BACKGROUND: Diagnostic errors are an underappreciated cause of preventable mortality in hospitals and pose a risk for severe patient harm and increase hospital length of stay.
OBJECTIVE: This study aims to explore the potential of machine learning and natural language processing techniques in improving diagnostic safety surveillance. We conducted a rigorous evaluation of the feasibility and potential to use electronic health records clinical notes and existing case review data.
METHODS: Safety Learning System case review data from 1 large health system composed of 10 hospitals in the mid-Atlantic region of the United States from February 2016 to September 2021 were analyzed. The case review outcome included opportunities for improvement including diagnostic opportunities for improvement. To supplement case review data, electronic health record clinical notes were extracted and analyzed. A simple logistic regression model along with 3 forms of logistic regression models (ie, Least Absolute Shrinkage and Selection Operator, Ridge, and Elastic Net) with regularization functions was trained on this data to compare classification performances in classifying patients who experienced diagnostic errors during hospitalization. Further, statistical tests were conducted to find significant differences between female and male patients who experienced diagnostic errors.
RESULTS: In total, 126 (7.4%) patients (of 1704) had been identified by case reviewers as having experienced at least 1 diagnostic error. Patients who had experienced diagnostic error were grouped by sex: 59 (7.1%) of the 830 women and 67 (7.7%) of the 874 men. Among the patients who experienced a diagnostic error, female patients were older (median 72, IQR 66-80 vs median 67, IQR 57-76; P=.02), had higher rates of being admitted through general or internal medicine (69.5% vs 47.8%; P=.01), lower rates of cardiovascular-related admitted diagnosis (11.9% vs 28.4%; P=.02), and lower rates of being admitted through neurology department (2.3% vs 13.4%; P=.04). The Ridge model achieved the highest area under the receiver operating characteristic curve (0.885), specificity (0.797), positive predictive value (PPV; 0.24), and F1-score (0.369) in classifying patients who were at higher risk of diagnostic errors among hospitalized patients.
CONCLUSIONS: Our findings demonstrate that natural language processing can be a potential solution to more effectively identifying and selecting potential diagnostic error cases for review and therefore reducing the case review burden.

Periodontal disease constitutes a widely prevalent category of non-communicable diseases and ranks among the top 10 causes of disability worldwide. Little however is known about diagnostic errors in dentistry. In this work, by retrospectively deploying an electronic health record (EHR)-based trigger tool, followed by gold standard manual review, we provide epidemiological estimates on the rate of diagnostic misclassification in dentistry through a periodontal use case.
An EHR-based trigger tool (a retrospective record review instrument that uses a list of triggers (or clues), i.e., data elements within the health record, to alert reviewers to the potential presence of a wrong diagnosis) was developed, tested and run against the EHR at the two participating sites to flag all cases having a potential misdiagnosis. All cases flagged as potentially misdiagnosed underwent extensive manual reviews by two calibrated domain experts. A subset of the non-flagged cases was also manually reviewed.
A total of 2,262 patient charts met the study\'s inclusion criteria. Of these, the algorithm flagged 1,124 cases as potentially misclassified and 1,138 cases as potentially correctly diagnosed. When the algorithm identified a case as potentially misclassified, compared to the diagnosis assigned by the gold standard, the kappa statistic was 0.01. However, for cases the algorithm marked as potentially correctly diagnosed, the review against the gold standard showed a kappa statistic of 0.9, indicating near perfect agreement. The observed proportion of diagnostic misclassification was 32 %. There was no significant difference by clinic or provider characteristics.
Our work revealed that about a third of periodontal cases are misclassified. Diagnostic errors have been reported to happen more frequently than other types of errors, and to be more preventable. Benchmarking diagnostic quality is a first step. Subsequent research endeavor will delve into comprehending the factors that contribute to diagnostic errors in dentistry and instituting measures to prevent them.
This study sheds light on the significance of diagnostic excellence in the delivery of dental care, and highlights the potential role of technology in aiding diagnostic decision-making at the point of care.

