Challenges to Transforming Unconventional Social Media Data into Actionable Knowledge for Public Health Systems During Disasters

Capturing Emerging Experiential Knowledge for Vaccination Guidelines Through Natural Language Processing: Proof-of-Concept Study

BACKGROUND

Experience-based knowledge and value considerations of health professionals, citizens, and patients are essential to formulate public health and clinical guidelines that are relevant and applicable to medical practice. Conventional methods for incorporating such knowledge into guideline development often involve a limited number of representatives and are considered to be time-consuming. Including experiential knowledge can be crucial during rapid guidance production in response to a pandemic but it is difficult to accomplish.

OBJECTIVE

This proof-of-concept study explored the potential of artificial intelligence (AI)-based methods to capture experiential knowledge and value considerations from existing data channels to make these insights available for public health guideline development.

METHODS

We developed and examined AI-based methods in relation to the COVID-19 vaccination guideline development in the Netherlands. We analyzed Dutch messages shared between December 2020 and June 2021 on social media and on 2 databases from the Dutch National Institute for Public Health and the Environment (RIVM), where experiences and questions regarding COVID-19 vaccination are reported. First, natural language processing (NLP) filtering techniques and an initial supervised machine learning model were developed to identify this type of knowledge in a large data set. Subsequently, structural topic modeling was performed to discern thematic patterns related to experiences with COVID-19 vaccination.

RESULTS

NLP methods proved to be able to identify and analyze experience-based knowledge and value considerations in large data sets. They provide insights into a variety of experiential knowledge that is difficult to obtain otherwise for rapid guideline development. Some topics addressed by citizens, patients, and professionals can serve as direct feedback to recommendations in the guideline. For example, a topic pointed out that although travel was not considered as a reason warranting prioritization for vaccination in the national vaccination campaign, there was a considerable need for vaccines for indispensable travel, such as cross-border informal caregiving, work or study, or accessing specialized care abroad. Another example is the ambiguity regarding the definition of medical risk groups prioritized for vaccination, with many citizens not meeting the formal priority criteria while being equally at risk. Such experiential knowledge may help the early identification of problems with the guideline's application and point to frequently occurring exceptions that might initiate a revision of the guideline text.

CONCLUSIONS

This proof-of-concept study presents NLP methods as viable tools to access and use experience-based knowledge and value considerations, possibly contributing to robust, equitable, and applicable guidelines. They offer a way for guideline developers to gain insights into health professionals, citizens, and patients' experience-based knowledge, especially when conventional methods are difficult to implement. AI-based methods can thus broaden the evidence and knowledge base available for rapid guideline development and may therefore be considered as an important addition to the toolbox of pandemic preparedness.

What Data Should Be Included in a Modern Public Health Data System

The public is inundated with data, both in where data are ubiquitously collected and in how organizations are using data to drive public sector and commercial decisions. The public health data system is no exception to this flood of data, both in growing data volume and variety. However, what are collected and analyzed about the health status of the nation, how particular data and measures are prioritized for parsimony, and how those data provide a signal for where to invest to address health inequities are in dire need of a reboot. As with other articles in this supplement, this article builds from a literature review, an environmental scan, and deliberations from the National Commission to Transform Public Health Data Systems. The article summarizes what data should be included and identifies where the technology and data sectors can contribute to fill current gaps to measure equity, positive health, and well-being.

COVID-19 Vaccine Perception in South Korea: Web Crawling Approach

BACKGROUND

The US Centers for Disease Control and Prevention and the World Health Organization emphasized vaccination against COVID-19 because physical distancing proved inadequate to mitigate death, illness, and massive economic loss.

OBJECTIVE

This study aimed to investigate Korean citizens' perceptions of vaccines by examining their views on COVID-19 vaccines, their positive and negative perceptions of each vaccine, and ways to enhance policies to increase vaccine acceptance.

METHODS

This cross-sectional study analyzed posts on NAVER and Instagram to examine Korean citizens' perception of COVID-19 vaccines. The keywords searched were "vaccine," "AstraZeneca," and "Pfizer." In total 8100 posts in NAVER and 5291 posts in Instagram were sampled through web crawling. Morphology analysis was performed, overlapping or meaningless words were removed, sentiment analysis was implemented, and 3 public health professionals reviewed the results.

RESULTS

The findings revealed a negative perception of COVID-19 vaccines; of the words crawled, the proportion of negative words for AstraZeneca was 71.0% (476/670) and for Pfizer was 56.3% (498/885). Among words crawled with "vaccine," "good" ranked first, with a frequency of 13.43% (312/2323). Meanwhile, "side effect" ranked highest, with a frequency of 29.2% (163/559) for "AstraZeneca," but 0.6% (4/673) for "Pfizer." With "vaccine," positive words were more frequently used, whereas with "AstraZeneca" and "Pfizer" negative words were prevalent.

