Introduction
At Evidence for Action, we support innovative research that can inform real-world changes with the potential to advance health and racial equity. We call this actionable research, or research that can drive action. While publishing in a peer-reviewed academic journal is undeniably valuable, these publications are geared toward very specific audiences with specialized knowledge. However, these publications can be inaccessible to some of the most critical audiences in many ways. There could be logistical or resource barriers, such as a lack of reliable internet or access to a computer or an inability to scale the paywall. There may be other barriers to understanding – from the use of jargon to terminology that requires an advanced degree to interpret. Ensuring evidence drives action requires that the people who need to use it to make decisions have access to it. In this blog, we delve into some considerations for ensuring research is shared through accessible language and platforms.
An Example for Consideration
Let’s explore an example of a fantastic scholarly piece by lead author Ellicott Matthay, written during her time as an E4A postdoctoral fellow. The paper, The Revolution Will Be Hard to Evaluate: How Co-Occurring Policy Changes Affect Research on the Health Effects of Social Policies, was initially published in Epidemiologic Reviews in December of 2022.
Here’s an excerpt from the abstract:
"Extensive empirical health research leverages variation in the timing and location of policy changes as quasi-experiments. Multiple social policies may be adopted simultaneously in the same locations, creating co-occurrence that must be addressed analytically for valid inferences. The pervasiveness and consequences of co-occurring policies have received limited attention. We analyzed a systematic sample of 13 social policy databases covering diverse domains including poverty, paid family leave, and tobacco use. We quantified policy co-occurrence in each database as the fraction of variation in each policy measure across different jurisdictions and times that could be explained by covariation with other policies."
Access the entire abstract and paper.
While none of this content is at all problematic and is, in fact, laudable in an academic journal article targeting highly educated social epidemiologists, the content should be developed and adjusted for other audiences to enhance understanding and engagement.
Using the Flesch-Kincaid Grade Level Readability Scoring System, the above text is appropriate for College Graduates and is rated as Extremely Difficult to read. As a college graduate, I would argue that this automated assessment overestimates the accessibility of the text to the typical college graduate, at least without the aid of additional resources and texts.
For example, using a Language Learning Model (LLM), Azure Open AI, and requesting a rewrite for a 5th-grade level yields the following:
“Scientists study how different policies impact people’s health. They use a method called quasi-experiments, where they look at policy changes that happen at different times and places. Sometimes, multiple policies are put into effect in the same places at the same time. This can make it tricky to understand the effects of each policy. We looked at 13 databases with information about policies like helping poor people, giving families time off from work, and reducing tobacco use. We measured how often these policies happened together and how they influenced each other.”
While this is a good start, we need additional editing. For example, the phrase “policies helping poor people” could sound demeaning and might be replaced with “poverty reduction policies.” We could replace words with more accessible terms, such as replacing “influenced” with “impacted.” We could also simplify it by removing references to the study design.
Putting this into Practice
Ideally, when communicating with community groups, particularly low-resourced or non-native English speakers, one should aim for 5th- to 8th-grade reading levels to ensure understandability.
Here are some tips for creating content for general audiences:
- Rather than attempting to translate your academic article to make it more accessible, try writing your findings out from scratch using the words you would use to describe your research and findings to your kids, siblings, neighbors, or a group of 5th graders.
- Get feedback from individuals who are part of your target audience. Workshop the content with children in your family, neighbors, community advisory boards, or other community members, etc.
- Use automated assessment tools to check the readability of your text. BUT, do not rely solely on such tools. Nothing compares to the human element.
- Use LLMs to help in drafting, but DO NOT simply copy and paste the results. Do a critical review of the final product.
Conclusion
For research to drive action in the communities it is meant to serve, the findings must be presented in a way that members of the community and decision-makers can understand. The findings must be communicated in straightforward and accessible language with clear calls to action. The reader should not be left to interpret the findings, forced to consult a dictionary, or use a crystal ball to figure out what you mean or what you want them to do with the findings.
Tools & Resources
- Messaging resources from the Public Health Communications Collaborative, particularly:
- Learn How to Use the Flesch-Kincaid Grade Level Formula
- ChatGPT (use with caution and care)