Codes For Pulmonary Embolism Research Articles

Abstract 14344: Pitfalls of ICD-10 Codes for Identifying Pulmonary Embolism in Electronic Records: Results From the Multicenter PE-EHR+ Study

Background: Many research studies from electronic databases and even the American Heart Association annual statistics for pulmonary embolism (PE) rely on the International Classification of Disease, 10 th modification (ICD-10) codes. However, the validity of ICD-10 codes for PE remains uncertain. Methods: Using a pre-specified protocol, we identified three groups of patients in the Mass General-Brigham Health system hospitals (MGB, 2016-2021) in equal distribution: Those with ICD-10 Principal discharge codes for PE, those with ICD-10 secondary codes for PE, and those without codes for PE (N=578 each). The accuracy of ICD-10 codes for identification of PE was assessed in reference to review of each chart by two independent physicians who used pre-specified criteria for presence of PE. Weighted estimates were obtained by considering the total of number of hospitalizations at MGB in each group. Results: After excluding duplicates, 1712 entered the analysis (age: 60.6 years, 52.3% female). Using the ICD-10 PEs only in the Principal discharge diagnosis, sensitivity and specificity were 60.3% and 94.6%, respectively. Although use of Principal-or-secondary ICD-10 PE codes had a sensitivity of 99.8% in the study sample, the weighted sensitivity in the parent cohort was reduced to 83.2%. The positive predictive value of Principal ICD-10 discharge codes and Principal-or-secondary ICD-10 codes were 91.9% and 79.2%, respectively (Table). Conclusions: Although the Principal discharge diagnosis codes for PE have an excellent positive predictive value, they miss nearly 40% of new PE events in hospitalized patients. Integration of secondary discharge codes only partially improves the sensitivity at the cost of reducing the positive predictive value, highlighting the need for modified strategies, such as use combination with procedure codes, or use of natural language processing -when feasible -for identification of patients with PE.

Abstract 17709: ICD-10 Codes Have Low Sensitivity and Suboptimal Positive Predictive Value for Identifying Cases of Pulmonary Embolism With Cor Pulmonale in Electronic Healthcare Records

Introduction: The International Classification of Diseases, 10 th revision (ICD-10) codes were updated in 2015 to include designation of pulmonary embolism (PE) with cor pulmonale . These codes can be advantageous to identify higher-risk patients with PE for research or quality improvement. However, the validity of these codes is unknown. Methods: We reviewed patient data from Mass General Brigham (MGB) health system (01/2016-12/2021). Patients were randomly selected from 3 groups: Patients with Principal discharge diagnosis codes for PE, patients with secondary discharge diagnosis of PE, and patients with no PE codes. Accuracy of codes for PE with cor pulmonale were assessed with reference to manual review by 2 independent physicians using pre-specified criteria. Weighted estimates were obtained by considering the total number of hospitalizations. Results: 1,712 patients were included (age: 60.6±17.8 years, 52.3% female). Chart review found 865 patients with PE, of whom 360 (41.6%) had PE with cor pulmonale according to predefined clinical criteria. Of these 360 PE patients, 358 (99.4%) had ICD-10 codes for PE. However, only in 36 (10.1%) cases, the specific cor pulmonale codes were used. There were 24 additional cases with ICD-10 codes for PE with cor pulmonale in whom the diagnosis was not confirmed per manual review. Sensitivity and specificity for PE with cor pulmonale codes in Principal-or-secondary position were 10.0% and 98.2% (Table, panel A). The positive predictive value (PPV) for PE with cor pulmonale in Principal-or-secondary discharge position was 63.2% in the weighted sample (Table, panel B). Conclusions: ICD-10 codes for identification of acute PE with cor pulmonale have excellent specificity, but very poor sensitivity and suboptimal PPV. These results are informative for investigators who may use such ICD-10 codes for research or quality improvement purposes.

Abstract 16342: Age- and Sex-Differences in Accuracy of ICD-10 Codes for Identifying Adults With Pulmonary Embolism: PE-EHR+ Study

Background Age and sex-differences exist in presentation of adults with pulmonary embolism (PE). However, it remains unclear if these differences influence the accuracy of International Classification of Diseases, Tenth Revision (ICD-10) codes used for identification of PE. Methods: Using multicenter data from Mass General-Brigham Health System (2016-2021), patients were randomly selected from 3 groups: 578 patients with Principal Discharge codes for PE, 578 with Secondary Discharge codes for PE, and 578 without codes for PE. The reference standard for identification of PE was manual chart review by 2 independent physicians. Sensitivity, specificity, and positive and negative predictive values were determined by age (<65 vs ≥65 years) and sex (female vs male). Weighted estimates were obtained by using the total number of hospitalizations. Results: Among 1712 patients included in final analyses, 896 (52.3%) were female and 775 (45.3%) were ≥65 years. Use of Principal-or-Secondary ICD-10 codes had a similar sensitivity by sex (99.5% vs 100% among females vs males) and age (99.8% among both groups) in the study sample. The weighted sensitivity among females was 72.8% (vs 100% for males) and 80.4% among patients ≥65 years (vs 85.2% for <65 years). In the weighted cohort, the positive predictive value was 78.2% among females (vs 80.3% for males) and 78.3% among those ≥65 years (vs 80.0% for <65 years) (Table). Age and sex-differences in the diagnostic accuracy were similarly observed in analyses restricted to Principal, or to Secondary codes. Conclusions: Principal-or-Secondary ICD-10 codes had lower sensitivity and positive predictive value among females and older adults, though the sex-differences in sensitivity may be influenced by limited number of false negatives in this study. Additionally, Secondary codes had higher false positives among older adults. These findings may have important implications for identification of patients with PE using ICD-10 codes.

