Diagnosis最新文献

Objectives: Context specificity occurs when a health professional sees two patients with identical signs and symptoms yet arrives at two different diagnoses due to other existing factors. For example, one patient speaks English as a first language, while the other patient has limited English proficiency. It is not known if context specificity extends beyond diagnosis and also affects management reasoning. Our study explored whether reduced sleep and other distracting contextual factors (e.g., limited English proficiency) lead to context specificity, resulting in suboptimal management reasoning.

Methods: Seventeen medical residents participated in a two-month study (consisting of one outpatient and one inpatient rotation), in which their sleep was tracked. After each rotation, participants watched two clinical encounter videos-one with and one without distracting contextual factors-and completed think-aloud interviews for each video discussing their management plans. Interviews were transcribed and assessed for management reasoning themes.

Results: Residents (n=17) on outpatient rotations received more sleep than those on inpatient rotations (450.5 min ± 7.13 vs. 425.6 min ± 10.78, p=0.023). Five management reasoning themes were identified: organized knowledge, disorganized knowledge, uncertainty, addressing non-pharmacologic interventions, and addressing patient needs and concerns. There was essentially no difference in the prevalence of utterances of organized knowledge themes between residents with more or less sleep (25 vs. 27 times, p=0.78) or those exposed to contextual factors vs. not exposed (24 vs. 28 times, p=0.58). However, disorganized knowledge themes were observed significantly more frequently in participants exposed to contextual factors (33 vs. 18 times, p=0.036).

Conclusions: Residents slept more during outpatient rotations. While sleep alone was not associated with the prevalence of management reasoning themes, residents exposed to videos with distracting contextual factors displayed significantly more instances of disorganized knowledge, supporting the phenomenon of context specificity in management reasoning.

Objectives: Emergency department (ED) encounters carry high risk of diagnostic error. Understanding how providers process information and reach diagnostic conclusions may identify interventions to reduce diagnostic errors. We aimed to determine if pediatric ED notes documenting a simple inventory of alternative diagnoses (inventorial differential diagnosis (DDx)) increased the odds of subsequent diagnostic error compared to encounters where the DDx was explicitly linked to specific data elements in the encounter (synthesized).

Methods: This is a cohort study of children 0-22 years who experienced unplanned admission within 10 days of an index pediatric ED or urgent care visit. Documented DDx (inventorial vs. synthesized) in the index visit notes served as the predictor variable. The primary outcome was presence of diagnostic error. Propensity scores were created using patient demographics and complexity and visit acuity. Propensity score matched patients were compared with multivariable conditional logistic regression to assess association between documented DDx and diagnostic error.

Results: Propensity scores matched 303 patient pairs of inventorial and synthesized DDx from 869 charts screened in for review. The adjusted odds ratio for diagnostic error at a subsequent unplanned admission was 1.79 (95 % CI 1.17-2.75) when an inventorial DDx was documented relative to synthesized. This finding includes adjustments for the number of diagnostic tests, obtaining a subspecialty consult and number of hospitalizations in the prior 6 months.

Conclusions: An inventorial DDx in pediatric emergency medical decision making is associated with significantly higher odds of subsequent diagnostic error, offering an actionable, simple opportunity for all providers to improve patient care.

Objectives: We performed a comparative evaluation of analytical performance between the novel bench-top Stago sthemO 301 hemostasis analyzer and the Werfen ACL TOP 750 coagulometer using routine plasma samples.

Methods: A minimum of 100 fresh plasma samples per test were analyzed using both analytical systems. The tests included prothrombin time (PT) expressed as international normalized ratio (INR) and seconds (s), activated partial thromboplastin time (APTT) in ratio and seconds, fibrinogen, antithrombin and D-dimer. Clotting assays were performed using mechanical clot detection on sthemO and optical detection on ACL TOP. The comparative evaluation included the calculation of Spearman's correlation, Passing-Bablok regression and Bland-Altman plots.

Results: Correlation coefficients ranged between 0.76 for APTT to 0.98 for PT-INR and D-dimer, indicating a generally acceptable/good agreement. The regression slopes varied from 0.82 for D-dimer to 1.17 for APTT-s. A significant bias was observed for all tests except antithrombin, with differences for sthemO ranging between -31 % for D-dimer and 13.7 % for PT-s.

