Automated Electronic Health Record Score to Predict Mortality Risk at the US Department of Veterans Affairs.

BACKGROUND: Terminally ill patients benefit from earlier engagement in palliative care. However, this does not always occur. This project assessed if an already available risk score, the Care Assessment Needs (CAN) score, would be able to identify patients at greatest risk for mortality within 30 days of hospital admission within the Veterans Health Administration (VHA). METHODS: The cohort of this retrospective analysis included all VA acute are patients over 18 years of age with a recent CAN score. The CAN score is an automatically calculated VA risk score that was repurposed to determine if it could predict risk of mortality after acute care admission. Univariate logistic regression was performed to obtain the probability of mortality within 30 days of admission, based on their CAN score. RESULTS: 298,467 patient records were assessed from January 1, 2019, to December 31, 2019. There was 6% mortality after 30 days of admissions, and 17% mortality within 1-year post-admission. Mean CAN score was 65 (SD: 29). On average, each incremental increase in the CAN score increased the probability of mortality by 7%. Patients with a CAN score of 90 had a 10% probability of 30-day post-admission mortality. CONCLUSION: A readily available risk score, automatically calculated from EHR data, was able to identify patients at high risk for 30-day mortality in the acute care setting. Next steps will be to assess how the CAN score can be utilized to in improve end of life care for high-risk hospitalized Veterans.

Towards global model generalizability: independent cross-site feature evaluation for patient-level risk prediction models using the OHDSI network.

BACKGROUND: Predictive models show promise in healthcare, but their successful deployment is challenging due to limited generalizability. Current external validation often focuses on model performance with restricted feature use from the original training data, lacking insights into their suitability at external sites. Our study introduces an innovative methodology for evaluating features during both the development phase and the validation, focusing on creating and validating predictive models for post-surgery patient outcomes with improved generalizability. METHODS: Electronic health records (EHRs) from 4 countries (United States, United Kingdom, Finland, and Korea) were mapped to the OMOP Common Data Model (CDM), 2008-2019. Machine learning (ML) models were developed to predict post-surgery prolonged opioid use (POU) risks using data collected 6 months before surgery. Both local and cross-site feature selection methods were applied in the development and external validation datasets. Models were developed using Observational Health Data Sciences and Informatics (OHDSI) tools and validated on separate patient cohorts. RESULTS: Model development included 41 929 patients, 14.6% with POU. The external validation included 31 932 (UK), 23 100 (US), 7295 (Korea), and 3934 (Finland) patients with POU of 44.2%, 22.0%, 15.8%, and 21.8%, respectively. The top-performing model, Lasso logistic regression, achieved an area under the receiver operating characteristic curve (AUROC) of 0.75 during local validation and 0.69 (SD = 0.02) (averaged) in external validation. Models trained with cross-site feature selection significantly outperformed those using only features from the development site through external validation (P < .05). CONCLUSIONS: Using EHRs across four countries mapped to the OMOP CDM, we developed generalizable predictive models for POU. Our approach demonstrates the significant impact of cross-site feature selection in improving model performance, underscoring the importance of incorporating diverse feature sets from various clinical settings to enhance the generalizability and utility of predictive healthcare models.

Upper Extremity Surgery in Cervical Spinal Cord Injury: A Prospective Comparative Mixed Methods Study.

BACKGROUND: Cervical spinal cord injury (SCI) is a devastating injury. Restoring upper extremity function is a top priority, which can be accomplished by tendon transfer (TT) and nerve transfer (NT) surgeries. The purpose of this prospective comparative study was to assess long-term changes in UE function between surgical (TT or NT) and non-surgical groups through a comprehensive mixed methods approach. METHODS: This multicenter, cohort study compared data among three groups: those undergoing 1) no surgery 2) TT surgery, or 3) NT surgery. Quantitative data, the Spinal Cord Independence Measure (SCIM) and Short Form Health Survey (SF-36), was collected at baseline and long-term follow-up (6-24 months). Qualitative semi-structured interview data was also obtained from these participants and their identified caregivers at baseline, early follow-up (1 month), and long-term follow-up (6-24 months). RESULTS: Thirty-one participants had quantitative data across all timepoints: no surgery (n=14), TT (n=7), and NT (n=10). SCIM scores improved in TT and NT groups compared to the no surgery group (p<0.05). SF-36 scores did not differ among groups. Qualitative data analysis (n=168 interviews) corroborated SCIM findings: surgical participants and their caregivers reported improvement in transfers and ability to perform activities of daily living, including grooming and self-catheterization. Improved use of electronics and ability to operate a motor vehicle were also reported. Post-operative therapy was identified as a critical component of achieving gains. CONCLUSION: Both TT and NT surgery leads to quantitative and qualitative functional gains as compared to the no surgery group. This comparative information should be used to help surgeons discuss treatment options.

