Incidence of diagnostic errors in unplanned hospitalized patients using an automated medical history-taking system with differential diagnosis generator: retrospective observational study (Preprint)

BACKGROUND Automated medical history-taking systems that generate differential diagnosis lists have been suggested to contribute to improved diagnostic accuracy. However, the effect of this system on diagnostic errors in clinical practice remains unknown. OBJECTIVE This study aimed to assess the incidence of diagnostic errors in an outpatient department, where an artificial intelligence (AI)-driven automated medical history-taking system that generates differential diagnosis lists was implemented in clinical practice. METHODS We conducted a retrospective observational study using data from a community hospital in Japan. We included patients aged 20 and older who used an AI-driven automated medical history-taking system that generates differential diagnosis lists in the outpatient department of internal medicine for whom the index visit was between July 1, 2019, and June 30, 2020, followed by unplanned hospitalization within 14 days. The primary endpoint was the incidence of diagnostic errors, which were detected using the Revised Safer Dx instrument by at least two independent reviewers. To evaluate the differential diagnosis list of AI on the incidence of diagnostic errors, we compared the incidence of diagnostic errors between the groups in which AI generated the final diagnosis in the differential diagnosis list and in which AI did not generate the final diagnosis in the differential diagnosis list; Fisher’s exact test was used for comparison between these groups. For cases with confirmed diagnostic errors, further review was conducted to identify the contributing factors of diagnostic errors via discussion among the three reviewers, using the Safer Dx Process Breakdown Supplement as a reference. RESULTS A total of 146 patients were analyzed. The final diagnosis was confirmed in 138 patients and the final diagnosis was observed in the differential diagnosis list of the AI in 69 patients. Diagnostic errors occurred in 16 of 146 patients (11.0%; 95% confidence interval, 6.4-17.2%). Although statistically insignificant, the incidence of diagnostic errors was lower in cases in which the final diagnosis was included in the differential diagnosis list of AI than in cases in which the final diagnosis was not included (7.2% vs. 15.9%, P=.18). Regarding the quality of clinical history taken by AI, the final diagnosis was easily assumed by reading only the clinical history taken by the system in 11 of 16 cases (68.8%). CONCLUSIONS The incidence of diagnostic errors in the internal medicine outpatients used an automated medical history-taking system that generates differential diagnosis lists seemed to be lower than the previously reported incidence of diagnostic errors. This result suggests that the implementation of an automated medical history-taking system that generates differential diagnosis lists could be beneficial for diagnostic safety in the outpatient department of internal medicine.

B.1 What Transition Skills Should be Targeted in Epilepsy Transition Clinics?

Background: Transition from pediatric to adult care can be a difficult time for adolescents with epilepsy. This period is often a period of extreme vulnerability and stress. As a result, research has recommended transition clinics to help these adolescents develop needed transition skills. However, the skills that need to be focused on remain unclear. Methods: Baseline transition skills in 113 adolescents with epilepsy, aged 14 to 18 (M= 16.46, male= 56) were analyzed. Results: Analyses showed that older adolescents showed significantly more transition skills than younger adolescents (F(4,108)=5.522, p=000). Although positive, older adolescents only scored, on average, 16.3/28 on the transition questionnaire; suggesting that many skills are still lacking, even at the time of transition. Specifically, although the majority of these older adolescents demonstrated being able to manage their condition independently (e.g., summarizing medical history, taking/knowing medications), these adolescents were less likely to demonstrate skills needed to be advocates for themselves and their health (e.g., asking questions, discussing concerns, speaking to the doctor instead of letting their parents). Conclusions: Results suggest it may be beneficial to restructure adolescent clinic visits; encouraging these patients to attend the initial portion of visits independently to help them feel more comfortable and confident championing for themselves.

