So Many Ways to Be Wrong: Completeness and Accuracy in a Prospective Study of OR-to-ICU Handoff Standardization.

BACKGROUND: Studies focused on improving handoffs often measure the quality of information exchange using information completeness without reporting on accuracy. The present investigation aimed to characterize changes in the accuracy of transmitted patient information after standardization of operating room (OR)-to-ICU handoffs. METHODS: Handoffs and Transitions in Critical Care (HATRICC) was a mixed methods study conducted in two US ICUs. From 2014 to 2016, trained observers captured the nature and content of information transmitted during OR-to-ICU handoffs, comparing this to the electronic medical record. Inconsistencies were compared before and after handoff standardization. Semistructured interviews initially conducted for implementation were reanalyzed to contextualize quantitative findings. RESULTS: A total of 160 OR-to-ICU handoffs were observed-63 before and 97 after standardization. Across seven categories of information, including allergies, past surgical history, and IV fluids, two types of inaccuracy were observed: incomplete information (for example, providing only a partial list of allergies) and incorrect information. Before standardization, an average of 3.5 information elements per handoff were incomplete, and 0.11 were incorrect. After standardization, the number of incomplete information elements per handoff decreased to 2.4 (-1.1, p < 0.001), and the number of incorrect items was similar, at 0.16 (p = 0.54). Interviews revealed that the familiarity of a transporting OR provider (for example, surgeon, anesthetist) with the patient's case was considered an important factor affecting information exchange. CONCLUSION: Handoff accuracy improved after standardizing OR-to-ICU handoffs in a two-ICU study. The improvement in accuracy was due to improved completeness rather than a change in the transmission of inaccurate information.

Systematic Approach for Comparing Team-based Care Processes.

Team-based care process modeling techniques have focused on understanding and designing solutions for a single site. Less is known about tailoring an effective team-based care process from one site to another, which is necessary for multi-site implementation efforts. We propose an approach for analyzing and comparing a team-based care process performed at two sites to inform redesign opportunities. Our approach includes abstracting the goals and strategies of each process by identifying whether sociotechnical system element differences exist. Element differences may exist for the phase, tasks, roles, information, and technology and tools. Differences in system elements may still support process goals and strategies and, thus, be irrelevant for redesign opportunities. We demonstrate the utility of the approach using an operating room to intensive care unit handoff protocol. This approach should be useful for researchers and practitioners that are tailoring and implementing a successful team-based care process at more than one site.

A machine-learning algorithm for distinguishing malignant from benign indeterminate thyroid nodules using ultrasound radiomic features.

Background: Ultrasound (US)-guided fine needle aspiration (FNA) cytology is the gold standard for the evaluation of thyroid nodules. However, up to 30% of FNA results are indeterminate, requiring further testing. In this study, we present a machine-learning analysis of indeterminate thyroid nodules on ultrasound with the aim to improve cancer diagnosis. Methods: Ultrasound images were collected from two institutions and labeled according to their FNA (F) and surgical pathology (S) diagnoses [malignant (M), benign (B), and indeterminate (I)]. Subgroup breakdown (FS) included: 90 BB, 83 IB, 70 MM, and 59 IM thyroid nodules. Margins of thyroid nodules were manually annotated, and computerized radiomic texture analysis was conducted within tumor contours. Initial investigation was conducted using five-fold cross-validation paradigm with a two-class Bayesian artificial neural networks classifier, including stepwise feature selection. Testing was conducted on an independent set and compared with a commercial molecular testing platform. Performance was evaluated using receiver operating characteristic analysis in the task of distinguishing between malignant and benign nodules. Results: About 1052 ultrasound images from 302 thyroid nodules were used for radiomic feature extraction and analysis. On the training/validation set comprising 263 nodules, five-fold cross-validation yielded area under curves (AUCs) of 0.75 [Standard Error (SE) = 0.04; P < 0.001 ] and 0.67 (SE = 0.05; P = 0.0012 ) for the classification tasks of MM versus BB, and IM versus IB, respectively. On an independent test set of 19 IM/IB cases, the algorithm for distinguishing indeterminate nodules yielded an AUC value of 0.88 (SE = 0.09; P < 0.001 ), which was higher than the AUC of a commercially available molecular testing platform (AUC = 0.81, SE = 0.11; P < 0.005 ). Conclusion: Machine learning of computer-extracted texture features on gray-scale ultrasound images showed promising results classifying indeterminate thyroid nodules according to their surgical pathology.