Inter-provider Variability Research Articles

Inter-rater agreement of CAUTI (catheter-associated urinary tract infections) diagnosis among Infectious disease physicians

Background: CAUTIs constitute forty percent of nosocomial infections, yet their direct link with mortality remains debated. In 2009, NHSN estimated the economic burden of CAUTIs in the U.S. to be over $340 million. Limited data exist on inter-physician concordance in diagnosing CAUTIs, especially in patients with abnormal genitourinary (GU) anatomy. Our study assessed inter-provider variability in diagnosing CAUTI in 50 such patients, including those meeting NHSN(National healthcare safety network) criteria. Methods: We included a random set of 50 adults (18+) with abnormal GU anatomy admitted to the University of Miami hospitals from January 2018 to November 2021 who had a urinary foley catheter and at least one positive urine culture during their hospitalization. Three Infectious disease fellows and three board-certified Infectious disease physicians independently reviewed each patient’s chart, classifying them as having or not having a CAUTI. Inter-physician reliability was assessed using kappa statistics. Results: Our findings highlight substantial variation in clinician-determined CAUTI incidence among the 50 patients with abnormal GU anatomy, ranging from 8% to 32% (Figures 2,3). Inter-rater agreement on CAUTI diagnosis was generally poor (Kappa Hollenbeak CS, et al. The attributable cost of catheter-associated urinary tract infections in the United States: A systematic review. Am J Infect Control. 2018 Jul;46(7):751-757. Trautner BW, et al. Development and validation of an algorithm to recalibrate mental models and reduce diagnostic errors associated with catheter-associated bacteriuria. BMC Med Inform Decis Mak. 2013 Apr 15;13:48. Gafary M, et al. Catheter Associated Urinary Tract Infections (CAUTI) in Bladder Cancer Patients Post Cystectomy With a Neobladder, Open Forum Infectious Diseases, Volume 2, Issue suppl_1, December 2015, 293.

Demonstration of Convolutional Neural Networks to Determine Patch Test Reactivity.

Background: Convolutional neural networks (CNNs) have the potential to assist allergists and dermatologists in the analysis of patch tests. Such models can help reduce interprovider variability and improve consistency of patch test interpretations. Objective: Our aim is to evaluate the performance of a CNN model as a proof of concept in discriminating between patch tests with reactions and patch tests without reactions. Methods: We performed a retrospective analysis of patch test images from March 2020 to March 2021. The CNN model was trained as a binary classifier to discriminate between reaction and nonreaction patches. Performance of the model was determined using summary statistics and receiver operator characteristics (ROC) curves. Results: In total, 13,622 images from 125 patients were recorded for analysis. The majority of patients in the cohort were female (81.6%) with Fitzpatrick skin types I-II (88.0%). The area under curve was 0.940, indicating a high discriminative performance of the model for this data set. This resulted in a total accuracy of 90.1%, sensitivity of 86.0%, and specificity of 90.2%. Conclusions: CNNs have the capacity to determine the presence of delayed-type reactions in patch tests. Future prospective studies are required to assess the generalizability of such models.

3D-printed phantoms to quantify accuracy and variability of goniometric and volumetric assessment of Peyronie’s disease deformities

Characterization of Peyronie’s disease (PD) involves manual goniometry and penile length measurement. These techniques neglect volume loss or hourglass deformities. Inter-provider variability complicates accuracy. Using 3D-printed models, we aimed to evaluate measurement accuracy and variability and establish computational assessment workflows. Five digital phantoms were created: 13.0 cm cylinder, 13.0 cm hourglass cylinder, 15.0 cm cylinder with 40° angulation, 12.0 cm straight penis, and 12.9 cm PD penis with 68° angulation and hourglass. Lengths, volumes, and angles were determined computationally. Each phantom was 3D-printed. Ten urology providers determined lengths, angles, and volumes with measuring tape, goniometer, and volume calculator. Provider versus computational measurements were compared to determine accuracy using t-tests or Wilcoxon rank-sum tests. No significant differences were observed between manual assessment of length of penile models and designed length in penile models. Average curvature angles from providers for bent cylinder and PD phantoms were 38.3° ± 3.9° (p = 0.25) and 57.5° ± 7.2° (p = 0.006), respectively. When assessing for volume, hourglass cylinder and bent cylinder showed significant differences between designed volume and provider averages. All assessments of length, angle, and volume showed significant provider variability. Our results suggest manual measurements suffer from inaccuracy and variability. Computational workflows are useful for improved accuracy and volume assessment.

