Opioid Prescribing Variation After Laparoscopic Cholecystectomy in the US Military Health System.

INTRODUCTION: After laparoscopic cholecystectomy (LC), there is a wide variation in opioid prescription miligram morphine equivalent dose (MED) and refills across US medical institutions. Given wide variation and opioid prescription guidelines, it is essential to conduct thorough health services research across medical, surgical, and patient-level factors that can be implemented to improve system-wide prescribing practices. Therefore, this study describes discharge MED variation and opioid refill probability after emergent and nonemergent LC. MATERIALS AND METHODS: This retrospective cohort study included medical record data of adult patients (N = 20,025) undergoing LC from January 2016 to June 2021 in the US Military Health System. Data visualizations and bivariate analyses examined prescription patterns across hospitals and evaluated the relationship between patient-level, care-level, and system-level factors and each outcome: discharge MED and opioid refill probability. Two generalized additive mixed models evaluated the relationship between predictors and each outcome. RESULTS: There was a significant variation in opioid and nonopioid pain medication prescribing practices across hospitals. While several factors were associated with discharge MED and opioid refill probability, the strongest effects were related to time period (before versus after a June 2018 Defense Health Agency policy release) and receipt of an opioid/nonopioid combination medication. Despite decreases in MED, the MED remained almost twice the recommended dose per prior research. CONCLUSIONS: Variation by hospital suggests the need for system-level changes that target genuine practice change and opioid stewardship. Inclusion of patient-reported outcomes, electronic health record decision support tools, and academic detailing programs may support system-level improvements.

Appendectomy Pain Medication Prescribing Variation in the U.S. Military Health System.

INTRODUCTION: Post-appendectomy opioid prescription practices may vary widely across and within health care systems. Although guidelines encourage conservative opioid prescribing and prescribing of non-opioid pain medications, the variation of prescribing practices and the probability of opioid refill remain unknown in the U.S. Military Health System. MATERIALS AND METHODS: This retrospective observational cohort study evaluated medical data of 11,713 patients who received an appendectomy in the Military Health System between January 2016 and June 2021. Linear-mixed and generalized linear-mixed models evaluated the relationships between patient-, care-, and system-level factors and the two primary outcomes; the morphine equivalent dose (MED) at hospital discharge; and the probability of 30-day opioid prescription refill. Sensitivity analyses repeated the generalized linear-mixed model predicting the probability of opioid (re)fill after an appendectomy, but with inclusion of the full sample, including patients who had not received a discharge opioid prescription (e.g., 0 mg MED). RESULTS: Discharge MED was twice the recommended guidance and was not associated with opioid refill. Higher discharge MED was associated with opioid/non-opioid combination prescription (+38 mg) relative to opioid-only, lack of non-opioid prescribing at discharge (+6 mg), care received before a Defense Health Agency opioid safety policy was released (+61 mg), documented nicotine dependence (+8 mg), and pre-appendectomy opioid prescription (+5 mg) (all P < .01). Opioid refill was more likely for patients with complicated appendicitis (OR = 1.34; P < .01); patients assigned female (OR = 1.25, P < .01); those with a documented mental health diagnosis (OR = 1.32, P = .03), an antidepressant prescription (OR = 1.84, P < .001), or both (OR = 1.54, P < .001); and patients with documented nicotine dependence (OR = 1.53, P < .001). Opioid refill was less likely for patients who received care after the Defense Health Agency policy was released (OR = 0.71, P < .001), were opioid naive (OR = 0.54, P < .001), or were Asian or Pacific Islander (relative to white patients, OR = 0.68, P = .04). Results from the sensitivity analyses were similar to the main analysis, aside from two exceptions. The probability of refill no longer differed by race and ethnicity or mental health condition only. CONCLUSIONS: Individual prescriber practices shifted with new guidelines, but potentially unwarranted variation in opioid prescribing dose remained. Future studies may benefit from evaluating patients' experiences with pain management, satisfaction, and patient-centered education after appendectomy within the context of opioid prescribing practices, amount of medications used, and refill probability. Such could pave a way for standardized patient-centered procedures that both decrease unwarranted prescribing pattern variability and optimize pain management regimens.