Comparing surgeries to restore upper extremity function in tetraplegia: Impact on function during the perioperative period.

CONTEXT/OBJECTIVE: To assess short-term changes in health outcomes in people with cervical-level spinal cord injury (SCI) who underwent upper extremity (UE) reconstruction via either novel nerve transfer (NT) or traditional tendon transfer (TT) surgery with individuals who did not undergo UE surgical reconstruction. DESIGN: Prospective, comparative cohort pilot study. PARTICIPANTS: 34 participants with cervical SCI met the following inclusion criteria: age 18 or older, greater than 6 months post-injury, and mid-cervical level SCI American Spinal Injury Association Impairment Scale (AIS) A, B or C. SETTING: Two tertiary academic hospitals and their affiliated veterans' hospitals. METHODS: Health outcomes were assessed using two previously validated measures, the Spinal Cord Independence Measure (SCIM) and Short-Form Health Survey (SF-36). Demographic, surgical, and survey data were collected at the initial evaluation and one month postoperatively/post-baseline. RESULTS: 34 participants with cervical SCI were recruited across three cohorts: no surgery (n = 16), NT (n = 10), and TT (n = 8). The TT group had a decline in SCIM and SF-36 scores whereas the NT and no surgery groups experienced little change in independence or health status in the immediate perioperative period. CONCLUSIONS: Surgeons and rehabilitation providers must recognize differences in the perioperative needs of people with cervical SCI who chose to have restorative UE surgery. Future work should focus on further investigation of health outcomes, change in function, and improving preoperative counseling and cross-disciplinary management.

Caregiving for People With Spinal Cord Injury Undergoing Upper Extremity Reconstructive Surgery: A Prospective Exploration of Lived Experiences, Perioperative Care, and Change Across Time.

Background: Nerve transfer (NT) and tendon transfer (TT) surgeries can enhance upper extremity (UE) function and independence in individuals with cervical spinal cord injury (SCI). Caregivers are needed to make this surgery possible, yet caregivers experience their own set of challenges. Objectives: This comparative study explored the perioperative and nonoperative experiences of caregivers of individuals with cervical SCI, focusing on daily life activities, burden, and mental health. Methods: Caregivers of individuals with cervical SCI were recruited and grouped by treatment plan for the person with SCI: (1) no surgery (NS), (2) TT surgery, and (3) NT surgery. Semistructured interviews were conducted at baseline/preoperative, early follow-up/postoperative, and late follow-up/postoperative. Caregivers were asked about their daily life, mental health, and challenges related to caregiving. Interviews were audio recorded, transcribed verbatim, and analyzed using thematic analysis. Quantitative, single-item standardized burden score (0-100) data were collected at each timepoint. Results: Participants included 23 caregivers (18 family members, 4 friends, 1 hired professional). The surgeries often brought hope and motivation for caregivers. Caregivers reported increased burden immediately following surgery (less for the NT compared to TT subgroup) yet no long-term changes in the amount and type of care they provided. NS caregivers discussed social isolation, relationship dysfunction, and everyday challenges. Conclusion: Health care providers should consider the changing needs of SCI caregivers during perioperative rehabilitation. As part of the shared surgical decision-making approach, providers should educate caregivers about the postoperative process and the extent and potential variability of short- and long-term care needs.

Peripheral nerve stimulation for saphenous neuralgia.

