“Hey assistant, how can I become a donor?” The case of a conversational agent designed to engage people in blood donation

A virtual assistant for primary care's food and nutrition surveillance system: Development and validation study in Brazil

OBJECTIVE

The study aimed to develop and validate a conversational agent (chatbot) designed to support Food and Nutrition Surveillance (FNS) practices in primary health care settings.

METHODS

This mixed-methods research was conducted in three stages. Initially, the study identified barriers and challenges in FNS practices through a literature review and feedback from 655 health professionals and FNS experts across Brazil. Following this, a participatory design approach was employed to develop and validate the chatbot's content. The final stage involved evaluating the chatbot's user experience with FNS experts.

RESULTS

The chatbot could accurately understand and respond to 60 different intents or keywords related to FNS. Themes such as training, guidance, and access emerged as crucial for guiding FNS initiatives and addressing implementation challenges, primarily related to human resources. The chatbot achieved a Global Content Validation Index of 0.88.

CONCLUSION

The developed chatbot represents a significant advancement in supporting FNS practices within primary health care.

PRACTICE IMPLICATION

By providing an innovative, interactive, educational tool that is both accessible and reliable, this digital assistant has the potential to facilitate the operationalization of FNS practices, addressing the critical need for effective training and counseling in developing countries.

Consent-GPT: is it ethical to delegate procedural consent to conversational AI?

Obtaining informed consent from patients prior to a medical or surgical procedure is a fundamental part of safe and ethical clinical practice. Currently, it is routine for a significant part of the consent process to be delegated to members of the clinical team not performing the procedure (eg, junior doctors). However, it is common for consent-taking delegates to lack sufficient time and clinical knowledge to adequately promote patient autonomy and informed decision-making. Such problems might be addressed in a number of ways. One possible solution to this clinical dilemma is through the use of conversational artificial intelligence using large language models (LLMs). There is considerable interest in the potential benefits of such models in medicine. For delegated procedural consent, LLM could improve patients' access to the relevant procedural information and therefore enhance informed decision-making.In this paper, we first outline a hypothetical example of delegation of consent to LLMs prior to surgery. We then discuss existing clinical guidelines for consent delegation and some of the ways in which current practice may fail to meet the ethical purposes of informed consent. We outline and discuss the ethical implications of delegating consent to LLMs in medicine concluding that at least in certain clinical situations, the benefits of LLMs potentially far outweigh those of current practices.

Development of a novel drug information provision system for Kampo medicine using natural language processing technology

BACKGROUND

Kampo medicine is widely used in Japan; however, most physicians and pharmacists have insufficient knowledge and experience in it. Although a chatbot-style system using machine learning and natural language processing has been used in some clinical settings and proven useful, the system developed specifically for the Japanese language using this method has not been validated by research. The purpose of this study is to develop a novel drug information provision system for Kampo medicines using a natural language classifier® (NLC®) based on IBM Watson.

METHODS

The target Kampo formulas were 33 formulas listed in the 17th revision of the Japanese Pharmacopoeia. The information included in the system comes from the package inserts of Kampo medicines, Manuals for Management of Individual Serious Adverse Drug Reactions, and data on off-label usage. The system developed in this study classifies questions about the drug information of Kampo formulas input by natural language into preset questions and outputs preset answers for the questions. The system uses morphological analysis, synonym conversion by thesaurus, and NLC®. We fine-tuned the information registered into NLC® and increased the thesaurus. To validate the system, 900 validation questions were provided by six pharmacists who were classified into high or low levels of knowledge and experience of Kampo medicines and three pharmacy students.

RESULTS

The precision, recall, and F-measure of the system performance were 0.986, 0.915, and 0.949, respectively. The results were stable even with differences in the amount of expertise of the question authors.

CONCLUSIONS

We developed a system using natural language classification that can give appropriate answers to most of the validation questions.

The Development and Use of Chatbots in Public Health: Scoping Review

BACKGROUND

Chatbots are computer programs that present a conversation-like interface through which people can access information and services. The COVID-19 pandemic has driven a substantial increase in the use of chatbots to support and complement traditional health care systems. However, despite the uptake in their use, evidence to support the development and deployment of chatbots in public health remains limited. Recent reviews have focused on the use of chatbots during the COVID-19 pandemic and the use of conversational agents in health care more generally. This paper complements this research and addresses a gap in the literature by assessing the breadth and scope of research evidence for the use of chatbots across the domain of public health.

OBJECTIVE

This scoping review had 3 main objectives: (1) to identify the application domains in public health in which there is the most evidence for the development and use of chatbots; (2) to identify the types of chatbots that are being deployed in these domains; and (3) to ascertain the methods and methodologies by which chatbots are being evaluated in public health applications. This paper explored the implications for future research on the development and deployment of chatbots in public health in light of the analysis of the evidence for their use.

METHODS

Following the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines for scoping reviews, relevant studies were identified through searches conducted in the MEDLINE, PubMed, Scopus, Cochrane Central Register of Controlled Trials, IEEE Xplore, ACM Digital Library, and Open Grey databases from mid-June to August 2021. Studies were included if they used or evaluated chatbots for the purpose of prevention or intervention and for which the evidence showed a demonstrable health impact.

RESULTS

Of the 1506 studies identified, 32 were included in the review. The results show a substantial increase in the interest of chatbots in the past few years, shortly before the pandemic. Half (16/32, 50%) of the research evaluated chatbots applied to mental health or COVID-19. The studies suggest promise in the application of chatbots, especially to easily automated and repetitive tasks, but overall, the evidence for the efficacy of chatbots for prevention and intervention across all domains is limited at present.

CONCLUSIONS

More research is needed to fully understand the effectiveness of using chatbots in public health. Concerns with the clinical, legal, and ethical aspects of the use of chatbots for health care are well founded given the speed with which they have been adopted in practice. Future research on their use should address these concerns through the development of expertise and best practices specific to public health, including a greater focus on user experience.

Security, privacy, and healthcare-related conversational agents: a scoping review

Health chatbots interview patients and collect health data. This process makes demands on data security and data privacy. To identify how and to what extent security and privacy are considered in current health chatbots. We conducted a scoping review by searching three bibliographic databases (PubMed, ACM Digital Library, IEEExplore) for papers reporting on chatbots in healthcare. We extracted which, how, and where data is stored by health chatbots and identified which external services have access to the data. Out of 1026 retrieved papers, we included 70 studies in the qualitative synthesis. Most papers report on chatbots that collect and process personal health data, usually in the context of mental health coaching applications. The majority did not provide any information regarding security or privacy aspects. We were able to determine limitations in literature and identified concrete challenges, including data access and usage of (third-party) services, data storage, data security methods, use case peculiarities and data privacy, as well as legal requirements. Data privacy and security in health chatbots are still underresearched and related information is underrepresented in scientific literature. By addressing the five key challenges in future, the transfer of theoretical solutions into practice can be facilitated.

Mobile Health Applications and Medication Adherence of Patients With Hypertension: A Systematic Review and Meta-Analysis

INTRODUCTION

Current evidence has revealed the beneficial effects of mobile health applications on systolic and diastolic blood pressure. However, there is still no solid evidence of the underlying factors for these outcomes, and hypertension treatment is performed mainly by medication intake. This study aims to analyze the impacts of health applications on medication adherence of patients with hypertension and understand the underlying factors.

METHODS

A systematic review and meta-analysis were conducted considering controlled clinical trials published, without year filter, through July 2020. The searches were performed in the electronic databases of Scopus, MEDLINE, and BVSalud. Study characteristics were extracted for qualitative synthesis. The meta-analysis examined medication-taking behavior outcomes using the generic inverse-variance method to combine multiple variables.

