scholarly journals Representation on feature and patient levels from structured Electronic Medical Records based on Skip-gram algorithm (Preprint)

2020 ◽  
Author(s):  
Yanqun Huang ◽  
Ni Wang ◽  
Zhiqiang Zhang ◽  
Honglei Liu ◽  
Xiaolu Fei ◽  
...  
Keyword(s):  
Data Representation ◽  
Primary Diagnosis ◽  
Machine Learning Algorithms ◽  
Categorical Variables ◽  
Continuous Variables ◽  
Stroke Patients ◽  
Feature Representations ◽  
Wide Range ◽  
Silhouette Index ◽  
Discrete Values

BACKGROUND The secondary utilization of the structured electronic medical record (sEMR) data has become a challenge due to the diversity, sparsity, and high-dimensionality of the data representation. OBJECTIVE We aimed to explore the feasibility of the embedding-based feature and patient representation for sEMR data and demonstrate the efficiency and superiority of the embedding-based patient representation. METHODS The entire training corpus consisted of records of 104752 hospitalized patients with 21 variables, including demographic characteristics, disease diagnoses, procedures, medications, laboratory tests, and other hospitalization indicators. Discrete values for original categorical variables and binned continuous variables were considered as words (concepts), and thus a record as a sentence in a text. To eliminate the influence the concept sequence played on the embedding algorithm, we randomly shuffled the concepts within a sentence 20 times. For a patient record, each feature concept was embedded into a 200-dimensional real number vector using the Skip-gram algorithm. Then the average of all the embedding concept vectors represented the patient. To assess the effectiveness of these embedding-based feature representations, we used the cosine distances among features’ embedding vectors to capture the latent relationship among the concepts of different features. We further conducted cluster analysis on stroke patients to evaluate and compare the efficiency and superiority of the embedding-based patient representation, where the embedding vectors were trained using the overall patients and just the stroke patients with and without the concept shuffling respectively. The representations of both multi-hot codes and one-hot codes plus original continuous numbers were used as the benchmark representations. RESULTS According to the Silhouette index, stroke patients were clustered into two groups, characterizing in patients with a primary diagnosis of hemorrhage stroke (HS) and ischemic stroke (IS), respectively. Cluster analyses conducted on patients with the embedding representations showed higher applicability (Hopkins Statistics, 0.925), higher aggregation (Silhouette index, 0.862), and lower dispersion (Davies Bouldin index, 0.551) than those conducted on patients with the benchmark representations. The two clusters for patients with the embedding-based representation learned from all the records after the concept shuffling achieved the highest F1-scores of 0.944 for IS and 0.717 for HS, respectively. CONCLUSIONS The feature-level embeddings can reflect the potential associations among medical concepts to some degree. The patient-level embeddings can be easily used as continuous input to standard machine learning algorithms and bring performance improvement. We expect that the embedding-based representation will be helpful in a wide range of the secondary use of the sEMR data.

scholarly journals 0798 “Uncovering My Feet Alleviates My Restless Legs”: Impact of Different Activities and Conditions on Restless Legs Symptoms

SLEEP ◽  
2020 ◽  
Vol 43 (Supplement_1) ◽  
pp. A303-A304
Author(s):  
E G Karroum ◽  
S Leu-Semenescu ◽  
R Amdur ◽  
I Arnulf
Keyword(s):  
Ambient Temperature ◽  
Categorical Variables ◽  
Sign Test ◽  
Fisher Exact Test ◽  
Continuous Variables ◽  
Exact Test ◽  
Restless Legs ◽  
Professional Activities ◽  
Wide Range ◽  
Activity Condition

