Opioid Prescribing Rate for Nonoperative Distal Radius Fractures and Clinician Response to a Clinical Decision Support Alert.

BACKGROUND: Opioids are commonly prescribed for the management of acute orthopedic trauma pain, including nonoperative distal radius fractures. OBJECTIVES: This prospective study aimed to determine if a clinical decision support intervention influenced prescribing decisions for patients with known risk factors. We sought to quantify frequency of opioid prescriptions for acute nonoperative distal radius fractures treated. METHODS: We performed a prospective study at one large health care system. Utilizing umbrella code S52.5, we identified all distal radius fractures treated nonoperatively, and the encounters were merged with the Prescription Reporting with Immediate Medication Mapping (PRIMUM) database to identify encounters with opioid prescriptions and patients with risk factors for opioid use disorder. We used multivariable logistic regression to determine patient characteristics associated with the prescription of an opioid. Among encounters that triggered the PRIMUM alert, we calculated the percentage of encounters where the PRIMUM alert influenced the prescribing decision. RESULTS: Of 2984 encounters, 1244 (41.7%) included an opioid prescription. Age increment is a significant factor to more likely receive opioid prescriptions (p < 0.0001) after adjusting for other factors. Among encounters where the physician received an alert, those that triggered the alert for early refill were more likely to influence physicians' opioid prescribing when compared with other risk factors (p = 0.0088). CONCLUSION: Over 90% of patients (106/118) continued to receive an opioid medication despite having a known risk factor for abuse. Additionally, we found older patients were more likely to be prescribed opioids for nonoperatively managed distal radius fractures.

Opioid and benzodiazepine use in gynecologic oncology patients.

OBJECTIVE: The goals of this study were to describe opioid and benzodiazepine prescribing practices in the gynecologic oncology patient population and determine risks for opioid misuse in these patients. METHODS: Retrospective study of opioid and benzodiazepine prescriptions for patients treated for cervical, ovarian (including fallopian tube/primary peritoneal), and uterine cancers within a single healthcare system from January 2016 to August 2018. RESULTS: A total of 7643 prescriptions for opioids and/or benzodiazepines were dispensed to 3252 patients over 5754 prescribing encounters for cervical (n=2602, 34.1%), ovarian (n=2468, 32.3%), and uterine (n=2572, 33.7%) cancer. Prescriptions were most often written in an outpatient setting (51.0%) compared with inpatient discharge (25.8%). Cervical cancer patients were more likely to have received a prescription in an emergency department or from a pain/palliative care specialist (p=0.0001). Cervical cancer patients were least likely to have prescriptions associated with surgery (6.1%) compared with ovarian cancer (15.1%) or uterine cancer (22.9%) patients. The morphine milligram equivalents prescribed were higher for patients with cervical cancer (62.6) compared with patients with ovarian and uterine cancer (46.0 and 45.7, respectively) (p=0.0001). Risk factors for opioid misuse were present in 25% of patients studied; cervical cancer patients were more likely to have at least one risk factor present during a prescribing encounter (p=0.0001). Cervical cancer was associated with a higher number of risk factors (p<0.001). CONCLUSIONS: Opioid and benzodiazepine prescribing patterns differ for cervical, ovarian, and uterine cancer patients. Gynecologic oncology patients are overall at low risk for opioid misuse; however, patients with cervical cancer are more likely to have risk factors present for opioid misuse.

Changes in opioid prescription duration for musculoskeletal injury associated with the North Carolina Strengthen Opioid Misuse Prevention (STOP) Act.

