Create Account | Sign In: Author or Forum

Search Symptoms

Category: Endocrinology | Internal Medicine | Nephrology | Pathology | Journal

Back to Journal Articles

Kt/V Use in Predicting Mortality Affected by Model

Last Updated: July 31, 2009.

In dialysis patients, Kt/V significantly predicts mortality when accelerated failure time models are used, but not proportional hazard models, according to research published online July 30 in the Journal of the American Society of Nephrology.

FRIDAY, July 31 (HealthDay News) -- In dialysis patients, Kt/V significantly predicts mortality when accelerated failure time models are used, but not proportional hazard models, according to research published online July 30 in the Journal of the American Society of Nephrology.

Christos Argyropoulos, M.D., of the University of Pittsburgh Medical Center, and colleagues analyzed data on all-cause mortality using proportional hazard and accelerated failure time models in 766 incident hemodialysis patients followed for nine years.

Kt/V -- a measurement of dialysis adequacy -- was not significantly associated with mortality using proportional hazards, the investigators found. But with accelerated failure time models, each 0.1-U increment of Kt/V was associated with a 3.5-percent improvement in adjusted patient survival. Proportional hazard models also provided less accurate estimates of median survival.

"The most accurate interpretation of our findings is that non-time-updated proportional hazards model may not be sensitive enough to detect survival benefits as a result of higher levels of Kt/V. Consequently, accelerated failure time models should be considered as adjuncts to the proportional hazard models when analyzing the role of dialysis dosage on survival. Most statistical packages used in outcomes research provide estimation procedures for accelerated failure time models (e.g., PROC LIFEREG in SAS, streg in STATA, survreg/psm in R/S-Plus), facilitating such analyses," the authors write.

This study was supported by the Renal Discoveries-Baxter Extramural Grant Program.

Abstract
Full Text (subscription or payment may be required)


Previous: Muscle Density Linked to Hospitalization in Aged Next: End-of-Life Decisions in Multidisciplinary Care Studied

Reader comments on this article are listed below. Review our comments policy.


Submit your opinion: