The rebound of platelet count could be a predictor of good prognosis of sepsis in the intensive care unit: a retrospective analysis of the large clinical database MIMIC-III

2021 ◽  
Author(s):  
Fa Huang ◽  
Jinghong Xu ◽  
Li Tong ◽  
Xiaoguang Hu ◽  
Ka Yin Lui ◽  
...  
Keyword(s):  
Platelet Count ◽  
Good Prognosis ◽  
Diverse Population ◽  
Proportional Hazard ◽  
Clinical Database ◽  
Risk Of Death ◽  
Cox Proportional Hazard ◽  
The Difference ◽  
Cox Proportional Hazard Regression ◽  
Mimic Iii

Abstract Background: The rise of platelets in sepsis patients with thrombocytopenia can lead to an improvement in clinical outcomes. We aimed to probe whether the rebound of platelet count could be a predictor of good prognosis in sepsis patients in a large, diverse population.Methods: All sepsis patients were initially stratified according to nadir platelet counts (very low<50´109/L, intermediate-low 50 ´ 109 to 99 ´ 109/L, low 100 ´ 109 to 149 ´ 109/L, normal 150 ´ 109 to 399 ´ 109/L, or thrombocytosis > 400 ´ 109/L). The delta platelet count (DPC) was defined as the difference between the last platelet count prior to transfer or death and the nadir platelet count after ICU admission. Results: A total of 3457 patients were enrolled in our study. The 28-day mortality in the very low (43.1%) and intermediate-low (36.9%) platelet count groups was higher than in the low (26.8%) and normal (23.2%) platelet count groups and thrombocytosis (18.2%) group (P <0.001). The patients in the ΔPC > 0 subgroup had lower 28-day morality (38.5% vs. 59.1%, P < 0.001, 33.3% vs. 44.7%, P = 0.015, 23.8% vs. 32.7% P = 0.01, 20.2 vs. 27.7, P = 0.001, respectively) except in the thrombocytosis group. The extended Cox proportional hazard regression model showed a decreased risk of death within 28 days in patients in the ΔPC > 0 subgroup (HR 0.570, 95% CI 0.498-0.651, P < 0.001).Conclusions: The rebound platelet count could be a biomarker of good prognosis in patients with sepsis.

Association of Dementia-Related Psychosis With Long-term Care Use and Death

Neurology ◽  
2021 ◽  
Vol 96 (12) ◽  
pp. e1620-e1631
Author(s):  
James B. Wetmore ◽  
Yi Peng ◽  
Heng Yan ◽  
Suying Li ◽  
Muna Irfan ◽  
...  
Keyword(s):  
Long Term Care ◽  
Fold Increase ◽  
Proportional Hazard ◽  
Comorbid Conditions ◽  
Term Care ◽  
Risk Of Death ◽  
Cox Proportional Hazard ◽  
Increased Risk ◽  
Cox Proportional Hazard Regression

ObjectiveTo determine the association of dementia-related psychosis (DRP) with death and use of long-term care (LTC); we hypothesized that DRP would be associated with increased risk of death and use of LTC in patients with dementia.MethodsA retrospective cohort study was performed. Medicare claims from 2008 to 2016 were used to define cohorts of patients with dementia and DRP. Outcomes were LTC, defined as nursing home stays of >100 consecutive days, and death. Patients with DRP were directly matched to patients with dementia without psychosis by age, sex, race, number of comorbid conditions, and dementia index year. Association of DRP with outcomes was evaluated using a Cox proportional hazard regression model.ResultsWe identified 256,408 patients with dementia. Within 2 years after the dementia index date, 13.9% of patients developed DRP and 31.9% had died. Corresponding estimates at 5 years were 25.5% and 64.0%. Mean age differed little between those who developed DRP (83.8 ± 7.9 years) and those who did not (83.1 ± 8.7 years). Patients with DRP were slightly more likely to be female (71.0% vs 68.3%) and white (85.7% vs 82.0%). Within 2 years of developing DRP, 16.1% entered LTC and 52.0% died; corresponding percentages for patients without DRP were 8.4% and 30.0%, respectively. In the matched cohort, DRP was associated with greater risk of LTC (hazard ratio [HR] 2.36, 2.29–2.44) and death (HR 2.06, 2.02–2.10).ConclusionsDRP was associated with a more than doubling in the risk of death and a nearly 2.5-fold increase in risk of the need for LTC.

