Principal Discharge Diagnosis Research Articles

Trends in COVID-19-Attributable Hospitalizations Among Adults With Laboratory-Confirmed SARS-CoV-2-COVID-NET, June 2020 to September 2023.

Screening for SARS-CoV-2 infection among hospital admissions made interpretation of COVID-19 hospitalization data challenging as SARS-CoV-2-positive persons with mild or asymptomatic infection may be incorrectly identified as COVID-19-associated hospitalizations. The study objective is to estimate the proportion of hospitalizations likely attributable to COVID-19 among SARS-CoV-2-positive hospitalized patients. A sample of laboratory-confirmed SARS-CoV-2-positive hospitalizations from the COVID-19-Associated Hospitalization Surveillance Network (COVID-NET) from June 2020 to September 2023 was analyzed, with a focus on July 2022 to September 2023. Likely COVID-19-attributable hospitalizations were defined as hospitalizations among SARS-CoV-2-positive non-pregnant adults ages ≥ 18 years with COVID-19-related presenting complaint, treatment, or discharge diagnosis. Among 44,816 sampled hospitalizations, 90% met the definition of likely COVID-19-attributable. Among the 9866 admissions occurring during July 2022 to September 2023, 86% were likely COVID-19-attributable; 87% had a COVID-19-related presenting complaint, 64% received steroids or COVID-19-related treatment, 47% had respiratory- and 10% had coagulopathy-related discharge diagnoses, and 39% had COVID-19 as the principal discharge diagnosis code. More than 70% met ≥ 2 criteria. Compared with likely COVID-19-attributable hospitalizations, SARS-CoV-2-positive patients who did not meet the case definition were more likely to be ages 18-49 years (27% vs. 13%), have no underlying medical conditions (14% vs. 4%), or be asymptomatic for COVID-19 upon admission (46% vs. 10%) (all p < 0.05). Most hospitalizations among SARS-CoV-2-positive adults in a recent period were likely attributable to COVID-19. COVID-19-attributable hospitalizations are less common among younger SARS-CoV-2-positive hospitalized adults but still account for nearly three quarters of all admissions among SARS-CoV-2-positive adults in this age group.

Ambient Air Pollution and Hospitalizations for Schizophrenia in China

Schizophrenia episodes may be triggered by short-term environmental stimuli. Short-term increases in ambient air pollution levels may elevate the risk of schizophrenia episodes, yet few epidemiologic studies have examined this association. To investigate whether short-term increases in air pollution levels are associated with an additional risk of schizophrenia episodes, independent of absolute air pollution concentrations, and whether sustained increases in air pollution levels for several days are associated with more pronounced risks of schizophrenia episodes. This nationwide, population-based, time-stratified case-crossover study was performed based on hospitalization records for schizophrenia across 295 administrative divisions of prefecture-level or above cities in China. Records were extracted from 2 major health insurance systems from January 1, 2013, to December 31, 2017. Thirty-six cities with a small number of schizophrenia hospitalizations (n < 50) were excluded. Data analysis for this study was performed from January to March 2024. Daily absolute concentrations of fine particulate matter (PM2.5), inhalable particulate matter (PM10), nitrogen dioxide, sulfur dioxide, ozone, and carbon monoxide were collected. Air pollution increases between neighboring days (APINs) were generated as the differences in absolute air pollution concentrations on the current day minus that on the previous day. Sustained increases (APIN ≥5 μg/m3 for PM2.5 and PM10, APIN ≥1 μg/m3 for nitrogen dioxide and sulfur dioxide, and APIN ≥0.05 mg/m3 for carbon monoxide) lasting for 1 or more to 4 or more days were defined for different air pollutants. Patients with schizophrenia episodes were identified by principal discharge diagnoses of schizophrenia. A conditional logistic regression model was used to capture the associations of absolute concentrations, APINs, and sustained increase events for different air pollutants with risks of schizophrenia hospitalizations. The study included 817 296 hospitalization records for schizophrenia across 259 Chinese cities (30.6% aged 0-39 years, 56.4% aged 40-64 years, and 13.0% aged ≥65 years; 55.04% male). After adjusting for the absolute concentrations of respective air pollutants, per-IQR increases in 6-day moving average (lag0-5) APINs of PM2.5, PM10, nitrogen dioxide, sulfur dioxide, and carbon monoxide were associated with increases of 2.37% (95% CI, 0.88%-3.88%), 2.95% (95% CI, 1.46%-4.47%), 4.61% (95% CI, 2.93%-6.32%), 2.16% (95% CI, 0.59%-3.76%), and 2.02% (95% CI, 0.39%-3.68%) in schizophrenia hospitalizations, respectively. Greater risks of schizophrenia hospitalizations were associated with sustained increases in air pollutants lasting for longer durations up to 4 or more days. This case-crossover study of the association between ambient air pollution increases and schizophrenia hospitalizations provides novel evidence that short-term increases in ambient air pollution levels were positively associated with an elevated risk of schizophrenia episodes. Future schizophrenia prevention practices should pay additional attention to APINs, especially sustained increases in air pollution levels for longer durations, besides the absolute air pollution concentrations.

