Background. Artificial intelligence (AI) has developed rapidly, and its application extends to clinical decision support system (CDSS) for improving healthcare quality. However, the interpretability of AI‐driven CDSS poses significant challenges to widespread application. Objective. This study is a review of the knowledge‐based and data‐based CDSS literature regarding interpretability in health care. It highlights the relevance of interpretability for CDSS and the area for improvement from technological and medical perspectives. Methods. A systematic search was conducted on the interpretability‐related literature published from 2011 to 2020 and indexed in the five databases: Web of Science, PubMed, ScienceDirect, Cochrane, and Scopus. Journal articles that focus on the interpretability of CDSS were included for analysis. Experienced researchers also participated in manually reviewing the selected articles for inclusion/exclusion and categorization. Results. Based on the inclusion and exclusion criteria, 20 articles from 16 journals were finally selected for this review. Interpretability, which means a transparent structure of the model, a clear relationship between input and output, and explainability of artificial intelligence algorithms, is essential for CDSS application in the healthcare setting. Methods for improving the interpretability of CDSS include ante‐hoc methods such as fuzzy logic, decision rules, logistic regression, decision trees for knowledge‐based AI, and white box models, post hoc methods such as feature importance, sensitivity analysis, visualization, and activation maximization for black box models. A number of factors, such as data type, biomarkers, human‐AI interaction, needs of clinicians, and patients, can affect the interpretability of CDSS. Conclusions. The review explores the meaning of the interpretability of CDSS and summarizes the current methods for improving interpretability from technological and medical perspectives. The results contribute to the understanding of the interpretability of CDSS based on AI in health care. Future studies should focus on establishing formalism for defining interpretability, identifying the properties of interpretability, and developing an appropriate and objective metric for interpretability; in addition, the user′s demand for interpretability and how to express and provide explanations are also the directions for future research.

Background: The internet has now become part of human life and is constantly changing people's way of life. With the increasing popularity of online health information (OHI), it has been found that OHI can affect the physician-patient relationship by influencing patient behaviors. Objective: This study aims to systematically investigate the impact of OHI-seeking behavior on the physician-patient relationship. Methods: Literature retrieval was conducted on 4 databases (Web of Science, PubMed, China National Knowledge Infrastructure, SinoMed), and the time limit for literature publication was before August 1, 2021. Results: We selected 53 target papers (42 [79%] English papers and 11 [21%] Chinese papers) that met the inclusion criteria. Of these, 31 (58%) papers believe that patients’ OHI behavior can enable them to participate in their own medical care, improve patient compliance, and improve the physician-patient relationship. In addition, 14 (26%) papers maintain a neutral attitude, some believing that OHI behavior has no significant effect on doctors and patients and others believing that due to changes in the factors affecting OHI behavior, they will have a negative or a positive impact. Furthermore, 8 (15%) papers believe that OHI search behavior has a negative impact on doctors and patients, while 6 (11%) papers show that OHI reduces Chinese patients’ trust in doctors. Conclusions: Our main findings showed that (1) OHI-seeking behavior has an impact on patients' psychology, behavior, and evaluation of doctors; (2) whether patients choose to discuss OHI with doctors has different effects on the physician-patient relationship; and (3) the negative impact of OHI on China’s internet users is worthy of attention. Due to the low quality of OHI, poor health information literacy, short physician-patient communication time, and various types of negative news, patients' trust in doctors has declined, thus affecting the physician-patient relationship. Improvement of people's health information literacy and the quality of OHI are important factors that promote the positive impact of OHI on the physician-patient relationship.

Aim: With the improvement in people's living standards, the incidence of chronic renal failure (CRF) is increasing annually. The increase in the number of patients with CRF has significantly increased pressure on China's medical budget. Predicting hospitalization expenses for CRF can provide guidance for effective allocation and control of medical costs. The purpose of this study was to use the random forest (RF) method and least absolute shrinkage and selection operator (LASSO) regression to predict personal hospitalization expenses of hospitalized patients with CRF and to evaluate related influencing factors.Methods: The data set was collected from the first page of data of the medical records of three tertiary first-class hospitals for the whole year of 2016. Factors influencing hospitalization expenses for CRF were analyzed. Random forest and least absolute shrinkage and selection operator regression models were used to establish a prediction model for the hospitalization expenses of patients with CRF, and comparisons and evaluations were carried out.Results: For CRF inpatients, statistically significant differences in hospitalization expenses were found for major procedures, medical payment method, hospitalization frequency, length of stay, number of other diagnoses, and number of procedures. The R² of LASSO regression model and RF regression model are 0.6992 and 0.7946, respectively. The mean absolute error (MAE) and root mean square error (RMSE) of the LASSO regression model were 0.0268 and 0.043, respectively, and the MAE and RMSE of the RF prediction model were 0.0171 and 0.0355, respectively. In the RF model, and the weight of length of stay was the highest (0.730).Conclusions: The hospitalization expenses of patients with CRF are most affected by length of stay. The RF prediction model is superior to the LASSO regression model and can be used to predict the hospitalization expenses of patients with CRF. Health administration departments may consider formulating accurate individualized hospitalization expense reimbursement mechanisms accordingly.

