Empirical Powers of MRPP Rank Tests for Symmetric Kappa Distributions

Compstat ◽  
1994 ◽  
pp. 515-520
Author(s):  
Derrick S. Tracy ◽  
Khushnood A. Khan
Keyword(s):  
2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 169.1-169
Author(s):  
R. Punekar ◽  
P. Lafontaine ◽  
J. H. Stone

Background:Polymyalgia rheumatica (PMR) is a chronic inflammatory condition characterized by aching and morning stiffness in the neck, shoulders and pelvic girdle. It is a common inflammatory rheumatic disease in patients age >50 years, particularly women. While giant cell arteritis (GCA) is present in 9–21% of PMR cases, many PMR patients have symptoms independent of GCA. Current treatment options are limited to long-term glucocorticoid (GC), however, with risks of GC-related complications, including cardiovascular disease, osteoporosis, and diabetes mellitus.Objectives:To compare GC use and subsequent GC-related complications in patients with PMR vs a general population (GnP) cohort.Methods:This retrospective, observational cohort study was based on Optum’s de-identified Clinformatics®Data Mart Database (study period 01Jan2006-30June2018). The PMR cohort included patients with ≥1 inpatient or ≥2 outpatient claims ≥30 days apart with PMR related diagnosis codes (ICD-9: 725.xx or ICD-10: M35.3x) between 01Jan2006–30June2017 (patient identification period) during which first occurrence of a PMR-related medical claim was set as the index date (ID). Patients with ≥1 medical claim related to rheumatoid arthritis (RA) or GCA during the study period were excluded. The GnP cohort included patients without any RA, GCA or PMR diagnosis codes during the study period, with their ID set as 12 months from the start of continuous health plan enrollment. Patients in both cohorts were required to be age ≥50 years (on ID) with continuous health plan enrollment ≥12 months pre- and post-ID. Cohorts were 1:1 propensity score matched. GC use and incidence of GC-related complications were assessed from GC initiation, starting from the baseline period (12-months pre-ID) through to the end of GC use during the post-index period (i.e. the end of data availability, end of the study period or death [whichever occurred first]). Mean, standard deviation (SD) and median values for continuous variables, and frequency (n and %) for categorical variables were compared between the matched cohorts. Wilcoxon sum rank tests andt-tests on continuous variables and Chi-square tests or Fisher’s exact tests on categorical variables between matched cohorts were conducted. Duration of GC use was analyzed using the Kaplan-Meier method and compared between matched cohorts using log-rank tests.Results:In each of the PMR and GnP cohorts, 16,865 patients were included. In both matched cohorts, median age was 76 years, median Elixhauser comorbidity index score was 2.0, and the majority (~65%) were women. The median follow-up duration was 45 months and 51 months in the PMR and GnP cohorts, respectively. A higher proportion of patients in the PMR cohort than the matched GnP cohort (90.4% vs 62.8%;p<0.001) used GC. The mean (SD) duration of GC therapy was significantly longer in the PMR cohort than in the matched GnP cohort (242.1 [±317.2] days vs 35.5 [±124.6] days;p<0.001). Although patients in the PMR cohort had a lower average daily dose of GC (prednisone equivalent) vs the GnP cohort (mean [SD] mg 16.3 [± 21.9] vs 27.8 [±24.5], respectively [p<0.0001)], the cumulative GC dose was significantly higher in the PMR cohort than the GnP cohort (2125.4 [±3689.5] mg vs 476.6 [±1450.9] mg;p<0.001). This indicates PMR patients used chronic low dose GC while the GnP patients utilized higher dose GC burst therapy less frequently. The number of incident complications associated with GC use were significantly greater in the PMR cohort, and included hypertension, diabetes, skin toxicity, infections, neuropsychiatric effects, endocrine abnormalities, renal dysfunction/ failure, ocular effects, and cardiovascular disease (p<0.05).Conclusion:The overall GC burden in patients with PMR is high. With a higher incidence of GC-related comorbidities among PMR patients, early onset of these complications may be a significant contributor to long-term healthcare costs in these patients.Acknowledgments:This study was funded by Sanofi, Inc. Medical writing, under the direction of authors, was provided by Gauri Saal, MA Economics, Prime, Knutsford, UK, and funded by Sanofi.Disclosure of Interests:Rajeshwari Punekar Shareholder of: Sanofi, Employee of: Sanofi, Patrick LaFontaine Shareholder of: Sanofi, Employee of: Sanofi, John H. Stone Grant/research support from: Roche, Consultant of: Roche


1971 ◽  
Vol 66 (336) ◽  
pp. 879 ◽  
Author(s):  
Robert B. Davies
Keyword(s):  

1963 ◽  
Vol 34 (2) ◽  
pp. 598-611 ◽  
Author(s):  
J. L. Hodges ◽  
E. L. Lehmann
Keyword(s):  

1998 ◽  
Vol 39 (2) ◽  
pp. 179-188 ◽  
Author(s):  
Ralf Runde
Keyword(s):  

Biometrika ◽  
1987 ◽  
Vol 74 (3) ◽  
pp. 615-624 ◽  
Author(s):  
F. LOMBARD
Keyword(s):  

2021 ◽  
pp. 096228022098857
Author(s):  
Yongqiang Tang

Log-rank tests have been widely used to compare two survival curves in biomedical research. We describe a unified approach to power and sample size calculation for the unweighted and weighted log-rank tests in superiority, noninferiority and equivalence trials. It is suitable for both time-driven and event-driven trials. A numerical algorithm is suggested. It allows flexible specification of the patient accrual distribution, baseline hazards, and proportional or nonproportional hazards patterns, and enables efficient sample size calculation when there are a range of choices for the patient accrual pattern and trial duration. A confidence interval method is proposed for the trial duration of an event-driven trial. We point out potential issues with several popular sample size formulae. Under proportional hazards, the power of a survival trial is commonly believed to be determined by the number of observed events. The belief is roughly valid for noninferiority and equivalence trials with similar survival and censoring distributions between two groups, and for superiority trials with balanced group sizes. In unbalanced superiority trials, the power depends also on other factors such as data maturity. Surprisingly, the log-rank test usually yields slightly higher power than the Wald test from the Cox model under proportional hazards in simulations. We consider various nonproportional hazards patterns induced by delayed effects, cure fractions, and/or treatment switching. Explicit power formulae are derived for the combination test that takes the maximum of two or more weighted log-rank tests to handle uncertain nonproportional hazards patterns. Numerical examples are presented for illustration.


Sign in / Sign up

Export Citation Format

Share Document