Leveraging Artificial Intelligence to Predict Health Belief Model and COVID-19 Vaccine Uptake Using Survey Text from US Nurses.

We investigated how artificial intelligence (AI) reveals factors shaping COVID-19 vaccine hesitancy among healthcare providers by examining their open-text comments. We conducted a longitudinal survey starting in Spring of 2020 with 38,788 current and former female nurses in three national cohorts to assess how the pandemic has affected their livelihood. In January and March-April 2021 surveys, participants were invited to contribute open-text comments and answer specific questions about COVID-19 vaccine uptake. A closed-ended question in the survey identified vaccine-hesitant (VH) participants who either had no intention or were unsure of receiving a COVID-19 vaccine. We collected 1970 comments from VH participants and trained two machine learning (ML) algorithms to identify behavioral factors related to VH. The first predictive model classified each comment into one of three health belief model (HBM) constructs (barriers, severity, and susceptibility) related to adopting disease prevention activities. The second predictive model used the words in January comments to predict the vaccine status of VH in March-April 2021; vaccine status was correctly predicted 89% of the time. Our results showed that 35% of VH participants cited barriers, 17% severity, and 7% susceptibility to receiving a COVID-19 vaccine. Out of the HBM constructs, the VH participants citing a barrier, such as allergic reactions and side effects, had the most associated change in vaccine status from VH to later receiving a vaccine.

Predicting Patient Satisfaction With Medications for Treating Opioid Use Disorder: Case Study Applying Natural Language Processing to Reviews of Methadone and Buprenorphine/Naloxone on Health-Related Social Media.

Background: Medication-assisted treatment (MAT) is an effective method for treating opioid use disorder (OUD), which combines behavioral therapies with one of three Food and Drug Administration-approved medications: methadone, buprenorphine, and naloxone. While MAT has been shown to be effective initially, there is a need for more information from the patient perspective about the satisfaction with medications. Existing research focuses on patient satisfaction with the entirety of the treatment, making it difficult to determine the unique role of medication and overlooking the views of those who may lack access to treatment due to being uninsured or concerns over stigma. Studies focusing on patients' perspectives are also limited by the lack of scales that can efficiently collect self-reports across domains of concerns. Objective: A broad survey of patients' viewpoints can be obtained through social media and drug review forums, which are then assessed using automated methods to discover factors associated with medication satisfaction. Because the text is unstructured, it may contain a mix of formal and informal language. The primary aim of this study was to use natural language processing methods on text posted on health-related social media to detect patients' satisfaction with two well-studied OUD medications: methadone and buprenorphine/naloxone. Methods: We collected 4353 patient reviews of methadone and buprenorphine/naloxone from 2008 to 2021 posted on WebMD and Drugs.com. To build our predictive models for detecting patient satisfaction, we first employed different analyses to build four input feature sets using the vectorized text, topic models, duration of treatment, and biomedical concepts by applying MetaMap. We then developed six prediction models: logistic regression, Elastic Net, least absolute shrinkage and selection operator, random forest classifier, Ridge classifier, and extreme gradient boosting to predict patients' satisfaction. Lastly, we compared the prediction models' performance over different feature sets. Results: Topics discovered included oral sensation, side effects, insurance, and doctor visits. Biomedical concepts included symptoms, drugs, and illnesses. The F-score of the predictive models across all methods ranged from 89.9% to 90.8%. The Ridge classifier model, a regression-based method, outperformed the other models. Conclusions: Assessment of patients' satisfaction with opioid dependency treatment medication can be predicted using automated text analysis. Adding biomedical concepts such as symptoms, drug name, and illness, along with the duration of treatment and topic models, had the most benefits for improving the prediction performance of the Elastic Net model compared to other models. Some of the factors associated with patient satisfaction overlap with domains covered in medication satisfaction scales (eg, side effects) and qualitative patient reports (eg, doctors' visits), while others (insurance) are overlooked, thereby underscoring the value added from processing text on online health forums to better understand patient adherence.

Visualizing the Interpretation of a Criteria-Driven System That Automatically Evaluates the Quality of Health News: Exploratory Study of 2 Approaches.

