Comorbidity Codes Research Articles

Relative effectiveness of high-dose versus standard-dose quadrivalent influenza vaccine against hospitalizations and mortality according to frailty score: a post-hoc analysis of the DANFLU-1 trial

Abstract Background Frailty is a major risk factor for adverse cardiovascular events. Influenza vaccination has been shown to protect against cardiovascular fatal and nonfatal events. However, limited data exist on the added benefit of high-dose versus standard-dose influenza vaccination according to frailty status. Purpose We sought to assess the relative effectiveness of high-dose (QIV-HD) versus standard-dose (QIV-SD) quadrivalent influenza vaccination according to frailty status in elderly individuals. Methods This is a post-hoc analysis of the randomized controlled feasibility trial of QIV-HD versus QIV-SD conducted during the 2021–2022 influenza season in adults aged 65–79 years. We assessed the following prespecified outcomes: hospitalizations for pneumonia or influenza, cardio-respiratory hospitalizations, cardiovascular hospitalizations, all-cause hospitalizations and all-cause mortality. Level of frailty was defined according to the Hospital Frailty Risk Score based on ICD-10 codes for relevant comorbidities any time prior to vaccination. We divided the participants into two categories according to Frailty Score (FS): low frailty (0-4 points) and intermediate or high frailty (≥5 points). We analyzed outcomes as time to first event using Cox proportional hazards regression. Results Among 12,473 randomly assigned participants to QIV-HD versus QIV-SD (mean age 71.7 years, 47.1% female), 10,711 (85.9%) were categorized as having low frailty and 1,762 (14.1%) had intermediate or high frailty. In total, 38 individuals were hospitalized for pneumonia or influenza during follow-up, 219 were hospitalized for cardiorespiratory illness, 1055 were hospitalized due to any cause and 62 died during follow-up. FS significantly modified the effect of QIV-HD versus QIV-SD for all-cause mortality (P for interaction 0.02). In participants with low frailty, QIV-HD was associated with a lower risk of all-cause mortality (Figure 1: HR 0.26, 95% CI 0.13 to 0.55), whereas there was no evidence of effect among those with intermediate or high frailty (HR 1.68, 95% CI 0.66 to 4.26). QIV-HD versus QIV-SD was associated with a lower incidence of hospitalizations for pneumonia and influenza, this effect was consistent regardless of frailty status (P for interaction 0.88). QIV-HD was not associated with a significantly lower incidence of hospitalizations due to cardiovascular, cardiorespiratory or respiratory disease or any cause regardless of frailty status. Conclusions In this post-hoc analysis of a large-scale pragmatic trial, QIV-HD was similarly associated with a lower incidence of hospitalizations for pneumonia and influenza among older adults irrespective of frailty status. The potential benefit of QIV-HD versus QIV-SD may therefore be applicable regardless of frailty status. However, our results suggest that QIV-HD is associated with a lower risk of all-cause mortality among adults with low frailty only.Figure 1.

Can researchers trust ICD-10 coding of medical comorbidities in orthopaedic trauma patients?

The 10th revision of the International Classification of Diseases (ICD-10) coding system may prove useful to orthopaedic trauma researchers to identify and document populations based on comorbidities. However, its use for research first necessitates determination of its reliability. The purpose of this study was to assess the reliability of electronic medical record (EMR) ICD-10 coding of nonorthopaedic diagnoses in orthopaedic trauma patients relative to the gold standard of prospective data collection. Nonexperimental cross-sectional study. Level 1 Trauma Center. Two hundred sixty-three orthopaedic trauma patients from 2 prior prospective studies from September 2018 to April 2022. Prospectively collected data were compared with EMR ICD-10 code abstraction for components of the Charlson Comorbidity Index (CCI), obesity, alcohol abuse, and tobacco use (retrospective data). Percent agreement and Cohen's kappa reliability. Percent agreement ranged from 86.7% to 96.9% for all CCI diagnoses and was as low as 72.6% for the diagnosis "overweight." Only 2 diagnoses, diabetes without end-organ damage (kappa = 0.794) and AIDS (kappa = 0.798) demonstrated Cohen's kappa values to indicate substantial agreement. EMR diagnostic coding for medical comorbidities in orthopaedic trauma patients demonstrated variable reliability. Researchers may be able to rely on EMR coding to identify patients with diabetes without complications or AIDS. Chart review may still be necessary to confirm diagnoses. Low prevalence of most comorbidities led to high percentage agreement with low reliability. Level 1 diagnostic.