A Deep-Learning Algorithm-Enhanced Electrocardiogram Interpretation for Detecting Pulmonary Embolism.

The early diagnosis of pulmonary embolism (PE) remains a challenge. Electrocardiograms (ECGs) and D-dimer levels are used to screen potential cases. To develop a deep learning model (DLM) to detect PE using ECGs and investigate the clinical value of false detections in patients without PE. Among patients who visited the emergency department between 2011 and 2019, PE cases were identified through a review of medical records. Non-PE ECGs were collected from patients without a diagnostic code for PE. There were 113 PE and 51,456 non-PE ECGs in the training and validation sets for developing the DLM, respectively, and 27 PE and 13,105 non-PE cases in an independent testing set for performance validation. A human-machine competition was conducted from the testing set to compare the performance of the DLM with that of physicians. Receiver operating characteristic (ROC) curves, sensitivity, and specificity were used to determine the diagnostic value. Survival analysis was used to assess the prognosis of the patients without PE, stratified by DLM prediction. The DLM was as effective as physicians in diagnosing PE, with 70.8% sensitivity and 69.7% specificity. The area under the ROC curve of DLM was 0.778 in the testing set and up to 0.9 with D-dimer and demographic data. The non-PE patients whose ECG was misclassified as PE by DLM had higher all-cause mortality [hazard ratio (HR) 2.13 (1.51-3.02)] and risk of non-cardiovascular hospitalization [HR 1.55 (1.42-1.68)] than those correctly classified. A DLM-enhanced ECG system may prompt PE recognition and provide prognostic outcomes in patients with false-positive predictions.

SAT464 Hyperthyroidism Does Not Negatively Impact Outcomes of Adult Hospitalizations for Pulmonary Embolism: A Nationwide Analysis

Abstract Disclosure: A.G. Olafimihan: None. V. Deenadayalan: None. P.E. Ojemolon: None. R. Litvin: None. U.A. Ogar: None. E. Krishnaraju: None. F. Sami: None. E.H. Olojakpoke: None. Introduction: Acute pulmonary embolism (PE) is a medical emergency associated with high mortality rates. Hyperthyroidism has been associated with an increased risk of venous thromboembolism and PE. However, there are limited studies on its impact on the outcomes of patients with PE. We aimed to assess hyperthyroidism’s effect on in-hospital PE outcomes using the National Inpatient Sample (NIS). Methods: Data were extracted from the US Nationwide Inpatient Sample (NIS) from 2016 to 2019. The NIS contains hospital discharge information estimating yearly US hospitalizations. We included hospitalizations with a principal discharge diagnostic code for PE using the ICD-10-CM code (E22.2). Hospitalizations of patients less than 18 years of age were excluded from the study. This cohort was divided based on the secondary diagnosis of hyperthyroidism.The primary outcome was inpatient mortality, and secondary outcomes were hospital length of stay (LOS), total hospital charge, odds of acute respiratory failure (ARF), acute renal failure (AKI), undergoing embolectomy, tissue plasminogen activator (tPA) use and Inferior vena cave (IVC) filter placement. Multivariate linear regression analysis was used to adjust for confounders. Statistical significance was set at a p-value < 0.05. Results: There were 745,360 hospitalizations involving PE as the reason for admission, and 20.1% had comorbidity with hyperthyroidism. Admissions for PE with comorbid hyperthyroidism had similar in-hospital mortality (adjusted OR:1.17, P=0.41), mean LOS (4.9 vs 4.3 days, P=0.11) and mean THC ($53,127 vs $48,236, P=0.43) compared to those without hyperthyroidism. The odds of developing ARF (aOR: 0.93, P=0.466) or AKI (aOR: 0.95, P=0.71) were similar. Odds of undergoing embolectomy (aOR: 1.07, p=0.83), receiving tPA (aOR=1.03, P=0.87) and placing IVC filter (aOR=0.80, P=0.19) were similar between both groups. Conclusion: Hospitalizations for pulmonary embolism with hyperthyroidism had similar inpatient mortality compared to those without hyperthyroidism. Also, mean LOS, total hospital charge, ARF and AKI complications, and odds of embolectomy, tPA administration and IVC filter placement were alike for both categories. Presentation Date: Saturday, June 17, 2023

