Dataset Information

A C-index for recurrent event data: Application to hospitalizations among dialysis patients.

ABSTRACT: We propose a C-index (index of concordance) applicable to recurrent event data. The present work addresses the dearth of measures for quantifying a regression model's ability to discriminate with respect to recurrent event risk. The data which motivated the methods arise from the Dialysis Outcomes and Practice Patterns Study (DOPPS), a long-running prospective international study of end-stage renal disease patients on hemodialysis. We derive the theoretical properties of the measure under the proportional rates model (Lin et al., 2000), and propose computationally convenient inference procedures based on perturbed influence functions. The methods are shown through simulations to perform well in moderate samples. Analysis of hospitalizations among a cohort of DOPPS patients reveals substantial improvement in discrimination upon adding country indicators to a model already containing basic clinical and demographic covariates, and further improvement upon adding a relatively large set of comorbidity indicators.

SUBMITTER: Kim S

PROVIDER: S-EPMC6647832 | biostudies-literature | 2018 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A C-index for recurrent event data: Application to hospitalizations among dialysis patients.

Kim Sehee S Schaubel Douglas E DE McCullough Keith P KP

Biometrics 20170803 2

We propose a C-index (index of concordance) applicable to recurrent event data. The present work addresses the dearth of measures for quantifying a regression model's ability to discriminate with respect to recurrent event risk. The data which motivated the methods arise from the Dialysis Outcomes and Practice Patterns Study (DOPPS), a long-running prospective international study of end-stage renal disease patients on hemodialysis. We derive the theoretical properties of the measure under the pr ...[more]

PMID: 28771674

Similar Datasets

Project description:Background:Hospitalizations of chronic dialysis patients have not been previously studied at a national level in Canada. Understanding the scope and variables associated with hospitalizations will inform measures for improvement. Objective:To describe the risk of all-cause and infection-related hospitalizations in patients on dialysis. Design:Retrospective cohort study using health care administrative databases. Setting:Provinces and territories across Canada (excluding Manitoba and Quebec). Patients:Incident chronic dialysis patients with a dialysis start date between January 1, 2005, and March 31, 2014. Patients with a prior history of kidney transplantation were excluded. Measurements:Patient characteristics were recorded at baseline. Dialysis modality was treated as a time-varying covariate. The primary outcomes of interest were all-cause and dialysis-specific infection-related hospitalizations. Methods:Crude rates for all-cause hospitalization and infection-related hospitalization were determined per patient year (PPY) at 7 and 30 days, and at 3, 6, and 12 months postdialysis initiation. A stratified, gamma-distributed frailty model was used to assess repeat hospital admissions and to determine the inter-recurrence dependence of hospitalizations within individuals, as well as the hazard ratio (HR) attributed to each covariate of interest. Results:A total of 38 369 incident chronic dialysis patients were included: 38 088 adults and 281 pediatric patients (age less than 18 years). There were 112 374 hospitalizations, of which 11.5% were infection-related hospitalizations. The all-cause hospitalization rate was similar for all adult age groups (age 65 years and older: 1.40, 1.35, and 1.18 admissions PPY at 7 days, 30 days, and 6 months, respectively). The all-cause hospitalization rate was higher for pediatric patients (1.67, 2.48, and 2.47 admissions PPY at 7 days, 30 days, and 6 months, respectively; adjusted HR: 2.73, 95% confidence interval [CI]: 2.37-3.15, referent age group: 45-64 years). Within the first 7 days after dialysis initiation, patients on peritoneal dialysis had a higher risk of all-cause hospitalization (HR: 1.27, 95% CI: 1.07-1.50) and infection-related hospitalization (HR: 2.05, 95% CI: 1.19-3.55) compared with patients on hemodialysis. Beyond 7 days, the risk did not differ significantly by dialysis modality. Female sex and Indigenous race were significant risk factors for all-cause hospitalization. Limitations:The cohort had too few home hemodialysis patients to examine this subgroup. The outcome of infection-related hospitalization was determined using diagnostic codes. Dialysis patients from Manitoba and Quebec were not included. Conclusions:In Canada, the rates of hospitalization were not influenced by dialysis modality beyond the initial 7-day period following dialysis initiation; however, the rate of hospitalization in pediatric patients was higher than in adults at every time frame examined.

Dataset Information

A C-index for recurrent event data: Application to hospitalizations among dialysis patients.

Publications

A C-index for recurrent event data: Application to hospitalizations among dialysis patients.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets