Comparative outcomes of AI-assisted ChatGPT and face-to-face consultations in infertility patients: a cross-sectional study.

BACKGROUND: With the advent of artificial intelligence (AI) in healthcare, digital platforms like ChatGPT offer innovative alternatives to traditional medical consultations. This study seeks to understand the comparative outcomes of AI-assisted ChatGPT consultations and conventional face-to-face interactions among infertility patients. METHODS: A cross-sectional study was conducted involving 120 infertility patients, split evenly between those consulting via ChatGPT and traditional face-to-face methods. The primary outcomes assessed were patient satisfaction, understanding, and consultation duration. Secondary outcomes included demographic information, clinical history, and subsequent actions post-consultation. RESULTS: While both consultation methods had a median age of 34 years, patients using ChatGPT reported significantly higher satisfaction levels (median 4 out of 5) compared to face-to-face consultations (median 3 out of 5; p < 0.001). The ChatGPT group also experienced shorter consultation durations, with a median difference of 12.5 minutes (p < 0.001). However, understanding, demographic distributions, and subsequent actions post-consultation were comparable between the two groups. CONCLUSIONS: AI-assisted ChatGPT consultations offer a promising alternative to traditional face-to-face consultations in assisted reproductive medicine. While patient satisfaction was higher and consultation durations were shorter with ChatGPT, further studies are required to understand the long-term implications and clinical outcomes associated with AI-driven medical consultations. Key Messages What is already known on this topic:  Artificial intelligence (AI) applications, such as ChatGPT, have shown potential in various healthcare settings, including primary care and mental health support. Infertility is a significant global health issue that requires extensive consultations, often facing challenges such as long waiting times and varied patient satisfaction. Previous studies suggest that AI can offer personalized care and immediate feedback, but its efficacy compared with traditional consultations in reproductive medicine was not well-studied. What this study adds:  This study demonstrates that AI-assisted ChatGPT consultations result in significantly higher patient satisfaction and shorter consultation durations compared with traditional face-to-face consultations among infertility patients. Both consultation methods were comparable in terms of patient understanding, demographic distributions, and subsequent actions postconsultation. How this study might affect research, practice, or policy:  The findings suggest that AI-driven consultations could serve as an effective and efficient alternative to traditional methods, potentially reducing consultation times and improving patient satisfaction in reproductive medicine. Further research could explore the long-term impacts and broader applications of AI in clinical settings, influencing future healthcare practices and policies toward integrating AI technologies.

Quality and Dependability of ChatGPT and DingXiangYuan Forums for Remote Orthopedic Consultations: Comparative Analysis.

BACKGROUND: The widespread use of artificial intelligence, such as ChatGPT (OpenAI), is transforming sectors, including health care, while separate advancements of the internet have enabled platforms such as China's DingXiangYuan to offer remote medical services. OBJECTIVE: This study evaluates ChatGPT-4's responses against those of professional health care providers in telemedicine, assessing artificial intelligence's capability to support the surge in remote medical consultations and its impact on health care delivery. METHODS: We sourced remote orthopedic consultations from "Doctor DingXiang," with responses from its certified physicians as the control and ChatGPT's responses as the experimental group. In all, 3 blindfolded, experienced orthopedic surgeons assessed responses against 7 criteria: "logical reasoning," "internal information," "external information," "guiding function," "therapeutic effect," "medical knowledge popularization education," and "overall satisfaction." We used Fleiss κ to measure agreement among multiple raters. RESULTS: Initially, consultation records for a cumulative count of 8 maladies (equivalent to 800 cases) were gathered. We ultimately included 73 consultation records by May 2023, following primary and rescreening, in which no communication records containing private information, images, or voice messages were transmitted. After statistical scoring, we discovered that ChatGPT's "internal information" score (mean 4.61, SD 0.52 points vs mean 4.66, SD 0.49 points; P=.43) and "therapeutic effect" score (mean 4.43, SD 0.75 points vs mean 4.55, SD 0.62 points; P=.32) were lower than those of the control group, but the differences were not statistically significant. ChatGPT showed better performance with a higher "logical reasoning" score (mean 4.81, SD 0.36 points vs mean 4.75, SD 0.39 points; P=.38), "external information" score (mean 4.06, SD 0.72 points vs mean 3.92, SD 0.77 points; P=.25), and "guiding function" score (mean 4.73, SD 0.51 points vs mean 4.72, SD 0.54 points; P=.96), although the differences were not statistically significant. Meanwhile, the "medical knowledge popularization education" score of ChatGPT was better than that of the control group (mean 4.49, SD 0.67 points vs mean 3.87, SD 1.01 points; P<.001), and the difference was statistically significant. In terms of "overall satisfaction," the difference was not statistically significant between the groups (mean 8.35, SD 1.38 points vs mean 8.37, SD 1.24 points; P=.92). According to how Fleiss κ values were interpreted, 6 of the control group's score points were classified as displaying "fair agreement" (P<.001), and 1 was classified as showing "substantial agreement" (P<.001). In the experimental group, 3 points were classified as indicating "fair agreement," while 4 suggested "moderate agreement" (P<.001). CONCLUSIONS: ChatGPT-4 matches the expertise found in DingXiangYuan forums' paid consultations, excelling particularly in scientific education. It presents a promising alternative for remote health advice. For health care professionals, it could act as an aid in patient education, while patients may use it as a convenient tool for health inquiries.