Variability in adherence to goal-directed transfusion protocols and inter-physician variability: an odd phenomenon

Introduction On May 15, 2019, we implemented an institutional goal-directed transfusion protocol based on thromboelastography (TEG) for cardiac surgeries in an effort to eliminate unjustifiably high non-RBC blood product usage [1]. Standardization of care delivery attenuates inter-provider variability and improves clinical, operational and financial outcomes for patients and the healthcare system [2,3]. The objective of this prospective single-center study is to describe overall guideline adherence and inter-physician variability to protocol adherence. Methods We prospectively collected transfusion data on all consecutive cardiac surgeries performed between May 2019 and January 2020. Based on our transfusion guideline recommendations, we calculated the observed to expected (O:E) transfusion of fresh frozen plasma (FFP), platelet, and cryoprecipitate for all 6 cardiac surgeons and 8 anesthesiologists on staff; O:E 1 refer to under-transfusion, adequate-transfusion, and over-transfusion per guideline recommendations, respectively. Descriptive data is described as percentages, means and standard deviations. Results Between May 2019 and January 2020, we collected 128 cases. Protocol adherence (O:E = 1) was 57% after adjusting to surgeons, and 65% after adjusting for anesthesiologists. Over-transfusion (O:E >1) is most common for FFP followed by platelets, and the variability is pronounced especially amongst surgeons, when stratified by specialty. Under-transfusion (O:E Discussion Herein, we observe two important phenomena related to implementing an institutional goal-directed transfusion protocol for cardiac surgery; 1. Despite an agreed-upon protocol, close to 30% of surgeons and anesthesiologists deviate, and 2. Despite efforts at standardization, inter-physician variability in care delivery remains. Future work will be directed at identifying barriers to physician adherence to recommended guidelines as they relate our institutional transfusion practices [3].

MP33-03 3D-PRINTED PHANTOMS TO QUANTIFY ACCURACY AND VARIABILITY OF GONIOMETRIC AND VOLUMETRIC ASSESSMENT OF PEYRONIE'S DISEASE DEFORMITIES

You have accessJournal of UrologySexual Function/Dysfunction: Peyronie's Disease (MP33)1 Apr 2020MP33-03 3D-PRINTED PHANTOMS TO QUANTIFY ACCURACY AND VARIABILITY OF GONIOMETRIC AND VOLUMETRIC ASSESSMENT OF PEYRONIE'S DISEASE DEFORMITIES Dyvon Walker*, Renea Sturm, Doug Daniels, Jesse Mills, and Sriram Eleswarapu Dyvon Walker*Dyvon Walker* More articles by this author , Renea SturmRenea Sturm More articles by this author , Doug DanielsDoug Daniels More articles by this author , Jesse MillsJesse Mills More articles by this author , and Sriram EleswarapuSriram Eleswarapu More articles by this author View All Author Informationhttps://doi.org/10.1097/JU.0000000000000877.03AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract INTRODUCTION AND OBJECTIVE: Characterization of Peyronie’s disease (PD) deformities involves manual goniometry and measurement of penile length. These techniques neglect volume loss or hourglass, a source of bother. Inter-provider variability complicates accuracy. Using standardized 3D-printed models, we aimed to evaluate accuracy and variability in measurement and establish a workflow for computational assessment including volumetrics. METHODS: Five 3D phantoms were created digitally (Fig. 1A): 13.0 cm cylinder, 13.0 cm hourglass cylinder, 15.0 cm cylinder with 40° angulation, 12.0 cm straight penis, and 12.9 cm PD penis with 68° angulation and hourglass. Lengths, volumes, and curvature angles were determined computationally to serve as standards. Each phantom was printed using a Makerbot Replicator+. Ten urology providers determined length, angle, and volume of phantoms using measuring tape, goniometer, and volume calculator. Means ± SD were calculated. To determine accuracy, depending on data distribution, a t-test or Wilcoxon rank sum test was used to compare provider-determined vs. computationally acquired measurements. RESULTS: Data are in Fig. 1B. Lengths for cylinder, hourglass cylinder, angled cylinder, straight penis, and PD penis were 12.9 ± 0.9 cm (p=0.0003), 12.9 ± 1.61 cm (p=0.058), 15.0 ± 4.3 cm (p=0.52), 12.0 ± 2.3 (p=0.68), and 12.7 ± 10.8 (p=0.36), respectively. Volumes were 174 ± 22 cc (p=0.003), 150 ± 24 cc (p=0.0008), 186 ± 33 cc (p=0.004), 101 ± 23 cc (p=0.16), and 87 ± 11 cc (p=0.23), respectively. Curvature angles from bent cylinder and PD phantoms were 38.3° ± 3.9° (p=0.25) and 57.5° ± 7.2° (p=0.006), respectively. Discrepancy between goniometry and computationally determined angle ranged 3° to 13°. CONCLUSIONS: Our results suggest urology providers’ measurements suffer from inaccuracy and variability, particularly in volume estimation and PD goniometry. A computational workflow may be useful for the clinical and research armamentarium when greater accuracy or volume assessment is needed. Further work using 3D image-capture techniques (Siapno et al., SPU 2019) may allow translation to patient evaluation. Source of Funding: Research Grant from Sexual Medicine Society of North America; Research Scholar Award from the Urology Care Foundation & American Urological Association. © 2020 by American Urological Association Education and Research, Inc.FiguresReferencesRelatedDetails Volume 203Issue Supplement 4April 2020Page: e495-e495 Advertisement Copyright & Permissions© 2020 by American Urological Association Education and Research, Inc.MetricsAuthor Information Dyvon Walker* More articles by this author Renea Sturm More articles by this author Doug Daniels More articles by this author Jesse Mills More articles by this author Sriram Eleswarapu More articles by this author Expand All Advertisement PDF downloadLoading ...

Non-clinical Drivers of Variation in Preoperative MRI Utilization for Breast Cancer.

Preoperative magnetic resonance imaging (MRI) utilization in breast cancer treatment has increased significantly over the past 2 decades, but its use continues to have interprovider variability and disputed clinical indications. The aim of this study was to evaluate non-clinical factors associated with preoperative breast MRI utilization. This study utilized TRICARE claims data from 2006 to 2015. TRICARE provides health benefits for active duty service members, retirees, and their dependents at both military (direct care with salaried physicians) and civilian (purchased care under fee-for-service structure) facilities. We studied patients aged 25-64years with a breast cancer diagnosis who had undergone mammogram/ultrasound (MMG/US) alone or with subsequent breast MRI prior to surgery. Facility characteristics included urban-rural location according to the National Center for Health Statistics classification. Adjusted multivariable logistic regression tests were used to identify independent factors associated with preoperative breast MRI utilization. Of the 25,392 identified patients, 64.7% (n = 16,428) received preoperative MMG/US alone, while 35.3% (n = 8964) underwent additional MRI. Younger age, Charlson Comorbidity Index score ≥ 2, active duty or retired beneficiary category, officer rank (surrogate for socioeconomic status), Air Force service branch, metropolitan location, and purchased care were associated with an increased likelihood of preoperative MRI utilization. Non-metropolitan location and Navy service branch were associated with decreased MRI use. After controlling for expected clinical risk factors, patients were more likely to receive additional MRI when treated at metropolitan facilities or through the fee-for-service system. Both associations may point toward non-clinical incentives to perform MRI in the treatment of breast cancer.

Lack of impact of nil-per-os (NPO) time on goal-directed fluid delivery in first case versus afternoon case starts: a retrospective cohort study

BackgroundGoal Directed Fluid Therapy (GDFT) represents an objective fluid replacement algorithm. The effect of provider variability remains a confounder. Overhydration worsens perioperative morbidity and mortality; therefore, the impact of the calculated NPO deficit prior to the operating room may reach harm.MethodsA retrospective single-institution study analyzed patients at UC Irvine Medical Center main operating rooms from September 1, 2013 through September 1, 2015 receiving GDFT. The primary study question asked if GDFT suggested different fluid delivery after different NPO periods, while reducing inter-provider variability. We created two patient groups distinguished by 0715 surgical start time or start time after 1200. We analyzed fluid administration totals with either a 1:1 crystalloid to colloid ratio or a 3:1 ratio. We performed direct group-wise testing on total administered volume expressed as total ml, total ml/hr., and total ml/kg/hr. between the first case start (AM) and afternoon case (PM) groups. A linear regression model included all baseline covariates that differed between groups as well as plausible confounding factors for differing fluid needs. Finally, we combined all patients from both groups, and created NPO time to total administered fluid scatterplots to assess the effect of patient-reported NPO time on fluid administration.ResultsWhether reported by total administered volume or net fluid volume, and whether we expressed the sum as ml, ml/hr., or ml/kg/hr., the AM group received more fluid on average than the PM group in all cases. In the general linear models, for all significant independent variables evaluated, AM vs PM case start did not reach significance in both cases at p = 0.64 and p = 0.19, respectively. In scatterplots of NPO time to fluid volumes, absolute adjusted and unadjusted R2 values are < 0.01 for each plot, indicating virtually non-existent correlations between uncorrected NPO time and fluid volumes measured.ConclusionsThis study showed NPO periods do not influence a patient’s volume status just prior to presentation to the operating room for surgical intervention. We hope this data will influence the practice of providers routinely replacing calculated NPO period volume deficit; particularly with those presenting with later surgical case start times.

Standardizing Opioid Prescribing Practices for Cancer-Related Pain Via a Novel Interactive Documentation Template at a Public Hospital

Opioid misuse during cancer pain management places patients at risk for harm and physicians for legal liability. Identifying and monitoring patients who are at risk is challenging given the lack of validated clinical tools and evidence-based guidelines. In the current study, we aimed to standardize opioid prescribing practices at a community oncology clinic to help ensure patient safety and physician compliance with Texas state regulations. We used the Plan-Do-Study-Act methodology. In the planning phase, current practices of assessing opioid efficacy, toxicity, and misuse were determined by surveying clinic physicians and reviewing patients' charts. We developed a new standardized process that incorporated published literature, the Texas Administrative Code, and expert opinion. Two interactive documentation templates (SmartPhrases) were designed to implement the standardized process. The intervention was studied using repeat physician surveys and chart reviews, which prompted action for refinement and sustainability. At baseline, 9% of providers followed a systematic approach to prescribing opioids and 86% expressed an interest in process standardization. We noted high interprovider variability in the opioid risk stratification and refill process. At 2 months and 6 months postimplementation, provider satisfaction with the intervention was 83% and 75%, whereas compliance with SmartPhrase use was 70% and 54%, respectively. The frequency of state database check improved from 36% to 94% at 6 months. Improvement was also noted in assessment and documentation of baseline risk, chemical coping, and toxicity. We implemented a systematic approach for assessing opioid misuse, toxicity, and efficacy during cancer pain management at a community oncology clinic. The approach resulted in notable improvement in provider practices and documentation compliance.

Abstract P5-13-13: Non-clinical drivers of variation in preoperative MRI utilization for breast cancer

Abstract Background: Preoperative MRI utilization in breast cancer treatment has increased significantly over the past two decades but its use continues to have inter-provider variability and disputed clinical indications. The objective of this study was to evaluate non-clinical factors associated with preoperative breast MRI utilization. Methods: This study utilized claims from the Military Health System Data Repository (MDR) on TRICARE Prime beneficiaries, from fiscal years 2006-2015. TRICARE provides health benefits for Active Duty service members, retirees, and their dependents at both military (direct care with salaried physicians) and civilian (purchased care with fee-for-service physicians) treatment facilities. We studied patients aged 25-64 years old with a breast cancer diagnosis who had undergone mammogram/breast ultrasound alone or with subsequent breast MRI prior to surgery. Patient demographics and treatment characteristics were abstracted from the data. The National Center for Health Statistics (NCHS) urban-rural classification was used to determine the urbanization level of the treatment facility. Adjusted multivariate logistic regression tests were used to identify independent factors associated with preoperative breast MRI utilization. Results: Of the 25,656 identified patients, 64.4% of patients (n=16,511) received preoperative mammogram/breast ultrasound alone while 35.6% of patients (n=9,145) underwent additional MRI after mammographic and/or ultrasound imaging. On multivariable analysis, younger age, presence of two or more comorbidities, active duty or retired beneficiary category, officer rank (surrogate for socioeconomic status), Air Force service branch, metropolitan location, and purchased care were associated with increased likelihood of preoperative MRI utilization. Nonmetropolitan location and Navy service branch were associated with decreased MRI use. Odds Ratio95% Confidence IntervalAge Group (Ref: 55-64 years)25-34 years1.851.60-2.15 35-44 years1.591.47-1.72 45-54 years1.271.19-1.35Charlson Comorbidity Index (Ref 0-1)2+2.472.33-2.61Beneficiary Category (Ref: Dependent)Active Duty1.201.04-1.38 Retired1.231.09-1.40Rank (Ref: Senior Enlisted)Junior Enlisted0.930.78-1.11 Junior Officer1.251.14-1.37 Senior Officer1.481.36-1.60 Warrant Officer1.231.06-1.42Service Branch (Ref: Army)Air Force1.101.03-1.18 Navy0.920.85-0.99 Marines0.950.84-1.07 Coast Guard1.070.89-1.29Urban-Rural Classification (Ref: Medium Metropolitan)Large Central Metropolitan1.801.68-1.93 Large Fringe Metropolitan1.591.47-1.71 Small Metropolitan0.650.59-0.71 Micropolitan0.400.34-0.46 Noncore0.250.18-0.34Treatment Facility Care Setting (Ref: Direct Care)Purchased Care1.601.48-1.73 Conclusions: After controlling for expected clinical risk factors, patients were more likely to receive additional MRI when treated at larger metropolitan facilities or through the purchased, fee-for-service system. Both associations may point toward non-clinical incentives to perform MRI in the treatment of breast cancer. Citation Format: Pak LM, Banaag A, Koehlmoos TP, Haider AH, Learn PA. Non-clinical drivers of variation in preoperative MRI utilization for breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-13-13.

