Search

Background: Machine learning (ML) analysis of biometric data in non-controlled environments is underexplored., Objective: To evaluate whether ML analysis of physical activity data can be employed to classify whether individuals have postural dysfunction in middle-aged and older individuals., Methods: A 1 week period of physical activity was measured by a waist-worn uni-axial accelerometer during the 2003-2004 National Health and Nutrition Examination Survey sampling period. Features of physical activity along with basic demographic information (42 variables) were paired with ML models to predict the success or failure of a standard 30 s modified Romberg test during which participants had their eyes closed and stood upon a 3-inch compliant surface. Model performance was evaluated by area under the receiver operating characteristic curve (AUC-ROC), balanced accuracy, and F1-score., Results: The cohort was comprised of 1625 participants ≥40 years (median age 61, IQR 51-71). Approximately half (47%) were diagnosed with postural dysfunction having failed the binarized (pass/fail) scoring mechanism of the modified Romberg exam. Five ML models were trained on the classification task, achieving AUC values ranging from 0.67 to 0.73. The support vector machine (SVM) and a gradient-boosted model, XGBoost, achieved the highest AUC of 0.73 (SD 0.71-0.75). Age was the most important variable for SVM classification, followed by four features that evaluated accelerometer counts at various thresholds, including those delineating total, moderate, and moderate-vigorous activity., Conclusions: ML analysis of accelerometer-derived physical activity data to classify postural dysfunction in middle-aged and older individuals is feasible in real-world environments such as the home., Level of Evidence: 3 Laryngoscope, 133:3529-3533, 2023., (© 2023 The Authors. The Laryngoscope published by Wiley Periodicals LLC on behalf of The American Laryngological, Rhinological and Otological Society, Inc.)

Objectives/hypothesis: To formally document online support community (OSC) use among patients with vestibular symptoms and gain an appreciation for the perceived influence of participation on psychosocial outcomes and the impact on medical decision-making., Study Design: Self reported internet-based questionnaire., Methods: The Facebook search function was paired with a comprehensive list of vestibular diagnoses to systematically collect publicly available information on vestibular OSCs. Next, a survey was designed to gather clinicodemographic information, OSC characteristics, participation measures, perceived outcomes, and influence on medical decision-making. The anonymous instrument was posted to two OSCs that provide support for patients with general vestibular symptoms., Results: Seventy-three OSCs were identified with >250,000 cumulative members and >10,000 posts per month. The survey was completed by 549 participants, a cohort of primarily educated middle-aged (median = 50, interquartile range 40-60), non-Hispanic white (84%), and female (89%) participants. The participants' most cited initial motivation and achieved goal of participants was to hear from others with the same diagnosis (89% and 88%, respectively). Daily users and those who reported seeing ≥5 providers before receiving a diagnosis indicated that OSC utilization significantly influenced their requested medical treatments (72% daily vs. 61% nondaily, P = .012; 61% <5 providers vs. 71% ≥5 providers P = .019, respectively). Most participants agreed that OSC engagement provides emotional support (74%) and helps to develop coping strategies (68%). Membership of ≥1 year was associated with a higher rate of learned coping skills (61% membership <1-year vs. 71% ≥1-year P = .016)., Conclusions: The use of OSCs is widespread among vestibular diagnoses. A survey of two OSCs suggests these groups provide a significant source of peer support and can influence users' ability to interface with the medical system., Level of Evidence: NA Laryngoscope, 132:1835-1842, 2022., (© 2021 The American Laryngological, Rhinological and Otological Society, Inc.)

Objectives: We implement a novel enhanced recovery after surgery (ERAS) protocol with pre-operative non-opioid loading, total intravenous anesthesia, multimodal peri-operative analgesia, and restricted red blood cell (pRBC) transfusions. 1) Compare differences in mean postoperative peak pain scores, opioid usage, and pRBC transfusions. 2) Examine changes in overall length of stay (LOS), intensive care unit LOS, complications, and 30-day readmissions., Methods: Retrospective cohort study comparing 132 ERAS vs. 66 non-ERAS patients after HNC tissue transfer reconstruction. Data was collected in a double-blind fashion by two teams., Results: Mean postoperative peak pain scores were lower in the ERAS group up to postoperative day (POD) 2. POD0: 4.6 ± 3.6 vs. 6.5 ± 3.5; P = .004) (POD1: 5.2 ± 3.5 vs. 7.3 ± 2.3; P = .002) (POD2: 4.1 ± 3.5 vs. 6.6 ± 2.8; P = .000). Opioid utilization, converted into morphine milligram equivalents, was decreased in the ERAS group (POD0: 6.0 ± 9.8 vs. 10.3 ± 10.8; P = .010) (POD1: 14.1 ± 22.1 vs. 34.2 ± 23.2; P = .000) (POD2: 11.4 ± 19.7 vs. 37.6 ± 31.7; P = .000) (POD3: 13.7 ± 20.5 vs. 37.9 ± 42.3; P = .000) (POD4: 11.7 ± 17.9 vs. 36.2 ± 39.2; P = .000) (POD5: 10.3 ± 17.9 vs. 35.4 ± 45.6; P = .000). Mean pRBC transfusion rate was lower in ERAS patients (2.1 vs. 3.1 units, P = .017). There were no differences between ERAS and non-ERAS patients in hospital LOS, ICU LOS, complication rates, and 30-day readmissions., Conclusion: Our ERAS pathway reduced postoperative pain, opioid usage, and pRBC transfusions after HNC reconstruction. These benefits were obtained without an increase in hospital or ICU LOS, complications, or readmission rates., Level of Evidence: 3 Laryngoscope, 131:E792-E799, 2021., (© 2020 The American Laryngological, Rhinological and Otological Society, Inc.)