A Dynamic Prognostic Model To Predict Survival in Post-Polycythemia Vera Myelofibrosis.

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 2547-2547
Author(s):  
Francesco F. Passamonti ◽  
Elisa E. Rumi ◽  
Marianna M. Caramella ◽  
Chiara C. Elena ◽  
Luca L. Arcaini ◽  
...  
Keyword(s):  
Platelet Count ◽  
Prognostic Factors ◽  
Polycythemia Vera ◽  
Median Survival ◽  
Prognostic Model ◽  
Time Dependent ◽  
Proportional Hazard ◽  
Cox Proportional Hazard ◽  
Cox Proportional Hazard Regression

Abstract Polycythemia vera (PV) is a chronic myeloproliferative disorder with a propensity to develop myelofibrosis, a condition named post polycythemia vera myelofibrosis (post-PV MF). Survival and prognostic factors after transition to MF remain to be defined. We studied 68 patients with post-PV MF to define survival and prognostic factors for survival at diagnosis of post-PV MF. We also developed a dynamic prognostic model to predict survival at any time from diagnosis of post-PV MF. The median interval between the diagnosis of PV and that of post-PV MF was 13 years (range, 4–29.6 years). Patients with post-PV MF were observed for 181 person-years of follow-up. At diagnosis of post-PV MF, 43 (63%) of 68 patients had less than 65 years. During the follow-up, the incidence of thrombosis was 42 × 1000 person-years (95% CI: 19–93.5) and the incidence of leukemia was 50.3 × 1000 person-years (95% CI: 26–115). The median survival was 5.7 years. Multivariable Cox proportional hazard regression including age, hemoglobin value, platelet count, leukocyte count, and spleen size, showed that hemoglobin &lt; 10 g/dL (P &lt; .001) and platelet count &lt; 100 × 109/L (P= .026) were independent risk factors for survival. We stratified patients at diagnosis of post-PV MF, according to these factors, obtaining two risk groups with significantly different survival (P = .003): low risk (Hb &gt; 10 g/dL and platelet count &gt; 100 × 109/L) with a median survival of 7 years, and high risk (Hb &lt; 10 g/dL or platelet count &lt; 100 × 109/L) with a median survival of 2 years. The prognostic model retained significance after adjustment for age in a multivariable Cox proportional hazard regression (HR: 4.3, 95% CI: 1.6–11.4; P= .003). To assess whether this prognostic model may predict survival at any time from diagnosis of post-PV MF, we evaluated in a time-dependent analysis 64 patients who had longitudinal blood cell counts during follow-up. As first step, we evaluated univariate survival analysis with hemoglobin value &lt; 10 g/dL and platelet count &lt; 100 ×109/L as time-dependent covariates. Both time-dependent parameters affected survival (HR for hemoglobin 5.8, 95% CI: 2.2–15.2, P &lt; 0.001; HR for platelets 4.5, 95% CI: 1.67-12, P=.002). As second step, we evaluated the prognostic model assessed at diagnosis as time-dependent covariate, to define whether the acquisition of one risk factor during follow-up may affect survival. The HR was 7.5 (95% CI: 2.4-23.4; P &lt; .001). The time-dependent prognostic model retained statistical significance after adjustment for age (P &lt; .001). In conclusion, in patients developing post-PV myelofibrosis, a prognostic model based on hemoglobin level &lt; 10 g/dL and platelet count &lt; 100 × 109/L may predict survival at diagnosis of post-PV MF and at any time thereafter.

scholarly journals Platelet Count Trajectory and Mortality in Septic Shock: A Retrospective Cohort Study

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 16-16
Author(s):  
Neelan Sriranjan ◽  
Brett L. Houston ◽  
Emily Rimmer ◽  
Chantalle Menard ◽  
Murdoch Leeies ◽  
...  
Keyword(s):  
Septic Shock ◽  
Platelet Count ◽  
Retrospective Cohort ◽  
Trajectory Analysis ◽  
Cox Proportional Hazard Model ◽  
Proportional Hazard ◽  
Proportional Hazard Model ◽  
Risk Of Death ◽  
Cox Proportional Hazard ◽  
Organ Failures