An observational study of end-tidal carbon dioxide trends in general anesthesia

PurposeDespite growing evidence supporting the potential benefits of higher end-tidal carbon dioxide (ETCO2) levels in surgical patients, there is still insufficient data to formulate guidelines for ideal intraoperative ETCO2 targets. As it is unclear which intraoperative ETCO2 levels are currently used and whether these levels have changed over time, we investigated the practice pattern using the Multicenter Perioperative Outcomes Group database.MethodsThis retrospective, observational, multicentre study included 317,445 adult patients who received general anesthesia for non-cardiothoracic procedures between January 2008 and September 2016. The primary outcome was a time-weighted average area-under-the-curve (TWA-AUC) for four ETCO2 thresholds (< 28, < 35, < 45, and > 45 mmHg). Additionally, a median ETCO2 was studied. A Kruskal-Wallis test was used to analyse differences between years. Random-effect multivariable logistic regression models were constructed to study variability.ResultsBoth TWA-AUC and median ETCO2 showed a minimal increase in ETCO2 over time, with a median [interquartile range] ETCO2 of 33 [31.0–35.0] mmHg in 2008 and 35 [33.0–38.0] mmHg in 2016 (P <0.001). A large inter-hospital and inter-provider variability in ETCO2 were observed after adjustment for patient characteristics, ventilation parameters, and intraoperative blood pressure (intraclass correlation coefficient 0.36; 95% confidence interval, 0.18 to 0.58).ConclusionsBetween 2008 and 2016, intraoperative ETCO2 values did not change in a clinically important manner. Interestingly, we found a large inter-hospital and inter-provider variability in ETCO2 throughout the study period, possibly indicating a broad range of tolerance for ETCO2, or a lack of evidence to support a specific targeted range. Clinical outcomes were not assessed in this study and they should be the focus of future research.

Quality Improvement and The Veterans Health Administration’s Palliative Care National Clinical Template (QI812)

•Identify opportunities to use existing process level EHR data in their practice settings for quality assessment.•Identify approaches to optimize quality efforts using process level data. Process measures provide evidence of care delivered at the bedside. The Veterans Health Administration’s (VHA) Palliative Care National Clinical Template (PC-NCT), writes process-level data from palliative care (PC) consultations to the Electronic Health Record (EHR) that are aggregated and used for quality improvement (QI). To describe experience using the PC-NCT across facilities for QI. Process-level data generated during PC-NCT consults (7/1/2014 – 6/30/2017) were extracted from the EHR. We obtained user feedback and descriptively analyzed patterns of use for QI. The PC-NCT was used 32,566 times (34% of palliative care consultations) at 82 facilities (67% of facilities) to document care. At the domain level, Clinical Presentation was used in 97%, Social History 84%, Symptom Review 78%, Psychological Assessment 17%, Palliative Performance Scale (PPS) 77%, Care Planning 79%, Spiritual Concerns 57%, Physical Examination 87%, and Assessment and Recommendations 94% of PC-NCT consults. Facility culture, preferential use of locally developed tools, and challenges with note formatting are reported as reasons for non-use. PC-NCT data identify clinical questions for teams that are reviewed during twice yearly QI calls. Sites with low PPS scores are prompted with, “How are you collaborating with other services to promote earlier PC consultation?” Those with large numbers of patients reporting dyspnea are asked to consider how opioids are used in dyspnea management, and sites with sparse documentation of spiritual concerns are asked to consider what questions clinicians ask about spiritual concerns and whether they are comfortable addressing those needs. In addition to template use for quality assessment, it is used to educate new PC trainees on salient content for comprehensive consults and reduce inter-provider variability in PC delivery. Competition for scarce resources in healthcare necessitates optimizing use of existing data. Process-level data from a standardized PC note template can be used for QI, training, and standardization among staff.

Quality improvement to reduce chronic lung disease rates in very-low birth weight infants: high compliance with a respiratory care bundle in a small NICU.

Reduce chronic lung disease by 10% among very-low birth weight infants by implementing a consistent respiratory care bundle. Prospective quality improvement study of infants below 32 weeks gestation in a small neonatal intensive care unit. A respiratory care bundle to eliminate inter-provider variability and minimize use of mechanical ventilation was implemented. This included: defining delivery room management with use of continuous positive airway pressure/nasal intermittent positive pressure ventilation, uniform intubation/extubation criteria, and standardizing ventilation/post-extubation support. A total of 107 very-low birth weight infants were included in this project. Compliance with the respiratory care bundle was >90%. Chronic lung disease rates at 36 weeks postmenstrual age fell from 43 to 12% (P = 0.0006), rates of combined chronic lung disease/death decreased from 50 to 20% (P = 0.002, OR = 0.25, 95% CI 0.1-0.6), rates of severe intraventricular hemorrhage decreased from 13 to 0% (P = 0.005), and surgical ligation of patent ductus arteriosus decreased from 35 to 3% (P < 0.0001). High compliance with the respiratory care bundle was achieved and a 73% reduction in chronic lung disease rates in very-low birth weight infants occurred.