RACIAL DISPARITIES IN PULMONARY EMBOLISM RISK FACTORS AND MORTALITY IN THE UNITED STATES: 1999-2020

RACIAL DISPARITIES IN PULMONARY EMBOLISM RISK FACTORS AND MORTALITY IN THE UNITED STATES: 1999-2020

Developing and validating natural language processing algorithms for radiology reports compared to ICD-10 codes for identifying venous thromboembolism in hospitalized medical patients

BackgroundIdentifying venous thromboembolism (VTE) from large clinical and administrative databases is important for research and quality improvement. ObjectiveTo develop and validate natural language processing (NLP) algorithms to identify VTE from radiology reports among general internal medicine (GIM) inpatients. MethodsThis cross-sectional study included GIM hospitalizations between April 1, 2010 and March 31, 2017 at 5 hospitals in Toronto, Ontario, Canada. We developed NLP algorithms to identify pulmonary embolism (PE) and deep venous thrombosis (DVT) from radiologist reports of thoracic computed tomography (CT), extremity compression ultrasound (US), and nuclear ventilation-perfusion (VQ) scans in a training dataset of 1551 hospitalizations. We compared the accuracy of our NLP algorithms, the previously-published “simpleNLP” tool, and administrative discharge diagnosis codes (ICD-10-CA) for PE and DVT to the “gold standard” manual review in a separate random sample of 4000 GIM hospitalizations. ResultsOur NLP algorithms were highly accurate for identifying DVT from US, with sensitivity 0.94, positive predictive value (PPV) 0.90, and Area Under the Receiver-Operating-Characteristic Curve (AUC) 0.96; and in identifying PE from CT, with sensitivity 0.91, PPV 0.89, and AUC 0.96. Administrative diagnosis codes and the simple NLP tool were less accurate for DVT (ICD-10-CA sensitivity 0.63, PPV 0.43, AUC 0.81; simpleNLP sensitivity 0.41, PPV 0.36, AUC 0.66) and PE (ICD-10-CA sensitivity 0.83, PPV 0.70, AUC 0.91; simpleNLP sensitivity 0.89, PPV 0.62, AUC 0.92). ConclusionsAdministrative diagnosis codes are unreliable in identifying VTE in hospitalized patients. We developed highly accurate NLP algorithms to identify VTE from radiology reports in a multicentre sample and have made the algorithms freely available to the academic community with a user-friendly tool (https://lks-chart.github.io/CHARTextract-docs/08-downloads/rulesets.html#venous-thromboembolism-vte-rulesets)

Effectiveness and Safety of Rivaroxaban in Patients with Cancer-Associated Venous Thromboembolism

Background: Venous thromboembolism (VTE), consisting of deep vein thrombosis (DVT) and pulmonary embolism (PE), is a common complication of both cancer and its treatment.Unlike low-molecular weight heparins (LMWHs) which represent the current standard-of-care for cancer-associated venous thromboembolism (CAT) management, rivaroxaban can be administered orally, avoiding the need for uncomfortable injections. While not guideline-recommended as first-line therapy for CAT, rivaroxaban has been used in routine clinical practice for this purpose.Objective: To estimate the cumulative incidence of recurrent VTE, major bleeding and mortality in patients with CAT treated with rivaroxaban as outpatients in routine clinical practice.Methods: Within US Truven MarketScan administrative claims data from 1/2012 to 6/2015, we identified adult patients with active cancer using National Cancer Institute-Surveillance, Epidemiology and End Results Program (NCI-SEER) coding who had ≥1 primary hospitalization or emergency department visit discharge diagnosis code for DVT and/or PE (the index event) and received a prescription of rivaroxaban as their first outpatient anticoagulant within 30-days of the index VTE. Eligible patients were required to have a minimum of 180-days continuous medical and prescription benefits prior to the index VTE (baseline period). Patients with a prior claim for VTE or receiving anticoagulation during the baseline period were excluded. We estimated the cumulative incidence with 95% confidence intervals (CIs) of recurrent VTE, major bleeding (identified per the Cunningham algorithm) and mortality (defined as in-hospital mortality or need for hospice care without subsequent claims) at 180-days assuming competing risks.Results: A total of 949 active cancer patients were initiated on rivaroxaban following their index VTE. The mean ± standard deviation (SD) time from index VTE to first outpatient rivaroxaban fill was 4.4±5.8 days. Mean±SD age of identified patients was 62.5±12.8 years, 45.6% were men and PE (with or without DVT) was present in 43.6% of patients. Time from the diagnosis of active cancer to index CAT was ≤90-days in 30.0%, 91-180 days for 18.9% and 181-days of greater for 54.3% of patients. Cancer sites for this rivaroxaban-treated cohort were as follows: 25.5% breast, 21.0% genitourinary (including prostate), 13.2% hematologic, 13.2% colorectal, 11.5% lung, 7.4% gynecologic, 5.8% pancreas, 4.1% stomach/esophagus and 39.5% other. Metastatic disease was present in 42.6% of patients. During follow-up, there were 37 cases of recurrent VTE, 22 cases of major bleeding and 105 deaths. The 180-day cumulative incidence estimate was 4.0% (95%CI 2.8-5.4%) for recurrent VTE; 2.7% (95%CI=1.7−4.0%) for major bleeding and 11.3% (95%CI=9.2-13.6%) for mortality (Figure).Conclusion: Event rates observed in this rivaroxaban treated cohort were consistent with those observed in prior studies of LMWH- and rivaroxaban-managed CAT patients. Our data may provide some reassurance to clinicians regarding the effectiveness and safety of rivaroxaban in treating CAT patients in routine outpatient practice. [Display omitted] DisclosuresBeyer-Westendorf:Pfizer: Honoraria, Research Funding; Bayer: Honoraria, Research Funding; Daiichi-Sankyo: Honoraria, Research Funding; Boehringer-Ingelheim: Honoraria, Research Funding. Spyropoulos:Boehringer Ingelheim: Consultancy; Janssen Pharmaceuticals: Consultancy, Membership on an entity's Board of Directors or advisory committees; Daiichi Sankyo: Consultancy; Bayer: Consultancy, Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees; Portola: Membership on an entity's Board of Directors or advisory committees. Coleman:Boehringer Ingelheim: Consultancy, Honoraria; Janssen Pharmaceuticals: Consultancy, Honoraria, Research Funding; Bayer AG: Consultancy, Honoraria, Research Funding.