RESULTS

A total of 1,199 records were identified, of which 10 studies met the inclusion criteria for qualitative synthesis, and 9 met the criteria for meta-analysis with 1,495 participants. The analysis of mean changes revealed significant improvements in medication adherence (standardized mean difference=0.41, 95% CI=0.02, 0.79, I²=82%, p=0.04) as well as the analysis of the values measured after follow-up (standardized mean difference=0.60, 95% CI=0.30, 0.90, I²=77%, p<0.0001). Ancillary improvements were also identified, such as patients' perceived confidence, treatment self-efficacy and self-monitoring, acceptance of technology, and knowledge about the condition and how to deal with health issues.

DISCUSSION

There is evidence that mobile health applications can improve medication adherence in patients with hypertension, with broad heterogeneity between studies on the topic. The use of mobile health applications conceivably leads to ancillary improvements inherent to better medication adherence.

Kinesiotherapy With Exergaming as a Potential Modulator of Epigenetic Marks and Clinical Functional Variables of Older Women: Protocol for a Mixed Methods Study

Background

Kinesiotherapy is an option to mitigate worsening neuropsychomotor function due to human aging. Moreover, exergames are beneficial for the practice of physical therapy by older patients. Physical exercise interventions are known to alter the epigenome, but little is known about their association with exergames.

Objective

We aim to evaluate the effects of kinesiotherapy with exergaming on older women’s epigenetic marks and cognitive ability, as well as on their clinical functional variables. Our hypothesis states that this kind of therapy can elicit equal or even better outcomes than conventional therapy.

Methods

We will develop a virtual clinic exergame with 8 types of kinesiotherapy exercises. Afterward, we will conduct a 1:1 randomized clinical trial to compare the practice of kinesiotherapy with exergames (intervention group) against conventional kinesiotherapy (control group). A total of 24 older women will be enrolled for 1-hour sessions performed twice a week, for 6 weeks, totaling 12 sessions. We will assess outcomes using epigenetic blood tests, the Montreal Cognitive Assessment test, the Timed Up and Go test, muscle strength grading in a hydraulic dynamometer, and the Game Experience Questionnaire at various stages.

Results

The project was funded in October 2019. Game development took place in 2020. Patient recruitment and a clinical trial are planned for 2021.

Conclusions

Research on this topic is likely to significantly expand the understanding of kinesiotherapy and the impact of exergames. To the best of our knowledge, this may be one of the first studies exploring epigenetic outcomes of exergaming interventions.

Trial Registration

Brazilian Clinical Trials Registry/Registro Brasileiro de Ensaios Clínicos (ReBEC) RBR-9tdrmw; https://ensaiosclinicos.gov.br/rg/RBR-9tdrmw.

International Registered Report Identifier (IRRID)

DERR1-10.2196/32729

Dietary Intake and Health Status of Elderly Patients With Type 2 Diabetes Mellitus: Cross-sectional Study Using a Mobile App in Primary Care

Background

Healthy dietary intake reduces the risk of complications of diabetes mellitus. Using assessment methods helps to understand these circumstances, and an electronic application may optimize this practice.

Objective

In this study, we aimed to (1) assess the dietary intake and health status of elderly patients with type 2 diabetes mellitus (T2DM) in primary care, (2) use a mobile app as a tool for data collection and analysis in the context of primary care, and (3) verify the perceptions of multidisciplinary health professionals regarding app use.

Methods

First, we developed a mobile app comprised of the questions of the Food and Nutrition Surveillance System (SISVAN) of Brazil, which includes a food frequency questionnaire of food categories with a recall of the previous 7 days. Thereafter, we used the app to collect data on the health status and dietary intake of 154 participants, aged 60-96 years, diagnosed with T2DM, and under treatment in primary care centers in the northern region of Rio Grande do Sul, Brazil. We also collected participants’ demographic, anthropometric, biochemical, and lifestyle variables. The associations between dietary intake and other variables were tested using chi-square tests with a 5% significance level. Regarding the app, we assessed usability and acceptance with 20 health professionals.