Online Medical Consultation Service-Oriented Recommendations: Systematic Review.

BACKGROUND: Online health communities have given rise to a new e-service known as online medical consultations (OMCs), enabling remote interactions between physicians and patients. To address challenges, such as patient information overload and uneven distribution of physician visits, online health communities should develop OMC-oriented recommenders. OBJECTIVE: We aimed to comprehensively investigate what paradigms lead to the success of OMC-oriented recommendations. METHODS: A literature search was conducted through e-databases, including PubMed, ACM Digital Library, Springer, and ScienceDirect, from January 2011 to December 2023. This review included all papers directly and indirectly related to the topic of health care-related recommendations for online services. RESULTS: The search identified 611 articles, of which 26 (4.3%) met the inclusion criteria. Despite the growing academic interest in OMC recommendations, there remains a lack of consensus among researchers on the definition of e-service-oriented recommenders. The discussion highlighted 3 key factors influencing recommender success: features, algorithms, and metrics. It advocated for moving beyond traditional e-commerce-oriented recommenders to establish an innovative theoretical framework for e-service-oriented recommenders and addresses critical technical issues regarding 2-sided personalized recommendations. CONCLUSIONS: This review underscores the essence of e-services, particularly in knowledge- and labor-intensive domains such as OMCs, where patients seek interpretable recommendations due to their lack of domain knowledge and physicians must balance their energy levels to avoid overworking. Our study's findings shed light on the importance of customizing e-service-oriented personalized recommendations to meet the distinct expectations of 2-sided users considering their cognitive abilities, decision-making perspectives, and preferences. To achieve this, a paradigm shift is essential to develop unique attributes and explore distinct content tailored for both parties involved.

Exploring the potential of ChatGPT in medical dialogue summarization: a study on consistency with human preferences.

BACKGROUND: Telemedicine has experienced rapid growth in recent years, aiming to enhance medical efficiency and reduce the workload of healthcare professionals. During the COVID-19 pandemic in 2019, it became especially crucial, enabling remote screenings and access to healthcare services while maintaining social distancing. Online consultation platforms have emerged, but the demand has strained the availability of medical professionals, directly leading to research and development in automated medical consultation. Specifically, there is a need for efficient and accurate medical dialogue summarization algorithms to condense lengthy conversations into shorter versions focused on relevant medical facts. The success of large language models like generative pre-trained transformer (GPT)-3 has recently prompted a paradigm shift in natural language processing (NLP) research. In this paper, we will explore its impact on medical dialogue summarization. METHODS: We present the performance and evaluation results of two approaches on a medical dialogue dataset. The first approach is based on fine-tuned pre-trained language models, such as bert-based summarization (BERTSUM) and bidirectional auto-regressive Transformers (BART). The second approach utilizes a large language models (LLMs) GPT-3.5 with inter-context learning (ICL). Evaluation is conducted using automated metrics such as ROUGE and BERTScore. RESULTS: In comparison to the BART and ChatGPT models, the summaries generated by the BERTSUM model not only exhibit significantly lower ROUGE and BERTScore values but also fail to pass the testing for any of the metrics in manual evaluation. On the other hand, the BART model achieved the highest ROUGE and BERTScore values among all evaluated models, surpassing ChatGPT. Its ROUGE-1, ROUGE-2, ROUGE-L, and BERTScore values were 14.94%, 53.48%, 32.84%, and 6.73% higher respectively than ChatGPT's best results. However, in the manual evaluation by medical experts, the summaries generated by the BART model exhibit satisfactory performance only in the "Readability" metric, with less than 30% passing the manual evaluation in other metrics. When compared to the BERTSUM and BART models, the ChatGPT model was evidently more favored by human medical experts. CONCLUSION: On one hand, the GPT-3.5 model can manipulate the style and outcomes of medical dialogue summaries through various prompts. The generated content is not only better received than results from certain human experts but also more comprehensible, making it a promising avenue for automated medical dialogue summarization. On the other hand, automated evaluation mechanisms like ROUGE and BERTScore fall short in fully assessing the outputs of large language models like GPT-3.5. Therefore, it is necessary to research more appropriate evaluation criteria.

