Use of Routine Emergency Department Care Practices with Deaf American Sign Language Users.

BACKGROUND: Deaf individuals who communicate using American Sign Language (ASL) seem to experience a range of disparities in health care, but there are few empirical data. OBJECTIVE: To examine the provision of common care practices in the emergency department (ED) to this population. METHODS: ED visits in 2018 at a U.S. academic medical center were assessed retrospectively in Deaf adults who primarily use ASL (n = 257) and hearing individuals who primarily use English, selected at random (n = 429). Logistic regression analyses adjusted for confounders compared the groups on the provision or nonprovision of four routine ED care practices (i.e., laboratories ordered, medications ordered, images ordered, placement of peripheral intravenous line [PIV]) and on ED disposition (admitted to hospital or not admitted). RESULTS: The ED encounters with Deaf ASL users were less likely to include laboratory tests being ordered: adjusted odds ratio 0.68 and 95% confidence interval 0.47-0.97. ED encounters with Deaf individuals were also less likely to include PIV placement, less likely to result in images being ordered in the ED care of ASL users of high acuity compared with English users of high acuity (but not low acuity), and less likely to result in hospital admission. CONCLUSION: Results suggest disparate provision of several types of routine ED care for adult Deaf ASL users. Limitations include the observational study design at a single site and reliance on the medical record, underscoring the need for further research and potential reasons for disparate ED care with Deaf individuals.

Applying machine learning classifiers to automate quality assessment of paediatric dynamic susceptibility contrast (DSC-) MRI data.

OBJECTIVE: Investigate the performance of qualitative review (QR) for assessing dynamic susceptibility contrast (DSC-) MRI data quality in paediatric normal brain and develop an automated alternative to QR. METHODS: 1027 signal-time courses were assessed by Reviewer 1 using QR. 243 were additionally assessed by Reviewer 2 and % disagreements and Cohen's κ (κ) were calculated. The signal drop-to-noise ratio (SDNR), root mean square error (RMSE), full width half maximum (FWHM) and percentage signal recovery (PSR) were calculated for the 1027 signal-time courses. Data quality thresholds for each measure were determined using QR results. The measures and QR results trained machine learning classifiers. Sensitivity, specificity, precision, classification error and area under the curve from a receiver operating characteristic curve were calculated for each threshold and classifier. RESULTS: Comparing reviewers gave 7% disagreements and κ = 0.83. Data quality thresholds of: 7.6 for SDNR; 0.019 for RMSE; 3 s and 19 s for FWHM; and 42.9 and 130.4% for PSR were produced. SDNR gave the best sensitivity, specificity, precision, classification error and area under the curve values of 0.86, 0.86, 0.93, 14.2% and 0.83. Random forest was the best machine learning classifier, giving sensitivity, specificity, precision, classification error and area under the curve of 0.94, 0.83, 0.93, 9.3% and 0.89. CONCLUSION: The reviewers showed good agreement. Machine learning classifiers trained on signal-time course measures and QR can assess quality. Combining multiple measures reduces misclassification. ADVANCES IN KNOWLEDGE: A new automated quality control method was developed, which trained machine learning classifiers using QR results.