Competency-Based Assessments: Leveraging Artificial Intelligence to Predict Subcompetency Content.

PURPOSE: Faculty feedback on trainees is critical to guiding trainee progress in a competency-based medical education framework. The authors aimed to develop and evaluate a Natural Language Processing (NLP) algorithm that automatically categorizes narrative feedback into corresponding Accreditation Council for Graduate Medical Education Milestone 2.0 subcompetencies. METHOD: Ten academic anesthesiologists analyzed 5,935 narrative evaluations on anesthesiology trainees at 4 graduate medical education (GME) programs between July 1, 2019, and June 30, 2021. Each sentence (n = 25,714) was labeled with the Milestone 2.0 subcompetency that best captured its content or was labeled as demographic or not useful. Inter-rater agreement was assessed by Fleiss' Kappa. The authors trained an NLP model to predict feedback subcompetencies using data from 3 sites and evaluated its performance at a fourth site. Performance metrics included area under the receiver operating characteristic curve (AUC), positive predictive value, sensitivity, F1, and calibration curves. The model was implemented at 1 site in a self-assessment exercise. RESULTS: Fleiss' Kappa for subcompetency agreement was moderate (0.44). Model performance was good for professionalism, interpersonal and communication skills, and practice-based learning and improvement (AUC 0.79, 0.79, and 0.75, respectively). Subcompetencies within medical knowledge and patient care ranged from fair to excellent (AUC 0.66-0.84 and 0.63-0.88, respectively). Performance for systems-based practice was poor (AUC 0.59). Performances for demographic and not useful categories were excellent (AUC 0.87 for both). In approximately 1 minute, the model interpreted several hundred evaluations and produced individual trainee reports with organized feedback to guide a self-assessment exercise. The model was built into a web-based application. CONCLUSIONS: The authors developed an NLP model that recognized the feedback language of anesthesiologists across multiple GME programs. The model was operationalized in a self-assessment exercise. It is a powerful tool which rapidly organizes large amounts of narrative feedback.

The effect of the perfluorocarbon emulsion Oxycyte™ in an ovine model of severe decompression illness.

Background: The treatment of decompression sickness (DCS) with hyperbaric oxygen (HBO2) serves to decrease intravascular bubble size, increase oxygen (O2) delivery to tissue and enhance the elimination of inert gas. Emulsified perfluorocarbons (PFC) combined with breathing O2 have been shown to have similar effects animal models. We studied an ovine model of severe DCS treated with the intravenous PFC Oxycyte™ while breathing O2 compared to saline control also breathing O2. Methods: Juvenile male sheep (N=67; weight 24.4±2.10kg) were compressed to 257 feet of sea water (fsw) in our multiple large-animal chamber where they remained under pressure for 31 minutes. Animals then were decompressed to surface pressure and randomized to receive either Oxycyte at 5mL/kg intravenously (IV) or 5mL/kg saline IV (both receiving 100% O2) 10 minutes after reaching surface pressure. Mortality was recorded at two hours, four hours, and 24 hours after receiving the study drug. Surviving animals underwent perfusion fixation and harvesting of the spinal cord at 24 hours. Spinal cord sections were assessed for volume of lesion area and compared. Results: There was no significant difference in survival at two hours (p=0.2737), four hours (p=0.2101), or 24 hours (p=0.3171). Paralysis at 24 hours was not significantly different. However, spinal cord lesion area was significantly smaller in the Oxycyte group as compared to the saline group, with median spinal cord lesion areas 0.65% vs. 0.94% (p=0.0107). Conclusion: In this ovine model of severe DCS the intravenous PFC Oxycyte did not reduce mortality but did ameliorate spinal cord injury when used after the onset of DCS.

Leveraging the Existing Anesthesia Information Management System to Improve Anesthesia Quality Assurance Outcome Reporting.