OBJECTIVES: To assess whether implementation of the Strengthen Opioid Misuse Prevention (STOP) Act was associated with an increase in the percentage of opioid prescriptions written for 7 days or fewer among patients with acute or postsurgical musculoskeletal conditions. DESIGN: An interrupted time-series study was conducted to determine the change in duration of opioid prescriptions associated with the STOP Act. SETTING: Data were extracted from the electronic health record of a large health care system in North Carolina. SUBJECTS: Patients presenting from 2016 to 2020 with an acute musculoskeletal injury and the clinicians treating them were included in an interrupted time-series study (n = 12 839). METHODS: Trends were assessed over time, including the change in trend associated with implementation of the STOP Act, for the percentage of prescriptions written for ≤7 days. RESULTS: Among patients with acute musculoskeletal injury, less than 30% of prescriptions were written for ≤7 days in January of 2016; by December of 2020, almost 90% of prescriptions were written for ≤7 days. Prescriptions written for ≤7 days increased 17.7% after the STOP Act was implemented (P < .001), after adjustment for the existing trend. CONCLUSIONS: These results demonstrate significant potential for legislation to influence opioid prescribing behavior.

Impact of clinical decision support on controlled substance prescribing.

BACKGROUND: Prescription drug overdose and misuse has reached alarming numbers. A persistent problem in clinical care is lack of easy, immediate access to all relevant information at the actionable time. Prescribers must digest an overwhelming amount of information from each patient's record as well as remain up-to-date with current evidence to provide optimal care. This study aimed to describe prescriber response to a prospective clinical decision support intervention designed to identify patients at risk of adverse events associated with misuse of prescription opioids/benzodiazepines and promote adherence to clinical practice guidelines. METHODS: This study was conducted at a large multi-center healthcare system, using data from the electronic health record. A prospective observational study was performed as clinical decision support (CDS) interventions were sequentially launched (January 2016-July 2019). All data were captured from the medical record prospectively via the CDS tools implemented. A consecutive series of all patient encounters including an opioid/benzodiazepine prescription were included in this study (n = 61,124,172 encounters; n = 674,785 patients). Physician response to the CDS interventions was the primary outcome, and it was assessed over time using control charts. RESULTS: An alert was triggered in 23.5% of encounters with a prescription (n = 555,626). The prescriber decision was influenced in 18.1% of these encounters (n = 100,301). As the number of risk factors increased, the rate of decision being influenced also increased (p = 0.0001). The effect of the alert differed by drug, risk factor, specialty, and facility. CONCLUSION: The delivery of evidence-based, patient-specific information had an influence on the final prescription in nearly 1 in 5 encounters. Our intervention was sustained with minimal prescriber fatigue over many years in a large and diverse health system.

Orthopaedic Trauma Never Sleeps: Resource Allocation Even During a Non-trauma Crisis.

This study aimed to describe hospital resource utilization of an orthopaedic trauma service and the injury epidemiology during the 2019-2020 coronavirus pandemic to help plan future non-trauma crises. A retrospective chart review was performed on adult patients > 18 years of age who presented to our Level I Trauma Center for musculoskeletal trauma from March 30, 2020 to May 8, 2020 (stay-at-home order) and from March 30, 2019 to May 8, 2019 (comparison group). There were 182 patient encounters and 274 fractures in the 2020 stay-at-home period, and there were 210 patient encounters and 337 fractures in the 2019 control group. There was no statistical difference found comparing the proportion of patient encounters in the stay-at-home period to the control period (p > 0.05). The similar volume of consultations and surgeries justifies maintenance of standard resource allocation. (Journal of Surgical Orthopaedic Advances 32(2):102-106, 2023).

Impact of an opioid prescribing alert system on patients with posttraumatic stress disorder.