BACKGROUND: Injury to saphenous nerve branches is frequent during knee surgery and can result in chronic pain. This saphenous neuralgia remains challenging to treat. Peripheral nerve stimulation (PNS) is a new potential non-pharmacologic treatment option. We present our outcomes experience using this technology in 12 patients. METHODS: We retrospectively reviewed PNS placement for saphenous neuralgia between 2000 and 2022 at a single institution. Demographic information was collected as well as response to the device. Four-question short-form Patient-Reported Outcome Measurement Information System (PROMIS) Scores were collected before and 2 weeks, 6 weeks, and 6 months postprocedure. Specific scores included pain interference and behavior, functional mobility, depression, anxiety, and sleep impairment. Change in pain interference measured by the short-form PROMIS tool at 6 months was chosen as the primary outcome. RESULTS: Twelve patients met inclusion criteria, with 10 patients having the full 6-month follow-up. In these 10 patients, the mean change from baseline in the short-form adjusted pain interference score (greater difference means improved pain) at 6 months was 5.8 (SD 6.5). Among all patients, average follow-up was 11.5 months (range 3-35 months). Most patients' symptoms developed after knee surgery (84%). Prior to PNS, patients underwent other treatments including cryoablation (8%), radiofrequency ablation (16%), saphenous neurectomy (16%), or surgical release of adjacent nerves (25%). Ten patients (83%) reported any improvement in symptoms while two reported no benefit. Complications occurred in four patients (33%). Two patients had the device removed and a third discontinued use. PROMIS Scores for pain, functional mobility, mood, and sleep impairment all improved. DISCUSSION: Limited effective treatments exist for saphenous neuralgia. Our case series demonstrates the potential of PNS as a treatment for saphenous neuralgia. Comparative effectiveness studies are warranted to assess whether our effect size is clinically relevant.

Prediction of opioid-related outcomes in a medicaid surgical population: Evidence to guide postoperative opiate therapy and monitoring.

BACKGROUND: Treatment of surgical pain is a common reason for opioid prescriptions. Being able to predict which patients are at risk for opioid abuse, dependence, and overdose (opioid-related adverse outcomes [OR-AE]) could help physicians make safer prescription decisions. We aimed to develop a machine-learning algorithm to predict the risk of OR-AE following surgery using Medicaid data with external validation across states. METHODS: Five machine learning models were developed and validated across seven US states (90-10 data split). The model output was the risk of OR-AE 6-months following surgery. The models were evaluated using standard metrics and area under the receiver operating characteristic curve (AUC) was used for model comparison. We assessed calibration for the top performing model and generated bootstrap estimations for standard deviations. Decision curves were generated for the top-performing model and logistic regression. RESULTS: We evaluated 96,974 surgical patients aged 15 and 64. During the 6-month period following surgery, 10,464 (10.8%) patients had an OR-AE. Outcome rates were significantly higher for patients with depression (17.5%), diabetes (13.1%) or obesity (11.1%). The random forest model achieved the best predictive performance (AUC: 0.877; F1-score: 0.57; recall: 0.69; precision:0.48). An opioid disorder diagnosis prior to surgery was the most important feature for the model, which was well calibrated and had good discrimination. CONCLUSIONS: A machine learning models to predict risk of OR-AE following surgery performed well in external validation. This work could be used to assist pain management following surgery for Medicaid beneficiaries and supports a precision medicine approach to opioid prescribing.

Postoperative opioid prescribing patients with diabetes: Opportunities for personalized pain management.