INTRODUCTION: Tracking measures of quality over time has been shown to improve care within institutions and across health systems. Perioperative quality assurance (QA) tracking by anesthesia departments in the Military Health System (MHS) has not used a uniform system integrated into the workflow of anesthesia providers. The purpose of this study was to demonstrate that the use of the embedded QA outcome reporting feature in the anesthesia information management system (AIMS) increased the rate of reporting compared to the current paper reporting system in a military anesthesia department. MATERIALS AND METHODS: An electronic outcome reporting mechanism embedded in the AIMS was activated as an alternative to paper QA outcome reporting. The proportion of anesthesia cases per month in a 12-month period with a reported QA outcome was compared to the previous year in which only the paper reporting system was used. The total number of cases in each time period with an outcome reported was compared using chi square for proportions, and systems were evaluated using the Statistical Process Control methodology. This project was evaluated and determined to be exempt from review by our institutional review board. RESULTS: There was a 389.8% increase in the number of cases with a QA outcome reported after the implementation of the outcome reporting function integrated into the AIMS (χ2 = 207.72; P <.001, Table I). Systems before and after the intervention were stable, and special cause variation was noted only at the point of implementation of the electronic reporting system. Anesthesia providers were surveyed and felt that the addition of QA reporting to the AIMS made QA reporting more likely. CONCLUSIONS: The use of an electronic QA outcome reporting method integrated into the AIMS dramatically increased the likelihood that a QA outcome would be reported. The decreased administrative burden of the integrated outcome reporting system was likely the primary reason for this increase. This study was limited by the fact that it was done in a single institution; however, the size and timing of the increase clearly indicate that the intervention was the reason for improved reporting. Electronic health record upgrades should consider incorporating QA reporting into the AIMS across the MHS. These measures could allow for system-wide improvement, evaluation, and evidence-based education on their own, but also by facilitating participation in the American Society of Anesthesiologists' Anesthesia Quality Institute's National Anesthesia Clinical Outcomes Registry. This report serves as a valuable example to institutions and perioperative leaders in the MHS of how to improve the robustness of perioperative QA reporting such that it could be used to validate and improve the value of care.

Exhaled Volatile Organic Compounds Precedes Pulmonary Injury in a Swine Pulmonary Oxygen Toxicity Model.

PURPOSE: Inspiring high partial pressure of oxygen (FiO2 > 0.6) for a prolonged duration can lead to lung damage termed pulmonary oxygen toxicity (PO2T). While current practice is to limit oxygen exposure, there are clinical and military scenarios where higher FiO2 levels and partial pressures of oxygen are required. The purpose of this study is to develop a non-invasive breath-based biomarker to detect PO2T prior to the onset of clinical symptoms. METHODS: Male Yorkshire swine (20-30 kg) were placed into custom airtight runs and randomized to air (0.209 FiO2, n = 12) or oxygen (>0.95 FiO2, n = 10) for 72 h. Breath samples, arterial blood gases, and vital signs were assessed every 12 h. After 72 h of exposure, animals were euthanized and the lungs processed for histology and wet-dry ratios. RESULTS: Swine exposed to hyperoxia developed pulmonary injury consistent with PO2T. Histology of oxygen-exposed swine showed pulmonary lymphatic congestion, epithelial sloughing, and neutrophil transmigration. Pulmonary injury was also evidenced by increased interstitial edema and a decreased PaO2/FiO2 ratio in the oxygen group when compared to the air control group. Breath volatile organic compound (VOC) sample analysis identified six VOCs that were combined into an algorithm which generated a breath score predicting PO2T with a ROC/AUC curve of 0.72 defined as a of PaO2/FiO2 ratio less than 350 mmHg. CONCLUSION: Exposing swine to 72 h of hyperoxia induced a pulmonary injury consistent with human clinical endpoints of PO2T. VOC analysis identified six VOCs in exhaled breath that preceded PO2T. Results show promise that a simple, non-invasive breath test could potentially predict the risk of pulmonary injury in humans exposed to high partial pressures of oxygen.

Perfluorocarbon in Delayed Recompression with a Mixed Gender Swine Model of Decompression Sickness.

