General Characteristics and Design Taxonomy of Chatbots for COVID-19: Systematic Review

BACKGROUND

A conversational agent powered by artificial intelligence, commonly known as a chatbot, is one of the most recent innovations used to provide information and services during the COVID-19 pandemic. However, the multitude of conversational agents explicitly designed during the COVID-19 pandemic calls for characterization and analysis using rigorous technological frameworks and extensive systematic reviews.

OBJECTIVE

This study aims to describe the general characteristics of COVID-19 chatbots and examine their system designs using a modified adapted design taxonomy framework.

METHODS

We conducted a systematic review of the general characteristics and design taxonomy of COVID-19 chatbots, with 56 studies included in the final analysis. This review followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines to select papers published between March 2020 and April 2022 from various databases and search engines.

RESULTS

Results showed that most studies on COVID-19 chatbot design and development worldwide are implemented in Asia and Europe. Most chatbots are also accessible on websites, internet messaging apps, and Android devices. The COVID-19 chatbots are further classified according to their temporal profiles, appearance, intelligence, interaction, and context for system design trends. From the temporal profile perspective, almost half of the COVID-19 chatbots interact with users for several weeks for >1 time and can remember information from previous user interactions. From the appearance perspective, most COVID-19 chatbots assume the expert role, are task oriented, and have no visual or avatar representation. From the intelligence perspective, almost half of the COVID-19 chatbots are artificially intelligent and can respond to textual inputs and a set of rules. In addition, more than half of these chatbots operate on a structured flow and do not portray any socioemotional behavior. Most chatbots can also process external data and broadcast resources. Regarding their interaction with users, most COVID-19 chatbots are adaptive, can communicate through text, can react to user input, are not gamified, and do not require additional human support. From the context perspective, all COVID-19 chatbots are goal oriented, although most fall under the health care application domain and are designed to provide information to the user.

CONCLUSIONS

The conceptualization, development, implementation, and use of COVID-19 chatbots emerged to mitigate the effects of a global pandemic in societies worldwide. This study summarized the current system design trends of COVID-19 chatbots based on 5 design perspectives, which may help developers conveniently choose a future-proof chatbot archetype that will meet the needs of the public in the face of growing demand for a better pandemic response.

Nurse Informaticians Report Low Satisfaction and Multi-level Concerns with Electronic Health Records: Results from an International Survey

This study presents a qualitative content analysis of nurses' satisfaction and issues with current electronic health record (EHR) systems, as reflected in one of the largest international surveys of nursing informatics. Study participants from 45 countries (n=469) ranked their satisfaction with the current state of nursing functionality in EHRs as relatively low. Two-thirds of the participants (n=283) provided disconcerting comments when explaining their low satisfaction rankings. More than one half of the comments identified issues at the system level (e.g., poor system usability; non-integrated systems and poor interoperability; lack of standards; and limited functionality/missing components), followed by user-task issues (e.g., failure of systems to meet nursing clinical needs; non nursing-specific systems) and environment issues (e.g., low prevalence of EHRs; lack of user training). The study results call for the attention of international stakeholders (educators, managers, policy makers) to improve the current issues with EHRs from a nursing perspective.

Advancing Nursing Informatics in the Next Decade: Recommendations from an International Survey

In the summer of 2015, the International Medical Informatics Association Nursing Informatics Special Interest Group (IMIA NISIG) Student Working Group developed and distributed an international survey of current and future trends in nursing informatics. The survey was developed based on current literature on nursing informatics trends and translated into six languages. Respondents were from 31 different countries in Asia, Africa, North and Central America, South America, Europe, and Australia. This paper presents the results of responses to the survey question: "What should be done (at a country or organizational level) to advance nursing informatics in the next 5-10 years?" (n responders = 272). Using thematic qualitative analysis, responses were grouped into five key themes: 1) Education and training; 2) Research; 3) Practice; 4) Visibility; and 5) Collaboration and integration. We also provide actionable recommendations for advancing nursing informatics in the next decade.

Current Trends in Nursing Informatics: Results of an International Survey

Nursing informatics (NI) can help provide effective and safe healthcare. This study aimed to describe current research trends in NI. In the summer 2015, the IMIA-NI Students Working Group created and distributed an online international survey of the current NI trends. A total of 402 responses were submitted from 44 countries. We identified a top five NI research areas: standardized terminologies, mobile health, clinical decision support, patient safety and big data research. NI research funding was considered to be difficult to acquire by the respondents. Overall, current NI research on education, clinical practice, administration and theory is still scarce, with theory being the least common. Further research is needed to explain the impact of these trends and the needs from clinical practice.