Prevalence and outcomes of chronic comorbid conditions in patients with sepsis in Korea: a nationwide cohort study from 2011 to 2016

BackgroundChronic comorbid conditions are common in patients with sepsis and may affect the outcomes. This study aimed to evaluate the prevalence and outcomes of common comorbidities in patients with sepsis.MethodsWe conducted a nationwide retrospective cohort study. Using data from the National Health Insurance Service of Korea. Adult patients (age ≥ 18 years) who were hospitalized in tertiary or general hospitals with a diagnosis of sepsis between 2011 and 2016 were analyzed. After screening of all International Classification of Diseases 10th revision codes for comorbidities, we identified hypertension, diabetes mellitus (DM), liver cirrhosis (LC), chronic kidney disease (CKD), and malignancy as prevalent comorbidities.ResultsOverall, 373,539 patients diagnosed with sepsis were hospitalized in Korea between 2011 and 2016. Among them, 46.7% had hypertension, 23.6% had DM, 7.4% had LC, 13.7% had CKD, and 30.7% had malignancy. In-hospital mortality rates for patients with hypertension, DM, LC, CKD, and malignancy were 25.5%, 25.2%, 34.5%, 28.0%, and 33.3%, respectively, showing a decreasing trend over time (P < 0.001). After adjusting for baseline characteristics, male sex, older age, use of mechanical ventilation, and continuous renal replacement therapy, LC, CKD, and malignancy were significantly associated with in-hospital mortality.ConclusionsHypertension is the most prevalent comorbidity in patients with sepsis, and it is associated with an increased survival rate. Additionally, liver cirrhosis, chronic kidney disease, and malignancy result in higher mortality rates than hypertension and DM, and are significant risk factors for in-hospital mortality in patients with sepsis.

Abstract WP108: Predictive Accuracy of Multivariable Models for 30-day and 1-year All Cause Readmission After Stroke Using Different Combinations of Registry, Hospital and Claims Based Data Sources

Introduction: Hospital readmissions are often used as an indicator of quality of care. However, identifying patients at risk of readmission after stroke is challenging and predictive models have historically not performed well, in part because they often rely on single data sources. Data linkage might offer a solution. Methods: We probabilistically linked data from the Michigan’s Get With The Guidelines Stroke registry and Michigan Value Collaborative multipayer claims database from Medicare and Blue Cross Blue Shield beneficiaries discharged alive following acute stroke (ICD-10 I61-I63) between 2016-2020. The registry dataset included 64 variables covering demographics, stroke presentation, medical history, procedures, and complications. The hospital dataset included 20 variables from the American Hospital Association’s database. The claims dataset included payer and 79 HCC comorbidity codes. Using combinations of the 3 data sources, we examined the performance of multivariable LASSO logistic regression models to predict all cause readmission at 30-days and 1-year post discharge. We generated hospital-specific testing and training models and reported the mean model discrimination (AUC) of all combinations of the 3 datasets. Results: Of 19,382 linked stroke discharges, 2,724 (14.1%) and 8,169 (42.2%) were readmitted within 30-days and 1-year, respectively. For 30-day readmission, the model based on only registry data produced the best performance (M1, Table). However, for prediction of 1-year readmission, the combination of registry and claims data produced the best performing model (M13, Table). Hospital level characteristics did not have any significant impact on prediction accuracy. Conclusions: Clinical registry data was the best data source for predicting 30-day readmission. However, claims based data were additive when predicting readmission within 1-year, probably because HCC codes add information about total comorbidity burden.

Psoriasis in Norway: a prescription-based registry study of psoriasis-associated comorbidities.

Psoriasis is an inflammatory skin disease, frequently associated with comorbidities. The aim of this study was to characterize comorbidities associated with skin psoriasis using real-world data. The study was conducted in Norway, taking advantage of the universal healthcare system with equal access to healthcare and a mandatory prescription reporting system that applies to all hospitals and clinics. The study focused on patients registered in the Norwegian Prescription Database (NorPD) who had either a diagnosis code indicating skin psoriasis alone or in combination with diagnosis codes for pre-selected comorbidities. Between 2004 and 2020, a total of 272,725 patients diagnosed with skin psoriasis were identified in NorPD, and 84,432 patients received their first prescription for skin psoriasis between 2015 and 2020. Among these, patients who received prescriptions for pre-selected comorbidities, either before or after initial prescription for psoriasis treatment, were included in the study. The most common prescriptions for patients with skin psoriasis were for primary prevention of atherosclerotic disease (22.6%), established atherosclerotic disease (19.3%), and depression (18.4%). These were closely followed by prescriptions for other comorbidities such as heart failure (16.1%), type II diabetes mellitus (9.4%), hypertension (9.4%), anxiety (7.8%), psoriatic arthritis (5.5%), acute myocardial infarction (4.2%), hypertensive cardiac disease (3.0%), ulcerative colitis (2.4%), type I diabetes mellitus (2.2%), secondary prophylaxis after myocardial infarction (1.6%), and Crohn's disease (1.0%). Patients with skin psoriasis have a high number of prescriptions not only for their psoriasis but also for many other conditions compared to reports in the literature on the general population.