Abstract Introduction The restless legs syndrome (RLS) is a resting wake state disorder with inactivity/decreased movement as an aggravating factor and activity/increased movement as an alleviating factor. Other activities and conditions may impact RLS symptoms but have not been systematically studied. Methods Fifty-six patients with primary severe RLS (age: 64.1±11.3; 66% women) responded about the effect of 20 activities/conditions on their RLS symptoms. Responses were assigned a numerical value: Aggravation (-1), No effect/Don’t know (0), Alleviation (+1), with calculating a mean effect score for each activity/condition and using a sign test to determine if that score was significantly above or below zero (no effect). Responses were further analyzed based on age, age at RLS onset, duration of RLS, RLS severity, gender, Familial/Non-familial RLS, and Painful/Non-Painful RLS. Association of continuous variables and categorical variables with each activity/condition was examined using Spearman correlation test and Fisher exact test, respectively. Bonferroni p threshold was set at p=0.00036. Results Activities/conditions with significant (p<0.0001) positive mean effect scores were: Feet uncovering (0.70); Leg massaging (0.63); Cold showers (0.54); and Manual activities (0.46). Activities with significant negative mean effect scores were: Vehicle passenger (-0.80); Show attendance (-0.70); Bedsheets weight on legs (-0.57); Watching TV (-0.54); High ambient temperature (-0.45); During meals (-0.39) (all p<0.0001); and Bedsheets rubbing on legs (-0.34; p=0.0002). Activities/conditions with no significant (all p>0.00036) mean effect scores were: Driving (0.00); Gambling (0.02); Professional activities (0.13); Hot showers (0.13); Using computer (0.14); Low ambient temperature (0.21); Sexual activities (0.27); Mental activities (0.29); and Sports activities (0.34). There was no significant association between each activity/condition and age, age at RLS onset, duration of RLS, RLS severity, gender, Familial/Non-familial RLS, or Painful/Non-Painful RLS. Conclusion There is a wide range of impact of different activities/conditions on RLS symptoms. These could be further considered in the non-pharmacological treatment or prevention of RLS symptoms. Support This study was not funded.

Abstract WMP12: Does IV tPA Improve Outcomes in Stroke Patients with Large Vessel Occlusions Treated with Intra-arterial Therapy?

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Joseph T Ho ◽  
Jason W Tarpley ◽  
Hsin-Fang Li
Keyword(s):  
Ischemic Stroke ◽  
Length Of Stay ◽  
Time Frame ◽  
Large Vessel ◽  
Categorical Variables ◽  
Complication Rates ◽  
Continuous Variables ◽  
Stroke Patients ◽  
Vessel Occlusion ◽  
Iv Tpa

Introduction: The benefit of endovascular therapy (IAT) for the treatment of emergent large vessel occlusion (ELVO) in stroke patients has been established. However, it is not known whether administration of IV tPA prior to IAT is beneficial in these patients. Methods: A retrospective review of ischemic stroke patients in the Providence Health & Services Get with the Guidelines (GWTG) database was performed from 01/2012 to 05/2016. The analysis was limited to patients who presented within 4.5 hours of last known well time (LKWT) and treatment included any form of IAT. End points were limited to data available in the GWTG database, including discharge mRS, discharge NIHSS, change in NIHSS from admission to discharge, and length of stay. Continuous variables were summarized using means and standard deviation while categorical variables were summarized using frequencies and percentages. To yield a more robust estimate against outliers for the time-related variables, medians and interquartile range (IQR) were computed and assessed using Wilcoxon rank sum tests. Chi-square tests and independent two-sample t-tests were used to evaluate the demographic and outcome differences for categorical and continuous variables, respectively. Results: A total of 10,868 patients with an ischemic stroke diagnosis were found in the specified time frame and presented within 4.5 hours of LKWT. Of these, 461 patients were treated with some form of IAT, 235 received IV tPA prior to IAT, 226 had IAT alone due to contraindication to IV tPA. There was no statistical difference in patient demographics, complication rates, TICI score, discharge NIHSS or mRS at discharge. There was a significantly higher NIHSS on admission (18.3 vs 16.7, p = 0.026), greater improvement in NIHSS (11.6 vs. 7.9, p=0.012), longer door to IAT (146 vs 101.5 min, p < 0.0001), and shorter length of stay (5 vs 6 days, p = 0.016) in the IV tPA group. Conclusions: These data suggest that IV tPA, when administered to eligible patients with ELVO, provided some benefit over IAT alone, even though it delayed IAT. Future prospective randomized trials are planned that may better address this question, but these results underscore the need for retrospective analysis of existing data.

