Background As the United States continues to tackle the opioid epidemic, it is imperative for digital healthcare organizations to provide Internet users with accurate and accessible online resources so that they can make informed decisions with regards to their health. Objective The primary objectives were to adapt and modify a previously established usability methodology from literature, apply this modified methodology in order to perform usability analysis of opioid-use-disorder (OUD)-related websites, and make important recommendations that OUD-related digital health organizations may utilize to improve their online presence. Methods A list of 208 websites (later refined) was generated for usability testing using a modified Google Search methodology. Four keywords were chosen and used in the search: “DEA-X Waiver Training”, “opioid-use-disorder (OUD) Initiatives”, “Buprenorphine Assisted Treatment”, and “Opioid-Use Disorder Websites”. Usability analysis was performed concurrently with optimization of the methodology. OUD websites were analyzed and scored on several usability categories established by previous literature. Results “DEA-X Waiver Training” yielded websites that scored the highest average in “Accessibility” (0.84), while “Opioid-Use Disorder Websites” yielded websites that scored the highest average in “Content Quality” (0.67). “Buprenorphine Assisted Treatment” yielded websites that scored the highest average across “Marketing” (0.52), “Technology” (0.89), “General Usability” (0.69), and “Overall Usability” (0.68). “Technology” and “Marketing” were the highest and lowest scoring usability categories, respectively. T-test analysis revealed that each usability, except “Marketing” had a pair of one or more keywords that were significantly different with a p-value that was equal to or less than 0.05. Conclusions Based on the study findings, we recommend that digital organizations in the OUD space should improve their “General Usability” score by making their websites easier to find online. Doing so, may allow users, especially individuals in the OUD space, to discover accurate information that they are seeking. Based on the study findings, we also made important recommendations that OUD-related digital organizations may utilize in order to improve website usability as well as overall reach.

No abstract

We aimed to study the simplification of biomedical text via large language models (LLMs). Specifically, we finetuned three language models to perform substitutions of complex words and word phrases for their respective hypernym in biomedical definitions. This process was then evaluated by readability metrics, and two measures of sentence complexity: the measure of lexical diversity (MLTD), and mean dependency distance (MDD) scoring. A sample of 1,000 biomedical definitions in the National Library of Medicine’s Unified Medical Language System (UMLS) was processed with three approaches, each with a different language models and analysis revealed an increase in FK score and a reduction in reading grade level across all metrics. Reading scores improved from a pre-processed collegiate reading level to a post-processed US high-school level. An inter-approach comparison showed that our GPT-J-6b approach had the best improvement in MLTD and MDD. This study demonstrates the merit of hypernym substitution in improving the readability and improving measures cognitive burden of biomedical content for the general public.

BACKGROUND Social media use by the public is at an all-time high and provides a convenient method of reaching broad audiences for public health agendas. Despite this, quantifying the impact of social media on health campaigns has proven to be a challenge, largely due to a lack of consistency in the metrics used, which can be unclear and variably effective. GetWaivered.com (GW) delivers and hosts free nationwide, remote DEA-X waiver training courses that enable clinicians to obtain their DEA-X so that they are able to prescribe Buprenorphine and effectively treat Opioid Use Disorder (OUD). Evaluating Get Waivered social media engagement can be key to increasing awareness of our course. OBJECTIVE This paper aims to evaluate the use of social media platforms such as Facebook, Twitter, and WordPress for raising awareness of obtaining a DEA-X Waiver by determining the reach and engagement of the social media program. METHODS The intended target of the social media postings were clinicians who were eligible to prescribe buprenorphine. For our preliminary evaluation, we analyzed Facebook ads, Twitter, and Google Analytics data to measure engagement with our course advertisements. Twitter data such as impressions, engagements, engagement rate, and interactions (tweets, retweets, replies, likes) was assessed using exported excel data from Twitter Analytics and analyzed on Microsoft Word Excel. Facebook Ads reach, engagement, reactions, comments, shares, post clicks, cost, and cost per post engagement/event response/link click were assessed with Python using data from the Facebook platform. Google analytics data from the Google Analytics platform was used to analyze engagement metrics for the Get Waivered website. RESULTS Twitter posts generated a total of 16,643 impressions and 291 engagements. Facebook Ads generated a total of 36,008 reach, 6,173 engagement, and 321 clicks. Google Analytics showed that the WordPress blog (Get Waivered website) had a 477.80% increase in total users, a 475.10% increase in new users, 390.61% increase in sessions, and a 219.59% increase in page views between comparison periods. CONCLUSIONS This evaluation shows that social media posting does garner engagement while being more convenient than traditional forms of advertising. Since methods for analyzing social media use in public health is varied, these preliminary findings set a basis for developing social media posting structure and evaluation for future projects. The use of Twitter analytics and Facebook Ad insights allows GW to track which posts have large engagements and how to better interact with Twitter and Facebook users. This evaluation provides preliminary data and a framework as a baseline for planned future studies.

No abstract