INTRODUCTION: Perfluorocarbons (PFC) are fluorinated hydrocarbons that dissolve gases to a much greater degree than plasma and hold promise in treating decompression sickness (DCS). The efficacy of PFC in a mixed gender model of DCS and safety in recompression therapy has not been previously explored. METHODS: Swine (25 kg; N = 104; 51 male and 53 female) were randomized into normal saline solution (NSS) or PFC emulsion treatment groups and subjected to compression on air in a hyperbaric chamber at 200 fsw for 31 min. Then the animals were decompressed and observed for signs of DCS. Afterwards, they were treated with oxygen and either PFC (4 cc · kg-1) or NSS (4 cc · kg-1). Surviving animals were observed for 4 h, at which time they underwent recompression therapy using a standard Navy Treatment Table 6. After 24 h the animals were assessed and then euthanized. RESULTS: Survival rates were not significantly different between NSS (74.04%) and PFC (66.67%) treatment groups. All swine that received recompression treatment survived to the end of the study and no seizures were observed in either PFC or NSS animals. Within the saline treated swine group there were no significant differences in DCS survival between male (75.00%, N = 24) and female (73.08%, N = 26) swine. Within the PFC treated swine, survival of females (51.85%, N = 27) was significantly lower than males (81.48%, N = 27). DISCUSSION: In this large animal mixed gender efficacy study in DCS, PFC did not improve mortality or spinal cord injury, but appears safe during recompressive therapy. Gender differences in DCS treatment with PFC will need further study.Cronin WA, Hall AA, Auker CR, Mahon RT. Perfluorocarbon in delayed recompression with a mixed gender swine model of decompression sickness. Aerosp Med Hum Perform. 2018; 89(1):14-18.

Propranolol Effects on Decompression Sickness in a Simulated DISSUB Rescue in Swine.

INTRODUCTION: Disabled submarine (DISSUB) survivors may face elevated CO2 levels and inert gas saturation, putting them at risk for CO2 toxicity and decompression sickness (DCS). Propranolol was shown to reduce CO2 production in an experimental DISSUB model in humans but its effects on DCS in a DISSUB rescue scenario are unknown. A 100% oxygen prebreathe (OPB) reduces DCS incidence and severity and is incorporated into some DISSUB rescue protocols. We used a swine model of DISSUB rescue to study the effect of propranolol on DCS incidence and mortality with and without an OPB. METHODS: In Experiment 1, male Yorkshire Swine (70 kg) were pressurized to 2.8 ATA for 22 h. Propranolol 1.0 mg · kg-1 (IV) was administered at 21.25 h. At 22 h, the animal was rapidly decompressed and observed for DCS type, onset time, and mortality. Experimental animals (N = 21; 69 ± 4.1 kg), PROP1.0, were compared to PROP1.0-OPB45 (N = 8; 69 ± 2.8 kg) with the same dive profile, except for a 45 min OPB prior to decompression. In Experiment 2, the same methodology was used with the following changes: swine pressurized to 2.8 ATA for 28 h; experimental group (N = 25; 67 ± 3.3 kg), PROP0.5 bis, propranolol 0.5 mg · kg-1 bis (twice) (IV) was administered at 22 h and 26 h. Control animals (N = 25; 67 ± 3.9 kg) received normal saline. RESULTS: OPB reduced mortality in PROP1.0-OBP45 compared to PROP1.0 (0% vs. 71%). PROP0.5 bis had increased mortality compared to CONTROL (60-% vs. 4%). DISCUSSION: Administration of beta blockers prior to saturation decompression appears to increase DCS and worsen mortality in a swine model; however, their effects in bounce diving remain unknown.Forbes AS, Regis DP, HallAA, Mahon RT, Cronin WA. Propranolol effects on decompression sickness in a simulated DISSUB rescue in swine. Aerosp Med Hum Perform. 2017; 88(4):385-391.

The effect of the perfluorocarbon emulsion Oxycyte on platelet count and function in the treatment of decompression sickness in a swine model.

