We were at a conference earlier this week that focused on implementing evidence-based practices to effectively address health equity. It’s a big problem because there’s a lot of evidence out there, and we need to draw conclusions. We decided to comb through the last 101 days of health equity and health equity-adjacent research to see what themes emerged.
To do this, we pulled every article abstract published in the last three-ish(?) months from the six journals below. This process yielded 210 unique articles. We know these are not the only places to find health equity research, but it’s a good place to start. Click on the journal names to go directly to go the journal’s home page.
- International Journal for Equity in Health
- Health Equity
- Journal of Health Disparities Research and Practice
- Journal of Racial and Ethnic Health Disparities
- Cancer Health Disparities
- Diversity and Equality in Health and Care
What We Found.
This first figure at right (or below on mobile) shows the most frequently occurring words across all 210 abstracts as a word cloud. The bigger the word; the more frequently the word appeared. This method isn’t too nuanced, but it provides a nice high-level overview of themes.
Unsurprisingly, health studies use the words, “health” and “study” a lot. We can see that COVID-19 plays (and will likely continue to play) a central role.
We also find it helpful to call out review articles. For those who are unfamiliar, review articles pose a specific question, then try to gather and evaluate all the published literature on the topic. In the past 101 days, there have been 18 such articles. We put them into a downloadable PDF at the link below.
Two hundred and ten articles are likely too many to read, so we clustered these according to topic. This was done via a method called Latent Dirichlet Allocation. Following this, we determined the words that distinguish each topic. The plot below appears complicated, but just focus on the axes on the left-hand side. These are the words that best represented each topic.
We assigned each article to the topic it most likely belonged to. We then counted up how many articles per topic to see what themes are coming up most frequently.
So, COVID-19 is clearly the big one, but we can also see topics around unmet insurance coverage, food and nutrition policy, and systemic racism.
One more step. After doing this clustering algorithm, we can pick the article that is most representative of each topic. We also exported this as a PDF and put it in the link below.
Still have too much to read? Check out Julia Moore’s 10 tips to help you catch up on the pile of articles you haven’t read.