BACKGROUND: Panoramic radiography quality can be impaired by some errors such as positioning errors. Palatoglossal air space shadow error is one of the most common positioning errors and it is due to the tongue not sticking to the roof of the palate. Techniques used to deal with this error might help prevent unnecessary radiation to patients and save them time and money. The study aimed to investigate the effects of using celluloid matrix and edible tapes (fruit leather and chewing gum) on reducing the palatoglossal air space shadow error in panoramic imaging.
METHODS: In our study, 270 patients referred to the Department of Radiology were randomised into three groups: a control group, a celluloid matrix group and an edible tapes group. Before panoramic imaging, all patients were instructed to adhere their tongues to the roof of their mouths, with the distinction that for the celluloid matrix and edible tapes groups, patients were asked to place celluloid tapes, fruit leathers, or chewing gums on their tongues before doing so. The routine imaging process was then performed, and the results were compared across groups to evaluate the incidence of palatoglossal air space shadow error.
RESULTS: The number of error-free images in each fruit leather, chewing gum and celluloid tape group were significantly higher than the control group (all cases P < 0.05). The chances of error-free images in the fruit leather groups were the highest (9.57 times). The age (P = 0.136) and gender (P = 0.272) of patients had no significant effect on the results of interventions.
CONCLUSIONS: The application of fruit leathers, chewing gums and celluloid tapes reduced the palatoglossal air space shadow error of panoramic imaging.

The 2021 US Cures Act may engage patients to help reduce diagnostic errors/delays. We examined the relationship between patient portal registration with/without note reading and test/referral completion in primary care.
Retrospective cohort study of patients with visits from January 1, 2018 to December 31, 2021, and order for (1) colonoscopy, (2) dermatology referral for concerning lesions, or (3) cardiac stress test at 2 academic primary care clinics. We examined differences in timely completion (\"loop closure\") of tests/referrals for (1) patients who used the portal and read ≥1 note (Portal + Notes); (2) those with a portal account but who did not read notes (Portal Account Only); and (3) those who did not register for the portal (No Portal). We estimated the predictive probability of loop closure in each group after adjusting for socio-demographic and clinical factors using multivariable logistic regression.
Among 12 849 tests/referrals, loop closure was more common among Portal+Note-readers compared to their counterparts for all tests/referrals (54.2% No Portal, 57.4% Portal Account Only, 61.6% Portal+Notes, P < .001). In adjusted analysis, compared to the No Portal group, the odds of loop closure were significantly higher for Portal Account Only (OR 1.2; 95% CI, 1.1-1.4), and Portal+Notes (OR 1.4; 95% CI, 1.3-1.6) groups. Beyond portal registration, note reading was independently associated with loop closure (P = .002).
Compared to no portal registration, the odds of loop closure were 20% higher in tests/referrals for patients with a portal account, and 40% higher in tests/referrals for note readers, after controlling for sociodemographic and clinical factors. However, important safety gaps from unclosed loops remain, requiring additional engagement strategies.

OBJECTIVE: Diagnostic errors are a source of morbidity and mortality in intensive care unit (ICU) patients. However, contextual factors influencing clinicians\' diagnostic performance have not been studied in authentic ICU settings. We sought to determine the accuracy of ICU clinicians\' diagnostic impressions and to characterize how various contextual factors, including self-reported stress levels and perceptions about the patient\'s prognosis and complexity, impact diagnostic accuracy. We also explored diagnostic calibration, i.e. the balance of accuracy and confidence, among ICU clinicians.
METHODS: We conducted an observational cohort study in an academic medical ICU. Between June and August 2019, we interviewed ICU clinicians during routine care about their patients\' diagnoses, their confidence, and other contextual factors. Subsequently, using adjudicated final diagnoses as the reference standard, two investigators independently rated clinicians\' diagnostic accuracy and on each patient on a given day (\"patient-day\") using 5-point Likert scales. We conducted analyses using both restrictive and conservative definitions of clinicians\' accuracy based on the two reviewers\' ratings of accuracy.
RESULTS: We reviewed clinicians\' responses for 464 unique patient-days, which included 255 total patients. Attending physicians had the greatest diagnostic accuracy (77-90 %, rated as three or higher on 5-point Likert scale) followed by the team\'s primary fellow (73-88 %). Attending physician and fellows were also least affected by contextual factors. Diagnostic calibration was greatest among ICU fellows.
CONCLUSIONS: Additional studies are needed to better understand how contextual factors influence different clinicians\' diagnostic reasoning in the ICU.

BACKGROUND: The gamification of learning increases student enjoyment, and motivation and engagement in learning tasks. This study investigated the effects of gamification using decision-making cards (DMCs) on diagnostic decision-making and cost using case scenarios.
METHODS: Thirty medical students in clinical clerkship participated and were randomly assigned to 14 small groups of 2-3 medical students each. Decision-making was gamified using DMCs with a clinical information heading and medical cost on the front, and clinical information details on the back. First, each team was provided with brief clinical information on case scenarios. Subsequently, DMCs depending on the case were distributed to each team, and team members chose cards one at a time until they reached a diagnosis of the case. The total medical cost was then scored based on the number and contents of cards drawn. Four case scenarios were conducted. The quantitative outcomes including confidence in effective clinical decision-making, motivation to learn diagnostic decision-making, and awareness of medical costs were measured before and after our gamification by self-evaluation using a 7-point Likert scale. The qualitative component consisted of a content analysis on the benefits of learning clinical reasoning using DMCs.
RESULTS: Confidence in effective clinical decision-making, motivation to learn diagnostic decision-making, and awareness of medical cost were significantly higher after the gamification. Furthermore, comparing the clinical case scenario tackled last with the one tackled first, the average medical cost of all cards drawn by students decreased significantly from 11,921 to 8,895 Japanese yen. In the content analysis, seven advantage categories of DMCs corresponding to clinical reasoning components were extracted (information gathering, hypothesis generation, problem representation, differential diagnosis, leading or working diagnosis, diagnostic justification, and management and treatment).
CONCLUSIONS: Teaching medical students clinical reasoning using DMCs can improve clinical decision-making confidence and learning motivation, and reduces medical cost in clinical case scenarios. In addition, it can help students to acquire practical knowledge, deepens their understanding of clinical reasoning, and identifies several important clinical reasoning skills including diagnostic decision-making and awareness of medical costs. Gamification using DMCs can be an effective teaching method for improving medical students\' diagnostic decision-making and reducing costs.

OBJECTIVE: Diagnostic errors pose a significant risk to patient safety and have substantial medical and economic consequences. Despite their importance, diagnostic error education is currently lacking in standard pre-graduate curricula. This study aimed to investigate the incidence of diagnostic errors and the frequency of recognition among medical students in Japan.
METHODS: A pilot survey was conducted immediately after the General Medicine In-Training Examination (GM-ITE), a comprehensive post-graduation test, administered to new residents right after graduation from medical school. The survey assessed whether they received education on diagnostic errors during their formal undergraduate medical education and whether they recognized diagnostic errors during their clinical training.
RESULTS: Of the 564 examinees, 421 participated in the study. The majority of participants (63.9 %) reported receiving education on diagnostic errors, and 15.7 % recognized diagnostic errors during their clinical training. Significantly, those who received education on diagnostic errors had a higher rate of recognizing such errors compared to those who did not (19.7 vs. 8.6 %; p=0.0017).
CONCLUSIONS: These findings suggest that the recognition rate of diagnostic errors increases with improved literacy in diagnostic error education. This highlights the importance of incorporating diagnostic error education into medical curricula to develop effective strategies to prevent and manage diagnostic errors, and thereby enhance medical and patient safety. However, this study did not examine the specific educational content of the errors or the details of the recognition, necessitating further investigation in the future.

