Use of natural language processing method to identify regional anesthesia from clinical notes.

INTRODUCTION: Accurate data capture is integral for research and quality improvement efforts. Unfortunately, limited guidance for defining and documenting regional anesthesia has resulted in wide variation in documentation practices, even within individual hospitals, which can lead to missing and inaccurate data. This cross-sectional study sought to evaluate the performance of a natural language processing (NLP)-based algorithm developed to identify regional anesthesia within unstructured clinical notes. METHODS: We obtained postoperative clinical notes for all patients undergoing elective non-cardiac surgery with general anesthesia at one of six Veterans Health Administration hospitals in California between January 1, 2017, and December 31, 2022. After developing and executing our algorithm, we compared our results to a frequently used referent, the Corporate Data Warehouse structured data, to assess the completeness and accuracy of the currently available data. Measures of agreement included sensitivity, positive predictive value, false negative rate, and accuracy. RESULTS: We identified 27,713 procedures, of which 9310 (33.6%) received regional anesthesia. 96.6% of all referent regional anesthesia cases were identified in the clinic notes with a very low false negative rate and good accuracy (false negative rate=0.8%, accuracy=82.5%). Surprisingly, the clinic notes documented more than two times the number of regional anesthesia cases that were documented in the referent (algorithm n=9154 vs referent n=4606). DISCUSSION: While our algorithm identified nearly all regional anesthesia cases from the referent, it also identified more than two times as many regional anesthesia cases as the referent, raising concerns about the accuracy and completeness of regional anesthesia documentation in administrative and clinical databases. We found that NLP was a promising alternative for identifying clinical information when existing databases lack complete documentation.

Opioid use in cancer patients compared with noncancer pain patients in a veteran population.

BACKGROUND: Opioid safety initiatives may secondarily impact opioid prescribing and pain outcomes for cancer care. METHODS: We reviewed electronic health record data at a tertiary Veterans Affairs system (VA Palo Alto) for all patients from 2015 to 2021. We collected outpatient Schedule II opioid prescriptions data and calculated morphine milligram equivalents (MMEs) using Centers for Disease Control and Prevention conversion formulas. To determine the clinical impact of changes in opioid prescription, we used the highest level of pain reported by each patient on the 0-to-10 Numeric Rating Scale in each year, categorized into mild (0-3), moderate (4-6), and severe (7 and above). RESULTS: Among 89 569 patients, 9073 had a cancer diagnosis. Cancer patients were almost twice as likely to have an opioid prescription compared with noncancer patients (69.0% vs 36.7%, respectively). The proportion of patients who received an opioid prescription decreased from 27.1% to 18.1% (trend P < .01) in cancer patients and from 17.0% to 10.2% in noncancer patients (trend P < .01). Cancer and noncancer patients had similar declines of MMEs per year between 2015 and 2019, but the decline was more rapid for cancer patients (1462.5 to 946.4, 35.3%) compared with noncancer patients (1315.6 to 927.7, 29.5%) from 2019 to 2021. During the study period, the proportion of noncancer patients who experienced severe pain was almost unchanged, whereas it increased among cancer patients, reaching a significantly higher rate than among noncancer patients in 2021 (31.9% vs 27.4%, P < .01). CONCLUSIONS: Our findings suggest potential unintended consequences for cancer care because of efforts to manage opioid-related risks.

Public Health Informatics and the Perioperative Physician: Looking to the Future.

The role of informatics in public health has increased over the past few decades, and the coronavirus disease 2019 (COVID-19) pandemic has underscored the critical importance of aggregated, multicenter, high-quality, near-real-time data to inform decision-making by physicians, hospital systems, and governments. Given the impact of the pandemic on perioperative and critical care services (eg, elective procedure delays; information sharing related to interventions in critically ill patients; regional bed-management under crisis conditions), anesthesiologists must recognize and advocate for improved informatic frameworks in their local environments. Most anesthesiologists receive little formal training in public health informatics (PHI) during clinical residency or through continuing medical education. The COVID-19 pandemic demonstrated that this knowledge gap represents a missed opportunity for our specialty to participate in informatics-related, public health-oriented clinical care and policy decision-making. This article briefly outlines the background of PHI, its relevance to perioperative care, and conceives intersections with PHI that could evolve over the next quarter century.