Risk assessment of acute pulmonary embolism utilizing coronary artery calcifications in patients that have undergone CT pulmonary angiography and transthoracic echocardiography.

To evaluate the relation of coronary artery calcifications (CAC) on non-ECG-gated CT pulmonary angiography (CTPA) with short-term mortality in patients with acute pulmonary embolism (PE). We retrospectively included all in-patients between May 2007 and December 2014 with an ICD-9 code for acute PE and CTPA and transthoracic echocardiography available. CAC was qualitatively graded as absent, mild, moderate, or severe. Relations of CAC with overall and PE-related 30-day mortality were assessed using logistic regression analyses. The independence of those relations was assessed using a nested approach, first adjusting for age and gender, then for RV strain, peak troponin T, and cardiovascular risk factors for an overall model. Four hundred seventy-nine patients were included (63 ± 16 years, 52.8% women, 47.2% men). In total, 253 (52.8%) had CAC-mild: 143 (29.9%); moderate: 89 (18.6%); severe: 21 (4.4%). Overall mortality was 8.8% (n = 42) with higher mortality with any CAC (12.6% vs. 4.4% without; odds ratio [OR] 3.1 [95%CI 2.1-14.5]; p = 0.002). Mortality with severe (19.0%; OR 5.1 [95%CI 1.4-17.9]; p = 0.011), moderate (11.2%; OR 2.7 [95%CI 1.1-6.8]; p = 0.031), and mild CAC (12.6%; OR 3.1 [95%CI 1.4-6.9]; p = 0.006) was higher than without. OR adjusted for age and gender was 2.7 (95%CI 1.0-7.1; p = 0.050) and 2.6 (95%CI 0.9-7.1; p = 0.069) for the overall model. PE-related mortality was 4.0% (n = 19) with higher mortality with any CAC (5.9% vs. 1.8% without; OR 3.5 [95%CI 1.1-10.7]; p = 0.028). PE-related mortality with severe CAC was 9.5% (OR 5.8 [95%CI 1.0-34.0]; p = 0.049), with moderate CAC 6.7% (OR 4.0 [95%CI 1.1-14.6]; p = 0.033), and with mild 4.9% (OR 2.9 [95%CI 0.8-9.9]; p = 0.099). OR adjusted for age and gender was 4.2 (95%CI 0.9-20.7; p = 0.074) and 3.4 (95%CI 0.7-17.4; p = 0.141) for the overall model. Patients with sub-massive PE showed similar results. CAC is frequent in acute PE patients and associated with short-term mortality. Visual assessment of CAC may serve as an easy, readily available tool for early risk stratification in those patients. • Coronary artery calcification assessed on computed tomography pulmonary angiography is frequent in patients with acute pulmonary embolism. • Coronary artery calcification assessed on computed tomography pulmonary angiography is associated with 30-day overall and PE-related mortality in patients with acute pulmonary embolism. • Coronary artery calcification assessed on computed tomography pulmonary angiography may serve as an additional, easy readily available tool for early risk stratification in those patients.

An update on pulmonary embolism-related mortality in Italy (2003-2015)

Data regarding pulmonary embolism (PE)-related mortality in Italy are scarce. We assessed PE-related mortality and its time trend in Italy by using the World Health Organization (WHO) Mortality Database. The vital registration data of Italy from the WHO Mortality Database were analyzed for the period between 2003 and 2015, and compared with time trends in Southern Europe. Death was defined as PE-related when classified with specific codes for PE or limb vein thrombosis listed as the primary cause of death. This coding was based on the International Classification of Diseases, tenth revision. Overall, 28 647 PE-related deaths (10 178 men and 18 469 women) were recorded between 2003 and 2015. The observed age-standardized annual PE-related mortality rates were 2.5 per 100 000 men and 2.8 per 100 000 women. Moreover, PE-related mortality increased with age with a seemingly exponential distribution. Joinpoint regression analysis demonstrated a statistically significant linear decrease in age-standardized PE-related mortality of -0.21 (95% confidence interval -0.27; -0.15) and -0.22 (95% confidence interval -0.28; -0.16) deaths per 100 000 population for men and women, respectively. The Italian age-adjusted mortality rates appeared lower compared to overall Southern Europe, despite a similar decreasing trend over time.

External Validation of ASPECT (Algorithm for Suspected Pulmonary Embolism Confirmation and Treatment).

Administrative health care databases are increasingly being used to study pulmonary embolism (PE), but the validity of single PE codes is variable. Using data from Quebec, Canada, we developed ASPECT (Algorithm for Suspected Pulmonary Embolism Confirmation and Treatment), combining 3 components to ascertain confirmed PE: emergency department (ED) diagnoses, imaging codes, and dispensed prescriptions or hospital diagnoses. Herein, we used unrelated administrative health care databases to externally validate ASPECT. We used ED electronic health records (ED-EHRs) to identify all residents of Calgary (Alberta, Canada) with PE codes between January and June, 2016. We applied ASPECT by identifying imaging studies in the ED-EHR, admission diagnoses in linked discharge abstract database and filled prescriptions in linked pharmacy information. Confirmed PE in ASPECT was validated against chart review in the ED-EHR. The cohort included 498 patients. Overall, 258 (51.9%) were managed as outpatients and 327 were adjudicated to have confirmed PE; the positive predictive value (PV) of single PE codes was 65.6%. With ASPECT the positive PV was 96.5% [95% confidence interval (CI), 94.4-98.5%] and positive likelihood ratio was 10.9 (95% CI, 6.8-15.1). The negative PV and negative likelihood ratio were 85.1% (95% CI, 80.0-90.2%) and 0.1 (95% CI, 0.0-0.1), respectively. Overall agreement of ASPECT with confirmed PE was 92.2%. Further, ASPECT was similarly robust in inpatients and outpatients and was more precise than any 2-component combination of ASPECT. Our findings reiterate the limitations of using single administrative codes for PE and suggest ASPECT as an acceptable tool to study PE.

Surveillance Ultrasound in the Neuro Intensive Care Unit: Time to Deep Vein Thrombosis Diagnosis.

Deep vein thrombosis (DVT) and pulmonary embolism (PE) are complications of hospitalization leading to increased morbidity and mortality. Routine surveillance ultrasound has become common practice in some intensive care units (ICU) to detect DVT early and initiate anticoagulation, preventing complications. However, initiating anticoagulants for asymptomatic DVT treatment may increase risk of hemorrhage. The objective of this study was to investigate the value of routine surveillance ultrasound in early DVT diagnosis in Neuro ICU patients. This is a retrospective review of patients diagnosed with DVT during admission to the Neuro ICU at University Hospital from January 1, 2012, through December 31, 2017. Patients were identified through International Classification of Diseases 9th and 10th Revision codes for DVT and PE, screened for inclusion criteria, and then classified as surveillance group or symptom-driven group based on intervention received. Primary outcome was time to DVT diagnosis. Secondary outcome included clinically significant hemorrhage identified by anticoagulation treatment discontinuation for suspected hemorrhage or new or expanding hemorrhage on head computerized tomography (CT). A total of 116 patients were identified, with 50 included: 27 were classified as surveillance and 23 as symptom-driven. Seven patients (surveillance = 3 and symptom-driven = 4) were diagnosed with only PE and were excluded from primary outcome. Median time to DVT diagnosis was similar at 148h for surveillance versus 172h for symptom driven (p = 0.2). There was no difference in treatment discontinuation rates (surveillance 21% vs symptom 31%; p = 0.4). Of the 27 patients with follow-up head CT, two in the surveillance group and two in the symptom-driven group showed a new or expanding hemorrhage. Routine surveillance ultrasound did not lead to significantly earlier DVT diagnosis. Hemorrhagic events were not different between groups. Utility of surveillance ultrasound in this population should be evaluated in large, prospective trials before routine use can be recommended.

Validity of Using Inpatient and Outpatient Administrative Codes to Identify Acute Venous Thromboembolism: The CVRN VTE Study.

Administrative data are frequently used to identify venous thromboembolism (VTE) for research and quality reporting. However, the validity of these codes, particularly in outpatients, has not been well-established. To determine how well International Classification of Diseases, Ninth Revision (ICD-9) codes for VTE predict chart-confirmed acute VTE in inpatient and outpatients. We selected 4642 adults with an incident ICD-9 diagnosis of VTE between years 2004 and 2010 from the Cardiovascular Research Network Venous Thromboembolism cohort study. Medical charts were reviewed to determine validity of events. Positive predictive values (PPVs) of ICD-9 codes were calculated as the number of chart-validated VTE events divided by the number with specific VTE codes. Analyses were stratified by VTE type [pulmonary embolism (PE), deep venous thrombosis (DVT)], code position (primary, secondary), and setting [hospital/emergency department (ED), outpatient]. The PPV for any diagnosis of VTE was 64.6% for hospital/ED patients and 30.9% for outpatients. Primary diagnosis codes from hospital/ED patients were more likely to represent acute VTE than secondary diagnosis codes (78.9% vs. 44.4%, P<0.001). Primary hospital/ED codes for PE and lower extremity DVT had higher PPV than for upper extremity DVT (89.1%, 74.9%, and 58.1%, respectively). Outpatient codes were poorly predictive of acute VTE: 28.0% for PE and 53.6% for lower extremity DVT. ICD-9 codes for VTE obtained from outpatient encounters or from secondary diagnosis codes do not reliably reflect acute VTE. More accurate ways of identifying VTE in outpatients are needed before these codes can be adopted for research or policy purposes.