Experiences of receiving a medical consultation - an interview study among hospitalized psychiatric patients.

PURPOSE: People with mental illness often experience more physical health problems, compared to the general population. Therefore, we conducted medical consultations to hospitalized psychiatric patients. The purpose of our study was to investigate experiences of medical consultations from a patient perspective. MATERIALS AND METHODS: We conducted 17 qualitative, semi-structured interviews with hospitalized psychiatric patients who received a medical consultation in the period of February to May 2023. The interviews were analyzed using Braun and Clarke's thematic analysis. RESULTS: Patients perceived the medical consultation as important in bridging the gap between psychiatric and somatic treatment. The consultation created a sense of security, with patients emphasizing the importance of acknowledgment by the medical doctor. However, some patients were affected by their state of mind and their psychiatric treatment, resulting in memory impairment, which prevented fully utilization of the medical consultation. The support from psychiatric staff proved crucial for patients in initiating somatic interventions.The results are presented in the two overarching themes: (1) Experiences of a medical consultation in a psychiatric setting and (2) communication and support as influential factors for benefits of the medical consultation. CONCLUSION: Our interview study reveals that hospitalized psychiatric patients experience a medical consultation as an important initiative. The relationship between the patient and the medical doctor, the patient's state of mind, and the support of the psychiatric staff play significant roles. Incorporating these factors in the medical consultation enables a beneficial outcome that can improve the health of people with mental illness.

Reactogenicity and safety of COVID-19 primary immunisation and booster vaccination regimens: a comparative observational cohort study.

BACKGROUND: Since the beginning of the COVID-19 vaccination campaigns, recommendations regarding the vaccination have been very dynamic. Although the safety and efficacy of different vaccines have been analysed, data were scarce for vaccine regimens combining different vaccines. We therefore aimed to evaluate and compare the perceived reactogenicity and need for medical consultation after the most frequently applied homologous and heterologous COVID-19 vaccination regimens. METHODS: In an observational cohort study, reactogenicity and safety were assessed within a maximum follow-up time of 124 days using web-based surveys. Reactogenicity was assessed for different vaccination regimens 2 weeks after a vaccination (short-term survey). The following surveys, long-term and follow-up surveys, focused on the utilisation of medical services, including those that were not suspected to be vaccine-related. RESULTS: Data of 17,269 participants were analysed. The least local reactions were seen after a ChAdOx1 - ChAdOx1 regimen (32.6%, 95% CI [28.2, 37.2]) and the most after the first dose with mRNA-1273 (73.9%, 95% CI [70.5, 77.2]). Systemic reactions were least frequent in participants with a BNT162b2 booster after a homologous primary immunisation with ChAdOx1 (42.9%, 95% CI [32.1, 54.1]) and most frequent after a ChAdOx1 - mRNA-1273 (85.5%, 95% CI [82.9, 87.8]) and mRNA-1273/mRNA-1273 regimen (85.1%, 95% CI [83.2, 87.0]). In the short-term survey, the most common consequences were medication intake and sick leave (after local reactions 0% to 9.9%; after systemic reactions 4.5% to 37.9%). In the long-term and follow-up surveys, between 8.2 and 30.9% of participants reported consulting a doctor and between 0% and 5.4% seeking hospital care. The regression analyses 124 days after the first and after the third dose showed that the odds for reporting medical consultation were comparable between the vaccination regimens. CONCLUSIONS: Our analysis revealed differences in reactogenicity between the COVID-19 vaccines and vaccination regimens in Germany. The lowest reactogenicity as reported by participants was seen with BNT162b2, especially in homologous vaccination regimens. However, in all vaccination regimens reactogenicity rarely led to medical consultations. Small differences in seeking any medical consultation after 6 weeks diminished during the follow-up period. In the end, none of the vaccination regimens was associated with a higher risk for medical consultation. TRIAL REGISTRATION: DRKS DRKS00025881 ( https://drks.de/search/de/trial/DRKS00025373 ). Registered on 14 October 2021. DRKS DRKS00025373 ( https://drks.de/search/de/trial/DRKS00025881 ). Registered on 21 May 2021. Registered retrospectively.

Patient-reported reactogenicity and safety of COVID-19 vaccinations vs. comparator vaccinations: a comparative observational cohort study.