Background: In Canada, septic shock accounts for approximately 30,000 hospitalizations annually and is associated with a mortality rate of 30%. Thrombocytopenia in septic shock is associated with a poor prognosis including increased length of stay, longer duration of organ support, increased major bleeding events and mortality. The trajectory of the platelet count over time in patients with septic shock has not been well-studied. We hypothesized that the platelet count trajectory in septic shock can identify distinct clinical groups and is an independent predictor of 30-day mortality. Objectives: 1) To identify groups of patients with distinct platelet count trajectories; 2) To evaluate patient and illness factors associated with platelet count trajectories; and 3) To estimate the association of platelet count trajectory with mortality patients with septic shock. Methods: We performed a retrospective cohort study of adult patients admitted with septic shock to an intensive care unit (ICU) in Winnipeg, Canada between 2006-2014. We used group-based trajectory analysis to analyze the trend of platelet count over the first seven days of ICU admission to group patients with similar platelet trajectories. Group-based trajectory analysis is a statistical method that analyzes the pattern of a variable over time and allows distinct groups with similar trajectories to arise from the data. We utilized both the Bayesian Information Criterion (BIC) and clinical validity characteristics to choose the most suitable trajectory model. We developed a multinomial logistic regression model to associate patient characteristics with platelet count trajectories. We created a multivariable Cox proportional hazard model adjusted for age, sex, Acute Physiology and Chronic Health Evaluation (APACHE) II score, comorbidities, site or source of infection, and time to first appropriate antimicrobial to examine the association between platelet count trajectory and 30-day mortality. Results: Our study cohort included 913 patients with septic shock. The favoured trajectory model identified six distinct trajectories (Figure 1) using the platelet count over the first 7 days of ICU admission. We found that the number of organ failures on day 1was independently associated with platelet count trajectory, while other characteristics were not. The 30-day mortality of the entire cohort was 26.2% and ranged from 16.4% in group 1 (rising platelet count) to 44.4% in group 6 (high platelet count throughout). In the multivariable Cox proportional hazard model, compared with group 2 (thrombocytopenia), group 4 (high normal platelet count) was independently associated with a reduced risk of death at 30 days (Hazard Ratio (HR) 0.33, p = 0.002). The trajectory group with thrombocytosis (group 6) was associated with an increased risk of death at 30 days (HR 3.24, p=0.48) however the small number in this group limits the generalizability of this finding. Conclusion: We identified 6 distinct and clinically relevant platelet count trajectories in critically ill patients with septic shock. Platelet count trajectory was associated with the number of organ failures on day 1. Our study confirms that thrombocytopenia is associated with a worse prognosis as other trajectories with higher platelet count were associated with a lower risk of death. While it is well recognized that thrombocytopenia is associated with adverse outcomes in patients with septic shock, it is not known whether other patterns of the platelet trajectory such as thrombocytosis are similarly clinically important. Further studies are needed to fully characterize the impact platelet count trajectory on outcomes in patients with septic shock. The interplay between platelet count trajectory and other parameters (such as the white blood cell count trajectory, or INR trajectory) may have a more predictive role in evaluating prognosis in sepsis. Disclosures No relevant conflicts of interest to declare.

scholarly journals Low serum parathyroid hormone is a risk factor for peritonitis episodes in incident peritoneal dialysis patients: a retrospective study

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Yuqi Yang ◽  
Jingjing Da ◽  
Yi Jiang ◽  
Jing Yuan ◽  
Yan Zha
Keyword(s):  
Risk Factor ◽  
Peritoneal Dialysis ◽  
Parathyroid Hormone ◽  
Proportional Hazard ◽  
Serum Parathyroid Hormone ◽  
Cox Proportional Hazard ◽  
Cox Proportional Hazard Regression ◽  
Gram Negative Organisms ◽  
Serum Pth