More Than a Perioperative Surgical Home: An Opportunity for Anesthesiologists to Advance Public Health.

Public health and the medical specialty of anesthesiology have been closely intertwined throughout history, dating back to the 1800s when Dr. John Snow used contact tracing methods to identify the Broad Street Pump as the source of a cholera outbreak in London. During the COVID-19 pandemic, leaders in anesthesiology and anesthesia patient safety came forward to develop swift recommendations in the face of rapidly changing evidence to help protect patients and healthcare workers. While these high-profile examples may seem like uncommon events, there are many common modern-day public health issues that regularly intersect with anesthesiology and surgery. These include, but are not limited to, smoking; chronic opioid use and opioid use disorder; and obstructive sleep apnea. As an evolving medical specialty that encompasses pre- and postoperative care and acute and chronic pain management, anesthesiologists are uniquely positioned to improve patient care and outcomes and promote long-lasting behavioral changes to improve overall health. In this article, we make the case for advancing the role of the anesthesiologist beyond the original perioperative surgical home model into promoting public health initiatives within the perioperative period.

Changing a clinical pathway to increase spinal anesthesia use for elective hip arthroplasty: a single-centre historical cohort study.

PURPOSE: International consensus recommendations support neuraxial anesthesia as the preferred anesthetic technique for total hip arthroplasty. We hypothesized that an institutional initiative to promote spinal anesthesia within a clinical pathway would result in increased use of this technique. METHODS: We reviewed primary unilateral total hip arthroplasty data between June 2017 and June 2019-one year before vs one year after implementation. The primary outcome was rate of spinal anesthesia use. Secondary outcomes included postoperative pain scores and opioid use, rates of postoperative complications, and unplanned resource use. We built a run chart-tracking rates of spinal anesthesia; compared postoperative outcomes based on anesthetic technique; and developed a mixed model, multivariable logistic regression with margins analysis evaluating the use of spinal anesthesia. RESULTS: The final sample included 172 patients (87 before and 85 after implementation) with no significant differences in baseline characteristics. For the primary outcome, 42/87 (48%) patients received spinal anesthesia before implementation compared with 48/85 (56%) postimplementation (unadjusted difference, 8%; 95% confidence interval, -7 to 23; P = 0.28). There were no differences in secondary outcomes. Factors associated with receipt of spinal anesthesia included American Society of Anesthesiologists Physical Status II (vs III), lower body mass index, and shorter case duration. Using a reduced mixed model, the average marginal effect was 10.7%, with an upper 95% confidence limit of 25.7%. CONCLUSION: Implementation of a clinical pathway change to promote spinal anesthesia for total hip arthroplasty may not have been associated with increased use of spinal anesthesia, but utilization rates can vary widely. Baseline spinal anesthesia usage at our institution was higher than the USA national average, and many factors may influence choice of anesthesia technique. Patients who receive spinal anesthesia have decreased opioid requirements and pain scores postoperatively.

