees and in fact, Miami-Dade County had planted over 200,000 trees since 2011.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine

Based on previous research, were you surprised with the association, or did you expect the obtained results? How does this study differ from other related research?

I think we may instinctively feel that trees have a good effect on us—they shade and cool and we appreciate their beauty—but we may not have specific evidence beyond our own feelings. This study reveals that there are measurable health benefits associated with higher levels of green, and so the findings provide tangible evidence of something that many of us may sense, which isn’t a surprise per se but is a new level of objective association.

—Joanna Lombard, professor of architecture at UM’s School of Architecture and public health sciences at the University of Miami Miller School of Medicine   

I was pleasantly surprised by how consistent these findings are. Three of the diagnoses showed relatively strong relationships of higher greenness to lower odds of heart disease, with the exception being atrial fibrillation. Instead of 19 percent overall reduced odds of heart disease, it was six percent reduction for A.Fib. We believe that the reason for that is that there is a form of A.Fib, called familial atrial fibrillation, that has more of a genetic component, so this kind is related more so to genetics than to the environment.

This is the first study in which looks at block-level greenness as a correlate or predictor of heart-disease, whereas prior work has looked at larger geographic areas, such as a quarter-mile buffer, or even bigger.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine

I was totally surprised by the strength of the associations, that is, how strongly greenness predicted heart disease. We were all also very surprised of the consistency of the findings. To the extent that greenness consistently impacts different types of heart disease greatly strengthens the credibility of the findings.

—José Szapocznik, Ph.D., professor of public health sciences, architecture, psychology, and educational & psychological Studies

What are the strengths and limitations of the study? Is there going to be a follow-up study?

The strength of the study is that the observations are based on objective measurements—health data from the Centers for Medicare and Medicaid Services and greenness measured by satellite data—so the correlations are not subject to individual perceptions, but really represent a clean association. Although not a weakness, the study is a snapshot of one moment in time. So expanding the study to look at those same study subjects and locations over time will give us an understanding of the impact of planting that took place after the initial study. Then the next step after that would be to expand to other cities to determine if the results are consistent across other climate zones and urban configurations. 

—Joanna Lombard, professor of architecture at UM’s School of Architecture and public health sciences at the University of Miami Miller School of Medicine   

One of the limitations is that this is an associational study. We are looking at where people lived in 2010 and 2011 and their heart disease diagnoses in those years. It is not certain if people with less heart disease pick greener environments to live in initially, as opposed to greenness causing less risk for heart disease. In other words, we do not know which came first.

That is why we are currently doing a follow-up study in the same sample that will be funded by Robert Wood Johnson Foundation Evidence for Action. The funding will allow us to look at relationships over time, between not only the initial levels of greenness but also changes in greenness as a result of tree planting. As mentioned, there were over 200,000 trees that were planted since 2011, but since Hurricane Irma in late 2017, we lost significant tree canopy. This new analysis will focus on the 2010 to 2016 timeframe to essentially use additional data on both the greenness at the block-level and also on cardiovascular disease diagnoses. We are focusing on low-income neighborhoods, where the median household income is $31,000 or less per year because that is where we are seeing the strongest relationship outcomes to greenness.

Across our studies so far, the relationships are approximately 50 percent stronger in low-income neighborhoods when compared to high-income neighborhoods. Within those low-income neighborhoods in Miami-Dade County, we are identifying which neighborhoods had low greenness in both 2010 and 2016, which are one of the comparisons we want to make. Another two are what happened to people who lived in high-greenness blocks both in 2010 and 2016, as well those who lived in low greenness in 2010 and high greenness in 2016, and so on.

We are using statistical matching methods, called propensity-score matching, to essentially identify beneficiaries who are equivalent in terms of certain characteristics, such as in their racial and ethnic compositions, as well other comorbidities or neighborhood characteristics as confounders. This is a three-year study and we are currently in the process of collecting both the greenness data from 2016 satellite imagery, as well as the 2016 Medicare data.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine

The strength is the consistency of findings across different heart disease. Also, methodologically we use nationally available data sets that makes it possible for others to replicate our findings within our own County and in communities across America.  Because we only conducted the study in Miami-Dade County we cannot generalize to other types of communities yet. However, my own sense is that greenness cover is likely to be a protective factor in other warm climates such as our own.

The current study is cross-sectional (one time point only, in 2010). The follow up study will look at data across time, which is methodologically a more rigorous research design. It will also look at the impact of greenness in blocks in which a greening interventions occurred, increasing further the rigor of the method.  If greening interventions impact heart disease, it increases the likelihood of a causal relationship between greenness and heart disease.

—José Szapocznik, Ph.D., professor of public health sciences, architecture, psychology, and educational & psychological Studies

How might the conclusions of this study provide potential protective measures to reduce heart disease risk?

The results could inspire greater investment in greenery in neighborhoods. Neighborhoods could work with their municipal or county parks and public works departments to plant street trees and encourage the development of parks within walking distances of residences. And community leaders could advocate support for ongoing programs such as the Parks Foundation of Miami-Dade and Million Trees Miami to green streets and neighborhoods.

—Joanna Lombard, professor of architecture at UM’s School of Architecture and public health sciences at the University of Miami Miller School of Medicine   

One possible takeaway from the findings would be that having more vegetation, such as street trees and green spaces, may be one strategy to reduce the risk for heart disease at the population level, possibly by increasing opportunities for physical activity, social interaction and/or stress mitigation.

Should findings of the follow-up study reveal that increases in greenness due to tree-planting are associated with reduced risk of cardiovascular disease over time, then this would provide further evidence that greening interventions may be one strategy to reduce the burden of cardiovascular disease, and hence lower health care costs, at the community level.

—Scott Brown, Ph.D., research associate professor of public health at the University of Miami Miller School of Medicine 

Published on March 5, 2019