Abstract B013: Sociodemographic differences and inpatient outcomes of opioid use disorder among patients hospitalized for head and neck cancer in the United States

Abstract Background: Opioid use disorder (OUD) is a rising public health concern in the United States. It poses substantial challenges in the management of head and neck cancers (HNC), given the pain associated with the disease. This study examines the sociodemographic differences in OUD among patients hospitalized for HNC in the United States as well as in-patient outcomes of HNC patients with OUD. Methods: We analyzed weighted retrospective data from the 2016 to 2020 National Inpatient Sample database. We included patients 18 years or older with principal discharge diagnoses of HNC with or without secondary diagnoses of OUD using specific ICD-10 codes. OUD was defined as opioid abuse, dependence, poisoning or adverse effects. Baseline sociodemographic and hospital characteristics of HNC patients with or without OUD were compared using χ2 for categorical variables and Student’s t-test for continuous variables. We examined sociodemographic characteristics associated with OUD using multivariable logistic regression models. We assessed three in-patient outcomes including mortality, length of stay and total hospitalization charge (THC) adjusted for inflation. Further we included interaction terms to explore the combined effects of relevant sociodemographic predictors with depression. Results: There were 117,085 hospitalizations for HNC between 2016 and 2020. Among HNC hospitalizations, 3.2% were opioid users and 2.36% had a secondary diagnosis of OUD. The mean age of HNC pts was 63.3 years (SD 0.08). 78.3% were Non-Hispanic White, 11.8% Non-Hispanic Black, 6.7% Hispanic and 3.2% Asians/Pacific Islanders. HNC patients with OUD had higher odds of being female (aOR 1.32; 95% CI 1.09, 1.59) and having more than one comorbidity (aOR 1.088; 95% CI 1.047, 1.132). There were lower odds of OUD with older age &amp;gt;70 years (aOR: 0.25; 95% CI 0.10, 0.59), private insurance (aOR 0.66; 95% CI 0.50, 0.87), higher income quartiles (aOR 0.66; 95% CI 0.50, 0.87) and with hospital regions in the South compared to the North-East United States (aOR 0.57; 95% CI 0.43, 0.74). There was no statistically significant interaction between MDD and gender (aOR 1.27; 95% CI 0.80, 2.04) among patients with HNC and OUD. OUD was not significantly associated with increased mortality in patients hospitalized for HNC (aOR 1.39, 95% CI 0.83, 2.33), however OUD was associated with a greater LOS by 3.6 days (95% CI 2.52, 4.60) and an increased THC of $26,947 (95% CI 13232, 40662). Conclusion: Our study demonstrates significant sociodemographic disparities in the prevalence of OUD among head and neck cancer hospitalizations in the US. Female patients and those with multiple comorbidities had higher odds of OUD. Additionally, OUD was significantly associated with an increased LOS and THC. These findings highlight the need for targeted interventions that address the unique needs of vulnerable populations with OUD and HNC. Citation Format: Ted O Akhiwu, Inemesit Akpan, Michael Imeh, Nosayaba Osazuwa-Peters. Sociodemographic differences and inpatient outcomes of opioid use disorder among patients hospitalized for head and neck cancer in the United States [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr B013.

Rule-based natural language processing to examine variation in worsening heart failure hospitalizations by age, sex, race and ethnicity, and left ventricular ejection fraction

BackgroundPrior studies characterizing worsening heart failure events (WHFE) have been limited in using structured healthcare data from hospitalizations, and with little exploration of sociodemographic variation. The current study examined the impact of incorporating unstructured data to identify WHFE, describing age-, sex-, race and ethnicity-, and left ventricular ejection fraction (LVEF)-specific rates. MethodsAdult members of Kaiser Permanente Southern California (KPSC) with a HF diagnosis between 2014-2018 were followed through 2019 to identify hospitalized WHFE. The main outcome was hospitalizations with a principal or secondary HF discharge diagnosis meeting rule-based Natural Language Processing (NLP) criteria for WHFE. In comparison, we examined hospitalizations with a principal discharge diagnosis of HF. Age-, sex-, and race and ethnicity-adjusted rates per 100 person-years (PY) were calculated among age, sex, race and ethnicity (non-Hispanic (NH) Asian/Pacific Islander [API], Hispanic, NH Black, NH White) and LVEF subgroups. ResultsAmong 44,863 adults with HF, 10,560 (23.5%) had an NLP-defined, hospitalized WHFE. Adjusted rates (per 100 PY) of WHFE using NLP were higher compared to rates based only on HF principal discharge diagnosis codes (12.7 and 9.3, respectively), and this followed similar patterns among subgroups, with the highest rates among adults ≥75 years (16.3 and 11.2), men (13.2 and 9.7), and NH Black (16.9 and 14.3) and Hispanic adults (15.3 and 11.4), and adults with reduced LVEF (16.2 and 14.0). Using NLP disproportionately increased the perceived burden of WHFE among API and adults with mid-range and preserved LVEF. ConclusionRule-based NLP improved the capture of hospitalized WHFE above principal discharge diagnosis codes alone. Applying standardized consensus definitions to EHR data may improve understanding of the burden of WHFE and promote optimal care overall and in specific sociodemographic groups.