BACKGROUND: In the course of the SARS-CoV-2 pandemic, multiple vaccines were developed. Little was known about reactogenicity and safety in comparison to established vaccines, e.g. influenza, pneumococcus, or herpes zoster. Therefore, the present study aimed to compare self-reported side effects in persons vaccinated against SARS-CoV-2 with the incidence of side effects in persons receiving one of the established vaccines. METHODS: A longitudinal observational study was conducted over a total of 124 days using web-based surveys. Persons receiving either a vaccination against SARS-CoV-2 or one of the established vaccines (comparator group) were included. In the first questionnaire (short-term survey), 2 weeks after vaccination, mainly local and systemic complaints were evaluated. The long-term survey (42 days after vaccination) and follow-up survey (124 weeks after vaccination) focused on medical consultations for any reason. Multivariate analyses were conducted to determine the influence of the vaccine type (SARS-CoV-2 vs. comparator) and demographic factors. RESULTS: In total, data from 16,636 participants were included. Self-reported reactogenicity was lowest in the comparator group (53.2%) and highest in the ChAdOx1 group (85.3%). Local reactions were reported most frequently after mRNA-1273 (73.9%) and systemic reactions mainly after vector-based vaccines (79.8%). Almost all SARS-CoV-2 vaccines showed increased odds of reporting local or systemic reactions. Approximately equal proportions of participants reported medical consultations. None in the comparator group suspected a link to vaccination, while this was true for just over one in 10 in the mRNA-1273 group. The multivariate analysis showed that people with SARS-CoV-2 vaccination were not more likely to report medical consultations; patients who had received a regimen with at least one ChAdOx1 were even less likely to report medical consultations. Younger age, female gender and higher comorbidity were mostly associated with higher odds of medical consultations. CONCLUSION: The rate of adverse reactions after established vaccinations was roughly comparable to previous studies. Two weeks after vaccination, participants in the SARS-CoV-2 vaccination group reported more local and systemic local reactions than participants in the comparator group. In the further course, however, there were no higher odds of medical consultations in either of the two groups. Thus, altogether, we assume comparable safety. TRIAL REGISTRATION: DRKS-ID DRKS00025881 and DRKS-ID DRKS00025373.

Graph-based clinical recommender: Predicting specialists procedure orders using graph representation learning.

OBJECTIVE: To determine whether graph neural network based models of electronic health records can predict specialty consultation care needs for endocrinology and hematology more accurately than the standard of care checklists and other conventional medical recommendation algorithms in the literature. METHODS: Demand for medical expertise far outstrips supply, with tens of millions in the US alone with deficient access to specialty care. Rather than potentially months long delays to initiate diagnostic workup and medical treatment with a specialist, referring primary care supported by an automated recommender algorithm could anticipate and directly initiate patient evaluation that would otherwise be needed at subsequent a specialist appointment. We propose a novel graph representation learning approach with a heterogeneous graph neural network to model structured electronic health records and formulate recommendation/prediction of subsequent specialist orders as a link prediction problem. RESULTS: Models are trained and assessed in two specialty care sites: endocrinology and hematology. Our experimental results show that our model achieves an 8% improvement in ROC-AUC for endocrinology (ROC-AUC = 0.88) and 5% improvement for hematology (ROC-AUC = 0.84) personalized procedure recommendations over prior medical recommender systems. These recommender algorithm approaches provide medical procedure recommendations for endocrinology referrals more effectively than manual clinical checklists (recommender: precision = 0.60, recall = 0.27, F1-score = 0.37) vs. (checklist: precision = 0.16, recall = 0.28, F1-score = 0.20), and similarly for hematology referrals (recommender: precision = 0.44, recall = 0.38, F1-score = 0.41) vs. (checklist: precision = 0.27, recall = 0.71, F1-score = 0.39). CONCLUSION: Embedding graph neural network models into clinical care can improve digital specialty consultation systems and expand the access to medical experience of prior similar cases.

The Effect of "Offline-to-Online" Trust Transfer on the Utilization of Online Medical Consultation Among Chinese Rural Residents: Experimental Study.