Conclusions: SthemO demonstrates acceptable global comparability with ACL TOP for routine coagulation testing. Nevertheless, reagent- and method-dependent bias has been observed, which highlight the need for additional harmonization efforts.

Objectives: Biomarkers are useful clinical tools but only a handful of them are used routinely for patient care. Despite intense efforts to discover new, clinically useful biomarkers, very few new circulating biomarkers were implemented in clinical practice in the last 40 years. This is mainly due to rather poor clinical performance. Here, our goal was to validate the performance of a group of newly discovered circulating biomarkers for glioma by comparing our data with data from a paper recently published in Science Advances.

Methods: We analyzed our own sets of clinical samples (gliomas (n=30), meningiomas (n=20)) and a different analytical assay (Proximity Extension Assay, OLINK Proteomics) to compare the results of Shen and colleagues.

Results: Despite the sophistication of the utilized discovery method by the original investigators, we found that the newly proposed biomarkers for glioma (the best one presumably being SERPINA6) did not perform as originally claimed.

Conclusions: Scientific irreproducibility has been extensively discussed in the literature. A large proportion of newly discovered candidate biomarkers likely represent "false discovery" and significantly contribute to the propagation of irreproducible results between investigators. One of the best ways to assess the value of any new biomarker is by independent and extensive validation. Based on our previous classification of irreproducible results, we believe that this new work likely represents another example of biomarker false discovery.

Objectives: Diagnostic and management reasoning of neurological disorders may present unique challenges and uncertainty for clinicians, particularly in emergency department settings. This study aims to assess the level of uncertainty emergency department physicians experience when diagnosing neurological versus non-neurological conditions, and whether this uncertainty extends to the management of neurological conditions. Additionally, the study explores whether clinical experience is related to perceived diagnostic and/or management uncertainty.

Methods: Fifty-three emergency department physicians completed a survey measuring diagnostic uncertainty, management uncertainty, and associated anxiety. The survey included clinical vignettes depicting neurological and non-neurological cases, as well as items which assessed perceived diagnostic and management uncertainty across eight different specialties. Statistical analyses included paired samples t-test for comparing uncertainty between neurology and non-neurology cases and a general linear model to assess relationship between clinical experience and uncertainty.

Results: Emergency department physicians reported greater diagnostic uncertainty for neurological vignettes compared to non-neurological vignettes (Cohen's d=1.37), as well as greater management uncertainty (Cohen's d=1.41). They also reported greater anxiety when diagnosing neurological cases compared to non-neurological cases (Cohen's d=1.33), as well as greater anxiety when managing them (Cohen's d=0.69). Exploratory analyses indicated that with greater experience, management uncertainty of neurology cases decreased, while diagnostic uncertainty remained unchanged.

Conclusions: The results suggest unique diagnostic and management challenges posed by neurological cases in emergency departments, particularly for less experienced providers. Future research could focus on developing interventions to reduce diagnostic and management uncertainty in neurological conditions.

Objectives: To compare proportions of pediatric emergency department (PED) patients with missed opportunities for diagnostic excellence (MODEs) by patient race and ethnicity, defined as either White non-Hispanic/Latino (WNH), or non-WNH. In addition, to assess the thoroughness of the PED patient evaluation by patient race and ethnicity.

Methods: Electronic trigger (E-trigger) followed by manual screening identified children with unplanned admission within 10 days of an index PED or pediatric urgent care (PUC) encounter from January 2018 through July 2022. Cases with disparate diagnoses at index encounter and hospital discharge were reviewed using the Revised Safer Dx tool to determine the presence of a MODE. Patient race and ethnicity were abstracted from the electronic record. The primary outcome was proportion of MODEs by race and ethnicity, analyzed using univariate comparisons; the secondary outcome was the completeness of the diagnostic evaluation. Independent predictors of MODEs were identified following multivariable logistic regression analysis.

Results: A total of 816 patients were screened in for Revised Safer Dx review, and a total of 183 potential MODEs were identified. Non-WNH populations did not differ significantly by proportion of potential MODEs when compared to WNH patients. WNH patients received a higher median number of diagnostic tests (p=0.02), more diagnostic workup (p=0.03), and more frequently had the eventual correct diagnosis initially considered (p=0.02) than non-WNH patients. Race and ethnicity did not significantly affect the odds of a MODE.