Race and Ethnicity and Diagnostic Testing for Common Conditions in the Acute Care Setting

Overuse of diagnostic testing is pervasive, but the extent to which it varies by race and ethnicity in the acute care setting is poorly understood. To use a previously validated diagnostic intensity index to evaluate differences in diagnostic testing rates by race and ethnicity in the acute care setting, which may serve as a surrogate for diagnostic test overuse. This was a cross-sectional study of emergency department (ED) discharges, hospital observation stays, and hospital admissions using administrative claims among EDs and acute care hospitals in Kentucky, Maryland, North Carolina, and New Jersey, from 2016 through 2018. The diagnostic intensity index pairs nonspecific principal discharge diagnoses (nausea and vomiting, abdominal pain, chest pain, and syncope) with related diagnostic tests to estimate rates of nondiagnostic testing. Adults with an acute care encounter with a principal discharge diagnosis of interest were included. Data were analyzed from January to February 2024. Race and ethnicity (Asian, Black, Hispanic, White, other [including American Indian, multiracial, and multiethnic], and missing). Receipt of a diagnostic test. Generalized linear models with a hospital-specific indicator variable were estimated to calculate the adjusted odds ratio of receiving a test related to the principal discharge diagnosis by race and ethnicity, controlling for primary payer and zip code income quartile. Of 3 683 055 encounters (1 055 575 encounters [28.7%] for Black, 300 333 encounters [8.2%] for Hispanic, and 2 140 335 encounters [58.1%] for White patients; mean [SD] age of patients with encounters, 47.3 [18.8] years; 2 233 024 encounters among females [60.6%]), most (2 969 974 encounters [80.6%]) were ED discharges. Black compared with White patients discharged from the ED with a diagnosis of interest had an adjusted odds ratio of 0.74 (95% CI, 0.72-0.75) of having related diagnostic testing. No other racial or ethnic disparities of a similar magnitude were observed in any acute care settings. In this study, White patients discharged from the ED with a nonspecific diagnosis of interest were significantly more likely than Black patients to receive related diagnostic testing. The extent to which this represents diagnostic test overuse in White patients vs undertesting and missed diagnoses in Black patients deserves further study.

Abstract Mo027: Impact of Ventricular Arrhythmias On Outcomes of Patients with Follicular Lymphoma

Background: Follicular lymphoma (FL) is typically a slow-growing form of non-Hodgkin lymphoma (NHL) that arises from B-lymphocytes. Large-scale studies are lacking for the outcomes of patients with ventricular arrhythmias and FL. We retrospectively reviewed FL patients with ventricular tachycardia utilizing the National Inpatient Sample (NIS) database. Our analysis aimed to identify risk factors that worsen outcomes in patients with FL and ventricular arrhythmias and evaluate the impact of ventricular arrhythmias on in-hospital outcomes of patients with FL. Methods: Data were extracted from the National Inpatient Sample (NIS) 2016-2019 Database. The NIS was searched for hospitalizations for adult patients with FL as the principal discharge diagnosis with and without ventricular arrhythmias as a secondary diagnosis using ICD-10 codes. The primary outcome was inpatient mortality. Multivariate logistic and linear regression analysis was used accordingly to adjust for confounders. STATA software was used for analysis. Results: This study included 15,795 patients with FL of which 225 had ventricular arrhythmias. Patients with both FL and an associated ventricular arrhythmia had higher inpatient mortality (OR 5.44, CI 2.48-11.93, p&lt;0.0001), hospital length of stay (LOS) (+5.48 days, CI 3.27-7.68, p&lt;0.0001) and total hospital charges (TOTHCG) (+$109,117.3, CI $65,702.9- $152,531.7, p&lt;0.0001). In this patient population, the secondary analysis revealed that mortality was increased if patients had a cardiac arrest (OR 25.62, CI 1.36-484.48, p&lt;0.031). LOS was increased in the presence of an acute kidney injury (+0.34 days, 0.04-0.64, p&lt;0.026). Conclusions: Our study revealed that patients with FL and ventricular arrhythmias had increased mortality, hospital LOS, and TOTHCG, as well as much higher rates of complications, including cardiac arrest and acute kidney injury, compared to those without ventricular arrhythmias. Efforts to prevent, monitor, and aggressively manage ventricular arrhythmias in patients with FL will improve overall outcomes.

Artificial intelligence approaches for phenotyping heart failure in U.S. Veterans Health Administration electronic health record.