BACKGROUND AND OBJECTIVES: Posttraumatic stress disorder (PTSD) is associated with higher rates of chronic pain and increased risk of developing Opioid use disorder. This paper evaluates the impact of PRIMUM, an electronic health record-embedded (EHR) clinical decision support intervention on opioid prescribing patterns for patients with diagnosis of PTSD. METHODS: Inpatient, emergency department (ED), urgent care, and outpatient encounters with ICD-10 codes F43.1 (PTSD), F43.10 (PTSD, unspecified), F43.11 (PTSD, acute), and F43.12 (PTSD, chronic) at Atrium Health between 1/1/2016 and 12/29/2018 were included in the study. RESULTS: A total of 3121 patients with a diagnosis of PTSD were seen in 37,443 encounters during the study period. Ten percent (n = 3761) of the encounters resulted in prescriptions for opioids and PRIMUM alerts were triggered in 1488 of these encounters. These alerts resulted in "decision influenced" for 17% of patients (n = 255) or no prescriptions for opioids or benzodiazepines for 5.8% (n = 86). The majority of the prescriptions were below 50 Morphine milligram equivalents (MME)/day, but there were 570 (15.5%) prescriptions for doses of 50-90 MME and 721 (19.6%) prescriptions for >90 MME/day. DISCUSSION AND CONCLUSION: The PRIMUM alert system helps improve patient safety. PRIMUM affected clinician decisions 17% of the time, and the effect was greater in patients with opioid overdose history and those presenting for early refills. SCIENTIFIC SIGNIFICANCE: The effectiveness of clinical support interventions for opioid prescribing for patients with PTSD has not been documented previously. Our findings provide novel evidence that the EHR can be used to improve patient safety among patients with PTSD in the context of substance use.

Orthopaedic Implant Coatings: Recent Approaches and Clinical Translation.

Despite improved surgical techniques and prophylactic procedures, orthopaedic implant-associated infections remain high with complications that can lead to devastating outcomes for the patient. Implant coatings and associated surface modification techniques represent a promising means to prevent infections. Various approaches have emerged to address the challenges associated with implant infections, such as antibacterial resistance, biofilm prevention, and appropriate efficacy kinetics. Methods including antibiotic and antimicrobial peptide surface tethering, use of osteo-conductive and -inductive materials, and altering hydrophobicity and hydrophilicity of the implant surface, have all demonstrated efficacy toward diminished infection risk. Though many of these techniques have shown great potential in in vitro and in vivo studies, clinical translation remains limited with very few commercially available implant coatings globally. This review summarizes recent advancements in orthopaedic implant coatings, pre-clinical studies, and clinical translation, as well as potential future marketed products. (Journal of Surgical Orthopaedic Advances 31(3):169-176, 2022).

A Randomized Trial of Preoperative Practice in a Simulated Operative Environment: 3D Printed Pilon Model Improves Resident Performance.

This study assessed the effect of preoperative planning using a 3D-printed periarticular fracture model on operative performance. A complex pilon fracture was 3D-printed, and a preoperative plan was developed. Orthopaedic surgery residents (n = 20) were randomized into two groups. Group 1 performed routine preoperative planning, while Group 2 was also practiced using a 3D-printed construct before performing fixation of the 3D-printed model. Resident performance was assessed using a video motion capture system and evaluated by blinded reviewers. Three residents (3D group) completed fixation within the allotted 45 minutes. The 3D group had less hand distance traveled for step 1 (89 m vs. 162 m, p = 0.04). The 3D group had better performance on three of the four components and more acceptable reductions (6 vs. 0, p = 0.009). Average global rating scale was higher in the 3D group (3.0 vs. 1.7, p = 0.0095). Use of 3D-printed models for preoperative planning improved resident performance. (Journal of Surgical Orthopaedic Advances 31(3):187-192, 2022).

Resident Education and Wellness: A Strategy for Future Pandemics.

The Coronavirus Disease 2019 (COVID-19) pandemic presented a novel challenge to modern healthcare systems and medical training. Resource allocation and risk mitigation dramatically affected resident training. The objective of this article is to develop new strategies to maintain a healthy, competent residency program while combating the unique challenges to resident education and wellness. In 2020, our institution implemented a revolving 3-Team system. While the "Inpatient-Team" delivered direct care, the "Back-up Team" and "Quarantine-Team" managed the telemedicine virtual clinic and education-wellness strategy, respectively. Our 3-Team system allowed delivery of safe, high-quality patient care while optimizing resident education, research, and wellness. The efficient use of technology led to both improved virtual education outside of the hospital and intentional wellness opportunities despite social distancing restrictions. Utilization of virtual platforms for patient care, education, research, and wellness grew out of necessity in this pandemic, yet represent an opportunity for lasting improvement. (Journal of Surgical Orthopaedic Advances 31(3):150-154, 2022).