BACKGROUND: Machine learning techniques have been shown to be efficient in identifying health misinformation, but the results may not be trusted unless they can be justified in a way that is understandable. OBJECTIVE: This study aimed to provide a new criteria-based system to assess and justify health news quality. Using a subset of an existing set of criteria, this study compared the feasibility of 2 alternative methods for adding interpretability. Both methods used classification and highlighting to visualize sentence-level evidence. METHODS: A total of 3 out of 10 well-established criteria were chosen for experimentation, namely whether the health news discussed the costs of the intervention (the cost criterion), explained or quantified the harms of the intervention (the harm criterion), and identified the conflicts of interest (the conflict criterion). The first step of the experiment was to automate the evaluation of the 3 criteria by developing a sentence-level classifier. We tested Logistic Regression, Naive Bayes, Support Vector Machine, and Random Forest algorithms. Next, we compared the 2 visualization approaches. For the first approach, we calculated word feature weights, which explained how classification models distill keywords that contribute to the prediction; then, using the local interpretable model-agnostic explanation framework, we selected keywords associated with the classified criterion at the document level; and finally, the system selected and highlighted sentences with keywords. For the second approach, we extracted sentences that provided evidence to support the evaluation result from 100 health news articles; based on these results, we trained a typology classification model at the sentence level; and then, the system highlighted a positive sentence instance for the result justification. The number of sentences to highlight was determined by a preset threshold empirically determined using the average accuracy. RESULTS: The automatic evaluation of health news on the cost, harm, and conflict criteria achieved average area under the curve scores of 0.88, 0.76, and 0.73, respectively, after 50 repetitions of 10-fold cross-validation. We found that both approaches could successfully visualize the interpretation of the system but that the performance of the 2 approaches varied by criterion and highlighting the accuracy decreased as the number of highlighted sentences increased. When the threshold accuracy was ≥75%, this resulted in a visualization with a variable length ranging from 1 to 6 sentences. CONCLUSIONS: We provided 2 approaches to interpret criteria-based health news evaluation models tested on 3 criteria. This method incorporated rule-based and statistical machine learning approaches. The results suggested that one might visually interpret an automatic criterion-based health news quality evaluation successfully using either approach; however, larger differences may arise when multiple quality-related criteria are considered. This study can increase public trust in computerized health information evaluation.

Development of a periodontitis risk assessment model for primary care providers in an interdisciplinary setting.

BACKGROUND: Periodontitis (PD), a form of gum disease, is a major public health concern as it is globally prevalent and harms both individual quality of life and economic productivity. Global cost in lost productivity is estimated at US$54 billion annually. Moreover, current PD assessment applies only after the damage has already occurred. OBJECTIVE: This study proposes and tests a new PD risk assessment model applicable at point-of-care, using supervised machine learning methods. METHODS: We compare the performance of five algorithms using retrospective clinical data: Naïve Bayes (NB), Logistic Regression (LR), Support Vector Machine (SVM), Artificial Neural Network (ANN), and Decision Tree (DT). RESULTS: DT and ANN demonstrated higher accuracy in classifying the patients with high or low PD risk as compared to NB, LR and SVM. The resultant model with DT showed a sensitivity of 87.08% (95% CI 84.12% to 89.76%) and specificity of 93.5% (95% CI 91% to 95.49%). CONCLUSIONS: A predictive model with high sensitivity and specificity to stratify individuals into low and high PD risk tiers was developed. Validation in other populations will inform translational value of this approach and its potential applicability as clinical decision support tool.

Assessing Unmet Information Needs of Breast Cancer Survivors: Exploratory Study of Online Health Forums Using Text Classification and Retrieval.

BACKGROUND: Patient education materials given to breast cancer survivors may not be a good fit for their information needs. Needs may change over time, be forgotten, or be misreported, for a variety of reasons. An automated content analysis of survivors' postings to online health forums can identify expressed information needs over a span of time and be repeated regularly at low cost. Identifying these unmet needs can guide improvements to existing education materials and the creation of new resources. OBJECTIVE: The primary goals of this project are to assess the unmet information needs of breast cancer survivors from their own perspectives and to identify gaps between information needs and current education materials. METHODS: This approach employs computational methods for content modeling and supervised text classification to data from online health forums to identify explicit and implicit requests for health-related information. Potential gaps between needs and education materials are identified using techniques from information retrieval. RESULTS: We provide a new taxonomy for the classification of sentences in online health forum data. 260 postings from two online health forums were selected, yielding 4179 sentences for coding. After annotation of data and training alternative one-versus-others classifiers, a random forest-based approach achieved F1 scores from 66% (Other, dataset2) to 90% (Medical, dataset1) on the primary information types. 136 expressions of need were used to generate queries to indexed education materials. Upon examination of the best two pages retrieved for each query, 12% (17/136) of queries were found to have relevant content by all coders, and 33% (45/136) were judged to have relevant content by at least one. CONCLUSIONS: Text from online health forums can be analyzed effectively using automated methods. Our analysis confirms that breast cancer survivors have many information needs that are not covered by the written documents they typically receive, as our results suggest that at most a third of breast cancer survivors' questions would be addressed by the materials currently provided to them.