RéSUMé: OBJECTIF: Les recommandations consensuelles internationales soutiennent l'anesthésie neuraxiale comme technique anesthésique à privilégier lors d'une arthroplastie totale de la hanche. Nous avons émis l'hypothèse qu'une initiative institutionnelle visant à promouvoir la rachianesthésie dans le cadre d'une trajectoire clinique entraînerait une utilisation accrue de cette technique. MéTHODE: Nous avons examiné les données sur l'arthroplastie totale unilatérale primaire de la hanche entre juin 2017 et juin 2019, soit un an avant vs un an après la mise en œuvre. Le critère d'évaluation principal était le taux d'utilisation de la rachianesthésie. Les critères d'évaluation secondaires comprenaient les scores de douleur postopératoire et la consommation d'opioïdes, les taux de complications postopératoires et l'utilisation imprévue des ressources. Nous avons construit un graphique de séquence faisant le suivi des taux de rachianesthésie, comparé les résultats postopératoires basés sur la technique anesthésique, et développé un modèle mixte et une régression logistique multivariée avec analyse des marges évaluant l'utilisation de la rachianesthésie. RéSULTATS: L'échantillon final comprenait 172 patients (87 avant et 85 après la mise en œuvre) sans différences significatives dans leurs caractéristiques initiales. Pour le critère d'évaluation principal, 42/87 (48 %) patients ont reçu une rachianesthésie avant la mise en œuvre, contre 48/85 (56 %) après la mise en œuvre de l'initiative (différence non ajustée, 8 %; intervalle de confiance à 95 %, -7 à 23; P = 0,28). Aucune différence significative n'a été observée dans les critères d'évaluation secondaires. Les facteurs associés à l'administrion d'une rachianesthésie comprenaient : l'état physique selon l'American Society of Anesthesiologists (II vs III), un indice de masse corporelle inférieur et une durée de cas plus courte. En utilisant un modèle mixte réduit, l'effet marginal moyen était de 10,7 %, avec une limite supérieure de confiance de 95 % de 25,7 %. CONCLUSION: La mise en œuvre d'un changement de trajectoire clinique pour promouvoir la rachianesthésie pour l'arthroplastie totale de la hanche n'a peut-être pas été associée à une utilisation accrue de la rachianesthésie, mais les taux d'utilisation peuvent varier considérablement. L'utilisation initiale de la rachianesthésie dans notre établissement était supérieure à la moyenne nationale des États-Unis, et de nombreux facteurs peuvent influencer le choix de la technique d'anesthésie. Les patients qui reçoivent une rachianesthésie ont diminué leurs besoins en opioïdes et leurs scores de douleur postopératoires.

A Comparative Utility Score for Digital Health Tools.

Digital health tools (DHT) are increasingly poised to change healthcare delivery given the Coronavirus Disease 2019 (COVID-19) pandemic and the drive to telehealth. Establishing the potential utility of a given DHT could aid in identifying how it could be best used and further opportunities for healthcare improvement. We propose a metric, a Utility Factor Score, which quantifies the benefits of a DHT by explicitly defining adherence and linking it directly to satisfaction and health goals met. To provide data for how the comparative utility score can or should work, we illustrate in detail the application of our metrics across four DHTs with two simulated users. The Utility Factor Score can potentially facilitate integration of DHTs into various healthcare settings and should be evaluated within a clinical study.

Naloxone Dispensing in Patients at Risk for Opioid Overdose After Total Knee Arthroplasty Within the Veterans Health Administration.

Background: Naloxone prescribing among patients undergoing surgery is not well described. This cohort study was designed to examine patients' risk factors for opioid overdose and their association with naloxone prescribing among veterans undergoing total knee arthroplasty (TKA) after a systemwide Overdose Education and Naloxone Distribution (OEND) initiative. Methods: A retrospective analysis of Veterans Health Administration (VHA) records was performed and consisted of 38,011 veterans undergoing primary TKA from 2013 to 2016. Patient overdose risk was determined using a validated risk index for overdose or serious opioid-induced respiratory depression (RIOSORD) based on patient diagnoses, health care utilization, and prescription drug use. Naloxone dispensing was examined from the year before surgery until 7 days after discharge. These rates were examined the year prior to implementation of a national OEND initiative (2013), the year of implementation (2014), and 2 years following implementation (2015-2016). Results: In 2013, 3.3% of patients presenting for TKA had moderate or high risk for overdose and none were prescribed naloxone. By 2016, after OEND implementation, 2.2% of patients presenting for TKA had moderate or high risk for overdose, but only 10.9% of the moderate-risk and 12.7% of the high-risk patients were prescribed naloxone. Conclusions: Patients presenting for TKA routinely have risk factors for opioid overdose and significant proportions are at moderate or high risk for overdose. Despite this, most patients at moderate and high risk do not receive perioperative naloxone. Risk mitigation strategies using validated tools such as RIOSORD may help identify surgical patients at greatest risk for opioid overdose who could benefit from OEND.

Effects of Buprenorphine Dose and Therapeutic Engagement on Illicit Opiate Use in Opioid Use Disorder Treatment Trials.