Heart failure (HF) is a clinical syndrome with no definitive diagnostic tests. HF registries are often based on manual reviews of medical records of hospitalized HF patients identified using International Classification of Diseases (ICD) codes. However, most HF patients are not hospitalized, and manual review of big electronic health record (EHR) data is not practical. The US Department of Veterans Affairs (VA) has the largest integrated healthcare system in the nation, and an estimated 1.5 million patients have ICD codes for HF (HF ICD-code universe) in their VA EHR. The objective of our study was to develop artificial intelligence (AI) models to phenotype HF in these patients. The model development cohort (n=20000: training, 16000; validation 2000; testing, 2000) included 10000 patients with HF and 10000 without HF who were matched by age, sex, race, inpatient/outpatient status, hospital, and encounter date (within 60days). HF status was ascertained by manual chart reviews in VA's External Peer Review Program for HF (EPRP-HF) and non-HF status was ascertained by the absence of ICD codes for HF in VA EHR. Two clinicians annotated 1000 random snippets with HF-related keywords and labelled 436 as HF, which was then used to train and test a natural language processing (NLP) model to classify HF (positive predictive value or PPV, 0.81; sensitivity, 0.77). A machine learning (ML) model using linear support vector machine architecture was trained and tested to classify HF using EPRP-HF as cases (PPV, 0.86; sensitivity, 0.86). From the 'HF ICD-code universe', we randomly selected 200 patients (gold standard cohort) and two clinicians manually adjudicated HF (gold standard HF) in 145 of those patients by chart reviews. We calculated NLP, ML, and NLP+ML scores and used weighted F scores to derive their optimal threshold values for HF classification, which resulted in PPVs of 0.83, 0.77, and 0.85 and sensitivities of 0.86, 0.88, and 0.83, respectively. HF patients classified by the NLP+ML model were characteristically and prognostically similar to those with gold standard HF. All three models performed better than ICD code approaches: one principal hospital discharge diagnosis code for HF (PPV, 0.97; sensitivity, 0.21) or two primary outpatient encounter diagnosis codes for HF (PPV, 0.88; sensitivity, 0.54). These findings suggest that NLP and ML models are efficient AI tools to phenotype HF in big EHR data to create contemporary HF registries for clinical studies of effectiveness, quality improvement, and hypothesis generation.

Comorbid opioid use disorder and resource utilization outcomes in patients with neoplasm-related pain: A population analysis.

11040 Background: Neoplasm-related pain (NRP) is one of the most debilitating sequalae of cancer, affecting patients’ (pts) quality of life. Its intersection with opioid use disorder (OUD) adds a layer of complexity to patient care in the United States (US). As the prevalence of both conditions continues to increase, understanding the association between OUD and NRP becomes important. This study aimed to shed light on this intricate relationship by examining outcomes in resource utilization among OUD pts with NRP. Methods: We analyzed weighted data from the 2016-2020 iterations of the National Inpatient Sample database. We included pts 18 years or older with a principal discharge diagnosis of NRP with or without a secondary diagnosis of OUD. The main outcomes for the study were length of stay (LOS) and total hospitalization charges (THC) adjusted for inflation using the Medicare Hospital Consumer Price Index. Baseline sociodemographic and hospital characteristics as well as comorbidities of patients with and without OUD were compared using χ2 for categorical variables and Student’s t-test for continuous variables. Multivariable Linear regression was used to assess the relationship between comorbid OUD and resource utilization among NRP hospitalizations. Statistical significance was set at p&lt;0.05. Results: There were 120,359 hospitalizations for NRP between 2016 and 2020. Among patients hospitalized for NRP, 3335 (2.78%) had a secondary diagnosis of OUD. The mean age of NRP pts was 60 years. Compared to patients without OUD, OUD pts were male (50.67% vs 45.47%; p = 0.0086), younger (mean age 56 vs 60 years; p&lt;0.0001), more likely to be on Medicaid (37.48% vs 21.00%; p &lt;0.0001) and hospitalized in urban hospitals (94.6% vs 91.37%; p = 0.0040). On multivariable linear regression, OUD was associated with an increased LOS by 0.93 days compared to non-OUD patients (95% CI: 0.265 – 1.600; p = 0.006), as well as increased THC by $9248 (95% CI: 1676.00 – 16820.41; p = 0.017) after adjusting for sociodemographic and hospital covariates. Conclusions: Our study highlights an increased resource utilization among pts with NRP and comorbid OUD. We surmise this could be due to increased tolerance to and dependence on the effects of opioids requiring more intricate multidisciplinary care in managing their pain. Recognizing this association is important for medical oncologists and other health care providers when managing NRP pts with OUD. Incorporating multimodality pain management early on admission could improve patient outcomes while optimizing healthcare resources.

