Long-Term Survival, Quality of Life, and Quality-Adjusted Survival in Critically Ill Patients With Cancer.

OBJECTIVES: To assess the long-term survival, health-related quality of life, and quality-adjusted life years of cancer patients admitted to ICUs. DESIGN: Prospective cohort. SETTING: Two cancer specialized ICUs in Brazil. PATIENTS: A total of 792 participants. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: The health-related quality of life before ICU admission; at 15 days; and at 3, 6, 12, and 18 months was assessed with the EQ-5D-3L. In addition, the vital status was assessed at 24 months. The mean age of the subjects was 61.6 ± 14.3 years, 42.5% were female subjects and half were admitted after elective surgery. The mean Simplified Acute Physiology Score 3 was 47.4 ± 15.6. Survival at 12 and 18 months was 42.4% and 38.1%, respectively. The mean EQ-5D-3L utility measure before admission to the ICU was 0.47 ± 0.43, at 15 days it was 0.41 ± 0.44, at 90 days 0.56 ± 0.42, at 6 months 0.60 ± 0.41, at 12 months 0.67 ± 0.35, and at 18 months 0.67 ± 0.35. The probabilities for attaining 12 and 18 months of quality-adjusted survival were 30.1% and 19.1%, respectively. There were statistically significant differences in survival time and quality-adjusted life years according to all assessed baseline characteristics (ICU admission after elective surgery, emergency surgery, or medical admission; Simplified Acute Physiology Score 3; cancer extension; cancer status; previous surgery; previous chemotherapy; previous radiotherapy; performance status; and previous health-related quality of life). Only the previous health-related quality of life and performance status were associated with the health-related quality of life during the 18-month follow-up. CONCLUSIONS: Long-term survival, health-related quality of life, and quality-adjusted life year expectancy of cancer patients admitted to the ICU are limited. Nevertheless, these clinical outcomes exhibit wide variability among patients and are associated with simple characteristics present at the time of ICU admission, which may help healthcare professionals estimate patients' prognoses.

Cutpoint selection for discretizing a continuous covariate for generalized estimating equations.

We consider the problem of dichotomizing a continuous covariate when performing a regression analysis based on a generalized estimation approach. The problem involves estimation of the cutpoint for the covariate and testing the hypothesis that the binary covariate constructed from the continuous covariate has a significant impact on the outcome. Due to the multiple testing used to find the optimal cutpoint, we need to make an adjustment to the usual significance test to preserve the type-I error rates. We illustrate the techniques on one data set of patients given unrelated hematopoietic stem cell transplantation. Here the question is whether the CD34 cell dose given to patient affects the outcome of the transplant and what is the smallest cell dose which is needed for good outcomes.

Effect of age on outcome of reduced-intensity hematopoietic cell transplantation for older patients with acute myeloid leukemia in first complete remission or with myelodysplastic syndrome.

PURPOSE Acute myelogenous leukemia (AML) and myelodysplastic syndrome (MDS) primarily afflict older individuals. Hematopoietic cell transplantation (HCT) is generally not offered because of concerns of excess morbidity and mortality. Reduced-intensity conditioning (RIC) regimens allow increased use of allogeneic HCT for older patients. To define prognostic factors impacting long-term outcomes of RIC regimens in patients older than age 40 years with AML in first complete remission or MDS and to determine the impact of age, we analyzed data from the Center for International Blood and Marrow Transplant Research (CIBMTR). PATIENTS AND METHODS We reviewed data reported to the CIBMTR (1995 to 2005) on 1,080 patients undergoing RIC HCT. Outcomes analyzed included neutrophil recovery, incidence of acute or chronic graft-versus-host disease (GVHD), nonrelapse mortality (NRM), relapse, disease-free survival (DFS), and overall survival (OS). RESULTS Univariate analyses demonstrated no age group differences in NRM, grade 2 to 4 acute GVHD, chronic GVHD, or relapse. Patients age 40 to 54, 55 to 59, 60 to 64, and > or = 65 years had 2-year survival rates as follows: 44% (95% CI, 37% to 52%), 50% (95% CI, 41% to 59%), 34% (95% CI, 25% to 43%), and 36% (95% CI, 24% to 49%), respectively, for patients with AML (P = .06); and 42% (95% CI, 35% to 49%), 35% (95% CI, 27% to 43%), 45% (95% CI, 36% to 54%), and 38% (95% CI, 25% to 51%), respectively, for patients with MDS (P = .37). Multivariate analysis revealed no significant impact of age on NRM, relapse, DFS, or OS (all P > .3). Greater HLA disparity adversely affected 2-year NRM, DFS, and OS. Unfavorable cytogenetics adversely impacted relapse, DFS, and OS. Better pre-HCT performance status predicted improved 2-year OS. CONCLUSION With these similar outcomes observed in older patients, we conclude that older age alone should not be considered a contraindication to HCT.

Methods for equivalence and noninferiority testing.

Classical hypothesis testing focuses on testing whether treatments have differential effects on outcome. However, sometimes clinicians may be more interested in determining whether treatments are equivalent or whether one has noninferior outcomes. We review the hypotheses for these noninferiority and equivalence research questions, consider power and sample size issues, and discuss how to perform such a test for both binary and survival outcomes. The methods are illustrated on 2 recent studies in hematopoietic cell transplantation.

A semi-Markov multistate model for estimation of the mean quality-adjusted survival for non-progressive processes.

We discuss the estimation of the expected value of the quality-adjusted survival, based on multistate models. We generalize an earlier work, considering the sojourn times in health states are not identically distributed, for a given vector of covariates. Approaches based on semiparametric and parametric (exponential and Weibull distributions) methodologies are considered. A simulation study is conducted to evaluate the performance of the proposed estimator and the jackknife resampling method is used to estimate the variance of such estimator. An application to a real data set is also included.