Promoting Bone Formation and Healing in Segmental Defects Through Ectopic Induced Membrane.

We aimed to determine whether addition of an in vivo ectopic induced membrane (EM) to the Masquelet Technique enhanced angiogenesis and bone formation in a segmental defect. After generating and stabilizing a diaphyseal femur defect, 10 rats received a polymethylmethacrylate (PMMA) spacer within the defect (control); 10 received another PMMA spacer implanted subcutaneously (EM). We removed the spacers and added autograft; the excised EM was added to their autograft (EM group). Post-mortem x-rays assessed bone formation and bridging. Osteogenesis in the proximal defect was significantly more uniform (p < 0.01), and there was greater amount of bone remodeling distally in the EM group (p < 0.05). There was no difference in bone formation (p = 0.19) but greater degrees of bridging in the EM group (2.20 vs. 1.20, p = 0.09). The EM resulted in more homogeneous proximal osteogenesis and increased bone remodeling distally. These findings could lead to more consistent and predictable bone healing. (Journal of Surgical Orthopaedic Advances 31(3):161-165, 2022).

Opioid Prescribing Risk Factors in Nonoperative Ankle Fractures: The Impact of a Prospective Clinical Decision Support Intervention.

Opioids are frequently used for acute pain management of musculoskeletal injuries, which can lead to misuse and abuse. This study aimed to identify the opioid prescribing rate for ankle fractures treated nonoperatively in the ambulatory and emergency department setting across a single healthcare system and to identify patients considered at high risk for abuse, misuse, or diversion of prescription opioids that received an opioid. A retrospective cohort study was performed at a large healthcare system. The case list included nonoperatively treated emergency department, urgent care and outpatient clinic visits for ankle fracture and was merged with the Prescription Reporting With Immediate Medication Mapping (PRIMUM) database to identify encounters with prescription for opioids. Descriptive statistics characterize patient demographics, treatment location and prescriber type. Rates of prescribing among subgroups were calculated. There were 1,324 patient encounters identified, of which, 630 (47.6%) received a prescription opioid. The majority of patients were 18-64 years old (60.3%). Patients within this age range were more likely to receive an opioid prescription compared to other age groups (p < .0001). Patients treated in the emergency department were significantly more likely to receive an opioid medication (68.3%) compared to patients treated at urgent care (33.7%) or in the ambulatory setting (16.4%) (p < .0001). Utilizing the PRIMUM tool, 14.2% of prescriptions were provided to patients with at least one risk factor. Despite the recent emphasis on opioid stewardship, 14.2% of patients with risk factors for misuse, abuse, or diversion received opioid analgesics in this study, identifying an area of improvement for prescribers.

Pilot randomized trial of pre-hospital advanced therapies for the control of hemorrhage (PATCH) using pelvic binders.

BACKGROUND: Pelvic fractures represent a small percent of all skeletal injuries but are associated with significant morbidity and mortality secondary to hemodynamic instability from bleeding bone surfaces and disrupted pelvic vasculature. Stabilization of the pelvis prior to arrival at a treatment facility may mitigate the hemodynamic consequences of pelvic ring injuries and improve morbidity and mortality. Whether pelvic compression devices such as pelvic binders or sheets can be safely applied in the prehospital setting has not been well-studied. This study aims to evaluate the safety of applying a pelvic binder to at-risk patients in the field after scalable training and the feasibility of conducting a randomized trial evaluating this practice in the prehospital setting. METHODS: A pilot study (prospective randomized trial design) was conducted in the pre-hospital environment in an urban area surrounding a level-one trauma center. Pre-hospital emergency medical (EMS) personnel were trained to identify patients at high-risk for pelvic fracture and properly apply a commercial pelvic binder. Adult patients with a high-energy mechanism, suspected pelvic fracture, and "Priority 1" criteria were prospectively identified by paramedics and randomized to pelvic binder placement or usual care. Medical records were reviewed for safety outcomes. Secondary outcomes were parameters of efficacy including interventions needed to control hemorrhage (such as angioembolization and surgical control of bleeding) and mortality. RESULTS: Forty-three patients were randomized to treatment (binder: N=20; nonbinder: N=23). No complications of binder placement were identified. Eight patients (40%) had binders placed correctly at the level of the greater trochanter. Two binders (10%) were placed too proximally and 10 (50%) binders were not visualized on x-ray. Two binder group patients and three nonbinder group patients required angioembolization. None required surgical control of pelvic bleeding. Two nonbinder group patients and one binder group patient were readmitted within 30 days and one nonbinder group patient died within 30 days. CONCLUSION: Identification of pelvic fractures in the field remains a challenge. However, a scalable training model for appropriate binder placement was successful without secondary injury to patients. The model for conducting prospective, randomized trials in the prehospital setting was successful.