P117 Machine-learning derived characteristics associated with tapering TNF inhibitors in individuals with rheumatoid arthritis

Abstract Background/Aims Tapering of TNF inhibitor (TNFi) drugs may be considered in some patients to reduce risks and costs. Selecting appropriate patients is not always straightforward and may be influenced by age, sex, comorbidity and disease activity state. We sought to identify predictors for dose tapering in a real-world clinical setting. Algorithmic extraction, selection and analysis of relevant patient sub-cohorts could enable identification of relevant predictors associated with TNFi dose tapering. Methods Our institution has a Rheumatology Biologics database running prospectively for over 15 years. Our approach for patients with RA receiving TNFi has been to dose-taper by one third and then 50% if remission achieved (defined as DAS28&amp;lt;2.6 on two occasions more than 6 months apart with no corticosteroid use). Prescribing, disease activity scores and demographics were extracted using SQL along with comorbidity coding, pathology results and anthropometric data. Data were anonymised and analysed in Python 3.8 within our institutions’ Trusted Research Environment (TRE). Pandas, NumPy and StatsModels python packages were used for the analysis in Jupyter notebooks. 49 covariates were considered clinically relevant and included in the regression analysis. Recursive feature elimination (RFE) was performed using logistic regression (LR) with threshold p-value of 0.05. The primary outcome was tapering of TNFi, algorithmically identified by a temporal increase in dosing interval or decrease in dosage. To avoid multiple-drug confounding, only the most recent TNFi data was included for each patient. Results 663 patients with RA were initiated on TNFi between 5th November 2001 and March 2nd 2020. 491 (74.1%) were female with a mean age of 65.5 (SD ± 14.2) years. 261 (39.4%) received adalimumab, 209 (31.5%) etanercept, 74 (11.2%) infliximab, 82 (12.4%) certolizumab and 37 (5.6%) golimumab. Concurrent methotrexate (MTX) was seen in 34.5% (n = 22), either oral or subcutaneous. There was no change in the likelihood of tapering associated with depression, hypothyroidism, obesity, smoking or seropositivity for RF or anti-CCP. Those taking MTX were more likely to taper their biologics (OR 3.33, 95%CI 1.83-6.09, p = &amp;lt;.000), as were patients who were coded as having type 1 or 2 diabetes (7.4% n = 49, OR 3.23, 95%CI 1.32-8.25, p = 0.011), Higher DAS28 CRP score (OR 0.548, 95%CI 0.38-0.78, p = 0.001), and DAS 28 ESR score (OR 0.71, 95%CI 0.53-0.95, p = 0.021) significantly decreased chance of tapering. Conclusion Concurrent methotrexate use increases likelihood of subsequent tapering in patients with RA receiving TNFi. Unexpectedly, patients with diabetes were also more likely to taper, however due to low numbers of patients in this group and the width of confidence intervals this should be interpreted cautiously. As expected, patients with high disease activity scores were less likely to taper. This algorithm driven approach produced results largely in keeping with clinical intuition, however these methods may aid in future selection of tapering cohorts. Disclosure T. Phillips: None. M. Bhandari: None. M. Stammers: None. S. Fraser: None. M. George: None. S. Lin: None. M. Lwin: None. C. Holroyd: None. C. Edwards: Other; CE has received fees for advisory boards, speaker’s bureau, research support from Abbvie, Astra Zeneca, BMS Celltrion Chugai, Gilead, Galapagos, GSK, Janssen, Eli Lilly, Pfizer, Roche, Sandoz, Samsung.

High baseline prevalence of atopic comorbidities and medication use in children treated with allergy immunotherapy in the REAl-world effeCtiveness in allergy immunoTherapy (REACT) study.