BACKGROUND: Opioids are commonly prescribed for postoperative pain, but may lead to prolonged use and addiction. Diabetes impairs nerve function, complicates pain management, and makes opioid prescribing particularly challenging. METHODS: This retrospective observational study included a cohort of postoperative patients from a multisite academic health system to assess the relationship between diabetes, pain, and prolonged opioid use (POU), 2008-2019. POU was defined as a new opioid prescription 3-6 months after discharge. The odds that a patient had POU was assessed using multivariate logistic regression controlling for patient factors (e.g., demographic and clinical factors, as well as prior pain and opiate use). FINDINGS: A total of 43,654 patients were included, 12.4% with diabetes. Patients with diabetes had higher preoperative pain scores (2.1 vs 1.9, p<0.001) and lower opioid naïve rates (58.7% vs 68.6%, p<0.001). Following surgery, patients with diabetes had higher rates of POU (17.7% vs 12.7%, p<0.001) despite receiving similar opioid prescriptions at discharge. Patients with Type I diabetes were more likely to have POU compared to other patients (Odds Ratio [OR]: 2.22; 95% Confidence Interval [CI]:1.69-2.90 and OR:1.44, CI: 1.33-1.56, respectively). INTERPRETATION: In conclusion, surgical patients with diabetes are at increased risk for POU even after controlling for likely covariates, yet they receive similar postoperative opiate therapy. The results suggest a more tailored approach to diabetic postoperative pain management is warranted.

Assessment of a wearable fall prevention system at a veterans health administration hospital.

Objective: In-hospital falls are a significant cause of morbidity and mortality. The Veterans Health Administration (VHA) has designated fall prevention as a major focus area. The objective of this report is to assess the performance of a new sensor-enabled wearable system to prevent patient falls. Methods: An integrated sensor-enabled wearable SmartSock system was utilized to prevent falls at the acute care wards of a large VA hospital. Individual patients were only provided the SmartSocks when they were determined to be at high risk of falling. All fall count rates, with and without using the SmartSock, were evaluated and compared for individual patients. SmartSock sensor and electronic health record data were combined to assess the system's performance from February 10, 2021, through October 31, 2021. Results: There were 20.7 falls per 1000 ward days of care (WDOC) for those not using the SmartSocks compared to 9.2 falls per 1000 WDOC for patients using the SmartSocks. This represents a reduction of falls by more than half. These findings are further confirmed with a negative binomial regression model, which showed the use of the SmartSock had a statistically significant effect on the rate of falls (p = 0.03) when length of stay was held constant and demonstrated the odds of fall incident rate of 0.48 (95% CI, 0.24-0.92), that is less than half compared to when patients were not wearing the SmartSock. Conclusion: The use of a sensor-enabled wearable SmartSock fall prevention system resulted in a clinically meaningful and statistically significant decrease in falls in the acute care setting.

Management of Brachial Plexus Birth Palsies: The Stanford Experience.

The start of Stanford's brachial plexus birth palsy (BPBP) experience dates back to 1983, when Dr. Vincent Rod Hentz visited Dr. Alain Gilbert on sabbatical. Since then, our principles of care for patients with BPBP have evolved based on our group's longitudinal experience caring for children with the entire spectrum of sequelae that arise in children with BPBP. We base our clinical decision making on frequent serial examinations and use intraoperative evoked potentials to guide surgical decisions. Here, we discuss our current principles on surgical indications, timing of surgery, and preferred techniques for secondary surgery in patients with BPBP.

Surgical Treatment of Symptomatic End-Neuroma With a New Bioresorbable Copolyester Nerve Capping Device: A Multicenter Prospective Cohort Study.

BACKGROUND: Neuroma-induced neuropathic pain is associated with loss of function and reduced quality of life. No consistently effective standard-of-care treatment has been defined. Neurocap, a bioresorbable nerve capping device, has been designed to isolate the nerve stump from surrounding tissues to reduce development of symptomatic end-neuromas. METHODS: Patients with peripheral symptomatic end-neuromas were included in a prospective, multicenter, single-arm design. Data were collected presurgery up till 24 months postsurgery. Eligible patients with neuromas were identified based on blocks using anesthetic. Intervention included surgical excision and capping of the transected proximal nerve end with the Neurocap. Main outcome measures were pain, function, recurrence of symptomatic neuroma, use of analgesics, and adverse events. RESULTS: In total, 73 patients with 50 upper-extremity and 23 lower-extremity end-neuromas were enrolled. End-neuromas were predominately located in the digits and lower leg. Statistical power of the study outcomes was preserved by 46 of 73 patients completing 24-month follow-up. The mean VAS-Pain score at baseline was 70.2 ± 17.8 (scale 0-100) and decreased significantly to 31 ± 32.5 (P < 0.001). Function significantly improved over time. The recurrence rate of confirmed symptomatic neuroma was low (2 of 98 capped nerves). Adverse event rate was low and included pain and infection; there were no unexpected device-related adverse events. Most patients reported lower use of nonsteroidal anti-inflammatory drugs, opioids, and antineuropathic medications at last follow-up compared with baseline. CONCLUSIONS: End-neuroma treatment with excision and capping resulted in long-term significant reduction in reported pain, disability, and analgesic medication use. Adverse event rate was low.