Redefining Mild Traumatic Brain Injury (mTBI) delineates cost effective triage.

OBJECTIVES: Mild traumatic brain injury (mTBI) is defined as Glasgow Coma Score (GCS) of 14 or 15. Despite good outcomes, patients are commonly transferred to trauma centers for observation and/or neurosurgical consultation. The aim of this study is to assess the value of redefining mTBI with novel radiographic criteria to determine the appropriateness of interhospital transfer for neurosurgical evaluation. METHODS: A retrospective study of patients with blunt head injury with GCS 13-15 and CT head from Jan 2014-Dec 2016 was performed. A novel criteria of head CT findings was created at our institution to classify mTBI. Outcomes included neurosurgical intervention and transfer cost. RESULTS: A total of 2120 patients were identified with 1442 (68.0%) meeting CT criteria for mTBI and 678 (32.0%) classified high risk. Two (0.14%) patients with mTBI required neurosurgical intervention compared with 143 (21.28%) high risk TBI (p < 0.0001). Mean age (55.8 years), and anticoagulation (2.6% vs 2.8%) or antiplatelet use (2.1% vs 3.0%) was similar between groups (p > 0.05). Of patients with mTBI, 689 were transferred without receiving neurosurgical intervention. Given an average EMS transfer cost of $700 for ground and $5800 for air, we estimate an unnecessary transfer cost of $733,600. CONCLUSION: Defining mTBI with the described novel criteria clearly identifies patients who can be safely managed without transfer for neurosurgical consultation. These unnecessary transfers represent a substantial financial and resource burden to the trauma system and inconvenience to patients.

Reduction in lead exposures with lead-free ammunition in an advanced urban assault course.

The training of soldiers for urban conflict involves marksmanship instruction on outdoor flat ranges and the teaching of close-quarter battle techniques in indoor facilities, referred to as shoot houses, where intense firing exercises can generate high air lead levels from small arms ammunition, flash bang grenades, and explosive devices. Levels of lead and copper in air were evaluated during five training activities of a 45-day training course using both stationary general area and breathing zone sampling over a 2-year period. Individual blood lead values were determined prior to and at course completion. Mean breathing zone lead concentrations for the five training activities ranged from 0.014 on the outdoor flat range to 0.064 mg/m3 inside shoot houses; with a change to lead-free ammunition the values were reduced to a range of 0.006-0.022 mg/m3. Isolated flash bang grenades generated very high general area lead concentrations (2.0 mg/m3), which in training were associated with the highest measured breathing zone concentration (0.16 mg/m3). For copper, mean breathing zone concentrations increased from 0.010 to 0.037 mg/m3 with the change to lead-free frangible ammunition on the outdoor range, but remained below the permissible exposure limit for copper fume. Inside shoot houses, mean breathing zone copper concentrations exceeded the permissible exposure limit with ball and lead-free frangible ammunition, ranging from 0.077-0.13 mg/m3. With the introduction of lead-free ammunition, when comparing the blood lead differences between start and finish of the course, there was a significant reduction in the mean blood lead difference from 13.3 µg/dL to 5.4 µg/dL. Options for mitigation of potentially high exposure areas using improved ventilation designs are discussed. These results advocate for improved designs for shoot house training facilities, stress the importance of removing lead from ammunition and explosive devices for training, and promote the continued need for implementation of controls to mitigate and manage metal exposures during training.