Respiratory allergy, commonly manifesting as allergic rhinitis (AR) and asthma, is a chronic progressive disease that frequently starts in childhood. Allergy immunotherapy (AIT) is the only causal treatment for respiratory allergy with the potential to modify the underlying cause of allergy and, ultimately, prevent disease progression. This analysis aimed to determine if AIT is received sufficiently early to halt the progression of allergic disease, by characterizing the burden and progression of disease in children prior to AIT initiation in real-life clinical practice. The REAl-world effeCtiveness in allergy immunoTherapy (REACT) study was a large retrospective cohort study using German claims data between 2007 and 2017. Characteristics of two pre-defined AIT age cohorts from the REACT study - children (aged <18 years) and adults (aged ≥18 years) - were evaluated during the 1-year period before the first AIT prescription. For comparison, a control group of all subjects with a confirmed diagnosis of AR and without prescriptions for AIT was included. Burden of disease was assessed using diagnostic codes for atopic comorbidities [e.g., atopic dermatitis (AD), asthma, and acute allergic conjunctivitis] and non-atopic comorbidities (e.g., migraine, headache); medication use, recorded as prescriptions for symptom-relieving AR medication and reliever/controller medication for asthma, was also assessed. Data were analyzed descriptively, using summary statistics. Both children (n = 11,036) and adults (n = 30,037) showed a higher prevalence of atopic comorbidities and a greater drug burden prior to AIT initiation compared to AR patients not treated with AIT (n = 1,003,332). In the two age-specific AIT cohorts, children consistently showed the highest prevalence of atopic comorbidities compared to adults (AIT children, AIT adults - asthma: 41.4%, 34.5%; AD: 19.9%, 10.2%; acute allergic conjunctivitis: 13.6%, 10.2%). Generally, prescriptions per year for symptom-relieving AR and asthma treatments were also higher for children initiating AIT vs. adults (AIT children, AIT adults - AR prescriptions per subject: 1.72, 0.73; asthma prescriptions per subject: 1.42, 0.79). Children with AR who are offered AIT in real-life show considerable disease burden prior to initiation. As AIT may alleviate the burden and halt the progression of allergic disease, considering AIT earlier in the disease course may be warranted.

A Machine Learning Model for Prediction of Amputation in Diabetics.

Diabetic foot ulcer (DFU) and the resulting lower extremity amputation are associated with a poor survival prognosis. The objective of this study is to generate a model for predicting the probability of major amputation in hospitalized patients with DFU. The National Inpatient Sample (NIS) database from 2008 to 2014 was used to select patients with DFU, who were then further divided by major amputation status. International Classification of Diseases, Ninth Edition, Clinical Modification (ICD-9-CM) and Agency for Healthcare Research and Quality (AHRQ) comorbidity codes were used to compare patient characteristics. For the descriptive statistics, the Student t test, the χ2 test, and the Spearman correlation were utilized. The five most predictive variables were identified. A decision tree model (CTREE) based on conditional inference framework algorithm and a random forest model were used to develop the algorithm. A total of 326 853 inpatients with DFU were identified, and 5.9% underwent major amputation. The top five contributory variables (all with P < .001) were gangrene (odds ratio [OR] = 11.8, 95% confidence interval [CI] = 11.5-12.2), peripheral vascular disease (OR = 2.9, 95% CI = 2.8-3.0), weight loss (OR = 2.6, 95% CI = 2.5-2.8), systemic infection (OR = 2.5, 95% CI = 2.4-2.53), and osteomyelitis (OR = 1.7, 95% CI = 1.6-1.73). The model performance of the training data was 77.7% (76.1% sensitivity and 79.3% specificity) and of the testing data was 77.8% (76.2% sensitivity and 79.4% specificity). The model was further validated with boosting and random forest models which demonstrated similar performance and area under the curve (AUC) (0.84, 95% CI = 0.83-0.85). Utilizing machine learning methods, we have developed a clinical algorithm that predicts the risk of major lower extremity amputation for inpatients with diabetes with 77.8% accuracy.

Comorbidity Coding and Its Impact on Hospital Complexity

Comorbidity Coding and Its Impact on Hospital Complexity

Comorbidity Coding and Its Impact on Hospital Complexity: Reply

Comorbidity Coding and Its Impact on Hospital Complexity: Reply

Identification of cognitively impaired patients at risk for development of Alzheimer’s disease dementia: an analysis of US Medicare claims data

ABSTRACT Background Identifying factors associated with transitioning from mild cognitive impairment (MCI) to dementia due to Alzheimer’s disease (AD dementia) or dementia due to any cause (all-cause dementia) may inform economic assessments of disease and early care planning. Research Design and Methods A multivariate logistic regression approach identified potential predictors of progression to AD dementia or all-cause dementia in individuals with MCI or cognitive impairment (CI). Eligible patients and variables of interest were identified using claims data from the Medicare Advantage Patient Database, by Optum. Results Predictors of an AD dementia diagnosis included age (odds ratio [OR], 1.71) and use of antipsychotics (OR, 2.50) and hypertension medication (OR, 1.25). Medication use for comorbid conditions was a better indicator of risk than comorbidity coding. Diagnosis of CI by a neurologist increased the odds of an AD dementia diagnosis. Possible protective factors for progression included the use of anxiolytics (OR, 0.76), inpatient status at time of diagnosis (OR, 0.49), and a history of stroke (OR, 0.87). None of these factors differentiated AD dementia from all-cause dementia. Conclusions Identifying patients at risk for AD dementia allows for improved system-level planning to guide policy and optimize economic and clinical outcomes for patients, caregivers, and society.