Median and Ulnar Nerve Compressions: Simplifying Diagnostics and Surgery at the Elbow and Hand.

LEARNING OBJECTIVES: After studying this article, the participant should be able to: 1. Understand the anatomy of the median and ulnar nerves. 2. Perform clinical examination of the upper limb. 3. Analyze examination results to diagnose level of nerve compression. SUMMARY: Numbness and loss of strength are common complaints in the hand surgery clinic. Two nerves that are commonly entrapped (median and ulnar nerves) have several potential sites of entrapment, and in busy clinical practice, the less common sites may be overlooked, leading to wrong or missed diagnoses. This article reviews the anatomy of the median and ulnar nerves, provides tips to assist the busy clinician in diagnosis of site of entrapment(s), and discusses how to simplify surgery. The goal is to help the clinician be as efficient and accurate as possible when evaluating the patient with numbness or loss of strength in their hand.

The Impact of Virtual Reality on the Patients and Providers Experience in Wide-Awake, Local-Only Hand Surgery.

Purpose: Virtual reality (VR) is an emerging technology with the potential to enhance patient care by reducing pain and anxiety for a variety of medical procedures. The aim of this study was to evaluate an immersive VR program as a nonpharmacologic intervention to reduce anxiety and increase satisfaction in patients undergoing wide-awake, local-only hand surgery. The secondary aim was to assess providers' experience with the program. Methods: An implementation evaluation was employed to assess the experience of 22 patients who used VR during outpatient, wide-awake hand surgery at a veterans affairs hospital. We assessed the patients' anxiety scores and vital signs before and after the procedure as well as postprocedural satisfaction measures. The providers' experience was also assessed. Results: Patients who used VR exhibited lower anxiety scores after the procedure compared with what they exhibited before the procedure and had high satisfaction levels with their VR experience. Surgeons who used the system reported that VR improved their ability to teach learners and better focus on the procedure. Conclusions: Virtual reality, as a nonpharmacologic intervention, reduced anxiety and contributed to the patients' perioperative satisfaction with wide-awake, local-only hand surgery. As a secondary finding, VR positively impacted the providers' experience by increasing their ability to concentrate on tasks during the surgery. Clinical relevance: Virtual reality represents a novel technology that can reduce anxiety and contribute to a positive experience for both patients and providers during wide-awake, local-only hand procedures.

Is Hand Surgery in the Procedure Room Setting Associated With Increased Surgical Site Infection? A Cohort Study of 2,717 Patients in the Veterans Affairs Population.

PURPOSE: Procedure rooms (PRs) are increasingly used for hand surgeries, but few studies have directly compared surgical site infection (SSI) rates between the PR and operating room. We tested the hypothesis that procedure setting is not associated with an increased SSI incidence in the VA population. METHODS: We identified carpal tunnel, trigger finger, and first dorsal compartment releases performed at our VA institution from 1999 to 2021 of which 717 were performed in the main operating room and 2,000 were performed in the PR. The incidence of SSI, defined as signs of wound infection within 60 days of the index procedure, which was treated with oral antibiotics, intravenous antibiotics, and/or operating room irrigation and debridement, was compared. We constructed a multivariable logistic regression analysis to assess the association between procedure setting and SSI incidence, adjusting for age, sex, procedure type, and comorbidities. RESULTS: Surgical site infection incidence was 55/2,000 (2.8%) in the PR cohort and 20/717 (2.8%) in the operating room cohort. In the PR cohort, five (0.3%) cases required hospitalization for intravenous antibiotics of which two (0.1%) cases required operating room irrigation and debridement. In the operating room cohort, two (0.3%) cases required hospitalization for intravenous antibiotics of which one (0.1%) case required operating room irrigation and debridement. All other SSIs were treated with oral antibiotics alone. The procedure setting was not independently associated with SSI (adjusted odds ratio, 0.84 [95% confidence interval, 0.49, 1.48]). The only risk factor for SSI was trigger finger release (odds ratio, 2.13 [95% confidence interval, 1.32, 3.48] compared with carpal tunnel release), which was independent of setting. CONCLUSIONS: Minor hand surgeries can be performed safely in the PR without an increased rate of SSI. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic II.

