Evaluating HIAs: A Q&A with Andrew Dannenberg
A key session at the Second National Health Impact Assessment Annual Meeting held recently in Washington, D.C., was a panel discussion on several evaluations of the value and benefits of health impact assessments (HIAs). Andrew Dannenberg, MD, MPH, an affiliate professor at the University Of Washington School of Public Health, was a consultant on a recent evaluation of HIAs funded by the Robert Wood Johnson Foundation and a member of the evaluation panel at the HIA meeting. NewPublicHealth spoke with Professor Dannenberg about some recent findings.
NewPublicHealth: What have the recent evaluations of HIA as a tool told us about the value conducting health impact assessments?
Andrew Dannenberg: Essentially, HIA works. The tool does seem to promote health, and does have influence in some cases but not others. HIAs can influence the health component of [policy] decisions.
There are also indirect HIA benefits: by getting public health professionals talking with decision makers in other sectors—such as transportation and housing—HIAs create partnerships and collaborations for longer-term value. So a transportation department building a highway may then always realize that there are health implications of what gets constructed.
We also came away with a list of factors that influence HIAs to make them successful. The list includes:
- Timeliness is often a factor when doing an HIA (in that the HIA must be completed and recommendations made in time to support or influence the policy decision).
- Involving stakeholders and decision makers gives a better chance that the recommendations will be considered.
- It is important to have community engagement and feedback, or, particularly when it is an HIA being done rapidly, it is critical to have a well-informed health leader at the helm.
- It is critical to screen the topic to be sure it is appropriate for an HIA.
- Dissemination to stakeholders, decision makers and media is very important, using methods, length and language appropriately customized for those audiences.
- HIA recommendations need to be clear and actionable.
- The Australian evaluation found that a key to successful HIAs was getting the right people at the right time to work together.
NPH: Do you have an example of an HIA that showed that using the tool leads to better decision making?
Dannenberg: An HIA conducted in San Francisco several years ago is one of our clearest examples. A developer wanted to tear down some low-income housing to build more expensive apartments that would have displaced the low-income people living at the site. The Department of Health conducted an HIA, which made it clear that it is bad for health to take low income people in an expensive city and throw them out in the street with no housing.
Ultimately because of the HIA and because the Health Department could influence the building permit process, the developer agreed to build one-to-one replacement housing for the residents who were going to be displaced. The findings of the HIA were a critical factor in being able to convince decision makers to build the replacement housing. It almost certainly wouldn’t have happened otherwise.
But there are also less-clear examples. When some changes happen, it can be difficult to prove that the good outcome was because of the HIA. In Baltimore, for example, an HIA recommended that zoning codes be updated to improve walkability and to reduce the density of alcohol outlets in neighborhoods. What is difficult to prove is whether one can show any difference in the health of people 10 years later because the zoning code was improved, even though the zoning code impacts what gets built eventually.
NPH: And what about barriers to conducting HIAs?
Dannenberg: Lack of resources, of course, can be a barrier. Who is going to conduct the HIA and who is going to pay for it? There is also the potential regulatory barrier—a decision maker may say that if the law does not require it, why should we do it? Most HIAs have been voluntary, and not required, and in some settings that is fine when the decision makers are happy to have the input. But in other settings, a decision maker may say “I am not required to look at it, so why should I?”
There are also qualitative and quantitative issues that come up, with some decision makers saying that if we cannot measure how much health difference it will make and we cannot do a cost-benefit analysis, then why should we make the change?
>>Bonus link: The American Public Health Association’s guide to incorporating public health considerations into policymaking, “Health in All Policies: A Guide for State and Local Governments.”