BACKGROUND: Online medical consultation can serve as a valuable means for rural residents to access high-quality health care resources, thereby mitigating the geographic and economic disadvantages prevalent in rural areas. Nevertheless, due to lower cognitive abilities, rural residents often face challenges in trusting and making effective use of online medical consultations. More likely, adopting a bounded rational decision-making model that facilitates the "offline-to-online" trust transfer could prove to be a potentially effective approach. This strategy aims to encourage less technologically experienced rural residents to trust and make use of online medical consultations. OBJECTIVE: This study aims to characterize the status of "offline-to-online" trust transfer among rural residents in the context of internet health care, and analyze its direct impact on facilitating the utilization of online medical consultation. Additionally, we investigate the family spillover effect of "offline-to-online" trust transfer in promoting the use of online medical consultation among rural family members, considering its distributional effect across various education levels of the population. METHODS: A multistage stratified random sampling method was used to survey participants in rural areas of China from July to September 2021, encompassing a total of 2597 rural residents from 960 rural households. Propensity score values were estimated using logit regression, and the propensity score matching method, using the K-nearest neighbor matching, radius matching, and kernel matching methods, was applied to create matched treatment and control samples of rural residents based on their experience of "offline-to-online" trust transfer. Subsequently, we calculated average treatment effect scores to compare the differences in utilizing online medical consultation between the treatment and control rural samples. RESULTS: As many as 551/960 (57.4%) rural residents experienced an "offline-to-online" trust transfer, with a higher likelihood observed in the older population with lower levels of education and higher satisfaction with local health care services. Furthermore, rural residents who underwent "offline-to-online" trust transfer were 37%-40% more likely to utilize online medical consultation compared with those who did not experience this trust transfer. Additionally, family members of householders who underwent "offline-to-online" trust transfer were 25%-28% more likely to utilize online medical consultation than those whose householders did not experience this trust transfer. Notably, when compared with populations with high-level education, the "offline-to-online" trust transfer had more significant direct and spillover effects on the utilization of online medical consultation services among rural residents with low-level education. CONCLUSIONS: To enhance the "offline-to-online" trust transfer among rural residents and its facilitation in their utilization of online medical consultation, as well as other mobile health (mHealth) and ubiquitous health (uHealth) services, we recommend that online health care providers adopt a "patient-oriented" service model. This approach aims to elevate rural residents' satisfaction with local health care services and harness the trust-building functions inherent in physician-patient relationships and among family members.

[2022 Update of the Austrian Society for Rheumatology and Rehabilitation nutrition and lifestyle recommendations for patients with gout and hyperuricemia]. / Empfehlungen der Österreichischen Gesellschaft für Rheumatologie und Rehabilitation zu Ernährung und Lebensstil bei Gicht und Hyperurikämie ­ Update 2022.

BACKGROUND: Gout is the most frequent inflammatory joint disease in the western world and has a proven genetic background. Additionally, lifestyle factors, such as increasing life expectation and standard of living, sufficient or excess nutrition and a growing prevalence of obesity in the population as well as e.g. alcohol consumption, contribute to the rising incidence of hyperuricemia and gout. Apart from an adequate medication, medical consultation on nutrition and lifestyle is an essential part of the management of gout patients, who have a high risk of internal comorbidities. OBJECTIVE: In 2015 the Austrian Society for Rheumatology and Rehabilitation (ÖGR) working group for osteoarthritis and crystal arthropathies published nutrition and lifestyle recommendations for patients with gout and hyperuricemia. Since then, a multitude of studies have been published addressing this topic, which necessitated an update. METHODS: First, the authors performed a hierarchical literature search to screen for the literature published since 2015. Considering references given in the first publication, the relevant literature was selected and the recommendations from 2015 were either kept as published, reformulated or recreated. Finally, the evidence level and the level of agreement for each recommendation were added by the authors. RESULTS: Based on this process, 10 recommendations were generated instead of the initial 9. As in the original publication, a graphical presentation with symbols was constructed to complement the written text. CONCLUSION: The ÖGR recommendations on nutrition and lifestyle for patients with gout and hyperuricemia were updated in accordance with the most recent relevant literature. These are supposed to serve as information and education material for patients and updated information for physicians.

Patient Activeness During Online Medical Consultation in China: Multilevel Analysis.