Burden of dementia in atrial fibrillation hospitalizations- analysis of national inpatient sample 2016-2018

2021 ◽  
Vol 28 (Supplement_1) ◽  
Author(s):  
H Thyagaturu ◽  
K Shah ◽  
S Li ◽  
S Thangjui ◽  
B Shrestha
Keyword(s):  
Atrial Fibrillation ◽  
Hospital Mortality ◽  
Skilled Nursing Facility ◽  
The Elderly ◽  
Primary Diagnosis ◽  
Categorical Variables ◽  
Continuous Variables ◽  
National Inpatient Sample ◽  
Chi Square ◽  
Student’S T

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Atrial fibrillation is a common disorder in the elderly population and a known risk factor for stroke and dementia. Purpose  To study the burden of dementia in Afib hospitalizations and identify the predictors of in-hospital mortality in Afib with dementia hospitalizations. Methods We queried January 2016 to December 2018 National Inpatient Sample (NIS) to identify adult (≥18 yrs) hospitalizations with a primary diagnosis of Afib. Hospitalizations of Afib with dementia was compared with Afib without dementia. We used the Chi-square test for differences between categorical variables, and Student’s t-test for continuous variables. Multivariate logistic regression was used in outcomes analysis to adjust for potential hospital and patient-level confounders. Results  We identified 1,236,540 weighted Afib hospitalizations across three years. Of which, 79,405 (6.4%) of them were associated with dementia. Afib with dementia hospitalizations were associated with older age (mean age 83.2 vs 70.0 yrs; P &lt; 0.01), higher rate of chronic Afib (15.3% vs 7.5%; P &lt; 0.01), higher rate of comorbidity (% of &gt;3 Elixhauser comorbidity score 91.8% vs 83.6%; P &lt; 0.01). After adjusting for patient and hospital-level characteristics, we observed that Afib with dementia hospitalizations was associated with higher odds of in-hospital mortality compared to Afib without dementia [Odds Ratio (OR): 1.6 (1.4 – 1.9); P &lt; 0.01]. We also observed statistically significant association with increased LOS [4.7 vs 3.2 days; P &lt; 0.01], repeated falls [OR: 2.8 (2.5 – 3.1); P &lt; 0.01] and protein calorie malnutrition [OR: 1.9 (1.7 – 2.0); P &lt; 0.01] in Afib with dementia group. Conclusion Afib with dementia hospitalizations are not only associated with higher mortality, but they are also associated with higher repeated fall rates, and skilled nursing facility discharge dispositions. Co-morbidities like hypertension, CKD, obesity, HFrEF, HFpEF, OSA are associated with higher in-hospital mortality. Our study findings emphasize the burden of dementia in Afib hospitalizations and the need for prevention of poor outcomes in this population.

scholarly journals Self-harm on a specialist adult eating disorder unit: a retrospective cohort study of patient characteristics and outcomes

BJPsych Open ◽  
2021 ◽  
Vol 7 (S1) ◽  
pp. S257-S257
Author(s):  
Leah Holm-Mercer ◽  
Douglas Kohler ◽  
Agnes Ayton
Keyword(s):  
Eating Disorders ◽  
Eating Disorder ◽  
General Hospital ◽  
Patient Characteristics ◽  
Primary Diagnosis ◽  
Categorical Variables ◽  
Hospital Treatment ◽  
Continuous Variables ◽  
History Of ◽  
Self Harm