The impact of agonist dose and of physician, staff and patient engagement on treatment have not been evaluated together in an analysis of treatment for opioid use disorder. Our hypotheses were that greater agonist dose and therapeutic engagement would be associated with reduced illicit opiate use in a time-dependent manner. Publicly-available treatment data from six buprenorphine efficacy and safety trials from the Federally-supported Clinical Trials Network were used to derive treatment variables. Three novel predictors were constructed to capture the time weighted effects of buprenorphine dosage (mg buprenorphine per day), dosing protocol (whether physician could adjust dose), and clinic visits (whether patient attended clinic). We used time-in-trial as a predictor to account for the therapeutic benefits of treatment persistence. The outcome was illicit opiate use defined by self-report or urinalysis. Trial participants (N = 3022 patients with opioid dependence, mean age 36 years, 33% female, 14% Black, 16% Hispanic) were analyzed using a generalized linear mixed model. Treatment variables dose, Odds Ratio (OR) = 0.63 (95% Confidence Interval (95%CI) 0.59−0.67), dosing protocol, OR = 0.70 (95%CI 0.65−0.76), time-in-trial, OR = 0.75 (95%CI 0.71−0.80) and clinic visits, OR = 0.81 (95%CI 0.76−0.87) were significant (p-values < 0.001) protective factors. Treatment implications support higher doses of buprenorphine and greater engagement of patients with providers and clinic staff.

Variability and Costs of Low-Value Preoperative Testing for Cataract Surgery Within the Veterans Health Administration.

Importance: The Choosing Wisely guidelines indicate that preoperative testing is often unnecessary and wasteful for patients undergoing cataract operations. However, little is known about the impact of these widely disseminated guidelines within the US Veterans Health Administration (VHA) system. Objective: To examine the extent, variability, associated factors, and costs of low-value tests (LVTs) prior to cataract operations in the VHA. Design, Setting, and Participants: This cohort study examined records of all patients receiving cataract operations within the VHA in fiscal year 2017 (October 1, 2016, to September 31, 2017). Records from 135 facilities nationwide supporting both ambulatory and inpatient surgery were included. Exposures: A laboratory test occurring within 30 days prior to cataract surgery and within 30 days after clinic evaluation. Main Outcomes and Measures: Overall national and facility-level rates and associated costs of receiving any of 8 common LVTs in the 30 days prior to cataract surgery. The patient characteristics, procedure type, and facility-level factors associated with receiving at least 1 test, the number of tests received, and receipt of a bundle of 4 tests (complete blood count, basic metabolic profile, chest radiograph, and electrocardiogram). Results: A total of 69 070 cataract procedures were identified among 50 106 patients (66 282 [96.0%] men; mean [SD] age, 71.7 [8.1] years; 53 837 [77.9%] White, 10 292 [14.9%] Black). Most of the patient population had either overweight (23 292 [33.7%] patients) or obesity (27 799 [40.2%] patients). Approximately 49% of surgical procedures (33 424 procedures) were preceded by 1 or more LVT with an overall LVT cost of $2 597 623. Among patients receiving LVTs, electrocardiography (7434 patients [29.9%]) was the most common, with some patients also receiving more costly tests, including chest radiographs (489 patients [8.2%]) and pulmonary function tests (127 patients [3.4%]). For receipt of any LVT, the intraclass correlation coefficient was 0.61 (P < .001) at the facility level and 0.06 (P < .001) at the surgeon level, indicating the substantial contribution of the facility to amount of tests given. Conclusions and Relevance: Despite existing guidelines, use of LVTs prior to cataract surgery is both common and costly within a large, national integrated health care system. Our results suggest that publishing evidence-based guidelines alone-such as the Choosing Wisely campaign-may not sufficiently influence individual physician behavior, and that system-level efforts to directly deimplement LVTs may therefore necessary to effect sustained change.

Implementation of a patient-specific tapering protocol at discharge decreases total opioid dose prescribed for 6 weeks after elective primary spine surgery.

