China
Africa
Explore chapters and articles related to this topic
Inference—Statistical Tests
Published in Prabhanjan Narayanachar Tattar, H. J. Vaman, Survival Analysis, 2022
Prabhanjan Narayanachar Tattar, H. J. Vaman
Specification of the choice of the weight process allows us pick appropriate test statistics. For instance, if we choose , we obtain the log-rank test statistic, and in this case we have the following:
Survival Analysis
Published in Trevor F. Cox, Medical Statistics for Cancer Studies, 2022
In the lower plot, the survival curves cross at about three months. The data for Arm 1 have been simulated from an exponential distribution with and for Arm 2, from a Weibull distribution with and . Patients in Arm 2 do better than patients in Arm 1 early on from the time of randomisation, but then patients in Arm 1 do better than patients in Arm 2 later. The median survival times are fairly close at 2.39 and 2.85 months. Clearly the survival curves are different, but we cannot reject the null hypothesis of no difference in the survival curves, based on the log-rank test, as with . So, we must be careful when using the log-rank test. What is happening when these survival curves cross, is that the 's will all tend to be positive for the first part of the survival curves and then all tend to be negative for the second part of the survival curves, thus cancelling each other out and producing a low value of LR. It can be shown that the log-rank test is optimal when the proportional hazards assumption holds, i.e. , is when it is most powerful.
Prognosis: Studies of disease course and outcomes
Published in Milos Jenicek, Foundations of Evidence-Based Medicine, 2019
The log-rank test involves counting in each stratum the number of deaths observed in each group (0) and comparing this with the number of expected events (E) that would otherwise occur in each group under the null hypothesis of equivalence in survival in both groups. p-values can be estimated by comparing the sum of (0 – E)2/E with an appropriate chi-square distribution. Readers interested in obtaining more information on this topic should consult the article written by Peto et al.17 The wide acceptance of the log-rank test is due to its capacity to compare entire survival curves instead of survival rates at selected unique points in time.
The prognostic role of lymphocyte to monocyte ratio (LMR) in patients with Myelodysplastic Neoplasms
Published in Hematology, 2023
Chuanyang Lu, Qiuni Chen, Jiaxin Li, Chunling Wang, Liang Yu
The cut-off of LMR was determined by X-Tile (version 3.6.1, Yale University, New Haven, CT, United States) [18]. Statistical Package for social sciences (SPSS, version 23, IBM SPSS Statistics 23 software, IBM Corp., Armonk, NY, USA) was used to perform statistical analysis. Mann–Whitney U-test and chi-square test were utilized to evaluate the difference between the two groups and p-value < 0.05 (2-tailed) demonstrated a statistical significance. Kaplan-Meier method has been applied to assess the correlation between LMR and OS. Furthermore, the corresponding p-value was achieved through the log-rank test. Survival curves were graphed by GraphPad Prism (version 8.0.1, GraphPad Software, San Diego, California, USA). Univariate and multivariate Cox regression analyses were conducted to investigate the prognostic factors affecting OS. In univariate analysis, p-value < 0.05 was considered statistically significant, and the corresponding prognostic factors were included in multivariate Cox regression. P-value < 0.05 were considered statistically significant in multivariate analysis.
Prevalence, risk, and outcomes of venous thromboembolic events in kidney transplant recipients: a nested case-control study
Published in Renal Failure, 2023
Vinant Bhargava, Priti Meena, Anil Kumar Bhalla, Devinder Singh Rana, Ashwani Gupta, Manish Malik, Anurag Gupta, Vaibhav Tiwari
Of the 97 patients who experienced VTE, 15 (15.6%) died. Six patients with pulmonary thromboembolism died of acute massive episodes, five of lower respiratory tract infection, and four of myocardial infarction. Of the 388 patients without VTE, 37 (9.4%) died. Survival analysis was performed considering death as an event, and time to event was compared between two patient categories: VTE and no VTE. The estimated mean time to event in the VTE group was 46 months (95% CI: 34.26–57.74; SE: 5.99), with a median of 49 months (95% CI: 21.08–76.91; SE: 14.24). The Kaplan–Meier plot showing the cumulative survival in the two groups is presented in Figure 2. The difference in the survival rates between groups was statistically significant (p = 0.003), obtained using the Log-rank test. The patients experiencing VTE had significantly shorter survival time than patients without VTE.
Prescription patterns and therapeutic effects of second-line drugs in Japanese patients with type 2 diabetes mellitus: Analysis of claims data for metformin and dipeptidyl peptidase-4 inhibitors as the first-line hypoglycemic agents
Published in Expert Opinion on Pharmacotherapy, 2023
Rimei Nishimura, Tomomi Takeshima, Kosuke Iwasaki, Sumiko Aoi
To investigate the treatment effect, Kaplan-Meier curves of the events were computed for each endpoint after adjusting for confounding factors of the first-line treatment, with metformin or DPP4i, using propensity scores. The propensity scores were calculated using a logistic regression model with first-line metformin prescription as the explained variable, and age, sex, Charlson Comorbidity Index [20,21], hypertension (defined by the prescription of antihypertensive agents coded as C03, C07, C08, or C09 by ATC), dyslipidemia (defined by the prescription of statins or other antihyperlipidemic agents coded as C10 by ATC), and prescription of antithrombotic drugs (such as aspirin, novel oral anticoagulants defined by generic name, or other antithrombotic agents coded as B01 by ATC) at baseline as the explanatory variables. Based on the propensity score, the patients were divided into five quintiles, and the weights were adjusted for each quintile [22]. Statistical significance was defined as p < 0.05 using a log-rank test. The statistical analyses were performed using Microsoft Excel 2010 (Microsoft, Redmond, WA, U.S.A) and SAS version 9.4 (SAS Institute, Cary, NC, U.S.A).