AimsDeliberate self-harm (DSH) is common but rarely studied among inpatients with eating disorders. We sought to investigate the frequency of DSH among inpatients in a specialist adult eating disorders unit, and the association of DSH with comorbidities and treatment outcomes. We also investigated changes in these parameters during the pandemic.MethodWe included the records of 70 patients consecutively admitted to Cotswold House in Oxford between April 2018 and November 2020. Data were analysed using Microsoft Excel using descriptive statistics. For comparisons, student T-tests were used for continuous variables and Chi-square tests used for categorical variables.Result99% of patients were female; their ages ranged from 17 to 67 years (mean 30.7). 81% had a primary diagnosis of anorexia nervosa, and 67% had a history of DSH prior to admission. There was a total of 100 incidences of DSH, of which 12% required transfer to a general hospital for medical treatment.Frequency of self-harm decreased with time throughout admission (17% self-harming on admission, vs 7% at discharge, p = 0.043).Compared to those with no history of DSH, patients who self-harmed during admission were more likely to be detained under the Mental Health Act (45% vs 17.4%, p = 0.003), and to have psychiatric comorbidities (85% vs 35%, p = 0.001). Patients whose self-harm required transfer for general hospital treatment had a lower mean discharge BMI (18.18kg/m2 vs 20.23kg/m2, p = 0.039), longer admission (105.9 days vs 78.1 days, p = 0.037), and gained weight at a slower rate (0.26kg/m2/week vs 0.43kg/m2/week, p = 0.048) than those who did not require transfer.During the pandemic, the frequency of DSH doubled on the ward. Overall outcomes were similar, however mean length of admission was lower during the pandemic (67.83 vs 89.94 days, p = 0.046), and patients regained weight more rapidly (0.43kg/m2/week vs 0.28kg/m2/week, p = 0.003) than prior to it.ConclusionSelf-harm during admission was seen in 29% of patients and was associated with the presence of comorbid psychiatric diagnoses. The frequency of DSH much reduced between admission and discharge, suggesting a beneficial effect of treatment. Medical transfer for DSH, considered as a proxy measure for severity, predicted poorer outcomes in weight restoration. We also noted an increase in rates of DSH during the pandemic, which may have resulted from a combination of increased psychosocial stressors and a reduction in admission capacity in eating disorder units.

scholarly journals Characteristics of patients infected with SARS COVID-19 admitted at COVID health facilities in one of the aspirational district of India: a retrospective study

2020 ◽  
Vol 7 (11) ◽  
pp. 4435
Author(s):  
Ekta Sharma ◽  
Gurmeet Katoch ◽  
Rajesh Guleri ◽  
Jalam Bhardwaj
Keyword(s):  
Large Scale ◽  
Clinical Laboratory ◽  
Descriptive Analysis ◽  
Laboratory Data ◽  
Age Groups ◽  
The Elderly ◽  
Categorical Variables ◽  
Close Monitoring ◽  
Continuous Variables ◽  
Wide Range

Background: COVID-19 is the third corona virus that has emerged among the human population in the last two decades. The main aim of this study was to describe the epidemiologic features, clinical presentation of first 52 patients diagnosed with polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infection admitted at COVID health facilities.Methods: A retrospective descriptive study was conducted over a period of three months from 1st April 2020 to 30th June 2020. We obtained demographic, epidemiological, clinical, laboratory data from the medical records of patients infected with SARS-Cov-2. The categorical variables were expressed in terms of frequency and percentages and the continuous variables were expressed as mean and standard deviation. In addition to descriptive analysis, Pearson’s chi- square test was applied to ascertain the associations between certain variables.Results: The mean age of participants was 29±11.67 years with a male preponderance. Forty three (83%) patients had travel history within India in the previous 30 days i.e. from Delhi (35%), Haryana (15%), Tamilnadu (11%), Himachal Pradesh (8%), Maharashtra (1.9%), Punjab (8%), and Uttar Pradesh (4%). Majority of the patients (90%) were asymptomatic. The age group of 21-30 years was the most affected group (44%) as comparison to the other age groups. No mortality was reported and 100% recovery rate was found.Conclusions: In conclusion, COVID-19 affects a wide-range of patients, from youth to the elderly.  In this study, all the COVID-19 infected patients were classified as mild as most were asymptomatic. Close monitoring and large-scale control strategies will be needed to prevent widespread transmission within the community.