Opioid Prescribing and Patient Satisfaction Scores Across Practice Types.

US physicians prescribe opioids at a high rate relative to other countries. Of the US physicians surveyed, almost half report having prescribed an inappropriate opioid due to concerns about patient satisfaction scores. We investigated patterns in controlled substance prescribing practices, patient risk factors, and associated Press Ganey patient satisfaction scores at a sample of orthopaedic surgery and primary care clinics over a 6month time period. Primary care practices had higher proportions of prescriptions, and patient risk profiles varied across sites. However, overall satisfaction was high, with little variation between sites (78.3 81.3%). Satisfaction with pain control was lower and more varied (67.1 78.0%). A total of 4,229 Press Ganey survey responses were received, including 7,232 comments, of which only 10 (0.1%) expressed frustration for not receiving opioids. Opioid prescriptions had minimal association with Press Ganey scores among varied practices and patient populations. Prescribers should prescribe opioids appropriately without fear that this will negatively impact their satisfaction scores. (Journal of Surgical Orthopaedic Advances 29(1):59, 2020).

Tiered team research: A novel concept for increasing research productivity in the academic setting.

Background: Research has become a key pillar of academic medicine and a cornerstone of residency training; however, there continues to be significant barriers to ensuring research productivity for residents. We implemented a novel tiered team approach which aimed to increase research productivity and promote collaboration during residency training. Methods: This was a retrospective study that evaluated the implementation of a novel tiered team research approach at a single institution between 2009 and 2013. Analytical software was used to visualize and display the research interconnections among the authors of the captured publications. In addition to using Gephi to determine the research interconnections, the growth in research capability of the tiered team and its individual members were also graphically depicted. Results: The research team produced a total of 77 publications during the study period (2009-2013). Significant and frequent collaboration and coauthorship was noted as the years progressed following implementation of tiered team research. Discussion: Tiered team research can be readily implemented at most institutions and can lead to increases in productivity of published research. It can also promote collaboration and peer mentorship among those involved.

Development and Implementation of a Pediatric Trauma Survivors Network Program.

: The Trauma Survivors Network (TSN) was developed as a program of the American Trauma Society (ATS) to support recovery for adult trauma patients. However, the children of adult trauma patients, families of pediatric trauma patients, and pediatric trauma patients previously had scarce resources. Our institution, in collaboration with the ATS, sought to expand the TSN to support pediatric trauma patients, caregivers, and family members. We aimed to determine whether the TSN could be transferred to the pediatric population. Focus groups identified psychosocial needs of younger survivors, children of adult survivors, and caregivers. A Pediatric TSN Coordinator was hired, Pediatric TSN Peer Visitors were recruited and trained, and Pediatric TSN Activity Hour was implemented for pediatric patients and families. Since implementation 1 year ago, 26 peer visitors have been trained and have conducted approximately 200 visitations. In total, 93 patients and family members have attended Pediatric TSN Activity Hour. TSN services can be adapted to address psychosocial needs of pediatric trauma survivors, families, and children of trauma survivors. When possible, sharing resources between an existing adult TSN program and a pediatric program is valuable to facilitate expansion. TSN complements and strengthens the care offered at our institution by providing patient-centered and family-centered care services for the entire family at various stages of development. The implementation of this program might be different at sites without an existing adult TSN with established resources and support. This article describes the development and implementation of the program; we did not assess outcomes.

Opioid Prescribing in the Pediatric Orthopaedic Trauma Population.