Standardizing opioid prescribing practices for cancer-related pain via a novel interactive documentation template at a public hospital.

253 Background: Aberrant opioid use during treatment of cancer related pain poses serious risks for patients and society. Identifying and monitoring those at risk is challenging and time consuming. We aimed to standardize the process of managing opioids at a public hospital oncology clinic in compliance with Texas (TX) state regulations to improve patient safety. Methods: We gauged the current practices of assessing and documenting opioid efficacy, toxicity and misuse via provider survey and chart review. A process flowchart was then created and streamlined in the light of: published literature, TX Administrative Code on pain management, and practices of palliative care specialists at MD Anderson Cancer Center, TX. We made an interactive documentation template (SmartPhrase) in electronic medical record system (EPIC) to implement the standardized process. Results: Pre-intervention data showedhigh inter-provider variability in opioid misuse screening, safety monitoring and documentation. Of the 22 providers surveyed, 6 (22%) had a TX PMP Aware login; 8 (36%) ever checked the website; and 19 (86%) stated the need of a standardized process for prescribing opioids. Hence, we instituted baseline screening for aberrant behavior based on substance abuse history, TX PMP Aware query and urine drug screen; increased monitoring of high risk patients; universal screening for chemical coping at each visit; and timely referral of such patients to palliative care. We created a SmartPhrase that prompts providers to risk stratify and monitor for opioid misuse as above; assess analgesic efficacy; evaluate and manage uncontrolled pain and toxicities; and specify a refill plan. This SmartPhrase allows quick selection of options pertinent to a given patient from pre-populated lists before the encounter can be closed. Conclusions: We created a systematic approach to prescribing and monitoring opioid analgesics so as to ensure safe and judicious use of opioids to treat cancer related pain. We generated an interactive SmartPhrase for efficient documentation compliant with TX state regulations. Post-implementation review of efficacy and use of this intervention is ongoing to further refine it.

Variability in intraosseous flush practices of emergency physicians

ObjectiveIntramedullary pressure changes during intraosseous (IO) procedures have been implicated in the intravasation of bone marrow fat and with pain in conscious patients. The objective of this study was to demonstrate inter-provider variability in pressures generated during initial flush procedures. MethodsIO cannulas were inserted into the proximal tibiae and humeri by study personnel. A second cannula was placed in the mid diaphysis of each bone to record intramedullary pressures. Fifteen emergency physicians performed 60 flushes in random order in two cadavers while flush duration and IO pressure were continuously recorded. Providers were blinded to the flush pressures they generated and the flush techniques of others. ResultsThe median IO pressure (IOP) generated by providers was 903 mm Hg (range, 83-2941 mm Hg) and the median flush duration was 5.2 seconds (range, 1.0-13.4 seconds). Significant differences were noted among providers in peak IOP generated (analysis of variance P < .001). Providers were consistent in the forces they generated relative to each other. An inverse, nonlinear relationship was observed between flush duration and the peak IOP generated. Significant differences were noted in intramedullary flush pressures at flush sites within cadavers (analysis of variance P: cadaver #1 P < .001; cadaver #2 P = .012). ConclusionsThe IO compartment pressures generated by physicians demonstrated significant interoperator variability with greater than 35-fold difference in flush forces, and an inverse relationship between intraosseous pressure and flush duration. It may be prudent practice for providers to extend the flush over several seconds, thus limiting maximal pressures.

Risk Factors for Non-Relapse Mortality in Chronic Graft Versus Host Disease: Predicting Outcomes At Disease Onset

AbstractAbstract 1941Chronic graft versus host disease (cGVHD) is a major contributor to morbidity and mortality in long-term survivors of allogeneic hematopoietic cell transplantation (HCT). The treatment course is heterogeneous and unpredictable, and a prognostic system at the time of disease onset is highly desirable. Current systems were not developed for disease onset (Arai, Blood, 2011), or used the limited/extensive classification (Arora, Blood, 2011). In addition, utilization of these systems is time consuming, complex, and practice variation can lead to inconsistent application. We developed a simple system that should minimize inter-provider variability, and can be applied at disease onset. We studied 341 consecutive adult patients (age 18–71y; median 53y) who received HCT at the University of Michigan from 2007 to 2010. Data was prospectively collected under an IRB approved protocol. All cGVHD data, including NIH global severity score, was adjudicated by 2 clinicians with expertise in cGVHD. Cumulative incidence of cGVHD was 47%, with a median time to onset of 190d (range 37–806d). Patients presented with de novo (n=64, 41%), quiescent (n=60, 38%), or progressive onset disease (n=34, 21%). The global NIH severity at onset was mild (n=23, 15%), moderate (n=79, 50%) or severe (n=56, 35%). We performed univariable analysis on multiple potential risk factors (RF) (Table) for non-relapse mortality (NRM) within 4 years of cGVHD onset, with relapse considered a competing risk. RF considered included characteristics present prior to cGVHD onset, and several cGVHD specific factors, such as number of organs involved and individual organ severity in multiple combinations. History of severe acute GVHD grades III/IV (severe aGVHD) (Hazard Ratio (HR)=4.0, p<0.001), Karnofsky performance score (KPS) ≤ 70% (HR 4.6, p<0.001), and grade 2/3 lung (HR=3.3, p=0.04) were the strongest predictors of NRM, while presence of grade 1 liver cGVHD was protective (HR=0.3, p=0.03). We next tested combinations of these RF to develop a simple scheme for risk stratification. The simplest stratification grouped patients into 3 categories: High risk (KPS ≤ 70% and/or severe aGVHD), low risk (grade 1 liver, without high risk features), and intermediate risk (all others). Four year NRM in our model was 52% in high risk (n=44, ref), 4% in low risk (n=28, p=0.01), and 18% in intermediate patients (n=86, p=0.001, Figure). Within the high risk group, both low KPS and prior severe aGVHD remained independent predictors after adjustment for each other (HR=4.5, p=0.02 and HR=3.4, p=0.02, respectively), and in the entire cohort. Low KPS patients were more likely to have lung involvement, be on steroids, and have more severe cGVHD, implying that KPS serves as a composite of multiple other RF, including both HCT-related morbidities and cGVHD burden. Severe aGVHD was prevalent in 18% of patients with quiescent onset and 26% of patients with progressive onset, which was not statistically significant (p=0.35), and the risk of NRM in patients with a history of severe aGVHD did not differ based on type of onset (HR 1.02, p=0.9). Platelet count at onset did not predict NRM, but was a RF 3 months post-onset, emphasizing the time dependence of some RF. In conclusion, at onset of cGVHD a history of severe aGVHD, low KPS and grade 1 liver are important predictors of NRM. Factors known to affect morbidity or quality of life did not appear to impact NRM in this study. The combination of the 3 RF into a simple, empirically derived stratification system broadly categorized patients into 3 meaningful NRM risk groups at onset of cGVHD. If validated in an independent population, this classification system may be clinically useful and facilitate referral and management in chronic GVHD clinics.UnivariableRisk FactornHR95% CIp valueBefore OnsetAge at HCT, y18–4029Ref41–55701.80.5–6.40.3> 55592.30.7–8.050.2Related donor790.60.3–1.20.2Reduced Intensity Conditioning511.10.5–2.40.7History of aGVHD III–IV204.01.9–8.5<0.001OnsetKarnofsky Performance Score ≤ 70%314.62.5–9.5<0.001NIH Onset SeverityMild23RefModerate790.60.2–1.80.4Severe561.20.4–3.40.7Progressive onset vs De novo/Quiescent342.11.0–4.50.05Platelets < 100,000/mm3 at onset421.40.7–3.10.3On steroids at onset282.00.9–4.50.09NIH grade 1 liver at onset400.30.1–0.90.03NIH grade 2/3 lung at onset173.31.03–10.30.04 [Display omitted] Disclosures:No relevant conflicts of interest to declare.

Motion Capture Measures Variability in Laryngoscopic Movement During Endotracheal Intubation

Success rates with emergent endotracheal intubation (ETI) improve with increasing provider experience. Few objective metrics exist to quantify differences in ETI technique between providers of various skill levels. We tested the feasibility of using motion capture videography to quantify variability in the motions of the left hand and the laryngoscope in providers with various experience. Three providers with varying levels of experience [attending physician (experienced), emergency medicine resident (intermediate), and postdoctoral student with no previous ETI experience (novice)] each performed ETI 4 times on a mannequin. Vicon, a 16-camera system, tracked the 3-dimensional orientation and movement of markers on the providers, handle of the laryngoscope, and mannequin. Attempt duration, path length of the left hand, and the inclination of the plane of the laryngoscope handle (mean square angular deviation from vertical) were calculated for each laryngoscopy attempt. We compared interattempt and interprovider variability of each measure. All ETI attempts were successful. Mean (SD) duration of laryngoscopy attempts differed between experienced [5.50 (0.68) seconds], intermediate [6.32 (1.13) seconds], and novice [12.38 (1.06) seconds] providers (P = 0.021). Mean path length of the left hand did not differ between providers (P = 0.37). Variability of the plane of the laryngoscope differed between providers: 8.3 (experienced), 28.7 (intermediate), and 54.5 (novice) degrees squared. Motion analysis can detect interprovider differences in hand and laryngoscope movements during ETI, which may be related to provider experience. This technology has potential to objectively measure training and skill in ETI.

Why surgeons can say “No”: Exploring “Unilateral Withholding”

To explore why it is permissible for surgeons to "unilaterally withhold" surgery, whereas it is not commonplace (in the United States) to unilaterally withhold cardiopulmonary resuscitation (CPR) for clinical situations with similar degrees of uncertainty and prognosis. The medical literature was sampled using PubMed and Google search engines, employing a variety of search strategies to capture articles relating to medical/surgical decision-making, risk aversion, acute care surgery, and withholding life-saving therapies. These topics are used to highlight interprovider variability that affects all practitioners-not just surgeons-and to consider why we deem it permissible for surgeons to withhold surgery, whereas-in the United States, at least-it is not routinely permissible for clinicians to unilaterally withhold mechanical ventilation and CPR for cases with similar prognoses. While there are no published research studies that deal directly with this topic, knowledge, heuristics, experience, variable aversion to risk, and other features inherent in medical-surgical education likely impact decisions to offer or withhold potentially life-saving therapies of all kinds. Both surgeons and clinicians, who request surgical consultation for hospitalized patients, should consider these issues and politely pursue second opinions when there is any doubt whether forgoing surgery is in the patient's best interests. Similarly, while unilateral withholding of CPR is not commonly employed in some medical cultures, including the United States, beneficence can be facilitated through robust informed consent.

Massachusetts Cardiac Surgery Report Card: Implications of Statistical Methodology

Choice of statistical methodology may significantly impact the results of provider profiling, including cardiac surgery report cards. Because of sample size and clustering issues, logistic regression may overestimate systematic interprovider variability, leading to false outlier classification. Theoretically, the use of hierarchical models should result in more accurate representation of provider performance. Extensively validated and audited data were available for all 4,603 isolated coronary artery bypass grafting procedures performed at 13 Massachusetts hospitals during 2002. To produce the official Massachusetts cardiac surgery report card, a 19-variable predictor set and a hierarchical generalized linear model were employed. For the current study, this same analysis was repeated with the 14 predictors used in the New York Cardiac Surgery Reporting System. Two additional analyses were conducted using each set of predictor variables and applying standard logistic regression. For each of the four combinations of predictors and models, the point estimates of risk-adjusted 30-day mortality, 95% confidence or probability intervals, and outlier status were determined for each hospital. Overall unadjusted mortality for coronary bypass operations was 2.19%. For most hospitals, there was wide variability in the point estimates and confidence or probability intervals of risk-adjusted mortality depending on statistical model, but little variability relative to the choice of predictors. There were no hospital outliers using hierarchical models, but there was one outlier using logistic regression with either predictor set. When used to compare provider performance, logistic regression increases the possibility of false outlier classification. The use of hierarchical models is recommended.

Cardiac surgery report cards: comprehensive review and statistical critique

Public report cards and confidential, collaborative peer education represent distinctly different approaches to cardiac surgery quality assessment and improvement. This review discusses the controversies regarding their methodology and relative effectiveness. Report cards have been the more commonly used approach, typically as a result of state legislation. They are based on the presumption that publication of outcomes effectively motivates providers, and that market forces will reward higher quality. Numerous studies have challenged the validity of these hypotheses. Furthermore, although states with report cards have reported significant decreases in risk-adjusted mortality, it is unclear whether this improvement resulted from public disclosure or, rather, from the development of internal quality programs by hospitals. An additional confounding factor is the nationwide decline in heart surgery mortality, including states without quality monitoring. Finally, report cards may engender negative behaviors such as high-risk case avoidance and “gaming” of the reporting system, especially if individual surgeon results are published. The alternative approach, continuous quality improvement, may provide an opportunity to enhance performance and reduce interprovider variability while avoiding the unintended negative consequences of report cards. This collaborative method, which uses exchange visits between programs and determination of best practice, has been highly effective in northern New England and in the Veterans Affairs Administration. However, despite their potential advantages, quality programs based solely on confidential continuous quality improvement do not address the issue of public accountability. For this reason, some states may continue to mandate report cards. In such instances, it is imperative that appropriate statistical techniques and report formats are used, and that professional organizations simultaneously implement continuous quality improvement programs. The statistical methodology underlying current report cards is flawed, and does not justify the degree of accuracy presented to the public. All existing risk-adjustment methods have substantial inherent imprecision, and this is compounded when the results of such patient-level models are aggregated and used inappropriately to assess provider performance. Specific problems include sample size differences, clustering of observations, multiple comparisons, and failure to account for the random component of interprovider variability. We advocate the use of hierarchical or multilevel statistical models to address these concerns, as well as report formats that emphasize the statistical uncertainty of the results.