Abstract 20: Patient Characteristics and Outcomes among Early-Arriving Acute Stroke Patients for Whom Alteplase Was Dispensed but Not Administered

2015 ◽  
Vol 8 (suppl_2) ◽  
Author(s):  
Khawja A Siddiqui ◽  
Syed F Ali ◽  
Christopher D Anderson ◽  
Natalia Rost ◽  
Lee H Schwamm
Keyword(s):  
Best Practice ◽  
Patient Characteristics ◽  
Clinical Decision ◽  
Categorical Variables ◽  
Continuous Variables ◽  
High Likelihood ◽  
Stroke Patients ◽  
Rapid Improvement ◽  
Chi Square ◽  
Chi Square Test

Introduction: Best practice recommendations such as the AHA Target Stroke campaign suggest that IV tPA be dispensed as early as possible prior to final clinical decision when evaluating stroke patients at high likelihood of tPA eligibility. This practice may reduce door to needle times. We sought to determine the patient characteristics, outcomes and reasons for non-treatment among this selected group of individuals presenting to our stroke service. Methods: During the period of April 2007 till June 2014, 4170 patients presented to our hospital with acute Ischemic stroke, and 1152/4180 (28%) presented within 4 hours of last seen well (LKW). Using our hospital’s GWTG database and pharmacy logs, we identified all tPA treated cases as well as those in whom tPA was dispensed but not given. Categorical variables were compared using chi- square test and continuous variables were compared using Student t-test and Wilcoxon rank sum test. Results: Of 382 patients for whom tPA was dispensed, 239 (62.6%) ultimately received tPA. Compared to tPA treated patients, the 143 untreated patients in this cohort were more likely to have a history of previous stroke, present with weakness or language disturbance, and had a lower median NIHSS (Table). Untreated patients were more likely to be transferred from another hospital than to present directly from the scene, had a shorter median time from ED arrival to tPA dispensed, were less likely to undergo intra-arterial thrombolysis but had similar rates of home discharge and mortality. The most common reasons for non-treatment were mild symptoms (28%), rapid improvement (15%) and arriving after “drip and ship” (8%). In multivariate analysis, only lower NIHSS [aOR 0.93 per point (0.90, 0.97), p<0.001] and arrival via emergency medical services from scene [aOR 0.40 (0.24, 0.66), p <0.001] were independently associated with non-treatment. Conclusion: Delays in tPA initiation are common, and reduce health outcomes after thrombolysis. In our study, early dispensing of tPA prior to the final eligibility assessment produced an overall rate of tPA treatment of >60%. This conversion rate suggests that early tPA dispensing in high likelihood candidates is appropriate and supports the use of this strategy. Reasons for non-treatment were similar to those in unselected patients.

Abstract 1122‐000126: Mobile Stroke Unit Process Metrics in Large Vessel Occlusion Stroke Patients: BEST‐MSU Substudy

2021 ◽  
Vol 1 (S1) ◽  
Author(s):  
Alexandra L Czap ◽  
Anne W Alexandrov ◽  
May Nour ◽  
Noopur Singh ◽  
Mengxi Wang ◽  
...  
Keyword(s):  
Stroke Unit ◽  
Large Vessel ◽  
Categorical Variables ◽  
Continuous Variables ◽  
Stroke Patients ◽  
Vessel Occlusion ◽  
Stroke Treatment ◽  
Symptomatic Intracerebral Hemorrhage ◽  
Mobile Stroke Unit ◽  
Process Metrics