BACKGROUND: Online medical consultation is an important complementary approach to offline health care services. It not only increases patients' accessibility to medical care, but also encourages patients to actively participate in consultation, which can result in higher shared decision making, patient satisfaction, and treatment adherence. OBJECTIVE: This study aims to explore multilevel factors that influence patient activeness in online medical consultations. METHODS: A data set comprising 40,505 patients from 300 physicians in 10 specialties was included for multilevel analysis. Patient activeness score (PAS) was calculated based on the frequency and the proportion of patient discourses to the total frequency of doctor-patient interactions. Intraclass correlation coefficients were calculated to identify between-group variations, and the final multilevel regression model included patient- and physician-level factors. RESULTS: Patients were not equally active in online medical consultations, with PASs varying from 0 to 125.73. Patient characteristics, consultation behavioral attributes, and physician professional characteristics constitute 3 dimensions that are associated with patient activeness. Specifically, young and female patients participated more actively. Patients' waiting times online (ß=-.17; P<.001) for physician responses were negatively correlated with activeness, whereas patients' initiation of conversation (ß=.83; P<.001) and patient consultation cost (ß=.52; P<.001) in online medical consultation were positively correlated. Physicians' online consultation volumes (ß=-.10; P=.01) were negatively associated with patient activeness, whereas physician online consultation fee (ß=.03; P=.01) was positively associated. The interaction effects between patient- and physician-level factors were also identified. CONCLUSIONS: Patient activeness in online medical consultation requires more scholarly attention. Patient activeness is likely to be enhanced by reducing patients' waiting times and encouraging patients' initiation of conversation in online medical consultation. The findings have practical implications for patient-centered care and the improvement of online medical consultation services.

The Influence of Perceived Social Presence on the Willingness to Communicate in Mobile Medical Consultations: Experimental Study.

BACKGROUND: With the rise of online health care service, there is growing discussion on the relationship between physicians and patients online, yet few researchers have paid attention to patients' perception of social presence, especially its influence on their willingness to communicate (WTC). OBJECTIVE: The goal of the research is to investigate the influence of perceived social presence (PSP) on WTC in mobile medical consultations. METHODS: Participants living in Yunnan province during the period of middle to high risk of COVID-19 infection were recruited via the internet. They were assigned randomly into 2 groups interacting with a virtual physician presenting high and low levels of social presence and then asked to complete a questionnaire. Based on the theoretical framework, the study puts forward a model evaluating the relationships among participants' PSP, communication apprehension (CA), self-perceived communication competence (SPCC), and willingness to communicate about health (WTCH) in the computer-mediated communication between virtual physicians and patients. RESULTS: In total 206 (106 in group 1 and 100 in group 2) valid samples were gathered (from 276 log-ins) and 88.8% (183/206) of them were aged 18 to 44 years, which approximately resembles the age distribution of the main population engaging in online medical consultation in China. Independent t test shows that there is significant difference between the PSP of the 2 groups (P=.04), indicating a successful manipulation of social presence. The total effect of PSP on WTCH is 0.56 (P<.001), among which 74.4% is direct effect (P<.001). Among the indirect effects between PSP and WTCH, the mediating effect of SPCC accounts for 68.8% (P<.001) and the sequential mediating effect of CA→SPCC accounts for 19.2% (P<.001), while the mediating effect of CA alone is not significant (P=.08). CONCLUSIONS: This study provides a comprehensible model, demonstrating that PSP is an important antecedent of WTCH, and the sequential mediating effect of CA and SPCC found in this study also proves that in the environment of online mobile medical services, CA cannot affect communication directly. The findings will provide some practical inspiration for the popularization of online medical service, especially for the promotion of online physician-patient communication.

A Teleconsultation Device, Consult Station, for Remote Primary Care: Multisite Prospective Cohort Study.

BACKGROUND: Telemedicine technology is a growing field, especially in the context of the COVID-19 pandemic. Consult Station (Health for Development) is the first telemedicine device enabling completely remote medical consultations, including the concurrent collection of clinical parameters and videos. OBJECTIVE: Our aim was to collect data on the multisite urban and suburban implementation of the Consult Station for primary care and assess its contribution to health care pathways in areas with a low density of medical services. METHODS: In a proof-of-concept multisite prospective cohort study, 2134 consecutive patients had teleconsultations. Consultation characteristics were analyzed from both the patient and practitioner perspective. RESULTS: In this study, the main users of Consult Station were younger women consulting for low-severity seasonal infections. Interestingly, hypertension, diabetes, and preventive medical consultations were almost absent, while they accounted for almost 50% of consultations with a general practitioner (GP). We showed that for all regions where the Consult Station was implemented, the number of consultations increased as GP density decreased. The study of practitioner characteristics showed GPs from metropolitan areas are motivated to work with this device remotely, with a high level of technology acceptability. CONCLUSIONS: The multisite implementation of Consult Station booths is suitable for primary care and could also address the challenge of "medical deserts." In addition, further studies should be performed to evaluate the possible contribution of Consult Station booths to limiting work absenteeism.

Person-environment fit of environmental support provided during medical consultations with older patients.