Nursing Informatics Research Priorities for the Future: Recommendations from an International Survey

We present one part of the results of an international survey exploring current and future nursing informatics (NI) research trends. The study was conducted by the International Medical Informatics Association Nursing Informatics Special Interest Group (IMIA-NISIG) Student Working Group. Based on findings from this cross-sectional study, we identified future NI research priorities. We used snowball sampling technique to reach respondents from academia and practice. Data were collected between August and September 2015. Altogether, 373 responses from 44 countries were analyzed. The identified top ten NI trends were big data science, standardized terminologies (clinical evaluation/implementation), education and competencies, clinical decision support, mobile health, usability, patient safety, data exchange and interoperability, patient engagement, and clinical quality measures. Acknowledging these research priorities can enhance successful future development of NI to better support clinicians and promote health internationally.

Occupational traumatic injuries among workers in health care facilities - United States, 2012-2014

In 2013, one in five reported nonfatal occupational injuries occurred among workers in the health care and social assistance industry, the highest number of such injuries reported for all private industries. In 2011, U.S. health care personnel experienced seven times the national rate of musculoskeletal disorders compared with all other private sector workers. To reduce the number of preventable injuries among health care personnel, CDC's National Institute for Occupational Safety and Health (NIOSH), with collaborating partners, created the Occupational Health Safety Network (OHSN) to collect detailed injury data to help target prevention efforts. OHSN, a free, voluntary surveillance system for health care facilities, enables prompt and secure tracking of occupational injuries by type, occupation, location, and risk factors. This report describes OHSN and reports on current findings for three types of injuries. A total of 112 U.S. facilities reported 10,680 OSHA-recordable* patient handling and movement (4,674 injuries); slips, trips, and falls (3,972 injuries); and workplace violence (2,034 injuries) injuries occurring from January 1, 2012-September 30, 2014. Incidence rates for patient handling; slips, trips, and falls; and workplace violence were 11.3, 9.6, and 4.9 incidents per 10,000 worker-months,† respectively. Nurse assistants and nurses had the highest injury rates of all occupations examined. Focused interventions could mitigate some injuries. Data analyzed through OHSN identify where resources, such as lifting equipment and training, can be directed to potentially reduce patient handling injuries. Using OHSN can guide institutional and national interventions to protect health care personnel from common, disabling, preventable injuries.

Confidence and Information Access in Clinical Decision-Making: An Examination of the Cognitive Processes that affect the Information-seeking Behavior of Physicians

Clinical decision-making involves the interplay between cognitive processes and physicians' perceptions of confidence in the context of their information-seeking behavior. The objectives of the study are: to examine how these concepts interact, to determine whether physician confidence, defined in relation to information need, affects clinical decision-making, and if information access improves decision accuracy. We analyzed previously collected data about resident physicians' perceptions of information need from a study comparing abstracts and full-text articles in clinical decision accuracy. We found that there is a significant relation between confidence and accuracy (φ=0.164, p<0.01). We also found various differences in the alignment of confidence and accuracy, demonstrating the concepts of underconfidence and overconfidence across years of clinical experience. Access to online literature also has a significant effect on accuracy (p<0.001). These results highlight possible CDSS strategies to reduce medical errors.

Optimizing the txt2MEDLINE search portal for low-resource clinical decision support

txt2MEDLINE provides access to high-quality medical evidence via text-messaging in settings with inadequate Internet access. We optimized the txt2MEDLINE search technique by parsing queries for MeSH (Medical Subject Heading) terms and searching MEDLINE for articles containing these terms in their titles or abstracts. We compared our results to the existing txt2MEDLINE tool by compiling benchmark queries from low-income and low-middle-income countries, and asking doctors and nurses with practice experience in low-resource areas to evaluate them. The median scores on a 5-point Likert scale were 2.9 for the existing txt2MEDLINE vs. 3.8 for the modified version (p=0.015). This reached our predefined criterion for clinical significance, a difference of 0.5 standard deviations. Improving this technology could improve clinical information resources in the world's most medically underserved communities.

Comparing data accuracy between structured abstracts and full-text journal articles: implications in their use for informing clinical decisions

BACKGROUND

The abstract is the most frequently read section of a research article. The use of 'Consensus Abstracts', a clinician-oriented web application formatted for mobile devices to search MEDLINE/PubMed, for informing clinical decisions was proposed recently; however, inaccuracies between abstracts and the full-text article have been shown. Efforts have been made to improve quality.

METHODS

We compared data in 60 recent-structured abstracts and full-text articles from six highly read medical journals.

RESULTS

Data inaccuracies were identified and then classified as either clinically significant or not significant. Data inaccuracies were observed in 53.33% of articles ranging from 3.33% to 45% based on the IMRAD format sections. The Results section showed the highest discrepancies (45%) although these were deemed to be mostly not significant clinically except in one. The two most common discrepancies were mismatched numbers or percentages (11.67%) and numerical data or calculations found in structured abstracts but not mentioned in the full text (40%). There was no significant relationship between journals and the presence of discrepancies (Fisher's exact p value =0.3405). Although we found a high percentage of inaccuracy between structured abstracts and full-text articles, these were not significant clinically.

CONCLUSIONS

The inaccuracies do not seem to affect the conclusion and interpretation overall. Structured abstracts appear to be informative and may be useful to practitioners as a resource for guiding clinical decisions.