Impact of Adding Carpal Tunnel Release or Trigger Finger Release to Carpometacarpal Arthroplasty on Postoperative Complications.

BACKGROUND: This study assessed whether adding trigger finger or carpal tunnel release at the time of thumb carpometacarpal (CMC) arthroplasty would increase postoperative opioid use, readmissions, complications, or development of complex regional pain syndrome (CRPS). METHODS: Using the IBM MarketScan Research Databases from 2012 through 2016, the authors identified two CMC arthroplasty groups. The CMC-only group only had a CMC arthroplasty on the day of operation; the multiple-procedures group had a CMC arthroplasty and concurrent carpal tunnel or trigger finger release. Between the two groups, the authors compared persistent opioid use, 30-day readmissions, 30-day complications, and diagnosis of CRPS. RESULTS: The CMC-only group consisted of 18,010 patients; the multiple-procedures group consisted of 4064 patients. The patients in the multiple-procedures group received a CMC arthroplasty and a carpal tunnel release (74%), a trigger finger release (20%), or both (6%). CMC-only patients had lower rates of persistent opioid use compared with patients who underwent multiple procedures (16% versus 18%). Readmission rates were also lower for CMC-only patients (3% versus 4%). CMC-only patients had decreased odds of persistent opioid use (OR, 0.85; 95% CI, 0.75 to 0.97; P = 0.013) and readmissions (OR, 0.80; 95% CI, 0.67 to 0.96; P = 0.016). The most common reason for readmission was pain (16%). CONCLUSIONS: Adding another procedure to a CMC arthroplasty slightly increases the odds of adverse outcomes such as persistent opioid use and readmission. Patients and providers should weigh the efficiency of performing these procedures concurrently against the risk of performing multiple procedures at once. CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.

Thinking Outside the Operating Room: Guidance on Designing a Safe and Effective Minor Procedure Room.

The minor procedure room (MPR) offers numerous advantages over the traditional operating room for performing many common hand surgeries. MPRs require less space, are subject to more practical architectural design standards, and facilitate more judicious use of disposable materials and unnecessary instruments than common hand surgeries. MPRs reduce costs to the system and patient at every step of the surgical workflow and improve efficiency by removing preoperative and postoperative monitoring requirements. Hand surgeons sometimes face resistance when attempting surgery in MPRs, often because of confusion about their design characteristics and capabilities. This article aims to clarify many of the major requirements for establishing an MPR and provide a guide to hand surgeons for performing safe, efficient surgery outside the operating room.

Surgery to restore upper extremity function in tetraplegia-Preferences for early and frequent access to information.