Introduction : Mobile Stroke Units (MSUs) speed thrombolytic treatment for acute ischemic stroke and improve clinical outcomes compared to standard management by Emergency Medical Services (EMS). However, MSU process metrics in the subset of patients with large vessel occlusions (LVOs) having endovascular thrombectomy (EVT) have yet to be optimized. Methods : A pre‐specified Benefits of Stroke Treatment Using a Mobile Stroke Unit (BEST‐MSU) substudy of tPA‐eligible stroke patients with imaging evident LVOs was conducted. The primary outcome was process metrics related to treatment times from stroke onset and first medical alert. Safety outcomes included rates of symptomatic intracerebral hemorrhage and procedural complications. Groups were compared using Chi‐square or Fisher’s exact tests for categorical variables, and Wilcoxon rank‐sum tests for continuous variables. Results : A total of 295 patients were included, 169 in the MSU group and 126 in the EMS group. Baseline characteristics were comparable between the groups, with the exception of baseline NIHSS (MSU mean 19.0 [IQR 13.0,23.0] vs EMS 16.0 [11.0, 20.0], p = 0.003). 92% of MSU and 87% of EMS LVO patients received tPA, and 78% and 85% went on to have EVT. Process metrics are detailed in Table 1. MSU LVO patients had faster tPA bolus from 911‐alert (MSU 45.0 minutes [40.0, 53.5] vs EMS 76.0 [64.0, 87.8], p<0.001), however the two groups had similar alert to groin puncture (MSU 142.5 [116.8, 171.0] versus EMS 131.5 [114.0, 159.8], p = 0.15). MSU patients spent more time on‐scene, (EMS arrival to ED arrival, 53.0 [45.0, 62.0] vs 27.0 [22.0, 33.0], p<0.001) however less time prior to EVT (door to groin puncture, 76.5 [54.8, 108.5] vs 94.0 [72.0, 123.0], p<0.001) with variable use of field CTAs and direct cath lab admission with ED bypass, yielding a net neutral result. The variability among site protocols is reflected in the range of median alert to groin puncture times (minimum 107.0 minutes, maximum 152.0). In the 222 patients undergoing EVT, median alert to recanalization time was 181.5 minutes [146.8, 225.5] in the MSU group and 190.5 [157.5, 227.5] in the EMS group (p = 0.47). Recanalization (Thrombolysis In Cerebral Infarction [TICI] 2b/3) was achieved in 76% of MSU and 70% of EMS (p = 0.32) with comparable rates of EVT complications (including hemorrhage, perforation, dissection, hematoma). 54% MSU and 44% of EMS LVO patients achieved good functional outcome (modified Rankin Scale [mRS] ≤ 2) at 90 days (p = 0.11). Conclusions : In tPA‐eligible LVO stroke patients, MSU management did not increase or expedite EVT treatment times as compared to standard EMS management. Future MSU processes should include field CTA with direct admission to cath labs to maximize the early treatment advantage this technology provides.

scholarly journals Substantial Variation in Peripherally Inserted Central Catheter Use and Outcomes in Patients with Hematologic Malignancies: A Multi-Center Study

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 4038-4038
Author(s):  
Marcus Geer ◽  
Urvashi Mitbander ◽  
Knut Taxbro ◽  
Qisu Zhang ◽  
Megan O'Malley ◽  
...  
Keyword(s):  
Vascular Access ◽  
Hematologic Malignancy ◽  
Hospital Medicine ◽  
Hematologic Malignancies ◽  
Categorical Variables ◽  
Continuous Variables ◽  
Catheter Occlusion ◽  
Major Complications ◽  
Primary Indication ◽  
Wide Range