Scientific authorship and collaboration network analysis on malaria research in Benin: papers indexed in the web of science (1996-2016).

BACKGROUND: To sustain the critical progress made, prioritization and a multidisciplinary approach to malaria research remain important to the national malaria control program in Benin. To document the structure of the malaria collaborative research in Benin, we analyze authorship of the scientific documents published on malaria from Benin. METHODS: We collected bibliographic data from the Web Of Science on malaria research in Benin from January 1996 to December 2016. From the collected data, a mulitigraph co-authorship network with authors representing vertices was generated. An edge was drawn between two authors when they co-author a paper. We computed vertex degree, betweenness, closeness, and eigenvectors among others to identify prolific authors. We further assess the weak points and how information flow in the network. Finally, we perform a hierarchical clustering analysis, and Monte-Carlo simulations. RESULTS: Overall, 427 publications were included in this study. The generated network contained 1792 authors and 116,388 parallel edges which converted in a weighted graph of 1792 vertices and 95,787 edges. Our results suggested that prolific authors with higher degrees tend to collaborate more. The hierarchical clustering revealed 23 clusters, seven of which form a giant component containing 94% of all the vertices in the network. This giant component has all the characteristics of a small-world network with a small shortest path distance between pairs of three, a diameter of 10 and a high clustering coefficient of 0.964. However, Monte-Carlo simulations suggested our observed network is an unusual type of small-world network. Sixteen vertices were identified as weak articulation points within the network. CONCLUSION: The malaria research collaboration network in Benin is a complex network that seems to display the characteristics of a small-world network. This research reveals the presence of closed research groups where collaborative research likely happens only between members. Interdisciplinary collaboration tends to occur at higher levels between prolific researchers. Continuously supporting, stabilizing the identified key brokers and most productive authors in the Malaria research collaborative network is an urgent need in Benin. It will foster the malaria research network and ensure the promotion of junior scientists in the field.

Would mobile health be a solution to rehospitalization?

It has been well-established that social environmental factors can increase the risk of rehospitalization for people receiving home healthcare services. For caregivers who might be challenged to keep up with sometimes unfamiliar health monitoring tasks or to know when to seek help, mobile health technology offers the potential to enhance the skills of informal caregivers and to improve the communication between home and clinical care. This paper described our recent work to determine the usability, functionality, and style of interaction that would be needed to provide an effective and well-accepted tool. Caregivers would likely adopt new mobile health tools, as long as care is taken to eliminate potential barriers, for example, by providing adequate training, and to include design aspects that enhance one's motivation to use a tool, such as by supporting autonomy and engagement.

Toward automated classification of consumers' cancer-related questions with a new taxonomy of expected answer types.

This article examines methods for automated question classification applied to cancer-related questions that people have asked on the web. This work is part of a broader effort to provide automated question answering for health education. We created a new corpus of consumer-health questions related to cancer and a new taxonomy for those questions. We then compared the effectiveness of different statistical methods for developing classifiers, including weighted classification and resampling. Basic methods for building classifiers were limited by the high variability in the natural distribution of questions and typical refinement approaches of feature selection and merging categories achieved only small improvements to classifier accuracy. Best performance was achieved using weighted classification and resampling methods, the latter yielding an accuracy of F1 = 0.963. Thus, it would appear that statistical classifiers can be trained on natural data, but only if natural distributions of classes are smoothed. Such classifiers would be useful for automated question answering, for enriching web-based content, or assisting clinical professionals to answer questions.

Information gathering and technology use among low-income minority men at risk for prostate cancer.

Health communication researchers, public health workers, and health professionals must learn more about the health information-gathering behavior of low-income minority men at risk for prostate cancer in order to share information effectively with the population. In collaboration with the Milwaukee Health Department Men's Health Referral Network, a total of 90 low-income adult men were recruited to complete a survey gauging information sources, seeking behavior, use of technology, as well as prostate cancer awareness and screening behavior. Results indicated participants primarily relied on health professionals, family, and friends for information about general issues of health as well as prostate cancer. The Internet was the least relied on source of information. A hierarchical regression indicated interpersonal information sources such as family or friends to be the only significant predictor enhancing prostate cancer awareness, controlling for other sources of information. Prostate screening behaviors were predicted by reliance on not only medical professionals but also the Internet. Practical implications of the study are discussed.