Results

Between August 2018 and December 2018, participants had an intake in line with recommended guidelines for raw salads (57.1%), fruits (76.6%), milk products (68.2%), fried foods (72.7%), savory biscuits (60.4%), cookies or sweets (72.1%), and sugary drinks (92.9%) Meanwhile, the consumption of beans (59.7%), pulses and cooked vegetables (73.4%), and processed meat products (59.7%) was not in line with the guidelines. There were statistically significant differences in meeting the recommended guidelines among participants of different genders (P=.006 and P=.035 for the intake of fried foods and sugary drinks, respectively), place of residence (P=.034 for the intake of cookies and sweets), family history of diabetes (P<.001 for the intake of beans), physical activity engagement (P=.003 for the intake fresh fruits), history of smoking (P=.001 for the intake of raw salads), and presence of coronary disease (P=.050 for the intake of pulses and cooked vegetables). The assessment of usability resulted in a mean score of 71.75 points. Similarly, the assessment of the 15 acceptance questions revealed high scores, and the qualitative questions revealed positive perceptions.

Conclusions

We identified that most participants complied with recommended intake guidelines for 7 of 10 categories in the SISVAN guidelines. However, most participants were overweight and had nutritional and clinical disorders, which justifies further investigations in this population. The app was well-rated by health professionals and considered a useful and promising tool for collecting and analyzing data in primary care settings.

Using a mobile health app to improve patients' adherence to hypertension treatment: a non-randomized clinical trial

Poor adherence to hypertension treatment increases complications of the disease and is characterized by a lack of awareness and acceptance of ongoing treatment. Mobile health (mHealth) apps can optimize processes and facilitate access to health information by combining treatment methods with attractive solutions. In this study, we aimed at verifying the influence of using an mHealth app on patients' adherence to hypertension treatment, also examining how user experience toward the app influenced the outcomes. A total of 49 participants completed the study, men and women, diagnosed with hypertension and ongoing medical treatment. For 12 weeks, the control group continued with conventional monitoring, while the experimental group used an mHealth app. From the experimental group, at baseline, 8% were non-adherent, 64% were partial adherents and 28% were adherent to the treatment. Baseline in the control group indicated 4.2% non-adherents, 58.3% partial adherents, and 37.5% adherents. After follow-up, the experimental group had an increase to 92% adherent, 8% partially adherent, and 0% non-adherent (P < 0.001). In the control group, adherence after follow-up remained virtually the same (P ≥ 0.999). Results of user experience were substantially positive and indicate that the participants in the experimental group had a satisfactory perception of the app. In conclusion, this study suggests that using an mHealth app can empower patients to manage their own health and increase adherence to hypertension treatment, especially when the app provides a positive user experience.

Multi-source Seq2seq guided by knowledge for Chinese healthcare consultation

Online healthcare consultation offers people a convenient way to consult doctors. In this paper, we aim at building a generative dialog system for Chinese healthcare consultation. As the original Seq2seq architecture tends to suffer the issue of generating low-quality responses, the multi-source Seq2seq architecture generating more informative responses is much more preferred in this task. The multi-source Seq2seq architecture takes advantage of retrieval techniques to obtain responses from the database, and then takes these responses alongside the user-issued question as input. However, some of the retrieved responses might be not much related to the user-issued question, resulting in the generation of unsatisfying responses that are not correct in diagnosis or instead provide inappropriate advice on prevention or treatment. Therefore, this paper proposes multi-source Seq2seq guided by knowledge (MSSGK) to handle this problem. MSSGK differs from the multi-source Seq2seq architecture in that domain knowledge, including disease labels and topic labels about prevention and treatment, is introduced into the response generation via a multi-task learning framework. To better exploit the domain knowledge, we propose three attention mechanisms to provide more appropriate guidance for response generation. Experimental results on a dataset of real-world healthcare consultation show the effectiveness of the proposed method.