Abstract Background: Use of peripherally inserted central catheters (PICCs) has grown rapidly in patients with hematologic malignancies. Studies demonstrating the safety of PICCs in hematologic malignancies, however, are largely single-center, retrospective designs focused on the outpatient setting. Little is known about inpatient PICC use in patients with hematologic malignancies and how such use varies across hospitals. Methods: Data was prospectively collected between November 2013 and December 2019 from a cohort of patients admitted at one of 42 Michigan hospitals participating in the Michigan Hospital Medicine Safety Consortium (HMS). Patients had a diagnosis of a hematologic malignancy and had a PICC placed during their hospital stay. Adult medical patients admitted to a general ward or intensive care unit were eligible for data collection and data were collected from the medical record using a standardized template. The indications for PICC placement, catheter characteristics, and associated complications from 17 hospitals with more than 25 discrete patients with hematologic malignancy were included in a comparative analysis. Major complications were defined as central line associated bloodstream infection (CLABSI), catheter occlusion, and venous thromboembolism (VTE). Differences across hospitals were tested using the Kruskal-Wallis test for continuous variables and Pearson chi-square test for categorical variables. Results: A total of 2092 PICCs placed in 1798 patients were included in the analysis with a median (IQR) dwell time of 15 (6-30) days. Most patients were male (n=1242, 59.4%), white (n=1720, 82.2%), and 50 to 69 years old (n=950, 45.4%). Patients were primarily admitted to large hospitals (≥375 beds: n=1429, 68.8%) that were teaching centers (n=1600, 76.5%) in metropolitan locations (n=2000, 95.6%). Leukemia (n=752, 36.0%) and non-Hodgkin's lymphoma (n=409, 19.6%) were the most represented malignancies. The most common primary indication for PICC placement was administration of chemotherapy (n=1180, 56.4%). The majority of PICCs were double lumen (n=1457, 69.6%), most often placed by vascular access nurses (n=1520, 72.7%). A concurrent central venous catheter was present at the time of PICC placement in 12.2% (n=264) of patients. A major complication event occurred in over 1 in 4 PICCs placed (n=562, 26.9%). Catheter occlusion occurred in 17.8% (n=372), CLABSI in 8.2% (n=171), and VTE in 3.8% (n=80) of PICCs respectively. There was wide variation in PICC indications, characteristics, and outcomes across hospitals. Placement of PICCs for chemotherapy varied from 33.3% to 85.3% (p&lt;0.001). Similarly, there was a wide range of placement for antibiotics (2.8% to 32.4%, p&lt;0.001), transfusion of blood products (0.0% to 9.7%, p&lt;0.001), medications requiring central access (0% to 26.7%, p&lt;0.001), difficult access and blood draws (1.8% to 33.3%, p&lt;0.001), and administration of total parenteral nutrition (0.0% to 14.3%, p=0.002). There was significant variation in the number of lumens used: single (5.6% to 37.9%, p&lt;0.001), double (51.9% to 93.0%, p&lt;0.001), or triple (0.0% to 30.8%, p&lt;0.001). The incidence of major complications spanned from 9.7% to 40.8% (p=0.001). Rate of catheter occlusion had the widest range (0.0% to 36.7%, p&lt;0.001). The differences between rates of VTE (0.0% to 8.7%, p=0.29) and CLABSI (1.4% to 15.9%, p&lt; 0.20) were not statistically significant. Patient mortality ranged from 2.8% to 19.4% (p&lt;0.001). Conclusion: Appropriate venous access is critical to the care of patients with hematologic malignancy. This study demonstrates wide variation in the practice patterns and outcomes for PICCs in patients with hematologic malignancies across hospitals in Michigan. Further work is necessary to further understand and improve decision making around choosing vascular access in this vulnerable population. Figure 1 Figure 1. Disclosures Zhang: SIMR, Inc: Ended employment in the past 24 months, Research Funding; AmerisourceBergen: Current Employment. Sood: Bayer: Consultancy.

scholarly journals Obesity paradox in HFrEF hospitalizations: analysis of national inpatient sample 2016-2018

2021 ◽  
Vol 28 (Supplement_1) ◽  
Author(s):  
H Thyagaturu ◽  
K Shah ◽  
S Li ◽  
S Thangjui ◽  
B Shrestha ◽  
...  
Keyword(s):  
Hospital Mortality ◽  
Systolic Heart Failure ◽  
Obesity Paradox ◽  
Primary Diagnosis ◽  
Categorical Variables ◽  
Continuous Variables ◽  
National Inpatient Sample ◽  
Chi Square ◽  
Significant Difference ◽  
Student’S T