Using health administrative data to identify patients with pulmonary hypertension: A single center, proof of concept validation study in Ontario, Canada.

Real‐world identification of pulmonary hypertension (PH) is largely based on the use of administrative databases identified by ICD codes. This approach has not been validated. The aim of this study was to validate a diagnosis of PH and its comorbidities using ICD 9/10 codes. Health records from Kingston Health Sciences Centre (2010 to 2012) were abstracted to identify a diagnosis of PH. Cohort 1 patients (n = 300) were selected because they had attended a cardiology or respirology clinic without knowledge of PH status. Cohort 2 patients (n = 200) were patients with a diagnosis of PH, identified using International Classification of Diseases (ICD) codes at the time of hospitalizations (CIHI‐DAD) or emergency department (ED) visits (CIHI‐NACRS). These cohorts were combined and reviewed to validate the diagnosis of PH. These data were securely transferred to the Institute of Clinical Evaluative Sciences (ICES). The diagnosis of PH from chart abstraction was used as the gold standard. The classification of PH into WHO groups, based on chart abstraction, was also compared to classification based on ICD code‐defined comorbidities. Cohort 1 and Cohort 2 were merged to yield 449 unique patients in the combined cohort. In the combined cohort, 248 of 449 (55.2%) had a diagnosis of PH by ICD code criteria. The mean age of this PH group was 70 years, and the majority were females (65.5%). One hospitalization or ED visit resulting in a diagnostic code for PH had a sensitivity of 73% and a specificity of 99% for a confirmed PH diagnosis on chart abstraction. When WHO classification by chart abstraction and ICD codes for comorbidities were compared, there was 87% agreement. Identification of PH and its comorbidities using ICD codes is a valid approach, and this single‐center study supports its application to identify PH.

Potential considerations in decision making on laparoscopic colorectal resections in Hungary based on administrative data

BackgroundLaparoscopic colorectal surgeries offer numerous advantages over their open counterparts. To compare these measurable short-time outcomes of open and laparoscopic resections in Hungary, data of colorectal surgeries were collected and analysed. The study focused on identifying patients’ characteristics that can influence the decision on laparoscopic colorectal resections and on comparing efficiency of Hungarian colorectal operations with international data.MethodsUsing patients’ data of laparoscopic and open colorectal surgery performed in 2015 and 2016 from the National Health Insurance Fund of Hungary, a countrywide retrospective comparative analysis was done. Logistic regression was used to explore main influencing factors for laparoscopic colorectal surgery.ResultsA total of 17,876 colorectal surgical cases, including 14,876 open and 3,000 laparoscopic resections were selected and analysed. Laparoscopy was used only in 16.78% of all cases. Comparison of age groups showed that odds ratio (OR) of laparoscopic colorectal resections was significantly lower in over 40 years than in younger patients (18–39 years). In university institutes patients had higher odds (OR: 2.23 p<0.0001) for laparoscopic colorectal resections. Presence of comorbidity codes and preoperative treatment in internal medicine department decreased odds for laparoscopic colorectal operations.ConclusionsPatients’ age, comorbidities and hospital type influenced the likelihood of decision on laparoscopic colorectal resection. Selection of patients contributed to improved laparoscopic outcomes.

