Misspecified regression model for the subdistribution hazard of a competing risk

2007 ◽  
Vol 26 (5) ◽  
pp. 965-974 ◽  
Author(s):  
A. Latouche ◽  
V. Boisson ◽  
S. Chevret ◽  
R. Porcher
Keyword(s):  
Regression Model ◽  
Competing Risk ◽  
Subdistribution Hazard

Gait speed and adverse outcomes following hospitalised exacerbation of COPD

2021 ◽  
pp. 2004047
Author(s):  
Jessica A. Walsh ◽  
Ruth E. Barker ◽  
Samantha S.C. Kon ◽  
Sarah E. Jones ◽  
Winston Banya ◽  
...  
Keyword(s):  
Hospital Discharge ◽  
Gait Speed ◽  
Surrogate Marker ◽  
Performance Measure ◽  
Cox Proportional Hazards ◽  
Hazard Ratio ◽  
Adverse Outcomes ◽  
Competing Risk ◽  
Subdistribution Hazard ◽  
Kaplan Meier

Four-metre gait speed (4MGS) is a simple physical performance measure and surrogate marker of frailty that is associated with adverse outcomes in older adults. We aimed to assess the ability of 4MGS to predict prognosis in patients hospitalised with acute exacerbations of COPD (AECOPD).213 participants hospitalised with AECOPD (52% male, mean age and FEV1, 72 years and 35% predicted) were enrolled. 4MGS and baseline demographics were recorded at hospital discharge. All-cause readmission and mortality were collected for 1 y after discharge, and multivariable Cox-proportional hazards regression were performed. Kaplan-Meier and Competing risk analysis was conducted comparing time to all-cause readmission and mortality between 4MGS quartiles.111 participants (52%) were readmitted, and 35 (16%) died during the follow-up period. 4MGS was associated with all-cause readmission, with an adjusted subdistribution hazard ratio of 0.868 (95% CI 0.797–0.945; p=0.001) per 0.1 m·s−1 increase in gait speed, and with all-cause mortality with an adjusted subdistribution hazard ratio of 0.747 (95% CI: 0.622–0.898; p=0.002) per 0.1 m·s−1 increase in gait speed. Readmission and mortality models incorporating 4MGS had higher discrimination than age or FEV1% predicted alone, with areas under the receiver operator characteristic curves of 0.73 and 0.80 respectively. Kaplan-Meier and Competing Risk curves demonstrated that those in slower gait speed quartiles had reduced time to readmission and mortality (log rank both p<0.001).4MGS provides a simple means of identifying at-risk patients with COPD at hospital discharge. This provides valuable information to plan post-discharge care and support.

scholarly journals Risk factors and survival outcomes of metachronous contralateral upper tract urothelial carcinoma

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Kan Wu ◽  
Jiayu Liang ◽  
Yiping Lu
Keyword(s):  
Risk Factors ◽  
Bladder Cancer ◽  
Regression Model ◽  
Urothelial Carcinoma ◽  
Large Population ◽  
Multivariable Analysis ◽  
Population Based ◽  
Upper Tract Urothelial Carcinoma ◽  
Competing Risk ◽  
Upper Tract

Abstract Because population-based risk estimates for metachronous contralateral UTUC are lacking. In this study, we aimed to evaluate the risk and survival of metachronous contralateral upper tract urothelial carcinoma (UTUC) on a large population-based level. A total of 23,075 patients were identified from the Surveillance, Epidemiology, and End Results database (1973–2015), 144 (0.6%) patients developed metachronous contralateral UTUC (median of 32 months after diagnosis). The cumulative incidence at 10, 20, and 30 years of follow-up was 1.1%, 1.6%, and 2.6%, respectively. We applied Fine and Gray’s competing risk regression model to determine the risk factors of a new contralateral, metachronous UTUC. The competing risk regression model demonstrated that older age (hazard ratio [HR] 0.75; 95% CI 0.67–0.85) and larger tumor size (HR 0.61; 95% CI 0.39–0.97) were associated with a significantly decreased risk of metachronous contralateral UTUC. However, bladder cancer presence was an independent risk factor for the development of contralateral tumors (HR 2.42; 95% CI 1.73–3.37). In addition, we demonstrated developing contralateral UTUC was not associated with poor prognosis by using Kaplan–Meier and multivariable analysis. Our findings suggest that metachronous contralateral UTUC is comparatively rare, and has not impact on survival. Importantly, patients with younger age, small tumours, and the presence of bladder cancer were more likely to develop a contralateral tumor, which may provide a rationale for lifelong surveillance in high-risk patients.

scholarly journals Mortality on the UNOS Waitlist for Patients with Autoimmune Liver Disease

2020 ◽  
Vol 9 (2) ◽  
pp. 319
Author(s):  
Jaspreet S. Suri ◽  
Christopher J. Danford ◽  
Vilas Patwardhan ◽  
Alan Bonder
Keyword(s):  
Liver Disease ◽  
Risk Analysis ◽  
Liver Transplant ◽  
Autoimmune Hepatitis ◽  
Competing Risk ◽  
Clinical Deterioration ◽  
Primary Biliary Cholangitis ◽  
Competing Risk Analysis ◽  
Subdistribution Hazard ◽  
Similar Risk