CONCLUSIONS

There is a negative perception of AstraZeneca and Pfizer vaccines in Korea, with 1 in 4 people refusing vaccination. To address this, accurate information needs to be shared about vaccines including AstraZeneca, and the experiences of those vaccinated. Furthermore, government communication about risk management is required to increase the AstraZeneca vaccination rate for herd immunity before the vaccine expires.

School Attendance, Asthma Risk, and COVID-19 in Children

In the face of tremendous uncertainty during the current pandemic, there is a need for clear and consistent recommendations and an understanding of the evidence in general, and for families of children with allergic conditions. A common concern of parents of children with asthma is the risk that in-person learning poses during the pandemic. This Rostrum examines the actual risk of in-person learning among children with asthma during novel coronavirus disease 2019 (COVID-19), the discrepancy between perceived and actual risk, the contributing factors to this discrepancy, and possible solutions to narrow this divide. Overall, the evidence does not support that children with asthma are at an increased risk of COVID-19 morbidity or mortality compared with children without asthma. Asthma medications do not appear to contribute to incidence or severity of COVID-19 disease. However, there is a high perceived risk of in-person learning that is partially related to how it is portrayed in the media. There is little guidance regarding transitioning asthmatic children back to school and how to properly counsel on mediation of risk. There are differences regionally and locally around school reopening, exemptions, and their implementation. To narrow the divide between perceived and actual risk, clear consistent and ongoing communication will be necessary.

Large-Scale, Language-Agnostic Discourse Classification of Tweets During COVID-19

Quantifying the characteristics of public attention is an essential prerequisite for appropriate crisis management during severe events such as pandemics. For this purpose, we propose language-agnostic tweet representations to perform large-scale Twitter discourse classification with machine learning. Our analysis on more than 26 million coronavirus disease 2019 (COVID-19) tweets shows that large-scale surveillance of public discourse is feasible with computationally lightweight classifiers by out-of-the-box utilization of these representations.

Reaction of Juntendo Shizuoka Hospital at Izu Peninsula to Typhoon Hagibis (2019) and an Analysis of Twitter Concerning Izunokuni City

OBJECTIVE

This research was carried out to report the activity of the headquarters for disaster control at our hospital, and investigate the usefulness of obtaining information from Twitter when Typhoon Hagibis hit Izu Peninsula.

METHODS

First, we recounted the activity of the headquarters for disaster control. We then collected information from Twitter from October 12-14, 2019, using the keyword Izunokuni.' We took into consideration the contents of Twitter user posts as well as the number of reactions ('retweets' and 'likes'). Twitter information was classified into photo (+) and (-) groups, depending on whether or not the post had a photo or video included. The number of reactions between the two groups was then analyzed.

RESULTS

We counted 122 Twitter posts containing Hagibis-related information for Izunokuni City. The average number of both 'retweets' and 'likes' in the photo (+) group were significantly greater than those in the photo (-) group. All photos and videos depicted specific places in Izunokuni City and included actual footage of disaster scenes or local warning signs.

CONCLUSION

Based on our experience at the headquarters for disaster control during Typhoon Hagibis, Twitter was considered to be a useful tool for obtaining local disaster information based on its timeliness, reality and specificity.

Prediction of outcome of treatment of acute severe ulcerative colitis using principal component analysis and artificial intelligence

Background and Aim

About 15% patients with acute severe ulcerative colitis (UC) fail to respond to medical treatment and may require colectomy. An early prediction of response may help the treating team and the patients and their family to prepare for alternative treatment options.

Methods

Data of 263 patients (mean age 37.0 ± 14.0-years, 176, 77% male) with acute severe UC admitted during a 12-year period were used to study predictors of response using univariate analysis, multivariate linear principal component analysis (PCA), and nonlinear artificial neural network (ANN).

Results

Of 263 patients, 231 (87.8%) responded to the initial medical treatment that included oral prednisolone (n = 14, 5.3%), intravenous (IV) hydrocortisone (n = 238, 90.5%), IV cyclosporine (n = 9, 3.4%), and inflixmab (n = 2, 0.7%), and 28 (10.6%) did not respond and the remaining 4 (1.5%) died, all of whom did were also nonresponders. Nonresponding patients had to stay longer in the hospital and died more often. On univariate analysis, the presence of complications, the need for use of cyclosporin, lower Hb, platelets, albumin, serum potassium, and higher C-reactive protein were predictors of nonresponse. Hb and albumin were strong predictive factors on both PCA and ANN. Though the nonlinear modeling using ANN had a good predictive accuracy for the response, its accuracy for predicting nonresponse was lower.

Conclusion

It is possible to predict the response to medical treatment in patients with UC using linear and nonlinear modeling technique. Serum albumin and Hb are strong predictive factors.