BACKGROUND AND OBJECTIVES: At our institution, we developed an individualized discharge opioid prescribing and tapering protocol for joint replacement patients and implemented the same protocol for neurosurgical spine patients. We then tested the hypothesis that this protocol will decrease the oral morphine milligram equivalent (MME) dose of opioid prescribed postdischarge after elective primary spine surgery. METHODS: In this retrospective cohort study, we identified all consecutive elective primary spine surgery cases 1 year before and after introduction of the protocol. This protocol used the patient's prior 24-hour inpatient opioid consumption to determine discharge opioid pill count and tapering schedule. The primary outcome was total opioid dose prescribed in oral MME from discharge through 6 weeks. Secondary outcomes included in-hospital opioid consumption in MME, hospital length of stay, MME prescribed at discharge, opioid refills, and rates of minor and major adverse events. RESULTS: Eighty-three cases comprised the final sample (45 preintervention and 38 postintervention). There were no differences in baseline characteristics. The total oral MME (median (IQR)) from discharge through 6 weeks postoperatively was 900 (420-1440) preintervention compared with 300 (112-806) postintervention (p<0.01, Mann-Whitney U test), and opioid refill rates were not different between groups. There were no differences in other outcomes. CONCLUSIONS: This patient-specific prescribing and tapering protocol effectively decreases the total opioid dose prescribed for 6 weeks postdischarge after elective primary spine surgery. Our experience also demonstrates the potential generalizability of this protocol, which was originally designed for joint replacement patients, to other surgical populations.

A Multidisciplinary Patient-Specific Opioid Prescribing and Tapering Protocol Is Associated with a Decrease in Total Opioid Dose Prescribed for Six Weeks After Total Hip Arthroplasty.

OBJECTIVE: This retrospective cohort study tested the hypothesis that implementing a multidisciplinary patient-specific discharge protocol for prescribing and tapering opioids after total hip arthroplasty (THA) will decrease the morphine milligram equivalent (MME) dose of opioids prescribed. METHODS: With institutional review board approval, we analyzed a Perioperative Surgical Home database and prescription data for all primary THA patients three months before (PRE) and three months after (POST) implementation of this new discharge opioid protocol based on patients' prior 24-hour inpatient opioid consumption. The primary outcome was total opioid dosage in MME prescribed and opioid refills for six weeks after surgery. Secondary outcomes included the number of tablets and MME prescribed at discharge, in-hospital opioid consumption, length of stay, and postoperative complications. RESULTS: Forty-nine cases (25 PRE and 24 POST) were included. Total median (10th-90th percentiles) MME for six weeks postoperatively was 900 (57-2082) MME PRE vs 295 (69-741) MME POST (mean difference = 721, 95% confidence interval [CI] = 127-1316, P = 0.007, Mann-Whitney U test). Refill rates did not differ. The median (10th-90th percentiles) initial discharge prescription in MME was 675 (57-1035) PRE vs 180 (18-534) POST (mean difference = 387, 95% CI = 156-618, P = 0.003, Mann-Whitney U test) MME. There were no differences in other outcomes. CONCLUSIONS: Implementation of a patient-specific prescribing and tapering protocol decreases the mean six-week dosage of opioid prescribed by 63% after THA without increasing the refill rate.

Perioperative Opioid Prescribing Patterns and Readmissions After Total Knee Arthroplasty in a National Cohort of Veterans Health Administration Patients.

OBJECTIVE: Among Veterans Health Administration (VHA) patients who undergo total knee arthroplasty (TKA) nationally, what are the underlying readmission rates and associations with perioperative opioid use, and are there associations with other factors such as preoperative health care utilization? METHODS: We retrospectively examined the records of 5,514 TKA patients (primary N = 4,955, 89.9%; revision N = 559, 10.1%) over one fiscal year (October 1, 2010-September 30, 2011) across VHA hospitals nationwide. Opioid use was classified into no opioids, tramadol only, short-acting only, or any long-acting. We measured readmission within 30 days and the number of days to readmission within 30 days. Extended Cox regression models were developed. RESULTS: The overall 30-day hospital readmission rate was 9.6% (N = 531; primary 9.5%, revision 11.1%). Both readmitted patients and the overall sample were similar on types of preoperative opioid use. Relative to patients without opioids, patients in the short-acting opioids only tier had the highest risk for 30-day hospital readmission (hazard ratio = 1.38, 95% confidence interval = 1.14-1.67). Preoperative opioid status was not associated with 30-day readmission. Other risk factors for 30-day readmission included older age (≥66 years), higher comorbidity and diagnosis-related group weights, greater preoperative health care utilization, an urban location, and use of preoperative anticonvulsants. CONCLUSIONS: Given the current opioid epidemic, the routine prescribing of short-acting opioids after surgery should be carefully considered to avoid increasing risks of 30-day hospital readmissions and other negative outcomes, particularly in the context of other predisposing factors.