Abstract P247: Rural-Urban Disparities in In-Hospital Stroke Mortality in North Carolina

Introduction/Hypothesis: Stroke remains a significant public health concern. North Carolina (NC) has the 11 th highest stroke mortality in the United States and an increasingly diverse population with respect to rurality and race-ethnicity. We assessed the hypothesis that rural stroke patients have higher in-hospital mortality than urban stroke patients in NC. We also explored the interaction between rurality and race-ethnicity on stroke mortality. Methods: We conducted a retrospective study of stroke hospitalizations at 115 NC hospitals from January 2019-September 2020. Data were obtained from an inpatient database of patient demographics, county of residence, insurance, admission date, and discharge disposition and diagnoses. Eligible patients included adults with principal discharge diagnosis codes for stroke (ICD-10-CM I60.xx-I63.xx) and were classified by rurality defined by the NC Office of Rural Health. Multivariable logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) comparing in-hospital mortality in rural versus urban patients, overall and stratified by race-ethnicity subgroups. Results: Among 32,808 patients, 37% resided in rural counties. Patients were mean 69 years old, 50% female, 30% Black, and 3% Hispanic. Unadjusted in-hospital mortality was lower in rural versus urban patients (4.2% and 5.0%, respectively). Adjusted model found a weak, non-significant association between rurality and stroke mortality (OR 1.06, 95% CI 0.95-1.19; Table 1). In race-ethnicity subgroups, rural-urban ORs were not statistically significant although the point estimates suggest modestly greater differences among racial/ethnic minorities compared to White patients. Conclusions: We found minimal, adjusted rural-urban differences in in-hospital stroke mortality in NC statewide and within race-ethnicity subgroups. Further research is needed to investigate potential disparities in long-term stroke mortality and morbidity and their underlying determinants.

Mortality in patients admitted to hospital with heart failure in China: a nationwide Cardiovascular Association Database-Heart Failure Centre Registry cohort study

China has the largest burden of heart failure worldwide. However, large-scale studies on heart failure mortality are scarce. We aimed to investigate mortality and identify risk factors for mortality among patients with heart failure in China. This prospective cohort study used data from the China Cardiovascular Association (CCA) Database-Heart Failure Centre Registry, which were linked to the National Mortality Registration Information Management System by the Chinese Centre for Disease Control and Prevention. We included patients enrolled from Jan 1, 2017, to Dec 31, 2021, across 572 CCA Database-Heart Failure Centre certified hospitals in 31 provinces of mainland China. Eligible patients were aged 18 years or older (younger than 100 years) with a principal discharge diagnosis of heart failure based on Chinese heart failure guidelines. All-cause mortality at 30 days, 1 year, and 3 years for patients with heart failure were calculated and the causes of death were recorded. Multivariable analysis was used to analyse factors associated with all-cause mortality and cardiovascular mortality. This study was registered with the Chinese Clinical Trial Registry, ChiCTR2200066305. Of the 327 477 patients in the registry, 230 637 eligible adults with heart failure were included in our analyses. Participant mean age was 69·3 years (SD 13·2), 94 693 (41·1%) participants were female, and 135 944 (58·9%) were male. The median follow-up time was 531 days (IQR 251-883). Post-discharge all-cause mortality of patients with heart failure at 30 days was 2·4% (95% CI 2·3-2·5), at 1 year was 13·7% (13·5-13·9), and at 3 years was 28·2% (27·7-28·6). Cardiovascular death accounted for 32 906 (71·5%) of 46 006 all-cause deaths. Patients with heart failure with reduced ejection fraction had the highest all-cause mortality. A lower guideline adherence score was independently associated with the increase of all-cause and cardiovascular mortality. In China, mortality for patients with heart failure is still high, especially in patients with reduced ejection fraction. Our findings suggest that guideline-directed medical therapy needs to be improved. National High Level Hospital Clinical Research Funding, the Capital's Funds for Health Improvement and Research, and the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences. For the Chinese translation of the abstract see Supplementary Materials section.

Nationwide analysis of neuromyelitis optica in systemic lupus erythematosus and Sjogren's syndrome.