The purpose of this article is to describe opioid prescribing patterns for children with orthopaedic injuries. A retrospective chart review was conducted on pediatric orthopaedic trauma patients (n = 124) who were discharged from the hospital or emergency department or had a clinic visit during a 1-month period. Patient demographics, prescription specifics, injury details, and fracture fixation information were collected. Results show that most children received opioids after injury (82.3%). While children undergoing operative fixation typically received opioids, only 39.5% with closed reduction did. Hydrocodone- acetaminophen accounted for 93% of prescriptions, but adolescents were more likely to receive other drugs. There was a significant trend of increasing daily dosage with increased age; 36.73% of adolescents received > 50 morphine milligram equivalents per day. Children with orthopaedic injuries are typically prescribed opioids; older children more commonly receive higher dosages. Further study is needed to define prescribing trends across facility and specialty types to aid in development of standardized prescribing guidelines. (Journal of Surgical Orthopaedic Advances 27(4):269-273, 2018).

Prescription reporting with immediate medication utilization mapping (PRIMUM): development of an alert to improve narcotic prescribing.

BACKGROUND: Prescription narcotic overdoses and abuse have reached alarming numbers. To address this epidemic, integrated clinical decision support within the electronic medical record (EMR) to impact prescribing behavior was developed and tested. METHODS: A multidisciplinary Expert Panel identified risk factors for misuse, abuse, or diversion of opioids or benzodiazepines through literature reviews and consensus building for inclusion in a rule within the EMR. We ran the rule "silently" to test the rule and collect baseline data. RESULTS: Five criteria were programmed to trigger the alert; based on data collected during a "silent" phase, thresholds for triggers were modified. The alert would have fired in 21.75 % of prescribing encounters (1.30 % of all encounters; n = 9998), suggesting the alert will have a low prescriber burden yet capture a significant number of at-risk patients. CONCLUSIONS: While the use of the EMR to provide clinical decision support is not new, utilizing it to develop and test an intervention is novel. We successfully built an alert system to address narcotic prescribing by providing critical, objective information at the point of care. The silent phase data were useful to appropriately tune the alert and obtain support for widespread implementation. Future healthcare initiatives can utilize similar methodology to collect data prospectively via the electronic medical record to inform the development, delivery, and evaluation of interventions.

Employing machine learning to enhance fracture recovery insights through gait analysis.

This study aimed to explore the potential of gait analysis coupled with supervised machine learning models as a predictive tool for assessing post-injury complications such as infection, malunion, or hardware irritation among individuals with lower extremity fractures. We prospectively identified participants with lower extremity fractures at a tertiary academic center. These participants underwent gait analysis with a chest-mounted inertial measurement unit device. Using customized software, the raw gait data were preprocessed, emphasizing 12 essential gait variables. The data were standardized, and several machine learning models, including XGBoost, logistic regression, support vector machine, LightGBM, and Random Forest, were trained, tested, and evaluated. Special attention was given to class imbalance, addressed using the synthetic minority oversampling technique (SMOTE). Additionally, we introduced a novel methodology to compute the post-injury recovery rate for gait variables, which operates independently of the time difference between the gait analyses of different participants. XGBoost was identified as the optimal model both before and after the application of SMOTE. Before using SMOTE, the model achieved an average test area under the ROC curve (AUC) of 0.90, with a 95% confidence interval (CI) of [0.79, 1.00], and an average test accuracy of 86%, with a 95% CI of [75%, 97%]. Through feature importance analysis, a pivotal role was attributed to the duration between the occurrence of the injury and the initial gait analysis. Data patterns over time revealed early aggressive physiological compensations, followed by stabilization phases, underscoring the importance of prompt gait analysis. χ2 analysis indicated a statistically significant higher readmission rate among participants with underlying medical conditions (p = 0.04). Although the complication rate was also higher in this group, the association did not reach statistical significance (p = 0.06), suggesting a more pronounced impact of medical conditions on readmission rates rather than on complications. This study highlights the transformative potential of integrating advanced machine learning techniques like XGBoost with gait analysis for orthopedic care. The findings underscore a shift toward a data-informed, proactive approach in orthopedics, enhancing patient outcomes through early detection and intervention. The χ2 analysis added crucial insights into the broader clinical implications, advocating for a comprehensive treatment strategy that accounts for the patient's overall health profile. The research paves the way for personalized, predictive medical care in orthopedics, emphasizing the importance of timely and tailored patient assessments.