The value of intrapartum factors in predicting maternal morbidity

The rates of severe maternal morbidity and mortality in the United States exceed those in other high-income nations. To aid providers and hospitals in recognizing the risk factors, there have been multiple attempts to develop stratification systems for morbidity based on maternal comorbidities. However, most women giving birth are healthy and do not have comorbidities to suggest that they are at an increased risk for severe maternal morbidity. There are small but inherent maternal risks to labor, and the events after admission may further influence a woman's risk for morbidity even for those initially at a low risk. To determine if the incorporation of intrapartum factors known at the start of the second stage of labor improves the predictive performance of a comorbidity-based risk tool for severe maternal morbidity. This is a retrospective cohort study of women at 8 hospitals in a single health system between July 1, 2016, and June 30, 2020. The women had term, singleton gestations and were admitted in labor and reached the second stage. The primary outcome was severe maternal morbidity. We compared logistic regression models using a validated risk-scoring tool (the Expanded Obstetric Comorbidity Score, which uses diagnosis codes for maternal comorbidities and pregnancy characteristics to predict maternal morbidity) with a model that included the Expanded Obstetric Comorbidity Score combined with parity and intrapartum factors. The intrapartum factors included labor induction or augmentation, length of labor, prolonged rupture of membranes, the presence of meconium-stained amniotic fluid, and gestational age. The hospitals were divided into a training (n=4) and testing (n=4) set to evaluate the predictive model performance. Discrimination was assessed by calculating the area under the receiver operating curve and calibration via calibration plots. Similar model comparisons were performed in a subgroup of women, who the Expanded Obstetric Comorbidity Score predicted to be at low risk for morbidity. This analysis included 33,770 deliveries from the 8 hospitals; severe maternal morbidity occurred in 498 (1.5%) deliveries. The model performance is reported among the testing set (n=15,350). Using the Expanded Obstetric Comorbidity Score alone, the area under the receiver operating curve was 0.676 (95% confidence interval, 0.636-0.716) and 155 (71%) events occurred among individuals above the median predicted risk. When combining intrapartum factors, the area under the receiver operating curve increased to 0.729, (95% confidence interval, 0.693-0.764) and 171 (78%) events occurred among individuals above the median predicted risk. The significant factors that were associated with severe maternal morbidity in this combined model included the Expanded Obstetric Comorbidity Score, length of labor, and the presence of meconium-stained amniotic fluid. The area under the receiver operating curve for the model with intrapartum factors was significantly higher than the models using the Expanded Obstetric Comorbidity Score alone (P<.001). The incorporation of intrapartum factors along with a validated risk tool (Expanded Obstetric Comorbidity Score) improved the ability to predict severe maternal morbidity at the start of the second stage. These findings emphasize the evolution of a woman's risk during her labor course and suggests that the prediction of maternal risk can be improved by considering intrapartum factors.

Machine learning prediction of diabetic foot ulcers in the inpatient population.

The objective of this study was to create an algorithm that could predict diabetic foot ulcer (DFU) incidence in the in-patient population. The Nationwide Inpatient Sample datasets were examined from 2008 to 2014. The International Classification of Diseases 9th Edition Clinical Modification (ICD-9-CM) and the Agency for Healthcare Research and Quality comorbidity codes were used to assist in the data collection. Chi-square testing was conducted, using variables that positively correlated with DFUs. For descriptive statistics, the Student T-test, Wilcoxon rank sum test, and chi-square test were used. There were six predictive variables that were identified. A decision tree model CTREE was utilized to help develop an algorithm. 326,853 patients were noted to have DFU. The major variables that contributed to this diagnosis (both with p < 0.001) were cellulitis (OR 63.87, 95% CI [63.87-64.49]) and Charcot joint (OR 25.64, 95% CI [25.09-26.20]). The model performance of the six-variable testing data was 79.5% (80.6% sensitivity and 78.3% specificity). The area under the curve (AUC) for the 6-variable model was 0.88. We developed an algorithm with a 79.8% accuracy that could predict the likelihood of developing a DFU.

Cancer Radiation Therapy May Be Associated With Atrial Fibrillation.

Background: The association of atrial fibrillation (AF) with cancer and cancer types is inconclusive. Similarly, data regarding the association of AF with different cancer therapies are controversial.Objectives: To study the association of AF with cancer subtypes and cancer therapies.Methods: We studied all patients aged 18–89 years who presented to the Feist Weiller Cancer Center, with or without a diagnosis of cancer, between January 2011 and February 2016. Electronic health records were systematically queried for baseline demographics and ICD-9 and ICD-10 codes for specific co-morbidities. Patients with a diagnosis of AF were tabulated based on cross-validation with the ECG database and/or by recorded history. We assessed the prevalence and risk of AF based on cancer diagnosis, specific cancer type, and cancer therapy.Results: A total of 14,600 patients were analyzed. Compared to non-cancer patients (n = 6,801), cancer patients (n = 7,799) had a significantly higher prevalence of AF (4.3 vs. 3.1%; p < 0.001). However, following correction for covariates in a multivariable logistic regression model, malignancy was not found to be an independent risk factor for AF (p = 0.32). While patients with solid tumors had a numerically higher prevalence of AF than those with hematological malignancies (4.3 vs. 4.1%), tumor type was not independently associated with AF (p = 0.13). AF prevalence was higher in patients receiving chemotherapy (4.1%), radiation therapy (5.1%), or both (6.9%) when compared to patients not receiving any therapy (3.6%, p = 0.01). On multivariable logistic regression, radiation therapy remained an independent risk factor for AF for the entire study population (p = 0.03) as well as for the cancer population (p < 0.01).Conclusions: Radiation therapy for cancer is an independent risk factor for AF. The known association between cancer and AF may be mediated, at least in part, by the effects of radiation therapy.