INTRODUCTION: People with cervical spinal cord injury (SCI) identify improving upper extremity (UE) function as a top priority. In addition to comprehensive rehabilitation, UE surgeries, including nerve and tendon transfers, enhance function. However, barriers exist to disseminating information about surgical options to enhance UE function. OBJECTIVE: To assess the experiences and preferences of people with cervical SCI and their caregivers in accessing information about surgery to enhance UE function. DESIGN: Prospective cohort study. Participants were followed up for 24 months and completed up to three interviews. SETTING: Tertiary care at academic and affiliated Veterans Administration Health Care Centers. PARTICIPANTS: Adults with cervical SCI (n = 35) ages 18 to 80 years with mid-cervical SCI American Spinal Injury Association Impairment Scale A, B, or C (at least 6 months post-injury) and their caregivers (n = 23) were eligible to participate. Participants were enrolled in three groups: nerve transfer, tendon transfer, or no UE reconstructive surgery. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURE: Semi-structured interviews about surgical knowledge and experiences. RESULTS: Data were analyzed and three themes were identified. First, providing information about UE surgical options early post-injury was recommended. The acute or inpatient rehabilitation phases of recovery were the preferred times to receive surgical information. Second, challenges with information dissemination were identified. Participants learned about UE surgery through independent research, medical provider interactions, or peers. Third, peers were identified as valuable resources for SCI needs and surgical information. CONCLUSIONS: Following cervical SCI, information about UE reconstructive surgeries should be a standard component of education during rehabilitation. An increased understanding of the reconstructive options available to improve UE function is necessary to educate stakeholders. Future research is needed to support the development of strategies to effectively present surgical information to individuals with SCI and health care providers.

Changes in postoperative opioid prescribing across three diverse healthcare systems, 2010-2020.

Objective: The opioid crisis brought scrutiny to opioid prescribing. Understanding how opioid prescribing patterns and corresponding patient outcomes changed during the epidemic is essential for future targeted policies. Many studies attempt to model trends in opioid prescriptions therefore understanding the temporal shift in opioid prescribing patterns across populations is necessary. This study characterized postoperative opioid prescribing patterns across different populations, 2010-2020. Data Source: Administrative data from Veteran Health Administration (VHA), six Medicaid state programs and an Academic Medical Center (AMC). Data extraction: Surgeries were identified using the Clinical Classifications Software. Study Design: Trends in average daily discharge Morphine Milligram Equivalent (MME), postoperative pain and subsequent opioid prescription were compared using regression and likelihood ratio test statistics. Principal Findings: The cohorts included 595,106 patients, with populations that varied considerably in demographics. Over the study period, MME decreased significantly at VHA (37.5-30.1; p = 0.002) and Medicaid (41.6-31.3; p = 0.019), and increased at AMC (36.9-41.7; p < 0.001). Persistent opioid users decreased after 2015 in VHA (p < 0.001) and Medicaid (p = 0.002) and increase at the AMC (p = 0.003), although a low rate was maintained. Average postoperative pain scores remained constant over the study period. Conclusions: VHA and Medicaid programs decreased opioid prescribing over the past decade, with differing response times and rates. In 2020, these systems achieved comparable opioid prescribing patterns and outcomes despite having very different populations. Acknowledging and incorporating these temporal distribution shifts into data learning models is essential for robust and generalizable models.

Prescription quantity and duration predict progression from acute to chronic opioid use in opioid-naïve Medicaid patients.

Opiates used for acute pain are an established risk factor for chronic opioid use (COU). Patient characteristics contribute to progression from acute opioid use to COU, but most are not clinically modifiable. To develop and validate machine-learning algorithms that use claims data to predict progression from acute to COU in the Medicaid population, Adult opioid naïve Medicaid patients from 6 anonymized states who received an opioid prescription between 2015 and 2019 were included. Five machine learning (ML) Models were developed, and model performance assessed by area under the receiver operating characteristic curve (auROC), precision and recall. In the study, 29.9% (53820/180000) of patients transitioned from acute opioid use to COU. Initial opioid prescriptions in COU patients had increased morphine milligram equivalents (MME) (33.2 vs. 23.2), tablets per prescription (45.6 vs. 36.54), longer prescriptions (26.63 vs 24.69 days), and higher proportions of tramadol (16.06% vs. 13.44%) and long acting oxycodone (0.24% vs 0.04%) compared to non- COU patients. The top performing model was XGBoost that achieved average precision of 0.87 and auROC of 0.63 in testing and 0.55 and 0.69 in validation, respectively. Top-ranking prescription-related features in the model included quantity of tablets per prescription, prescription length, and emergency department claims. In this study, the Medicaid population, opioid prescriptions with increased tablet quantity and days supply predict increased risk of progression from acute to COU in opioid-naïve patients. Future research should evaluate the effects of modifying these risk factors on COU incidence.