Decompression from elevated ambient pressure is associated with platelet activation and decreased platelet counts. Standard treatment for decompression sickness (DCS) is hyperbaric oxygen therapy. Intravenous perfluorocarbon (PFC) emulsion is a nonrecompressive therapy being examined that improves mortality in animal models of DCS. However, PFC emulsions are associated with a decreased platelet count. We used a swine model of DCS to study the effect of PFC therapy on platelet count, function, and hemostasis. Castrated male swine (n = 50) were fitted with a vascular port, recovered, randomized, and compressed to 180 feet of sea water (fsw) for 31 min followed by decompression at 30 fsw/min. Animals were observed for DCS, administered 100% oxygen, and treated with either emulsified PFC Oxycyte (DCS-PFC) or isotonic saline (DCS-NS). Controls underwent the same procedures, but were not compressed (Sham-PFC and Sham-NS). Measurements of platelet count, thromboelastometry, and coagulation were obtained 1 h before compression and 1, 24, 48, 96, 168 and 192 h after treatment. No significant changes in normalized platelet counts were observed. Prothrombin time was elevated in DCS-PFC from 48 to 192 h compared with DCS-NS, and from 96 to 192 h compared with Sham-PFC. Normalized activated partial thromboplastin time was also elevated in DCS-PFC from 168 to 192 h compared with Sham-PFC. No bleeding events were noted. DCS treated with PFC (Oxycyte) does not impact platelet numbers, whole blood clotting by thromboelastometry, or clinical bleeding. Late changes in prothrombin time and activated partial thromboplastin time associated with PFC use in both DCS therapy and controls warrant further investigation.

The autonomic effects of cardiopulmonary decompression sickness in swine using principal dynamic mode analysis.

Methods to predict onset of cardiopulmonary (CP) decompression sickness (DCS) would be of great benefit to clinicians caring for stricken divers. Principal dynamic mode (PDM) analysis of the electrocardiogram has been shown to provide accurate separation of the sympathetic and parasympathetic tone dynamics. Nine swine (Sus scrofa) underwent a 15-h saturation dive at 184 kPa (60 ft. of saltwater) in a hyperbaric chamber followed by dropout decompression, whereas six swine, used as a control, underwent a 15-h saturation dive at 15 kPa (5 ft. of saltwater). Noninvasive electrocardiograms were recorded throughout the experiment and autonomic nervous system dynamics were evaluated by heart rate series analysis using power spectral density (PSD) and PDM methods. We observed a significant increase in the sympathetic and parasympathetic tones using the PDM method on average 20 min before DCS onset following a sudden induction of decompression. Parasympathetic activities remained elevated, but the sympathetic modulation was significantly reduced at onset of cutis and CP DCS signs, as reported by a trained observer. Similar nonsignificant observations occurred during PSD analysis. PDM observations contrast with previous work showing that neurological DCS resulted in a >50% reduction in both sympathetic and parasympathetic tone. Therefore, tracking dynamics of the parasympathetic tones via the PDM method may allow discrimination between CP DCS and neurological DCS, and this significant increase in parasympathetic tone has potential use as a marker for early diagnosis of CP DCS.

The cost of pursuing a medical career in the military: a tale of five specialties.

PURPOSE: The physician payment system is a focus of potential reform in the United States. The authors explored the effects of the military's method of physician payment on physicians' returns on educational investment for several specialties. METHOD: This retrospective, observational study used national data from 2003 and standard financial techniques to calculate the net present value-the current value of an expected stream of cash flows at a particular rate of interest-of the educational investments of medical students in ten 30-year career paths: either military or civilian careers in internal medicine, psychiatry, gastroenterology, general surgery, or orthopedics. RESULTS: At a 5% discount rate, in the civilian world, the lowest return on an educational investment accrued to psychiatrists ($1.136 million) and the highest to orthopedists ($2.489 million), a range of $1.354 million. In the military, the lowest returns accrued to internists ($1.377 million) and the highest to orthopedists ($1.604 million); however, the range was only $0.227 million, one-sixth that found in the civilian sector. The authors also found that most military physicians do not remain in the military for their full careers. CONCLUSIONS: Choosing a military career substantially decreases the net present value of an educational investment for interventionalists, but it does so only modestly for primary care physicians. Further, a military career path markedly diminishes specialty-specific variation in the net present values of educational investment. Adopting a military structure for engaging medical students might help reverse the current trend of declining interest in primary care.