Abstract Background Serum parathyroid hormone (PTH) levels have been reported to be associated with infectious mortality in peritoneal dialysis (PD) patients. Peritonitis is the most common and fatal infectious complication, resulting in technique failure, hospital admission and mortality. Whether PTH is associated with peritonitis episodes remains unclear. Methods We examined the association of PTH levels and peritonitis incidence in a 7-year cohort of 270 incident PD patients who were maintained on dialysis between January 2012 and December 2018 using Cox proportional hazard regression analyses. Patients were categorized into three groups by serum PTH levels as follows: low-PTH group, PTH < 150 pg/mL; middle-PTH group, PTH 150-300 pg/mL; high-PTH group, PTH > 300 pg/mL. Results During a median follow-up of 29.5 (interquartile range 16–49) months, the incidence rate of peritonitis was 0.10 episodes per patient-year. Gram-positive organisms were the most common causative microorganisms (36.2%), and higher percentage of Gram-negative organisms was noted in patients with low PTH levels. Low PTH levels were associated with older age, higher eGFR, higher hemoglobin, calcium levels and lower phosphate, alkaline phosphatase levels. After multivariate adjustment, lower PTH levels were identified as an independent risk factor for peritonitis episodes [hazard ratio 1.643, 95% confidence interval 1.014–2.663, P = 0.044]. Conclusions Low PTH levels are independently associated with peritonitis in incident PD patients.

scholarly journals AB0292 EFFICACY AND SAFETY OF TWO BIOSIMILAR ETANERCEPT AFTER THE SWITCH FROM THEIR CORRESPONDING ORIGINATOR IN THE TREATMENT OF PATIENTS WITH AUTOIMMUNE ARTHRITIS; A RETROSPECTIVE ANALYSIS IN A REAL LIFE SETTING

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 1445.1-1445
Author(s):  
F. Girelli ◽  
A. Ariani ◽  
M. Bruschi ◽  
A. Becciolini ◽  
L. Gardelli ◽  
...  
Keyword(s):  
Regression Analysis ◽  
Disease Duration ◽  
Retention Rate ◽  
Real Life ◽  
Autoimmune Arthritis ◽  
Proportional Hazard ◽  
Hazard Regression ◽  
Cox Proportional Hazard ◽  
Real Life Setting ◽  
Cox Proportional Hazard Regression

Background:The available biosimilars of etanercept are as effective and well tolerated as their bio originator molecule in the naive treatment of chronic autoimmune arthritis. More data about the switching from the bio originator are needed.Objectives:To compare the clinical outcomes of the treatment with etanercept biosimilars (SB4 and GP2015) naïve and after the switch from their corresponding originator in patients affected by autoimmune arthritis in a real life settingMethods:We retrospectively analyzed the baseline characteristics and the retention rate in a cohort of patients who received at least a course of etanercept (originator or biosimilar) in our Rheumatology Units from January 2000 to January 2020. We stratified the study population according to biosimilar use. Descriptive data are presented by medians (interquartile range [IQR]) for continuous data or as numbers (percentages) for categorical data. Drug survival distribution curves were computed by the Kaplan-Meier method and compared by a stratified log-rank test. A Cox proportional hazards regression analysis stratified by indication, drug, age, disease duration, sex, treatment line, biosimilar use and prescription year was performed. P values≤0.05 were considered statistically significant.Results:477 patients (65% female, median age 56 [46-75] years, median disease duration 97 [40.25-178.75] months) treated with etanercept were included in the analysis. 257 (53.9%) were affect by rheumatoid arthritis, 139 (29.1%) by psoriatic arthritis, and 81 (17%) by axial spondylarthritis. 298 (62.5%) were treated with etanercept originator, 97 (20.3%) with SB4, and 82 (17.2%) with GP2015. Among the biosimilars 90/179 (50.3%) patients were naïve to etanercept treatment. Among the 89 switchers we observed 8 treatment discontinuations: one due to surgical infection complication, three due to disease flare, two due to subjective worsening and one due to remission. The overall 6- and 12-month retentions rate were 92.8% and 80.2%. The 6- and 12-month retention rate for etanercept, SB4 and GP2015 were 92.7%, 93.4% and 90.2%, and 82%, 74.5% and 88.1% respectively, without significant differences among the three groups (p=0.374). Patients switching from originator to biosimilars showed and overall higher treatment survival when compared to naive (12-month retention rate 81.2% vs 70.8%, p=0.036). The Cox proportional hazard regression analysis highlighted that the only predictor significantly associated with an overall higher risk of treatment discontinuation was the year of prescription (HR 1.08, 95% CI 1.04 to 1.13; p<0.0001).Conclusion:In our retrospective study etanercept originator and its biosimilars (SB4 and GP2015) showed the same effectiveness. Patients switching from originator to biosimilar showed an significant higher retention rate when compared to naive. The only predictor of treatment discontinuation highlighted by the Cox proportional hazard regression analysis was the year of treatment prescription.Disclosure of Interests:Francesco Girelli: None declared, Alarico Ariani: None declared, Marco Bruschi: None declared, Andrea Becciolini Speakers bureau: Sanofi-Genzyme, UCB and AbbVie, Lucia Gardelli: None declared, Maurizio Nizzoli: None declared