Neuromyelitis optica (NMO), also known as Devic's disease, is a rare inflammatory demyelinating disorder causing myelitis and optic neuritis. While there have been reports of systemic lupus erythematosus (SLE) and primary Sjogren's syndrome (SS) occurring with NMO, a formal association is not established. We aimed to investigate the occurrence of NMO in SLE and SS patients and study the clinical characteristics and outcomes of NMO and SLE/SS hospitalizations utilizing the national inpatient sample (NIS) database. The NIS database from 2016 to 2019 was used to extract data. Adult hospitalizations with the principal or secondary diagnosis of NMO were included. We classified NMO patients with and without concomitant diagnosis of SLE or Sjogren's syndrome. We evaluated and compared the clinical characteristics and outcomes of NMO hospitalizations with and without SLE or Sjogren's syndrome. STATA17 was used for data analysis. We also calculated the odds ratio of NMO in SLE and Sjogren's syndrome. There were a total of 16,360 adult hospitalizations with the principal or secondary discharge diagnosis of NMO. Among all NMO hospitalizations, 1425 (8.71%) had the primary or secondary diagnosis of SLE or SS. The odds of NMO in SLE and Sjogren's syndrome were noted to be 12.29 and 5.56, respectively. NMO with SLE/SS group had higher proportion of females (89.82% vs 79%, P value < 0.001), African Americans (56.63% vs 38.28, P value < 0.001), and Asians (5.73% vs 3.25, P value 0.04). The Charlson comorbidity index was higher for NMO-SLE/SS overlap (2.44 vs 1.28, P value < 0.001). There was no significant difference in overall mortality rates of both groups (2.11% vs 1.2%, P value 0.197). There were significantly higher reported seizures (14.73% vs 6.05, P value < 0.001) and paraplegia (21.75% vs 13.93%, P value < 0.001) in NMO-SLE/SS patients. These patients also had a longer length of stay in comparison to the reference group (7 vs 5 days, P value < 0.001) as well as higher total charges. NMO patients had a 12-fold higher risk of SLE and 5-fold higher risk of Sjogren's disease when compared to general population. Patients with overlap of NMO and SLE or Sjogren's were predominantly women and were more likely to be African-American. Co-existence of these autoimmune disorders was associated with poor prognosis in terms of higher morbidity for patients and increased health care burden. Key Points • NMO is a rare autoimmune disease seen predominantly in women in the middle age group with low overall mortality. • SLE and Sjogren's have increased odds of NMO in comparison to general population. • NMO patients have high rates of several complications such as paraplegia, quadriplegia, seizures, blindness, sepsis, and respiratory failure with even higher rates of seizures and paraplegia in those with concomitant SLE or Sjogren's.

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 16309: Impact of the Pandemic on AMI Hospitalizations and Outcomes: A Comprehensive Medicare Fee-for-Service Analysis

Introduction: Studies suggest hospitalizations for acute myocardial infarction (AMI) declined early in the COVID-19 pandemic, but national patterns during the entire pandemic are not well characterized. This study aims to characterize the changes in AMI hospitalizations and 30-day mortality rates among Medicare Fee-For-Service (FFS) beneficiaries during the pandemic. Methods: We used Medicare FFS Part A data from March 2014 to November 2021 consisting of people aged ≥65 years. The dataset was divided by quarters each representing the four seasons, March 2014 to February 2020 defined as the prepandemic phase, March 2020 to May 2020 defined as the early pandemic phase, and June 2020 to August 2021 as the late pandemic phase. Patients with a principal discharge diagnosis of AMI were identified using ICD-9 and 10 codes. The number of AMI hospitalizations and crude 30-day mortality was calculated for each quarter. Results: Before the pandemic, number of AMI hospitalizations in the US were gradually declining from 48,258 per quarter in Q2 2017 to 42,639 in Q2 2019 (12 % decrease compared with 2017) with regular seasonal increases during the winter months (Q1). Thereafter, a sharp decline to 29,446 in Q2 2020 (31% decrease compared with 2019) was observed followed by a return to 34,105 in Q2 2021 (20% decrease compared with 2019). The crude 30-day mortality rate was 12.3% in the 12 months before the pandemic but increased to 13.3% in Q2 2020 during the early pandemic and remained the same at 13.3% after Q3 2020. Conclusions: In the US Medicare fee-for-service population, the number of AMI hospitalizations was in a steady decline before the pandemic, but dramatically declined during the first 3 months of the pandemic. In contrast, the crude 30-day mortality increased during the pandemic. Why these changes occurred remains a source of continuing investigation.

Abstract 17700: Double Trouble With Identification of Isolated Sub-Segmental Pulmonary Embolism Using ICD-10 Discharge Diagnosis Codes: The PE-EHR+ Study

Introduction: International Classification of Diseases 10 th revision (ICD-10) discharge diagnosis codes are widely used for research studies that use electronic databases. The accuracy of the codes for identifying patients with isolated sub-segmental pulmonary embolism (issPE) is unknown. Methods: We used data from the Mass General Brigham (MGB) health system (01/2016-12/2021) and included 1,734 patients randomly selected from 3 groups: Patients with ICD-10 Principal Discharge Diagnosis codes for PE, patients with secondary Discharge Diagnosis codes for PE, and patients with no ICD-10 codes for PE. Weighted estimates were obtained by considering the total number of hospitalizations in each category at MGB hospitals. Accuracy of the codes was assessed by two independent physicians who reviewed discharge summaries, daily notes, and radiology reports. Results: Of 1,734 records reviewed, 1,712 were included (age: 60.6±17.8 years, 52.3% female). Manual chart review found 865 patients with acute PE, including 104 (12.0%) with issPE. Of these 104 patients, 103 (98.3%) had ICD-10 codes for PE. However, only in 15 (14.6%) cases, PE codes were specified as being limited to subsegmental vessels (issPE). In turn, in another 43 cases, Principal-or-secondary ICD-10 codes for issPE were listed but the patients did not have acute issPE during index hospitalization per review of records. The sensitivity and specificity for issPE codes in Principal or secondary discharge position were 14.4% and 97.3% (Table, panel A). Positive predictive value in the weighted sample was only 26.2% (Table, panel B). Conclusions: ICD-10 codes are insensitive and have a poor positive predictive value for detection of isolated subsegmental PE and should not be used for case identification for research purposes or quality improvement initiatives.