Meatal Stenosis and Atopic Condition: A Pilot Study Demonstrating a Possible Unidentified Etiology

Meatal stenosis (MS) is known as one of the most frequent complications of circumcision. In the present study, we aimed to find any possible relationship between MS and allergic disorders. A total of 36 children with a mean±SD age of 5.84±2.03 years were referred with MS and an atopic background even in themselves or in one of their family members (Group A). There were also age-matched controls with a mean±SD age of 5.70±2.17 years who were referred to our center with allergic symptoms and no urinary complaints (Group B, n=17). The RIDA qLine allergy and allergy explorer (ALEX) tests were performed for all patients to find possible allergen sensitization. Laboratory findings revealed that IgE-sensitization to the main food and aeroallergens in Group A (with the chief complaint of MS in whom a mild atopic condition was found during concise medical history taking) were very similar to the control group with no significant difference (except for ryegrass which was higher in the control group). Although total IgE level was considerably higher in group B compared to group A, food sensitization to cow’s milk and ß-lactoglobulin was higher in asthmatic patients of group A compared to the controls. It seems that not all patients with MS should be considered as a complication of circumcision and undergo a surgical procedure for correction of the stenosis. Further investigations are required to determine the role of concise medical history taking and proper treatment of the allergic disorder to reduce failed surgical attempts in atopic boys with MS.

A RARE CAUSE OF CHRONIC CONSTIPATION IN THE FIELD OF FAMILY MEDICINE PRACTICE: DYSSYNERGIC DEFECATION

Constipation is a disorder that can result in infrequent defecation, pain, stiffness and difficult stool passage and pathogenesis is multifactorial. A 32-year-old female patient was admitted to family medicine clinic with the complaint of constipation that has been present since childhood. Dyssynergic defecation, which is one of the rare causes of chronic constipation, was detected in the patient whose etiological investigation was carried out. Chronic constipation is a condition that should be evaluated with a detailed medical history, taking into account the risk factors and etiological causes.

User Experience of Self-Reported Computerized Medical History Taking for Acute Chest Pain in the Clinical Expert Operating System - Chest Pain Danderyd Study (CLEOS-CPDS): Interview Study Among Patients (Preprint)

BACKGROUND Chest pain is one of the most common complaints in emergency departments (EDs). The effectiveness of care for patients with chest pain requires that individual circumstances and the complexity of the patient's clinical issues are determined by medical history taking. The knowledge base for history taking and for interpreting the clinical significance of the information collected can be formalized by software to enable computerized history taking (CHT) directly from patients and subsequently reporting the findings promptly to physicians. The adoption of CHT in clinical practice depends, among other issues, on reactions and attitudes to the technology from patients and their belief that the technology will have benefits for their medical care. OBJECTIVE The objective of this study was to explore the user experience of the self-reported CHT program CLEOS (Clinical Expert Operating System) in the setting of patients attending an ED for acute chest pain. METHODS This qualitative interview study is part of the ongoing CLEOS-Chest Pain Danderyd Study (CLEOS-CPDS), at a cardiology ED. A subset (n=84) of the larger sample who had taken part in self-reported history taking by the CLEOS program during waiting times at the ED were contacted by telephone and n=54 (64%) accepted participation. An interview guide with open-ended questions developed for this study was used. The text was analyzed using conventional content analysis in a deductive way by using the headings in the interview guide. RESULTS Six categories emerged: Clinical context, Individual context, Time aspect, Acceptability of the program, Usability of the program, and Perceptions of usefulness in a clinical context. A majority of the patients thought CLEOS had worked well. Patients also appreciated the opportunities, of interacting with the program and contributing to information about their own health while waiting for care. The program was generally perceived as easy to use and that the questions were mostly seen as relevant and straight forward. Many patients were of the opinion, however, that too many questions were asked and sometimes they mentioned there was insufficient time for responding to them. Some had found it difficult to find the strength to answer all the questions due to their condition. CONCLUSIONS The patients’ experience of the CLEOS program at a cardiology ED reflects an overall positive attitude. The CLEOS program was by some perceived as extensive, although most found the program user-friendly. Despite the busy ED environment, patients were highly motivated and felt that the program would be helpful in leading to a correct diagnosis. These findings suggest an important role for patient-entered CHT programs such as CLEOS in the setting of an ED from the perspective of the patients. CLINICALTRIAL ClinicalTrials.gov NCT03439449