Background: Outcomes on the liver transplant waitlist can vary by etiology. Our aim is to investigate differences in waitlist mortality of autoimmune hepatitis (AIH), primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC) using the United Network for Organ Sharing (UNOS) database. Methods: We identified patients who were listed for liver transplantation from 1987 to 2016 with a primary diagnosis of AIH, PBC, or PSC. We excluded patients with overlap syndromes, acute hepatic necrosis, missing data, and those who were children. The primary outcome was death or removal from the waitlist due to clinical deterioration. We compared waitlist survival using competing risk analysis. Results: Between 1987 and 2016, there were 7412 patients listed for liver transplant due to AIH, 8119 for PBC, and 10,901 for PSC. Patients with AIH were younger, more likely to be diabetic, and had higher listing model for end-stage liver disease (MELD) scores compared to PBC and PSC patients. Patients with PBC and AIH were more likely to be removed from the waitlist due to death or clinical deterioration. On competing risk analysis, AIH patients had a similar risk of being removed from the waitlist compared to those with PBC (subdistribution hazard ratio (SHR) 0.94, 95% CI 0.85–1.03) and higher risk of removal compared to those with PSC (SHR 0.8, 95% CI 0.72 to 0.89). Conclusion: Autoimmune hepatitis carries a similar risk of waitlist removal to PBC and a higher risk than PSC. The etiology of this disparity is not entirely clear and deserves further investigation.

Adherence and cumulative bisphosphonate dose in relation to risk of skeletal-related events among older patients with multiple myeloma.

2019 ◽  
Vol 37 (15_suppl) ◽  
pp. e18250-e18250
Author(s):  
Jifang Zhou ◽  
Karen Sweiss ◽  
Pritesh Rajni Patel ◽  
Edith Nutescu ◽  
Naomi Ko ◽  
...  
Keyword(s):  
Multiple Myeloma ◽  
Cumulative Dose ◽  
Proportional Hazards ◽  
Cox Proportional Hazards ◽  
Competing Risk ◽  
Rank Test ◽  
Subdistribution Hazard ◽  
Cox Proportional Hazards Models ◽  
Utilization Patterns ◽  
Skeletal Related Events

e18250 Background: Adjuvant intravenous bisphosphonates (IV BP) reduce the risk of skeletal-related events (SRE) in patients with multiple myeloma (MM). We examined the effects of bisphosphonate utilization patterns (adherence, cumulative dose and frequency) on risk of SRE. Methods: Patients aged 65 years or older and diagnosed with first primary MM between 2001 and 2011 were identified using the Surveillance, Epidemiology and End Results (SEER)-Medicare linked database. Patients receiving at least one dose of IV BP after MM diagnosis were identified and 5-year SRE-free survival was estimated using the Kaplan-Meier method stratified by demographic groups and compared with the log rank test. Cox proportional hazards models were fit to determine the association between IV BP utilization patterns and SRE after propensity score matching. We investigated the outcome of multiple recurrent SRE using the approach of Andersen-Gill, and estimated subdistribution hazard ratios (SHR) and 95% confidence intervals for risk of first SRE, accounting for death as competing risk. Results: The final cohort included 9176 MM patients with a median age of 76 years. The adjusted 5-year competing-risk SRE model showed a 48% reduction in risk of SRE (95% CI 0.49-0.55) with use of IV BP. In multivariable analyses taking into account competing risks, greater adherence to IV BP, higher cumulative IV BP dose and more frequent administration were all associated with a statistically significant reduction in SRE risks (See Table). Conclusions: Use of IV BP in patients with MM was associated with significant reduction in SRE risk over the 5-year period after MM diagnosis. The effectiveness of IV BP therapy was greater with increasing cumulative dose, adherence to and greater frequency of IV BP administration. [Table: see text]

scholarly journals Competing risk of mortality in association studies of non-fatal events

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0255313
Author(s):  
Petra Buzkova
Keyword(s):  
Cardiovascular Health ◽  
Association Studies ◽  
Population Based ◽  
Competing Risk ◽  
Health Study ◽  
Cardiovascular Health Study ◽  
Subdistribution Hazard ◽  
Hazard Regression ◽  
Risk Of Mortality

In geriatric research of non-fatal events, participants often die during the study follow-up without having the non-fatal event of interest. Cause-specific (CS) hazard regression and Fine-Gray (FG) subdistribution hazard regression are the two most common estimation approaches addressing such competing risk. We explain how the conventional CS approach and the FG approach differ and why many FG estimates of associations are counter-intuitive. Additionally, we clarify the indirect link between models for hazard and models for cumulative incidence. The methodologies are contrasted on data from the Cardiovascular Health Study, a population-based study in adults aged 65 years and older.

Sign in / Sign up

Export Citation Format

Share Document