Information needs, seeking behaviors, and support among low-income expectant women.

Previous studies have consistently found associations between low income and infant health outcomes. Moreover, although health information-seeking is a maternal behavior related to improved health outcomes, little is known about the health information-seeking behaviors and information needs of low-income pregnant women. The purpose of the current investigation was to examine the information needs, information-seeking behaviors, and perceived informational support of low-income pregnant women. Accordingly, the study recruited 63 expectant women enrolled in a subsidized prenatal care program in Milwaukee, Wisconsin, during two time periods: March-May 2011 and October-December 2011. Results indicated that participants relied heavily upon interpersonal sources of information, especially family and the father of the baby; rarely used the Internet for health-related information; and desired information beyond infant and maternal health, such as finding jobs and accessing community/government resources. Participants who used family members as primary sources of information also had significantly increased levels of perceived informational support and reduced uncertainty about pregnancy. Our findings have implications for the dissemination of pregnancy-related health information among low-income expectant women.

A two-way text-messaging system answering health questions for low-income pregnant women.

OBJECTIVE: The purpose of the study was to gauge the effectiveness of a low-cost, automated, two-way text-messaging system to distribute pregnancy and health-related information to low-income expectant women. METHODS: In total, 20 participants were recruited for a one-month intervention involving the use of cell phones to text pregnancy-related questions to the system. Participants received either a direct answer or encouragement to seek answers from health care providers. Pre- and post-tests as well as a focus group at the end of the intervention were conducted. RESULTS: Participants uniformly found the system easy to use and accessible. Using the system increased levels of perceived pregnancy-related knowledge and facilitated patient-provider communication. Moreover, participants reported significant reductions in stress and depression and improved mental health after using the system. The system responded to most known questions quickly and accurately, and also encountered many new topics and linguistic expressions. CONCLUSION: Overall, the data indicated that the text messaging system offered psychological benefits and promoted health communication by providing health information and encouraging patient-provider communication. PRACTICE IMPLICATIONS: An automated, two-way text messaging system is an efficient, cost-effective, and acceptable method for providing health information to low-income pregnant women.

The biomedical discourse relation bank.

BACKGROUND: Identification of discourse relations, such as causal and contrastive relations, between situations mentioned in text is an important task for biomedical text-mining. A biomedical text corpus annotated with discourse relations would be very useful for developing and evaluating methods for biomedical discourse processing. However, little effort has been made to develop such an annotated resource. RESULTS: We have developed the Biomedical Discourse Relation Bank (BioDRB), in which we have annotated explicit and implicit discourse relations in 24 open-access full-text biomedical articles from the GENIA corpus. Guidelines for the annotation were adapted from the Penn Discourse TreeBank (PDTB), which has discourse relations annotated over open-domain news articles. We introduced new conventions and modifications to the sense classification. We report reliable inter-annotator agreement of over 80% for all sub-tasks. Experiments for identifying the sense of explicit discourse connectives show the connective itself as a highly reliable indicator for coarse sense classification (accuracy 90.9% and F1 score 0.89). These results are comparable to results obtained with the same classifier on the PDTB data. With more refined sense classification, there is degradation in performance (accuracy 69.2% and F1 score 0.28), mainly due to sparsity in the data. The size of the corpus was found to be sufficient for identifying the sense of explicit connectives, with classifier performance stabilizing at about 1900 training instances. Finally, the classifier performs poorly when trained on PDTB and tested on BioDRB (accuracy 54.5% and F1 score 0.57). CONCLUSION: Our work shows that discourse relations can be reliably annotated in biomedical text. Coarse sense disambiguation of explicit connectives can be done with high reliability by using just the connective as a feature, but more refined sense classification requires either richer features or more annotated data. The poor performance of a classifier trained in the open domain and tested in the biomedical domain suggests significant differences in the semantic usage of connectives across these domains, and provides robust evidence for a biomedical sublanguage for discourse and the need to develop a specialized biomedical discourse annotated corpus. The results of our cross-domain experiments are consistent with related work on identifying connectives in BioDRB.