OBJECTIVE: Current autopsy practice guidelines do not provide a mechanism to identify potential causes of diagnostic error (DE). We used our autopsy data registry to ask if gender or race were related to the frequency of diagnostic error found at autopsy.
METHODS: Our autopsy reports include International Classification of Diseases (ICD) 9 or ICD 10 diagnostic codes for major diagnoses as well as codes that identify types of error. From 2012 to mid-2015 only 2 codes were used: UNDOC (major undocumented diagnoses) and UNCON (major unconfirmed diagnoses). Major diagnoses contributed to death or would have been treated if known. Since mid-2015, codes included specific diagnoses, i.e. undiagnosed or unconfirmed myocardial infarction, infection, pulmonary thromboembolism, malignancy, or other diagnosis as well as cause of death. Adult autopsy cases from 2012 to 2019 were assessed for DE associated with reported sex or race (nonwhite or white). 528 cases were evaluated between 2012 and 2015 and 699 between 2015 and 2019.
RESULTS: Major DEs were identified at autopsy in 65.9 % of cases from 2012 to 2015 and in 72.1 % from 2015 to 2019. From 2012 to 2015, female autopsy cases showed a greater frequency in 4 parameters of DE, i.e., in the total number of cases with any error (p=0.0001), in the number of cases with UNDOC errors (p=0.0038) or UNCON errors (p=0.0006), and in the relative proportions of total numbers of errors (p=0.0001). From 2015 to 2019 undocumented malignancy was greater among males (p=0.0065); no other sex-related error was identified. In the same period some DE parameters were greater among nonwhite than among white subjects, including unconfirmed cause of death (p=0.035), and proportion of total error diagnoses (p=0.0003), UNCON diagnoses (p=0.0093), and UNDOC diagnoses (p=0.035).
CONCLUSIONS: Coding for DE at autopsy can identify potential effects of biases on diagnostic error.

BACKGROUND: Accurate interpretation of radiographs is crucial for junior doctors in the accident and emergency (A&E) department (the emergency medicine department). However, it remains a significant challenge and a leading cause of diagnostic errors.
OBJECTIVE: This study aimed to evaluate the accuracy and confidence of foundation doctors (doctors within their first 2 years of qualifying) in correctly interpreting and managing forearm and hand fractures on plain radiographs.
METHODS: A total of 42 foundation doctors with less than 2 years of experience and no prior emergency medicine training who worked in a large district general hospital participated in a web-based questionnaire. The questionnaire consisted of 3 case studies: distal radius fracture, scaphoid fracture, and a normal radiograph. Respondents were required to identify the presence or absence of a fracture, determine the fracture location, suggest appropriate management, and rate their confidence on a Likert scale.
RESULTS: Overall, 48% (61/126) of respondents accurately identified the presence and location of fractures. The correct management option was chosen by 64% (81/126) of respondents. The median diagnostic confidence score was 4 of 10, with a mean diagnostic certainty of 4.4 of 10. Notably, respondents exhibited a significantly lower confidence score for the normal radiograph compared to the distal radius fracture radiograph (P=.01).
CONCLUSIONS: This study reveals diagnostic uncertainty among foundation doctors in interpreting plain radiographs, with a notable inclination toward overdiagnosing fractures. The findings emphasize the need for close supervision and senior support to mitigate diagnostic errors. Further training and educational interventions are warranted to improve the accuracy and confidence of junior doctors in radiographic interpretation. This study has several limitations, including a small sample size and reliance on self-reported data. The findings may not be generalizable to other health care settings or specialties. Future research should aim for larger, more diverse samples and explore the impact of specific educational interventions on diagnostic accuracy and confidence.

Diagnostic error has recently become a crucial clinical problem and an area of intense research. However, the reality of diagnostic errors in regional hospitals remains unknown. This study aimed to clarify the reality of diagnostic errors in regional hospitals in Japan. A 10-month retrospective cohort study was conducted from January to October 2021 at the emergency room of Oda Municipal Hospital in central Shimane Prefecture, Japan. Participants were divided into groups with or without diagnostic errors, and independent variables of patient, physician, and environmental factors were analyzed using Fisher\'s exact test, univariate (Student\'s t-test and Welch\'s t-test), and logistic regression analyses. Diagnostic errors accounted for 13.1% of all eligible cases. Remarkably, the proportion of patients treated without oxygen support and the proportion of male patients were significantly higher in the group with diagnostic errors. Sex bias was present. Additionally, cognitive bias, a major factor in diagnostic errors, may have occurred in patients who did not require oxygen support. Numerous factors contribute to diagnostic errors; however, it is important to understand the trends in the setting of each healthcare facility and plan and implement individualized countermeasures.