Five-year follow-up to assess long-term sustainability of changing clinical practice regarding anesthesia and regional analgesia for lower extremity arthroplasty.

Background: Long-term and sustainable clinical practice changes in anesthesia procedures have not previously been reported. Therefore, we performed a 5-year audit following implementation of a clinical pathway change favoring spinal anesthesia for total knee arthroplasty (TKA). We similarly evaluated a parallel cohort of patients undergoing total hip arthroplasty (THA), who did not undergo a clinical pathway change, and studied utilization rates of continuous peripheral nerve block (CPNB). METHODS: We identified all primary unilateral TKA and THA cases completed from January 2013 through December 2018, thereby including clinical pathway change data from one-year pre-implementation to 5-years post-implementation. Our primary outcome was the overall application rate of spinal anesthesia. Secondary outcomes included CPNB utilization rate, 30-day postoperative complications, and resource utilization variables such as hospital readmission, emergency department visits, and blood transfusions. RESULTS: The sample included 1,859 cases, consisting of 1,250 TKAs and 609 THAs. During the initial year post-implementation, 174/221 (78.7%) TKAs received spinal anesthesia compared to 23/186 (12.4%) cases the year before implementation (P < 0.001). During the following 4-year period, 647/843 (77.2%) TKAs received spinal anesthesia (P = 0.532 vs. year 1). The number of THA cases receiving spinal anesthesia the year after implementation was 78/124 (62.9%), compared to 48/116 (41.4%) pre-implementation (P = 0.001); however, the rate decreased over the following 4-year period to 193/369 (52.3%) (P = 0.040 vs. year 1). CPNB use was high in both groups, and there were no differences in 30-day postoperative complications, hospital readmission, emergency department visits, or blood transfusions. CONCLUSIONS: A clinical pathway change promoting spinal anesthesia for TKA can be effectively implemented and sustained over a 5-year period.

Variation in Surgeons' Requests for General Anesthesia When Scheduling Carpal Tunnel Release.

Background: Carpal tunnel release (CTR) can be performed with a variety of anesthesia techniques. General anesthesia is associated with higher risk profile and increased resource utilization, suggesting it should not be routinely used for CTR. The purpose of this study was to examine the patient factors associated with surgeons' requests for general anesthesia for CTR and the frequency of routine use of general anesthesia by Veterans Health Administration (VHA) surgeons and facilities. Methods: National VHA data for fiscal years 2015 and 2017 were used to identify patients receiving CTR. Mixed-effects logistic regression was used to evaluate patient, procedure, and surgeon factors associated with requests by the surgeon for general anesthesia versus other anesthesia techniques. Results: In all, 18 145 patients underwent CTR performed by 780 surgeons in 113 VHA facilities. Overall, there were 2218 (12.2%) requests for general anesthesia. Although some patient (eg, older age, obesity), procedure (eg, open vs endoscopic), and surgeon (eg, higher volume) factors were associated with lower odds of requests for general anesthesia, there was substantial facility- and surgeon-level variability. The percentage of patients with general anesthesia requested ranged from 0% to 100% across surgeons. Three facilities and 28 surgeons who performed at least 5 CTRs requested general anesthesia for more than 75% of patients. Conclusions: Where CTR is performed and by whom appear to influence requests for general anesthesia more than patient factors in this study. Avoidance of routine use of general anesthesia for CTR should be considered in future clinical practice guidelines and quality measures.