Manual occupations with high all-cause mortality: The contribution of socioeconomic and occupational characteristics

2020 ◽  
pp. 140349482096065
Author(s):  
Hanna Rinne ◽  
Mikko Laaksonen
Keyword(s):  
High Mortality ◽  
Excess Mortality ◽  
Proportional Hazard ◽  
Individual Level ◽  
Cox Proportional Hazard ◽  
Occupational Characteristics ◽  
Study Population ◽  
Cox Proportional Hazard Regression ◽  
Year 2000

Aims: Most high mortality-risk occupations are manual occupations. We examined to what extent high mortality of such occupations could be explained by education, income, unemployment or industry and whether there were differences in these effects among different manual occupations. Methods: We used longitudinal individual-level register-based data, the study population consisting of employees aged 30–64 at the end of the year 2000 with the follow-up period 2001–2015. We used Cox proportional hazard regression models in 31 male and 11 female occupations with high mortality. Results: There were considerable differences between manual occupations in how much adjusting for education, income, unemployment and industry explained the excess mortality. The variation was especially large among men: controlling for these variables explained over 50% of the excess mortality in 23 occupations. However, in some occupations the excess mortality even increased in relation to unadjusted mortality. Among women, these variables explained a varying proportion of the excess mortality in every occupation. After adjustment of all variables, mortality was no more statistically significantly higher than average in 14 occupations among men and 2 occupations among women. Conclusions: The high mortality in manual occupations was mainly explained by education, income, unemployment and industry. However, the degree of explanation varied widely between occupations, and considerable variation in mortality existed between manual occupations after controlling for these variables. More research is needed on other determinants of mortality in specific high-risk occupations.

Association between Preoperative Monocyte-to-Lymphocyte Ratio and Late Arteriovenous Fistula Dysfunction in Hemodialysis Patients: A Cohort Study

2021 ◽  
pp. 1-7
Author(s):  
Shouliang Hu ◽  
Dan Wang ◽  
Tean Ma ◽  
Fanli Yuan ◽  
Yong Zhang ◽  
...  
Keyword(s):  
Cohort Study ◽  
Arteriovenous Fistula ◽  
Vessel Diameter ◽  
Transluminal Angioplasty ◽  
Proportional Hazard ◽  
Clinical Value ◽  
Hazard Regression ◽  
Lymphocyte Ratio ◽  
Cox Proportional Hazard ◽  
Cox Proportional Hazard Regression

<b><i>Background:</i></b> Inflammation appears to be at the biological core of arteriovenous fistula (AVF) dysfunction, and the occurrence of AVF dysfunction is related to high death and disability in hemodialysis (HD) patients. Despite several studies on the correlations between AVF dysfunction and inflammatory indicators, how AVF dysfunction is related to the monocyte-to-lymphocyte ratio (MLR) is much unclear. We hypothesize that preoperative MLR is associated with AVF dysfunction in Chinese HD patients. <b><i>Methods:</i></b> In this single-center retrospective cohort study, totally 769 adult HD patients with a new AVF created between 2011 and 2019 were included. Association of preoperative MLR with AVF dysfunction (thrombosis or decrease of normal vessel diameter by &#x3e;50%, requiring either surgical revision or percutaneous transluminal angioplasty) was assessed by multivariable Cox proportional hazard regression. <b><i>Results:</i></b> The patients were aged 55.8 ± 12.2 years and were mostly males (55%). During the average 32-month follow-up (maximum 119 months), 223 (29.0%) patients had permanent vascular access dysfunction. In adjusted multivariable Cox proportional hazard regression analyses, the risk of AVF dysfunction was 4.32 times higher with 1 unit increase in MLR (hazard ratio [HR]: 5.32; 95% confidence interval [CI]: 3.1–9.11). Compared with patients with MLR &#x3c;0.28, HRs associated with an MLR of 0.28–0.41 and ≥0.41 are 1.54 (95% CI: 1.02–2.32) and 3.17 (2.18–4.62), respectively. <b><i>Conclusions:</i></b> A higher preoperative MLR is independently connected with a severer risk of AVF dysfunction in HD patients. Its clinical value should be determined in the future.