The objective of this study was to analyze knowledge mapping and demonstrate the status quo, intellectual base, and hotspots in the field of medication literacy. Using the data from Web of Science Core database, we constructed a knowledge map to visualize medication literacy using CiteSpace, which revealed the power of the studies, core authors and journals, intellectual base, and hotspots in this field. According to an analysis of 2025 literature reports, the stronger studies were mainly conducted at research institutions of higher education in the United States. Core author groups with a higher influence were not identified. The core journals included Patient Educ Couns and Fam Med. The health literacy studies served as the foundation for the medication literacy studies. The keywords formed 13 clusters including 5 major clusters. The topics in medication literacy study focused on instruments assessing medication literacy, measurement and assessment of medication literacy, medication literacy for the prevention of chronic disease and medication treatment adherence, medication literacy education, and family practice. This study provides an insight into medication literacy and valuable information for medication literacy researchers to identify new perspectives on potential collaborators and cooperative institutions and hotspots.

Cyberchondria is considered “the anxiety-amplifying effects of online health-related searches.” During the COVID-19 pandemic, people are likely to search health-related information online for reassurance because of fear and related physical symptoms, while cyberchondria may be triggered due to the escalation of health anxiety, different online seeking behavior preference, information overload, and insufficient e-health literacy. This study aimed to investigate the status and influencing factors of cyberchondria in residents in China during the epidemic period of COVID-19. The participants were 674 community residents of Nanyang city surveyed from February 1 to 15, 2020. We administered online measures, including the Chinese Short Form of the Cyberchondria Severity Scale (C-CSS-12), Short Health Anxiety Inventory (SHAI), eHealth Literacy Scale (eHEALS), Patient Health Questionnaire-15 (PHQ-15), and COVID-19-related online information seeking behavior questionnaire. In our study, the average C-CSS-12 total score of residents was 30.65 ± 11.53 during the virus epidemic; 25% of participants scored 22 or below, 50% scored 23 to 38, and 21.9% scored 39 to 60. The SHAI total score (β = 0.598 > 0, P < 0.001), the use of general search engines (β = 1.867 > 0, P = 0.039), and searching for information on how to diagnose COVID-19 (β = 2.280 > 0, P = 0.020) were independent risk factors for cyberchondria, while searching lasting less than 10 min each (β = −2.992 < 0, P = 0.048), the use of traditional media digital platforms (β = −1.650 < 0, P = 0.024) and professional medical communication platforms (β = −4.189 < 0, P = 0.007) were independent protective factors. Our findings showed that nearly a quarter of the participants scored 39 or higher on the C-CSS-12 in Nanyang city during the pandemic, which should be taken seriously. Health anxiety and COVID-19-related online information seeking behavior including online duration, topics and choice on different information channels were important influencing factors of cyberchondria. These findings have implications for further research and clinical practice on cyberchondria in China.

Background: In spite of strict regulation of coronavirus disease 2019 (COVID-19) preventive measures and containment in China, there are still confirmed cases sporadically occurring in many cities. College students live in groups and have active social activities so that it will trigger a serious public health event once an infection event occurs. Thus, identifying the status and related factors of protective behaviors among them after receiving vaccination will be crucial for epidemic control. This study aimed to gather information on the protective behaviors and to identify the associations of COVID-19 risk perception, eHealth literacy, and protective behaviors for Chinese college students following vaccination.Methods: A cross-sectional survey of college students engaged in protective behaviors post vaccination was conducted using the COVID-19 risk perception scale, eHealth literacy scale, and protective behaviors following vaccination questionnaire in one of the groups. Multiple linear regression analysis was used to confirm the correlation among the COVID-19 risk perception, eHealth literacy, and protective behaviors for Chinese college students.Results: A total of 5,641 Chinese college students were included. Male students comprised 59.01% with an average age of (21.39 ± 2.75) years and most students rating their health as very good (44.85%) or pretty good (46.98%). A smaller percentage (13.76%) believed that they would likely or most likely be infected with COVID-19 after getting vaccinated. In addition, more than 1 in 10 (10.35%) college students had ever suspected to suffer from post-vaccination reactions following the COVID-19 vaccination. The mean score of protective behaviors was 26.06 ± 3.97. Approximately one-third (30.42%) of the students always or often did not wear a mask when going out. Some college students (29.25%) did not maintain distance of at least 1 m from others in social situations. Older female college students who were in good health and perceived as being at a low risk of getting infected with COVID-19, and those never suspected to suffer from post-vaccination reactions expected to engage in post-vaccination protective measures. Those with a higher level of perceived risk, severe risk perception and eHealth literacy, and a lower level of unknown risk perception were more likely to engage in further protective behaviors after getting vaccinated.Conclusions: Overall, the level of protective behaviors among the Chinese college students following vaccination could be improved, especially for male, younger college students in poor health. This study revealed the predictive effects of risk perception and eHealth literacy on protective behaviors, recommending that the negative and positive effects of risk perception should be balanced in epidemic risk management, and eHealth literacy promotion should also be emphasized for public health and social measures.