External validation of a claims-based and clinical approach for predicting post-pulmonary embolism outcomes among United States veterans.

The In-hospital Mortality for PulmonAry embolism using Claims daTa (IMPACT) rule can accurately identify pulmonary embolism (PE) patients at low risk of early complications using claims data. We sought to externally validate the IMPACT and simplified Pulmonary Embolism Severity Index (sPESI) tools for predicting all-cause mortality and readmission. We used Veteran Health Administration data (10/1/2010-9/30/2015) to identify adults with ≥1 inpatient diagnosis code for acute PE, ≥12months continuous medical and pharmacy benefits prior to the index PE, ≥90days of post-event follow-up (unless death occurred) and ≥1 claim for an anticoagulant during the index PE stay. Prognostic accuracies of IMPACT and sPESI for 30- and 90-day all-cause mortality and 90-day readmission were estimated. Of 6,746 PE patients, 7.5 and 12.6% died at 30 and 90days. Within 90days, 20.1% were readmitted for any reason. Hospitalization for recurrent VTE and major bleeding occurred in 5.6 and 1.7% of patients. IMPACT classified 15.2% as low risk, while 28.4% were low risk per sPESI. Both tools displayed sensitivity >90% and negative predictive values (NPVs) >97% for 30-day mortality, but low specificity (range 16.2-30.0) and positive predictive values (PPVs) (range 8.7-9.5); with similar results observed for 90-day mortality. IMPACT's sensitivity for all-cause readmission was numerically higher than sPESI (88.2 vs. 79.0%), but both had comparable NPVs (85.1 vs. 84.2%). Similar trends were observed for VTE or major bleeding readmissions. IMPACT classified patients for post-PE outcomes with similar accuracy as sPESI. IMPACT appears useful for identifying PE patients at low risk for early mortality or readmission in claims-based studies.

Shortened hospital length of stay and lower costs associated with rivaroxaban in patients with pulmonary embolism managed as observation status.

Unlike rivaroxaban, treatment of patients with pulmonary embolism (PE) with warfarin requires parenteral bridging and coagulation monitoring that may prolong length-of-stay (LOS) and increase hospital costs. The aim of this study was to compare LOS, hospital costs and readmissions in PE patients managed through observation stays treated with rivaroxaban or parenterally bridged warfarin. Premier Hospital claims data from November 2012 to March 2015 were used to identify patients with a primary diagnosis code for PE managed through an observation stay and with ≥1 claim for a PE-related diagnostic test on day 0-2. Rivaroxaban users, allowing ≤2days of prior parenteral therapy, were 1:1 propensity-score matched to patients receiving parenterally bridged warfarin. LOS, the proportion of encounters lasting >2 midnights, total hospital costs of the index visit and risk of readmission for venous thromboembolism (VTE) or major bleeding during the same month or 2months subsequent to the index event were compared between matched cohorts using multivariable regression. A total of 312 rivaroxaban users were matched to 312 patients receiving parenterally bridged warfarin. Rivaroxaban was associated with an average of 0.27-day shorter LOS, a 52% decreased odds of an encounter lasting >2 midnights and a $403 mean reduction in costs vs parenterally bridged warfarin (P≤.002 for all). The readmission rate for VTE during the same or subsequent 2months following the index PE was similar between cohorts (P=.75). No patient in either cohort was readmitted for major bleeding. Rivaroxaban was associated with shortened LOS and lowered cost vs parenterally bridged warfarin in PE observation stay patients, without increases in the short-term rate of complications or readmission.

Reduced Costs and Length-of-Stay Associated with Rivaroxaban As Compared to Parenteral Bridging and Warfarin in Pulmonary Embolism Patients Managed in Observation Status

▪Background: Observation stays are intended to assess patients for short periods (i.e., <2-midnights) to determine need for inpatient admission or discharge. Unlike rivaroxaban, treatment of pulmonary embolism (PE) with parenteral bridging to warfarin requires frequent coagulation monitoring and dosing adjustments that may prolong length-of-stay (LOS) and increase hospital treatment costs.Objective: To compare LOS, hospital treatment costs, and readmissions in PE patients managed through observation stays treated with either rivaroxaban or parenterally-bridged warfarin.Methods: US Premier Hospital claims data spanning 11/2012-9/2015 was used to identify patients with a primary diagnosis code for PE (415.1x) managed through an observation stay and having ³1 claim for a PE-related diagnostic test on day 0-2. Rivaroxaban users, allowing ²2 days of prior parenteral therapy, were 1:1 propensity-score matched to patients parenterally bridged to warfarin. LOS, the proportion of encounters lasting >2-midnights, total hospital costs and risk of readmission for venous thromboembolism (VTE) or major bleeding during the same month or 2 months subsequent to the index event were compared between propensity-score-matched treatment cohorts.Results: A total of 401 rivaroxaban patients were matched to 401 patients receiving parenteral bridging to warfarin.Rivaroxaban use was associated with a shorter LOS (-0.25 days), fewer encounters lasting >2 midnights (21.1% vs. 32.7%) and lower total hospital costs (-$240) vs. parenteral bridging to warfarin (p²0.03 for all) (Table). Readmission was similar between cohorts (p>0.99 for both VTE and major bleeding readmission).Conclusion: Shorter LOS and lower hospital costs occurred with rivaroxaban vs. parenteral bridging to warfarin in PE observation stay patients. This was achieved without increasing short-term risk of VTE or major bleeding readmission. [Display omitted] DisclosuresPeacock:Portola: Consultancy, Research Funding; Janssen Pharmaceuticals: Consultancy, Research Funding. Fermann:Janssen Pharmaceuticals: Other: Advisory Board, Speakers Bureau; Pfizer: Research Funding. Baugh:Janssen Pharmaceuticals: Consultancy; Roche Diagnostics: Other: Advisory Board. Wells:Janssen Pharmaceuticals: Consultancy; Bristol Myers Squib: Research Funding; Pfizer: Research Funding; Bayer Healthcare: Other: Advisory Board, Speakers Bureau; Itreas: Other: Writing Committee. Ashton:Janssen Scientific Affairs, LLC, Raritan, New Jersey: Employment. Crivera:Janssen Scientific Affairs, LLC, Raritan, New Jersey: Employment, Equity Ownership. Wildgoose:Janssen Scientific Affairs, LLC, Raritan, New Jersey: Employment, Equity Ownership. Schein:Johnson & Johnson: Employment, Equity Ownership, Other: Own in excess of $10,000 of J&J stock. Coleman:Boehringer-Ingelheim Pharmaceuticals, inc.: Consultancy, Research Funding; Bayer Pharmaceuticals AG: Consultancy, Research Funding; Janssen Pharmaceuticals: Consultancy, Research Funding.

LO016: Can we use administrative data to define an emergency department population at risk for pulmonary embolism? Development and validation of an algorithm to identify a research population

Introduction: Pulmonary embolism (PE) is a potentially life-threatening condition that is in the differential diagnosis of many emergency department (ED) presentations. However, no diagnostic code for suspected PE exists. Thus, identifying the population of patients undergoing PE workup from administrative data for use as a denominator in clinical research and quality improvement can be difficult. To overcome this, we used standardized triage complaint codes and investigations to develop search algorithms useful to identify patients undergoing PE workup from an administrative dataset. Our objective was to quantify the sensitivity, specificity, and case yield of these search algorithms in order to identify a superior search strategy. Methods: Hospital administrative data for adult patients (age ≥18 years), which included standardized triage complaint codes and ICD-10 diagnostic codes for PE, were obtained from four urban EDs between July 2013 to January 2015. Standardized triage complaint codes were evaluated for the proportion of patients diagnosed with PE. Combinations of high-yield presenting complaints, in combination with D-dimer testing or imaging orders, were evaluated for sensitivity, specificity, and predictive values for PE. Results: Of 479,937 patients presenting with 174 different complaints, 1,048 were diagnosed with PE. The best-performing search strategy was the combination of standardized CEDIS complaints of Cardiac Pain, Chest Pain (Cardiac Features), Chest Pain (Non-Cardiac Features), Shortness of Breath, Syncope/Pre-syncope, Hemoptysis, and Unilateral Swollen Limb/Pain, along with with D-dimer testing and/or CTPA, or V/Q scan. This combination captured 808 PE diagnoses for a sensitivity of 77.1% (95%CI 74.4-79.5%) and specificity of 86.8% (95%CI 86.7-86.6%). Conclusion: We identified a high-yield combination of presenting complaints and test ordering that can be used to define an ED population with suspected PE. This population of patients can be used as a denominator in research or quality improvement work that evaluates the utilization of diagnostic testing for PE.