The Impact of Age on the Epidemiology and Cost of Atrial Fibrillation Hospitalizations

Atrial fibrillation (AF) is the most common arrhythmia and increases with age. This rising prevalence of AF is contributing to an increasing public health and economic burden. The 2018 Healthcare Cost and Utilization Project National Inpatient Sample dataset was used. All patients ≥15years with a principal discharge diagnosis of AF were included. The patient population was divided into an "older" cohort (aged ≥65years) and a "younger" (aged <65years). Desired outcomes included hospital length of stay, discharge disposition, hospital charges, and in-hospital mortality. A generalized linear mixed model was used to calculate hospitalization rates for the "younger" and "older" groups. We identified 896,328 AF hospitalizations. Younger patients (18.1%) were more likely to be male (65.5% vs 49.9%), to smoke (21.6% vs 6.1%), and to use alcohol (9.7% vs 2.1%). Older patients were more likely to have heart failure (49.6% vs 43.9%) and hypertension (84.6% vs 76.1%). Hospitalization rates increased with increasing age groups. Older patients had higher in-hospital mortality (4.6% vs 2.9%) and were more likely to be discharged to another facility (31.6% vs 13.2%). AF hospitalization rates vary between hospitals across the United States. Hospital divisions with greater than expected admissions for AF, when compared with the national mean, were driven by higher "older" patient hospitalizations. In conclusion, older patients account for most AF hospitalizations. Older patients have higher AF morbidity and mortality. Hospitalization rates for AF increase with increasing increments of age.

Developing Validated Tools to Identify Pulmonary Embolism in Electronic Databases: Rationale and Design of the PE-EHR+ Study.

Contemporary pulmonary embolism (PE) research, in many cases, relies on data from electronic health records (EHRs) and administrative databases that use International Classification of Diseases (ICD) codes. Natural language processing (NLP) tools can be used for automated chart review and patient identification. However, there remains uncertainty with the validity of ICD-10 codes or NLP algorithms for patient identification. The PE-EHR+ study has been designed to validate ICD-10 codes as Principal Discharge Diagnosis, or Secondary Discharge Diagnoses, as well as NLP tools set out in prior studies to identify patients with PE within EHRs. Manual chart review by two independent abstractors by predefined criteria will be the reference standard. Sensitivity, specificity, and positive and negative predictive values will be determined. We will assess the discriminatory function of code subgroups for intermediate- and high-risk PE. In addition, accuracy of NLP algorithms to identify PE from radiology reports will be assessed. A total of 1,734 patients from the Mass General Brigham health system have been identified. These include 578 with ICD-10 Principal Discharge Diagnosis codes for PE, 578 with codes in the secondary position, and 578 without PE codes during the index hospitalization. Patients within each group were selected randomly from the entire pool of patients at the Mass General Brigham health system. A smaller subset of patients will also be identified from the Yale-New Haven Health System. Data validation and analyses will be forthcoming. The PE-EHR+ study will help validate efficient tools for identification of patients with PE in EHRs, improving the reliability of efficient observational studies or randomized trials of patients with PE using electronic databases.

Abstract P410: Spontaneous Coronary Artery Dissection and Intervention With Percutaneous Coronary Intervention (PCI): A Population Based Study 2016-2019

Background: Spontaneous coronary artery dissection (SCAD) remains an important cause of myocardial infarction. A paucity of data exists regarding optimal management of SCAD due to lack of randomized trials comparing medical therapy and revascularization strategies. Methods: Data were extracted from the National Inpatient Sample (NIS) 2016-2019 Database. The NIS was searched for hospitalizations for adult patients with SCAD as principal discharge diagnosis with and without intervention with percutaneous coronary intervention (PCI) as secondary diagnosis using ICD-10 codes. The primary outcome was inpatient mortality. Multivariate logistic and linear regression analysis was used accordingly to adjust for confounders. STATA software was used for analysis. Results: 2,400 patients who had SCAD were divided into those that had PCI versus those who did not-- followed by matching. Majority of the patient population was noted to be females (76.9%) and mean age was a 54.63 years. The adjusted odds ratio (OR) for inpatient mortality for SCAD with PCI intervention showed significantly reduced mortality (OR 0.32, p&lt;0.01), which was statistically significant. Hospital length of stay (LOS) was less (-1.23 days, p&lt;0.04) and total hospital charges were lower (-$3,697.02, p&lt;0.03) in this patient population. The only independent positive predictor of increased mortality for the SCAD and PCI intervention group was acute kidney injury (AKI) (OR 4.99, p&lt;0.001). Independent positive predictors of hospital LOS were the following: cardiogenic shock (4.14 days, p&lt;0.01), AKI (4.99 days, p&lt;0.01), ventricular arrhythmias (3.53 days, p&lt;0.01), and systolic heart failure (2.84 days, p&lt;0.03). Independent positive predictors for total charges were the following: cardiogenic shock ($108,113.9, p&lt;0.015), ventricular arrhythmias ($109,009.6, p&lt;0.04), AKI ($107.635.7, p&lt;0.003), atrial flutter ($134,619.3, p&lt;0.042). Conclusion: Previous studies showed that the majority of patients with SCAD except those with high risk should have a conservative revascularization strategy. Our study shows decreased mortality, hospital LOS, and total charges for patients with SCAD who had intervention with PCI. Randomized clinical trials will need to further examine the role of PCI for SCAD in different populations.

