Vital sign measurements demonstrate terminal digit bias and boundary effects.

OBJECTIVE: The measurement and recording of vital signs may be impacted by biases, including preferences for even and round numbers. However, other biases, such as variation due to defined numerical boundaries (also known as boundary effects), may be present in vital signs data and have not yet been investigated in a medical setting. We aimed to assess vital signs data for such biases. These parameters are clinically significant as they influence care escalation. METHODS: Vital signs data (heart rate, respiratory rate, oxygen saturation and systolic blood pressure) were collected from a tertiary hospital electronic medical record over a 2-year period. These data were analysed using polynomial regression with additional terms to assess for underreporting of out-of-range observations and overreporting numbers with terminal digits of 0 (round numbers), 2 (even numbers) and 5. RESULTS: It was found that heart rate, oxygen saturation and systolic blood pressure demonstrated 'boundary effects', with values inside the 'normal' range disproportionately more likely to be recorded. Even number bias was observed in systolic heart rate, respiratory rate and blood pressure. Preference for multiples of 5 was observed for heart rate and blood pressure. Independent overrepresentation of multiples of 10 was demonstrated in heart rate data. CONCLUSION: Although often considered objective, vital signs data are affected by bias. These biases may impact the care patients receive. Additionally, it may have implications for creating and training machine learning models that utilise vital signs data.

Impact of perioperative direct oral anticoagulant assays: a multicenter cohort study.

BACKGROUND: There is little evidence to guide the perioperative management of patients on a direct oral anticoagulant (DOAC) in the absence of a last known dose. Quantitative serum titers may be ordered, but there is little evidence supporting this. AIMS: This multi-center retrospective cohort study of consecutive surgical in-patients with a DOAC assay, performed over a five-year period, aimed to characterize preoperative DOAC assay orders and their impact on perioperative outcomes. MATERIALS AND METHODS: Patients prescribed regular DOAC (both prophylactic and therapeutic dosing) with a preoperative DOAC assay were included. The DOAC assay titer was evaluated against endpoints. Further, patients with an assay were compared against anticoagulated patients who did not receive a preoperative DOAC assay. The primary endpoint was major bleeding. Secondary endpoints included perioperative hemoglobin change, blood transfusions, idarucizumab or prothrombin complex concentrate administration, postoperative thrombosis, in-hospital mortality and reoperation. Adjusted and unadjusted linear regression models were used for continuous data. Binary logistic models were performed for dichotomous outcomes. RESULTS: 1065 patients were included, 232 had preoperative assays. Assays were ordered most commonly by Spinal (11.9%), Orthopedics (15.4%), and Neurosurgery (19.4%). For every 10 ng/ml increase in titer, the hemoglobin decreases by 0.5066 g/L and the odds of a preoperative reversal increases by 13%. Compared to those without an assay, patients with preoperative DOAC assays had odds 1.44× higher for major bleeding, 2.98× higher for in-hospital mortality and 16.3× higher for receiving anticoagulant reversal. CONCLUSION: A preoperative DOAC assay order was associated with worse outcomes despite increased reversal administration. However, the DOAC assay titer can reflect the patient's likelihood of bleeding.

Lost in relocation: longitudinal surveys evaluating the effectiveness of ICU to ward handover after the introduction of an electronic patient record.

BACKGROUND: Poor communication and lack of standardized handover practices contribute to adverse events. Intensive care organizations recommend standardized, structured written and verbal handover. OBJECTIVES: Investigate the effectiveness of, and barriers to, Intensive Care Unit (ICU) patient handover at ward transfer. Screen for patient safety incidents related to poor handover and improve practice where deficiencies are identified. METHODS: A survey of ward doctors about specific ICU to ward transfers and online surveys ascertaining opinions of handover processes were sent to ward-based and ICU doctors at a large, adult, university affiliated, Australian quaternary hospital. We delivered departmental education and created then publicized a new electronic ICU transfer summary. The summary included a mandatory tick-box to confirm verbal handover completion. Surveys re-assessing practice were then performed. RESULTS: Forty ward-based doctors were surveyed about specific transfers, with 7 (18%) instances of issues related to handover identified. Eighty-seven ward doctors completed the pre-interventions survey; 48 (55%) were aware of the existing written transfer summary. Post-interventions, 47 (75%) of 63 ward doctor responders were aware of it (p < 0.05). Pre-interventions, 14 (16%) ward doctors rated ICU handovers as excellent or good, rising to 21 (34%) post-interventions (p < 0.05). Thirty-nine ICU doctors completed the pre-interventions survey; 5 (13%) rated ICU to ward handover as excellent or good, rising to 9 (35%) when re-surveyed (p = 0.097). CONCLUSIONS: The perceived quality of ICU to ward handover improved after our interventions. However, ICU doctors continue to transfer patients without verbally handing over, with contacting the ward team representing a significant handover barrier.

Automated information extraction from free-text medical documents for stroke key performance indicators: a pilot study.

Automated information extraction might be able to assist with the collection of stroke key performance indicators (KPI). The feasibility of using natural language processing for classification-based KPI and datetime field extraction was assessed. Using free-text discharge summaries, random forest models achieved high levels of performance in classification tasks (area under the receiver operator curve 0.95-1.00). The datetime field extraction method was successful in 29 of 43 (67.4%) cases. Further studies are indicated.

Improving the accuracy of stroke clinical coding with open-source software and natural language processing.

Clinical coding is an important task, which is required for accurate activity-based funding. Natural language processing may be able to assist with improving the efficiency and accuracy of clinical coding. The aims of this study were to explore the feasibility of using natural language processing for stroke hospital admissions, employed with open-source software libraries, to aid in the identification of potentially misclassified (1) category of Adjacent Diagnosis Related Groups (ADRG), (2) the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM) diagnoses, and (3) Diagnosis Related Groups (DRG). Data was collected for consecutive individuals admitted to the Royal Adelaide Hospital Stroke Unit over a five-month period for misclassification identification analysis. 152 admissions were included in the study. Using free-text discharge summaries, a random forest classifier correctly identified two cases classified as B70 ("Stroke and Other Cerebrovascular Disorders") that should be classified as B02 (having received endovascular thrombectomy). A regular expression-based analysis correctly identified 33 cases in which ataxia was present but was not coded. Two cases were identified that should have been classified as B70D, rather than B70A/B/C, based on transfer to another centre within five days of admission. A variety of techniques may be useful to help identify misclassifications in ADRG, ICD-10-AM and DRG codes. Such techniques can be implemented with open-source software libraries, and may have significant financial implications. Future studies may seek to apply open-source software libraries to the identification of misclassifications of all ICD-10-AM diagnoses in stroke patients.