LO013: Can you trust administrative data? Accuracy of ICD-10 codes for diagnosis of pulmonary embolism

Introduction: Administrative data is a useful tool for research and quality improvement; however, the validity of research findings based on these data depends on their reliability. Diagnoses are recorded using diagnostic codes, as defined by the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10). Several groups have reported coding errors associated with ICD-10 assignments to patient diagnoses; these errors have serious implications for research, quality improvement, and policymaking. As part of a quality improvement project targeting emergency department (ED) diagnostic appropriateness for pulmonary embolism (PE), we sought to validate the accuracy of ICD-10 codes for studying ED patients diagnosed with PE. Methods: Hospital administrative data for adult patients (age ≥18 years) with an ICD-10 code for PE (I26.0 and I26.9) were obtained from the records of four urban EDs between July 2013 to January 2015. A review of medical records and imaging reports was used to confirm the diagnosis of PE. In the case of discrepancy between ICD-10 coding and chart review, the diagnosis obtained from chart review was considered correct. The physicians’ discharge notes in the administrative database were also searched using ‘pulmonary embolism’ and ‘PE’, and patients who were diagnosed with PE but not coded as PE were identified. Coding discrepancies were quantified and described. Results: 1,453 ED patients had a PE ICD-10 code during our study period. 257 (17.7%) of these patients’ diagnoses were improperly coded. 211 patients assigned an ICD-10 PE code had ED discharge diagnoses of ‘rule-out PE’ or ‘query PE’. 64 other patients were miscoded as having a PE and should have been assigned an alternate code, such as chest pain, hypoxia, or dyspnea. The physician did not include a discharge diagnosis in 4 of the 64 miscoded patients; however, triage and physician assessment notes indicated no suspicion of PE. Furthermore, 117 patients who had an ED discharge diagnosis of PE were not assigned a PE code, meaning that 8.91% of true PEs were missed by using ICD-10 codes alone. Thus, 1,313 ED patients truly had a PE. Conclusion: Our work suggests the need for more accuracy in ICD-10 coding of ED diagnoses of PE. Caution should be exercised when using administrative data for studying PE, and validation of the accuracy of ICD-10 coding prior to research use is recommended.

Identifying venous thromboembolism and major bleeding in emergency room discharges using administrative data

BackgroundAdministrative data can be used to identify venous thromboembolism (VTE) and major bleeding (MB) events. However, the validity of this data in emergency room discharge records in Canada is unknown. MethodsWe conducted a single-institution retrospective chart re-abstraction study in London, Canada. We identified all adult patients with a VTE or MB code included in the mandatory Canadian Institute for Health Information National Ambulatory Care Reporting System seen at our institution between July 2002 and March 2014. VTE was defined using the International Classification of Diseases, 10th revision (ICD-10CM) codes for deep venous thrombosis (DVT), and pulmonary embolism (PE) whereas MB was defined using codes for intracerebral hemorrhage, subarachnoid hemorrhage, subdural hemorrhage, upper, and lower gastrointestinal bleeding. A random sample of 50 patients was obtained for each condition. Two abstractors independently conducted blinded diagnostic adjudication using standard criteria. Agreement was calculated using kappa statistics. Positive predictive values were calculated for VTE, MB and each diagnosis. ResultsOverall, ICD-10CM codes demonstrated very good ability to identify major bleeding events (PPV 88%). Diagnostic codes performed particularly well for all intracranial and lower gastrointestinal bleeds. In contrast, ICD-10CM codes for VTE had moderate ability (PPV 49%). Diagnostic codes for PE performed better than those for DVT. ConclusionSingle ICD-10CM codes for venous thromboembolism have moderate predictive value for identifying DVT and PE in emergency room discharges. In contrast, codes for MB events have very good ability and it would be adequate to use them for research purposes.

National Trends in Pulmonary Embolism Hospitalization Rates and Outcomes for Adults Aged ≥65 Years in the United States (1999 to 2010)

Little is known about national trends of pulmonary embolism (PE) hospitalizations and outcomes in older adults in the context of recent diagnostic and therapeutic advances. Therefore, we conducted a retrospective cohort study of 100% Medicare fee-for-service beneficiaries hospitalized from 1999 to 2010 with a principal discharge diagnosis code for PE. The adjusted PE hospitalization rate increased from 129/100,000 person-years in 1999 to 302/100,000 person-years in 2010, a relative increase of 134% (p <0.001). Black patients had the highest rate of increase (174 to 548/100,000 person-years) among all age, gender, and race categories. The mean (standard deviation) length of hospital stay decreased from 7.6 (5.7) days in 1999 to 5.8 (4.4) days in 2010, and the proportion of patients discharged to home decreased from 51.1% (95% confidence interval [CI] 50.5 to 51.6) to 44.1% (95% CI 43.7 to 44.6), whereas more patients were discharged with home health care and to skilled nursing facilities. The in-hospital mortality rate decreased from 8.3% (95% CI 8.0 to 8.6) in 1999 to 4.4% (95% CI 4.2 to 4.5) in 2010, as did adjusted 30-day (from 12.3% [95% CI 11.9 to 12.6] to 9.1% [95% CI 8.5 to 9.7]) and 6-month mortality rates (from 23.0% [95% CI 22.5 to 23.4] to 19.6% [95% CI 18.8 to 20.5]). There were no significant racial differences in mortality rates by 2010. There was no change in the adjusted 30-day all-cause readmission rate from 1999 to 2010. In conclusion, PE hospitalization rates increased substantially from 1999 to 2010, with a higher rate for black patients. All mortality rates decreased but remained high. The increase in hospitalization rates and continued high mortality and readmission rates confirm the significant burden of PE for older adults.