Identification of a Warm Autoimmune Hemolytic Anemia (wAIHA) Population Using Predictive Analytics of a Known Clinically Profiled Cohort

Introduction The clinical presentation of warm autoimmune hemolytic anemia (wAIHA), a rare, Ig-G mediated disorder, has not been well studied. There is no diagnosis code specific for wAIHA, presenting challenges in identifying wAIHA patients and assessing the clinical course of the disease. For this study, we aimed to 1) identify a wAIHA cohort using a collection of specific diagnostic codes, 2) evaluate whether a collection of diagnostic codes could bisect severe versus non-severe wAIHA patients, 3) observe the frequency of comorbidities and anemia symptoms in severe and non-severe wAIHA groups, and 4) use a predictive model to validate clinical variables and prevalence estimates. Methods A de-identified, longitudinal, patient-level claims database of &amp;gt;300 million US patients was used for this study. wAIHA patients were identified based on a diagnosis code of "autoimmune hemolytic anemia" (AIHA) with multiple distinct events within 36 months. Patients were required to have at least 30 days use of steroids over 36 months, and utilization of a known drug regimen for treatment of an autoimmune disorder. Patients were classified as severe if claims related to transfusion or blood composition testing or if high frequency interactions with a hematologist were observed in the 36-month period were. Codes for comorbidities, treatments, and procedures were grouped and analyzed within the most recent 12 months for each patient. Prevalence was estimated using AI/ML Lookalike modeling, using the known wAIHA patients as the positive training class. Results A cohort of 1,548 wAIHA patients was identified. Median patient age was &amp;gt;65 years, and patients were evenly distributed by gender. Patients showed evidence of anemia, anemia symptomology (such as shortness of breath, cough, and fatigue), and wAIHA-specific testing and treatments. The rate of disease-relevant claims was disproportionately higher in the severe cohort versus the non-severe cohort; over the 12 month study period, variances ranged from 61% higher (for anemia-based comorbidity codes) to as high as 570% higher (for anemia-based procedural codes). Primary hypertension, hyperlipidemia, gastro-esophageal reflux, and evidence of chemotherapy use were also present in wAIHA patients; all of these conditions were observed more frequently for severe patients. Of interest, lupus was observed more frequently in the non-severe wAIHA cohort. Almost 44% of wAIHA claims for the full cohort were associated with Hospital/Emergency care - 48% for the severe group. AI/ML modeling predicted patients using claims variables for hemolytic anemia, other blood count abnormalities, and medical procedure claims commonly used for the diagnosis and management of wAIHA patients (such as Coombs and haptoglobin testing). The predicted population supports reported US prevalence estimates of 30,000-49,000 patients. Conclusions We developed and validated a method for defining wAIHA patients using de-identified claims data based on AIHA ICD-10 codes and relevant treatments. We observed that while disease manifestations are generally the same, there is an increased rate of occurrence in the severe cohort during the same 12-month period. This increase was associated with higher utilization of healthcare resources. The comorbidity of lupus was more commonly associated with less severe wAIHA patients; this may indicate that, despite comorbidities, a sharper clinical focus is placed on the wAIHA condition itself when managing more severe patients. Disclosures McCrae: Dova: Consultancy; Momenta Pharmaceuticals: Consultancy; Novartis: Honoraria; Rigel: Consultancy. Beachell:Momenta Pharmaceuticals: Current Employment. Gooljarsingh:Momenta Pharmaceuticals: Current Employment. Tjoa:Momenta Pharmaceuticals: Current Employment. Jones:IPM.ai: Current Employment. Gubitosa:IPM.ai: Current Employment.

Liposomal Bupivacaine Infiltration Versus Bupivacaine Hydrochloride for the Management of Unilateral Total Knee Arthroplasty in Geriatric Patients at a 287-Bed Community Hospital.