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Obesity is a common and known risk factor for many cardiovascular diseases. Prior studies on chronic systolic heart failure have demonstrated that obesity is inversely associated with mortality, the so-called obesity paradox. Purpose  To study the phenomenon of obesity paradox in HFrEF hospitalizations Methods We queried January 2016 to December 2018 National Inpatient Sample (NIS) to identify adult (≥18 yrs) hospitalizations with a primary diagnosis of HFrEF. Patients with an associated diagnosis of obesity and higher BMI were also identified based on appropriate ICD-10 CM codes. We used the Chi-square test to evaluate the differences between binary or categorical variables, and Student’s t-test for continuous variables. Multivariate logistic regression was used in outcomes analysis to adjust for potential hospital and patient-level confounders. Results  We identified 639,944 weighted HFrEF hospitalizations across three years. Of which, 130,949 (20.4%) of them were associated with obesity. HFrEF with obesity hospitalizations were associated with younger age (mean age 62.0 vs 70.7 yrs; P &lt; 0.01), lesser CAD (55% vs 61%; P &lt; 0.01), higher rate of comorbidity (% of &gt;3 Elixhauser comorbidity score 99.2% vs 94.1%; P &lt; 0.01) and higher Medicaid primary payer (18.2% vs 12.7%, P &lt; 0.01). After adjusting for patient and hospital-level characteristics, we observe statistically significant difference in odds of in-hospital mortality when HFrEF with obesity hospitalizations was compared to HFrEF without obesity [Odds Ratio (OR): 1.1 (0.8 – 1.5); P = 0.52]. We observed statistically significant association with increased LOS [6.0 vs 5.3 days; P &lt; 0.01], increased total hospitalization charges [US$ 61524 vs 55677; P &lt; 0.01] and decreased coronary catheterizations [OR: 0.7 (0.5 – 0.9); P = 0.01] in HFrEF with obesity group compared to HFrEF without obesity. Conclusion In this retrospective cohort of hospitalized patients with HFrEF, higher BMI and obesity was not associated with in-hospital mortality. However, it was associated with longer LOS and higher total hospitalization charges. HFrEF with obesity hospitalizations are associated with lesser left coronary catheterizations. This may be explained by lesser burden of CAD in this patient population.

Preoperative MRI findings and prediction of diagnostic utility of foramen ovale electrodes

2020 ◽  
Vol 132 (3) ◽  
pp. 692-699 ◽  
Author(s):  
Sarah K. Bick ◽  
Marjan S. Dolatshahi ◽  
Benjamin L. Grannan ◽  
Andrew J. Cole ◽  
Daniel B. Hoch ◽  
...  
Keyword(s):  
Minimally Invasive ◽  
Treatment Decision ◽  
Temporal Lobectomy ◽  
Categorical Variables ◽  
Continuous Variables ◽  
Foramen Ovale ◽  
Mri Findings ◽  
Noninvasive Methods ◽  
Diagnostic Modality ◽  
Diagnostic Investigations

OBJECTIVEForamen ovale electrodes (FOEs) are a minimally invasive method to localize mesial temporal seizures in cases in which noninvasive methods are inconclusive. The objective of this study was to identify factors predicting the ability of FOEs to yield a diagnosis in order to determine optimal candidates for this procedure.METHODSAll cases of diagnostic investigations performed with FOEs at the authors’ institution between 2005 and 2017 were reviewed. FOE investigation was defined as diagnostic if it led to a treatment decision. Demographic and clinical variables for diagnostic and nondiagnostic investigations were compared using a Wilcoxon rank-sum test for continuous variables and Fisher’s exact test for categorical variables.RESULTSNinety-three patients underwent investigations performed with FOEs during the study period and were included in the study. FOE investigation was diagnostic in 75.3% of cases. Of patients who underwent anterior temporal lobectomy following diagnostic FOE evaluation, 75.9% were Engel class I at last follow-up (average 40.1 months). When the diagnostic and nondiagnostic FOE groups were compared, patients who had diagnostic investigations were more likely to be male (57.1% male vs 26.1% in the nondiagnostic group, p = 0.015). They were also more likely to have temporal lesions on preoperative MRI (p = 0.018).CONCLUSIONSFOEs are a useful, minimally invasive diagnostic modality resulting in a treatment decision in 75% of cases. Male patients and patients with temporal lesions on MRI may be most likely to benefit from FOE investigation.

Sign in / Sign up

Export Citation Format

Share Document