As some cognitive functions decline in old age, the ability to decide about important life events such as medical treatment is endangered. Environmental support to improve the comprehension of health-related information is therefore necessary. With a small-scale explorative approach, the present survey study aimed at investigating person-environment fit (PE-fit) of support provided during medical consultations. This fit was calculated by assessing the match between aids provided by five medical practitioners during medical consultations and aids most appreciated by the geriatric patients (N = 88). The results showed that the largest discrepancies of used and appreciated aids could be found concerning the opportunity to discuss decisions with relatives, the possibility to take notes, the use of objects, pictures and a keyword list. Female patients indicated a lower PE-fit. These findings highlight discrepancies between the use of specific aids and the wishes of patients and call for thoughtful use of aids during consultations with geriatric patients.

Composition-driven symptom phrase recognition for Chinese medical consultation corpora.

BACKGROUND: Symptom phrase recognition is essential to improve the use of unstructured medical consultation corpora for the development of automated question answering systems. A majority of previous works typically require enough manually annotated training data or as complete a symptom dictionary as possible. However, when applied to real scenarios, they will face a dilemma due to the scarcity of the annotated textual resources and the diversity of the spoken language expressions. METHODS: In this paper, we propose a composition-driven method to recognize the symptom phrases from Chinese medical consultation corpora without any annotations. The basic idea is to directly learn models that capture the composition, i.e., the arrangement of the symptom components (semantic units of words). We introduce an automatic annotation strategy for the standard symptom phrases which are collected from multiple data sources. In particular, we combine the position information and the interaction scores between symptom components to characterize the symptom phrases. Equipped with such models, we are allowed to robustly extract symptom phrases that are not seen before. RESULTS: Without any manual annotations, our method achieves strong positive results on symptom phrase recognition tasks. Experiments also show that our method enjoys great potential with access to plenty of corpora. CONCLUSIONS: Compositionality offers a feasible solution for extracting information from unstructured free text with scarce labels.

Turn to the Internet First? Using Online Medical Behavioral Data to Forecast COVID-19 Epidemic Trend.

The surveillance and forecast of newly confirmed cases are important to mobilize medical resources and facilitate policymaking during a public health emergency. Digital surveillance using data available online has increasingly become a trend with the advancement of the Internet. In this study, we assessed the predictive value of multiple online medical behavioral data, including online medical consultation (OMC), online medical appointment (OMA), and online medical search (OMS) for the regional outbreak of coronavirus disease 2019 in Shenzhen, China during January 1, 2020 to March 5, 2020. Multivariate vector autoregression models were used for the prediction. The results identified a novel predictor, OMC, which can forecast the disease trend up to 2 days ahead of the official reports of confirmed cases from the local health department. OMS data had relatively weaker predictive power than OMC in our model, and OMA data failed to predict the confirmed cases. This study highlights the importance of OMC data and has implication in providing evidence-based guidelines for local authorities to evaluate risks and allocate resources during the pandemic.

Parents' and children's decision-making and experiences in pediatric epilepsy surgery.

BACKGROUND: In pediatric epilepsy surgery, little research has been conducted on parents' decision-making for or against surgery, their satisfaction with the surgical outcome, as well as their children's personal experiences with the process. OBJECTIVE: This study explores (1) factors that may influence parents' decision-making, (2) factors associated with their postoperative satisfaction, and (3) their children's involvement in decision-making and their experiences with epilepsy surgery. METHODS: Self-developed questionnaires were provided to parents and their children in Germany, Switzerland, and Austria. Clinical and psychosocial differences between parents who decided for or against surgery, as well as associations between postoperative factors and the parents' satisfaction were statistically analyzed. RESULTS: Fifty-one questionnaires (42 parent and nine patient questionnaires) were evaluated. Parents who decided for epilepsy surgery reported significantly more frequently to have received a good medical consultation and a consistent recommendation. They made significantly less use of information websites, internet forums, and patient organizations. Their children were classified as more intelligent and resisted surgery less. Most of the parents were satisfied with the surgical outcome (83%). Parents were significantly more satisfied when their children had fewer medication side effects, their memory or concentration had improved, their character or behavior had changed in a positive direction, or when their children were more independent or less excluded. They were also significantly more content when they had more free time to themselves and their professional situation or their relationship with their children, friends, or other family members had improved. However, no significant association was found between parental satisfaction and reduced number of medications or improved seizure outcome. Although the children were reported to have been minimally involved in the decision-making process, they were satisfied with their involvement. CONCLUSIONS: A good medical consultation that involves the children and considers the family's living conditions is a crucial factor for parents' decision-making on pediatric epilepsy surgery. For parents' satisfaction, a positive change in their child's character or behavior and an improved psychosocial situation of the family are more important than postoperative seizure frequency or number of antiepileptic drugs (AEDs). Therefore, the medical consultation should not only focus on clinical factors, but also point out psychosocial and behavioral changes that may occur after the surgical treatment.