Introduction: Geriatric patients receiving total knee arthroplasty (TKA) are found to have similar postoperative complications, functional scores, and perioperative mortality, as compared to younger patients. Conversely, geriatric patients often have longer lengths of stay. Periarticular injection (PAI) of liposomal bupivacaine (LB) as part of the multimodal pain management strategy is thought to improve recovery, however, mixed comparative efficacy data exists for its use in TKA.2-5. Methods: A retrospective, chart review was conducted at a 287-bed community teaching hospital. Orthopedic surgical patients who received an infiltration with liposomal bupivacaine versus bupivacaine HCl for unilateral TKA were compared. Patients identified in the electronic medical record by Diagnosis Related Group (DRG) 470-major joint replacement or reattachment of lower extremity without major complication or comorbidity codes were utilized. Patients who meet the following criteria were included: age 65 and older who underwent a TKA between 8/1/2018 to 7/31/2019 were discharged to home. Patients who have contraindications or hypersensitivity to bupivacaine formulations or a history of opioid dependence were excluded. The primary outcome is to identify whether patients who received an infiltration with liposomal bupivacaine had a lower total opioid consumption during their hospital stay. Results: A total of 114 patients who had a DRG 470 code and were above the age of 65 years were studied. There was no statistically significant difference in mean total opioid consumption (oral morphine equivalents) between the bupivacaine HCl (n = 25) and liposomal groups (n = 85) respectively, 93.76 versus 83.72 mg; P = .569. In addition, patients in both groups had similar lengths of hospital stay, 2.5 versus 3 days; P = .529 and mean pain scores until discharge 3.7 versus 4.34 on VAS; P = .305. Conclusion: The results of this drug utilization evaluation do not support a strong clinical advantage with local infiltration of liposomal bupivacaine over bupivacaine HCl in geriatric patients undergoing primary TKA surgery at this institution. There was not a statistically significant difference in mean total opioid consumption between the 2 groups. Additionally, the use of non-opioid analgesics, mean pain scores, and hospital lengths of stay were similar in both groups.

Risk of stroke after new-onset seizures

PurposeObservational cohort studies have reported a potentially increased risk of stroke in patients with epileptic seizures. Whether late-onset seizures merit primary stroke prophylaxis is not known, and more information on stroke risk is needed for the planning of RCTs. We performed a case-control study based on Swedish national registers to quantify the risk of stroke after epileptic seizures. MethodsCases ≤100 years of age with a first-ever stroke 2001–2009 were identified through the Swedish Stroke Register, and stroke-free controls (matched for age and sex) were obtained from the Population Register. The National Patient Register provided information on diagnostic codes for seizures, epilepsy and comorbidities. 123 105 stroke cases and 250 506 controls were included. ResultsEpileptic seizures prior to index stroke date were detected in 1559 (1.27 %) cases and 1806 (0.72 %) controls, yielding an odds ratio (95 % confidence interval) for stroke of 1.77 (1.65–1.89). ORs were similar in men and women, but higher below the age of 75. An onset of seizures in the year preceding stroke date resulted in a higher risk for stroke (OR = 2.21, 95 % CI = 1.79–2.72) compared to when more than 5 years had passed since the first seizure (OR = 1.57, 95 % CI = 1.43–1.72). ConclusionA history of epileptic seizures was associated with an increased risk of subsequent stroke. The risk seems to be particularly high in the first year following seizure diagnosis, which supports the notion that unexplained late-onset seizures may merit swift assessment of vascular risk profile. The nature of stroke prevention requires further study.

Can the Charlson Comorbidity Index be used to predict the ASA grade in patients undergoing spine surgery?

The American Society of Anaesthesiologists' Physical Status Score (ASA) is a key variable in predictor models of surgical outcome and "appropriate use criteria". However, at the time when such tools are being used in decision-making, the ASA rating is typically unknown. We evaluated whether the ASA class could be predicted statistically from Charlson Comorbidy Index (CCI) scores and simple demographic variables. Using established algorithms, the CCI was calculated from the ICD-10 comorbidity codes of 11'523 spine surgery patients (62.3 ± 14.6y) who also had anaesthetist-assigned ASA scores. These were randomly split into training (N = 8078) and test (N = 3445) samples. A logistic regression model was built based on the training sample and used to predict ASA scores for the test sample and for temporal (N = 341) and external validation (N = 171) samples. In a simple model with just CCI predicting ASA, receiver operating characteristics (ROC) analysis revealed a cut-off of CCI ≥ 1 discriminated best between being ASA ≥ 3 versus < 3 (area under the curve (AUC), 0.70 ± 0.01, 95%CI,0.82-0.84). Multiple logistic regression analyses including age, sex, smoking, and BMI in addition to CCI gave better predictions of ASA (Nagelkerke's pseudo-R2 for predicting ASA class 1 to 4, 46.6%; for predicting ASA ≥ 3 vs. < 3, 37.5%). AUCs for discriminating ASA ≥ 3 versus < 3 from multiple logistic regression were 0.83 ± 0.01 (95%CI, 0.82-0.84) for the training sample and 0.82 ± 0.01 (95%CI, 0.81-0.84), 0.85 ± 0.02 (95%CI, 0.80-0.89), and 0.77 ± 0.04 (95%CI,0.69-0.84) for the test, temporal and external validation samples, respectively. Calibration was adequate in all validation samples. It was possible to predict ASA from CCI. In a simple model, CCI ≥ 1 best distinguished between ASA ≥ 3 and < 3. For a more precise prediction, regression algorithms were created based on CCI and simple demographic variables obtainable from patient interview. The availability of such algorithms may widen the utility of decision aids that rely on the ASA, where the latter is not readily available.