Abstract P583: Cardiac Sarcoidosis is an Under-Recognized Cause of Heart Failure Admissions in the United States: A Population Based Epidemiologic Study

Background: Progressive heart failure is the cause of mortality for almost 25% of patients with cardiac sarcoidosis (CA). However, there are limited studies on the impact of heart failure (HF) on mortality, hospital length of stay (LOS), and total hospital charges for patients with cardiac sarcoidosis. Methods: Data were extracted from the National Inpatient Sample (NIS) 2016-2019 Database. The NIS was searched for hospitalizations for adult patients with cardiac sarcoidosis as the principal discharge diagnosis with and without different types of heart failure as secondary diagnosis using ICD-10 codes. Multivariate analysis via STATA was used to adjust for confounders. Results: 6,560 patients were identified with a diagnosis of cardiac sarcoidosis. Only 0.08% of patients with heart failure were identified with a diagnosis of cardiac sarcoidosis, of which 15.8% had systolic HF, 3.7% diastolic had HF, and 14% had combined systolic and diastolic HF. There were more males than females (60.0% vs. 40.0 %, p&lt;0.01). Mortality was increased for combined systolic and diastolic HF (OR 1.98, p&lt;0.04). LOS was increased for patients with any type of HF(+2.48 days, &lt;0.001), systolic heart failure (+2.14 days, p&lt;0.017), combined systolic and diastolic HF (+1.96 days, p&lt;0.010). Total hospital charges were increased with any type of HF ($44,214.43, p&lt;0.004), and systolic heart failure ($58,298.90, p&lt;0.015). Interestingly, patients with cardiac sarcoidosis did not have increased mortality with all types of HF(OR 1.04, p=0.928), systolic HF (OR 0.98, p=0.973), diastolic HF (OR 0.35, p=0.179), and combined systolic and diastolic HF (OR 1.34 experienced a 51% increased odds of mortality (OR 1.51, 95%CI 1.41-1.62, p&lt;0.01), a longer LOS with a mean difference of 1.4 days (95%CI, 1.16-1.57, p&lt;0.01), and a more costly overall hospitalization with a mean difference of $12,841 (95%CI $9,705-$15,976, p&lt;0.01). Conclusion: Cardiac sarcoidosis remains a widely under-recognized cause of HF hospitalizations in the United States.

Abstract P528: Acute Kidney Injury Increases Mortality in Patients With Upper Gastrointestinal Hemorrhage (UGIB) and Acute Coronary Syndrome: A Population Based Study

Background: Upper GI Bleeding (UGIB) is a common emergency with a relatively high incidence in the general population. Acute Coronary Syndrome (ACS) is often overlooked in gastrointestinal bleeding as its symptoms and signs are often subtle. Purpose: This study aims to compare the outcomes of patients admitted for ACS with and without UGIB. Methods: Data were extracted from the National Inpatient Sample (NIS) 2016-2019 Database. The NIS was searched for hospitalizations for adult patients with ACS as the principal discharge diagnosis with and without UGIB as the secondary diagnosis using ICD-10 codes. The primary outcome was inpatient mortality. Multivariate logistic and linear regression analysis was used accordingly to adjust for confounders. STATA software was used for analysis. Results: Among 35,740 patients who had ACS, 463 (1.3%) patients had UGIB. The adjusted odds ratio (OR) for inpatient mortality for ACS with UGIB compared to those without UGIB was higher (OR 1.94, p&lt;0.0001), length of stay was 4.27 days longer (p&lt;0.0001), and total charges were $55,479 higher (p&lt;0.0001). The Independent positive predictors of increased mortality were: Acute kidney injury (AKI) (p&lt;0.0001). Independent positive predictors of LOS and total charges were the following: Coronary artery bypass graft surgery (CABG), Intra-aortic balloon pump (IABP) insertion, AKI (p&lt;0.0001) Conclusion: UGIB patients with ACS did have higher inpatient mortality. A positive predictor of mortality included the presence of concomitant AKI. Thus, the prevention of AKI Is of utmost importance. Further, larger studies are warranted to identify new risk factors as predictors of mortality and improve awareness of GI bleeding risk and decrease LOS.