Conclusions: While race and ethnicity did not predict higher odds of a MODE, non-WNH PED/PUC populations received disparate levels of diagnostic consideration.

Diagnostic error is a pervasive problem in healthcare with approximately one-third of adverse events in hospitals attributed to a failure in the diagnostic process. Cognitive biases are systematic, often unconscious, automatic patterns of thought that sometimes skew thinking and are considered a major contributor to diagnostic error. More than 100 different biases have been described that affect clinical decision-making, and the challenge for educators and clinicians is bringing the conceptual knowledge of cognitive bias to the bedside in an applicable and useful way to mitigate the effects of cognitive bias in diagnosis. The language that is commonly used around cognitive bias is technical in nature, often with complicated and nuanced descriptions, so developing a clear understanding of cognitive bias is a task that needs sophisticated language and memory skills as well as clinical reasoning skills. A novel language approach to learning and talking about biases in medicine is to use idioms, short phrases with a particular meaning that differs from the meaning of each word on their own, to simplify the terminology and improve recognition of cognitive bias at the frontline. We present 'The Idiom's Guide to Cognitive Bias', a Table that lists 21 common cognitive biases in the diagnostic process, and defines each, offering a healthcare example and possible explanation for why each occurs. The benefit of The Guide is its practical approach to reinforcing cognitive and medical concepts through the synergy of language and imagery and to demystify cognitive bias in the diagnostic process.

Objectives: Diagnostic errors represent the most common and costly preventable patient safety events, with historically marginalized populations disproportionately impacted due to systemic inequities in healthcare. Addressing these disparities requires embedding equity into every facet of the diagnostic process. The aim was to develop, refine, and validate a competency framework for Equity-Driven Diagnostic Excellence (DxEqEx).

Methods: A modified Delphi method was used, involving transdisciplinary diverse healthcare system participants, including patient advocates, physicians, nurses, and other healthcare professionals. Participants were guided through multiple rounds of feedback and ratings, assessing the importance, disciplinary relevance, feasibility, skill acquisition level required, granularity, and representativeness of the DxEqEx framework.

Results: Sixteen essential competencies have been identified, categorized into three domains: Intrapersonal, Team-based, and Structural. Participants rated the framework with high importance and strong relevance to their respective disciplines. However, the feasibility of implementing the framework varied, largely due to broader challenges within the healthcare system. The competencies were assessed as requiring a proficient skill level according to Dreyfus' model. The final round maintained strong ratings for granularity and representativeness, which supported the final version of the framework.

Conclusions: The DxEqEx framework holds significant potential to proactively address the needs of historically marginalized patients throughout the diagnostic process. Future research should focus on participatory, resource-efficient implementation.

Objectives: This study aimed to develop and implement autoverification (AV) system for routine coagulation assays, specifically prothrombin time (PT) and activated partial thromboplastin time (APTT), in tertiary care hospital. The efficiency, accuracy, and impact on turnaround time (TAT) were evaluated.

Methods: AV rules were developed using historical data from 70,865 coagulation test results. The rules included pre-analytical, analytical, and post-analytical checks. The system underwent validation through data simulations, pilot phase, go-live implementation. Performance metrics included sensitivity, specificity, predictive values, passing rates, error rates, TAT.

Results: The AV system achieved 63.3 % overall passing rate (analyzed from 159,183 data), with outpatient settings showing higher rate (69.2 %) than inpatient settings (56.3 %). Final performance evaluation showed sensitivity, specificity, PPV, and NPV of 93.0 , 65.0, 59.7, and 94.4 %, respectively. Manual verification was required for 36 % of cases, mainly due to defective sample volumes (21.5 %). False negatives, primarily from partial clots, occurred in 0.1 % of cases. Integrating CBC clot alerts into AV rules halved the errors. The system increased tests completed within guaranteed TAT of 90 min by 2.4 %, from 89.7 to 92.1 % and reduced median TAT by 5 min. Outpatient TAT improved significantly, with a reduction over 19 min.

Conclusions: The AV system for APTT and PT tests was successfully implemented, reducing manual verification, improving TAT, particularly in outpatient settings. This study highlights AV systems' potential to enhance laboratory performance for routine coagulation panels, which rely only on APTT and PT assays. Ongoing rule refinement and monitoring remain crucial for enhancing system accuracy and effectiveness.