Cross domains adversarial learning for Chinese named entity recognition for online medical consultation.

Deep learning methods have been applied to Chinese named entity recognition for the online medical consultation. They require a large number of marked samples. However, no such database is available at present. This paper begins with constructing a larger labelled Chinese texts database for the online medical consultation. Second, a basic framework unit is proposed, which is pre-trained by the transfer learning from both Bidirectional language model and Mask language model trained on the larger unlabelled data. Finally, cross domains adversarial learning (CDAL) for Chinese named entity recognition is proposed to further improve the performance, which not only uses the pre-trained basic framework unit, but also uses the adversarial multi-task learning on both electronic medical record texts and online medical consultation texts. Experimental results validate the effectiveness of CDAL.

Investigating Patients' Intention to Continue Using Teleconsultation to Anticipate Postcrisis Momentum: Survey Study.

BACKGROUND: The COVID-19 crisis has drastically changed care delivery with teleconsultation platforms experiencing substantial spikes in demand, helping patients and care providers avoid infections and maintain health care services. Beyond the current pandemic, teleconsultation is considered a significant opportunity to address persistent health system challenges, including accessibility, continuity, and cost of care, while ensuring quality. OBJECTIVE: This study aims at identifying the determinants of patients' intention to continue using a teleconsultation platform. It extends prior research on information technology use continuance intention and teleconsultation services. METHODS: Data was collected in November 2018 and May 2019 with Canadian patients who had access to a teleconsultation platform. Measures included patients' intention to continue their use; teleconsultation usefulness; teleconsultation quality; patients' trust toward the digital platform, its provider. and health care professionals; and confirmation of patients' expectations toward teleconsultation. We used structural equation modeling employing the partial least squares component-based technique to test our research model and hypotheses. RESULTS: We analyzed a sample of 178 participants who had used teleconsultation services. Our findings revealed that confirmation of expectations had the greatest influence on continuance intention (total effects=0.722; P<.001), followed by usefulness (total effects=0.587; P<.001) and quality (total effects=0.511; P<.001). Usefulness (ß=.60; P<.001) and quality (ß=.34; P=.01) had direct effects on the dependent variable. The confirmation of expectations had direct effects both on usefulness (ß=.56; P<.001) and quality (ß=.75; P<.001) in addition to having an indirect effect on usefulness (indirect effects=0.282; P<.001). Last, quality directly influenced usefulness (ß=.34; P=.002) and trust (ß=.88; P<.001). Trust does not play a role in the context under study. CONCLUSIONS: Teleconsultation is central to care going forward, and it represents a significant lever for an improved, digital delivery of health care in the future. We believe that our findings will help drive long-term teleconsultation adoption and use, including in the aftermath of the current COVID-19 crisis, so that general care improvement and greater preparedness for exceptional situations can be achieved.

The Impact of Monetary Incentives on Physician Prosocial Behavior in Online Medical Consulting Platforms: Evidence From China.

BACKGROUND: In online medical consulting platforms, physicians can get both economic and social returns by offering online medical services, such as answering questions or sharing health care knowledge with patients. Physicians' online prosocial behavior could bring many benefits to the health care industry. Monetary incentives could encourage physicians to engage more in online medical communities. However, little research has studied the impact of monetary incentives on physician prosocial behavior and the heterogeneity of this effect. OBJECTIVE: This study aims to explore the effects of monetary incentives on physician prosocial behavior and investigate the moderation effects of self-recognition and recognition from others of physician competence. METHODS: This study was a fixed-effect specification-regression model based on a difference-in-differences design with robust standard errors clustered at the physician level using monthly panel data. It included 26,543 physicians in 3851 hospitals over 133 months (November 2006-December 2017) from a leading online health care platform in China. We used the pricing strategy of physicians and satisfaction levels to measure their own and patients' degree of recognition, respectively. Physicians' prosocial behavior was measured by free services offered. RESULTS: The introduction of monetary incentives had a positive effect on physician prosocial behavior (ß=1.057, P<.01). Higher self-recognition and others' recognition level of physician competence increased this promotion effect (γ=0.275, P<.01 and γ=0.325, P<.01). CONCLUSIONS: This study explored the positive effect of the introduction of monetary incentives on physician prosocial behavior. We found this effect was enhanced for physicians with a high level of self-recognition and others' recognition of their competence. We provide evidence of the effect of monetary incentives on physicians' prosocial behaviors in the telemedicine markets and insight for relevant stakeholders into how to design an effective incentive mechanism to improve physicians' prosocial engagements.