A Human-Organisation-Technology (HOT) perspective on patients’ experiences of a chat- and automated medical history-taking service in primary healthcare: qualitative study (Preprint)

BACKGROUND Worldwide, the use of e-consultations in healthcare is progressing fast. So far, studies on the advantages and disadvantages of e-consultations in the form of chat services for all enquiries in primary care have focused on the perspective of the healthcare professionals rather than the end-users (patients). OBJECTIVE To explore patients´ experiences of using a chat- and automated medical history-taking service in Swedish regular tax-based not-for-profit primary care. METHODS In this qualitative study, 25 individual interviews were conducted with patients in the catchment areas of five primary care centres (PCCs) in Sweden that tested a chat- and automated medical history-taking service for all kinds of patient enquiries. The semi-structured interviews were transcribed verbatim prior to content analysis, using inductive and deductive strategies, the latter including an unconstrained matrix of Human, Organisation and Technology (HOT) perspectives. RESULTS The service provided an easily managed way for patients to make written contact, which was considered beneficial for some patients and issues, but less suitable for others (like acute or more complex cases). The automated medical history-taking service was perceived as having potential, but still derived from what healthcare professionals need to know and how they address and communicate health and healthcare issues. Technical skills were not considered as necessary for a mobile phone chat as for handling a computer, for example, but patients still expressed concern for people with less digital literacy. The opportunity for patients to take their time and reflect before answering questions from the healthcare professionals was found to be stress reducing and error preventing, and patients speculated that it might be the same for the healthcare professionals on the other end of the system. Patients appreciated the ability to have a conversation from almost anywhere, even from places not suitable for telephone calls. The asynchronicity of the chat service let the patients take more control of the conversation and initiate a chat at any time at their own convenience, but it could also lead to lengthy conversations where a single issue in the worst cases could take days to close. The opportunity to upload photographs made some visits to the PCC redundant which would otherwise have been necessary if a telephone service had been used, saving patients both time and money. CONCLUSIONS Patients generally had a positive attitude towards e-consultations in primary care and were generally pleased with the prospects of the digital tool tested, somewhat more with the actual chat than with the automated history-taking system preceding the chat. While patients expect their PCC to offer a range of different means of communication, the HOT analysis revealed a need for a more extensive (end) user-experience design in the further development of the PCC chat service.

Efficacy of Artificial-Intelligence-Driven Differential-Diagnosis List on the Diagnostic Accuracy of Physicians: An Open-Label Randomized Controlled Study

Background: The efficacy of artificial intelligence (AI)-driven automated medical-history-taking systems with AI-driven differential-diagnosis lists on physicians’ diagnostic accuracy was shown. However, considering the negative effects of AI-driven differential-diagnosis lists such as omission (physicians reject a correct diagnosis suggested by AI) and commission (physicians accept an incorrect diagnosis suggested by AI) errors, the efficacy of AI-driven automated medical-history-taking systems without AI-driven differential-diagnosis lists on physicians’ diagnostic accuracy should be evaluated. Objective: The present study was conducted to evaluate the efficacy of AI-driven automated medical-history-taking systems with or without AI-driven differential-diagnosis lists on physicians’ diagnostic accuracy. Methods: This randomized controlled study was conducted in January 2021 and included 22 physicians working at a university hospital. Participants were required to read 16 clinical vignettes in which the AI-driven medical history of real patients generated up to three differential diagnoses per case. Participants were divided into two groups: with and without an AI-driven differential-diagnosis list. Results: There was no significant difference in diagnostic accuracy between the two groups (57.4% vs. 56.3%, respectively; p = 0.91). Vignettes that included a correct diagnosis in the AI-generated list showed the greatest positive effect on physicians’ diagnostic accuracy (adjusted odds ratio 7.68; 95% CI 4.68–12.58; p < 0.001). In the group with AI-driven differential-diagnosis lists, 15.9% of diagnoses were omission errors and 14.8% were commission errors. Conclusions: Physicians’ diagnostic accuracy using AI-driven automated medical history did not differ between the groups with and without AI-driven differential-diagnosis lists.