Significance of Platelet and AFP Levels and Liver Function Parameters for HCC Size and Survival

2014 ◽  
Vol 29 (3) ◽  
pp. 215-223 ◽  
Author(s):  
Brian I. Carr ◽  
Vito Guerra ◽  
Edoardo G. Giannini ◽  
Fabio Farinati ◽  
Francesca Ciccarese ◽  
...  
Keyword(s):  
Tumor Size ◽  
Blood Platelet ◽  
Tumor Diameter ◽  
Proportional Hazard ◽  
Gamma Glutamyl Transpeptidase ◽  
Vein Thrombosis ◽  
Risk Of Death ◽  
Cox Proportional Hazard ◽  
Maximum Tumor Diameter ◽  
Glutamyl Transpeptidase

Background Hepatocellular carcinoma (HCC) is a heterogeneous disease with both tumor and liver factors being involved. Aims To investigate HCC clinical phenotypes and factors related to HCC size. Methods Prospectively-collected HCC patients' data from a large Italian database were arranged according to the maximum tumor diameter (MTD) and divided into tumor size terciles, which were then compared in terms of several common clinical parameters and patients' survival. Results An higer MTD tercile was significantly associated with increased blood alpha-fetoprotein (AFP), gamma-glutamyl transpeptidase (GGTP), and platelet levels. Patients with higher platelet levels had larger tumors and higher GGTP levels, with lower bilirubin levels. However, patients with the highest AFP levels had larger tumors and higher bilirubin levels, reflecting an aggressive biology. AFP correlation analysis revealed the existence of 2 different groups of patients: those with higher and with lower AFP levels, each with different patient and tumor characteristics. The Cox proportional-hazard model showed that a higher risk of death was correlated with GGTP and bilirubin levels, tumor size and number, and portal vein thrombosis (PVT), but not with AFP or platelet levels. Conclusions An increased tumor size was associated with increased blood platelet counts, AFP and GGTP levels. Platelet and AFP levels were important indicators of tumor size, but not of survival.

scholarly journals PERBANDINGAN MODEL REGRESI COX PROPORTIONAL HAZARD MENGGUNAKAN METODE BRESLOW DAN EFRON (Studi Kasus: Penderita Stroke di RSUD Tugurejo Kota Semarang)

2019 ◽  
Vol 8 (1) ◽  
pp. 93-105
Author(s):  
Eri Setiani ◽  
Sudarno Sudarno ◽  
Rukun Santoso
Keyword(s):  
Blood Pressure ◽  
Systolic Blood Pressure ◽  
Regression Model ◽  
Diastolic Blood Pressure ◽  
Proportional Hazard ◽  
Hazard Regression ◽  
Cox Proportional Hazard ◽  
History Of ◽  
Cox Proportional Hazard Regression ◽  
Sugar Levels

Cox proportional hazard regression is a regression model that is often used in survival analysis. Survival analysis is phrase used to describe analysis of data in the form of times from a well-defined time origin until occurrence of some particular even or end-point. In analysis survival sometimes ties are found, namely there are two or more individual that have together event. This study aims to apply Cox model on ties event using two methods, Breslow and Efron and determine factors that affect survival of stroke patients in Tugurejo Hospital Semarang. Dependent variable in this study is length of stay, then independent variables are gender, age, type of stroke, history of hypertension, systolic blood pressure, diastolic blood pressure, blood sugar levels, and BMI. The two methods give different result, Breslow has four significant variables there are type of stroke, history of hypertension, systolic blood pressure, and diastolic blood pressure, while Efron contains five significant variables such as type of stroke, history of hypertension, systolic blood pressure, diastolic blood pressure and blood sugar levels. From the smallest AIC criteria obtained the best Cox proportional hazard regression model is Efron method. Keywords: Stroke, Cox Proportional Hazard Regression model, Breslow method, Efron method.

Sign in / Sign up

Export Citation Format

Share Document