Center for Medical Statistics


Center for Medical Statistics

Time filter
Source Type

Tango T.,Center for Medical Statistics
Journal of Biopharmaceutical Statistics | Year: 2017

Tango (Biostatistics 2016) proposed a new repeated measures design called the S:T repeated measures design, combined with generalized linear mixed-effects models and sample size calculations for a test of the average treatment effect that depend not only on the number of subjects but on the number of repeated measures before and after randomization per subject used for analysis. The main advantages of the proposed design combined with the generalized linear mixed-effects models are (1) it can easily handle missing data by applying the likelihood-based ignorable analyses under the missing at random assumption and (2) it may lead to a reduction in sample size compared with the simple pre-post design. In this article, we present formulas for calculating power and sample sizes for a test of the average treatment effect allowing for missing data within the framework of the S:T repeated measures design with a continuous response variable combined with a linear mixed-effects model. Examples are provided to illustrate the use of these formulas. © 2017 Taylor & Francis Group, LLC

Dunkler D.,Center for Medical Statistics
Haematologica | Year: 2012

Background Systemic activation of hemostasis is frequently observed in cancer patients, even in the absence of thrombosis. Moreover, this activation has been implicated in tumor progression, angiogenesis and metastatic spread. Increased levels of D-dimer, which is a degradation product of cross-linked fibrin, indicate a global activation of hemostasis and fibrinolysis. Design and Methods In a prospective and observational cohort study, we assessed the prognostic value of D-dimer levels for overall survival and mortality risk in 1178 cancer patients included in the Vienna Cancer and Thrombosis Study (CATS). Patients were followed over 2 years at regular intervals until occurrence of symptomatic venous thromboembolism or death. D-dimer levels were measured with a quantitative D-dimer latex agglutination assay Results The main solid tumors were malignancies of the lung (n=182), breast (n=157), lower gastrointestinal tract (n=133), pancreas (n=74), stomach (n=50), kidney (n=37), prostate (n=133), and brain (n=148); 201 of the patients had hematologic malignancies; 63 had other tumors. During a median follow-up of 731 days, 460 (39.0%) patients died. The overall survival probabilities for patients with D-dimer levels categorized into four groups based on the 1st, 2nd and 3rd quartiles of the D-dimer distribution in the total study population were 88%, 82%, 66% and 53% after 1 year, and 78%, 66%, 50% and 30% after 2 years, respectively (P<0.001). The univariate hazard ratio of Ddimer (per double increase) for mortality was 1.5 (95% confidence interval: 1.4-1.6, P<0.001) and remained increased in multivariable analysis including tumor subgroups, age, sex and venous thromboembolism. Conclusions High D-dimer levels were associated with poor overall survival and increased mortality risk in cancer patients. © 2012 Ferrata Storti Foundation.

Matsuda A.,Center for Cancer Control and Information Services | Yamaoka K.,Teikyo University | Tango T.,Center for Medical Statistics
Experimental and Therapeutic Medicine | Year: 2012

For advanced non-small cell lung cancer (NSCLC) patients, the only treatment option is palliative therapy, with the aim of prolonging overall survival and improving disease-related symptoms and quality of life (QOL). However, to date, the effect of palliative care on QOL has not yet been thoroughly examined, and there has been no meta-analysis of previous studies reporting QOL outcomes following palliative care. We consider that it is important to evaluate not only survival and/or response rates, but also QOL in patients with advanced NSCLC receiving palliative chemotherapy. The aim of the present study was to obtain useful information for the selection of suitable chemotherapy regimens for advanced NSCLC patients, taking into consideration QOL, and to demonstrate the importance of QOL assessments during treatment. We performed a meta-analysis of QOL outcomes following treatments that compared carboplatin- to cisplatin-based chemotherapy. Trials were eligible for analysis if they had compared carboplatin- to cisplatin-based chemotherapy in advanced NSCLC patients who had not received prior chemotherapy, and if these studies reported QOL data. In the six trials eligible for analysis, 2,405 patients were randomized to receive cisplatin-based or carboplatin-based chemotherapy. The patients who received carboplatin-based chemotherapy had higher global QOL and less severe symptoms than those who received cisplatin-based chemotherapy. The survival rate, which was the primary outcome in clinical trials, and the response rate did not differ significantly between the two treatment groups. It is important to evaluate QOL in addition to the survival and response rates for advanced NSCLC, particularly when the treatment is palliative.

Tango T.,Center for Medical Statistics | Takahashi K.,Japan National Institute of Public Health
Statistics in Medicine | Year: 2012

Spatial scan statistics are widely used tools for detection of disease clusters. Especially, the circular spatial scan statistic proposed by Kulldorff (1997) has been utilized in a wide variety of epidemiological studies and disease surveillance. However, as it cannot detect noncircular, irregularly shaped clusters, many authors have proposed different spatial scan statistics, including the elliptic version of Kulldorff's scan statistic. The flexible spatial scan statistic proposed by Tango and Takahashi (2005) has also been used for detecting irregularly shaped clusters. However, this method sets a feasible limitation of a maximum of 30 nearest neighbors for searching candidate clusters because of heavy computational load. In this paper, we show a flexible spatial scan statistic implemented with a restricted likelihood ratio proposed by Tango (2008) to (1) eliminate the limitation of 30 nearest neighbors and (2) to have surprisingly much less computational time than the original flexible spatial scan statistic. As a side effect, it is shown to be able to detect clusters with any shape reasonably well as the relative risk of the cluster becomes large via Monte Carlo simulation. We illustrate the proposed spatial scan statistic with data on mortality from cerebrovascular disease in the Tokyo Metropolitan area, Japan. © 2012 John Wiley & Sons, Ltd.

Hida E.,Japan National Institute of Public Health | Tango T.,Center for Medical Statistics
Journal of Biopharmaceutical Statistics | Year: 2013

The design of a three-arm trial including the experimental treatment, an active reference treatment, and a placebo is recommended as a useful approach to the assessment of noninferiority of the experimental treatment. The inclusion of the placebo arm enables the assessment of assay sensitivity and internal validation, in addition to testing the noninferiority of the experimental treatment to the reference. Generally, the acceptable noninferiority margin has been defined as the maximum clinically irrelevant difference between treatments in many two-arm noninferiority trials. However, many articles have considered a design in which the noninferiority margin is relatively defined as a prespecified fraction f of the unknown effect size of the reference treatment. Therefore, these methods cannot be applied to cases where the margin is defined as a prespecified difference between treatments. In this article, we propose score-based statistical procedures for a three-arm noninferiority trial with a prespecified margin for inference of the difference in the proportions of binary endpoints. In addition, we derive the approximate sample size and optimal allocation to minimize the total sample size and that of the placebo arm. A randomized controlled trial on major depressive disorder based on the difference in the proportions of remission is used to demonstrate our proposed method. © Taylor & Francis Group, LLC.

Yamaoka K.,Teikyo University | Watanabe M.,Showa Women's University | Nishikawa M.,Japan National Institute of Public Health | Kobayashi I.,Tokyo Polytechnic University | And 2 more authors.
BMC Public Health | Year: 2013

Background: The prevalence of type 2 diabetes is rising worldwide, as has been the global mean fasting plasma glucose level. This study aimed to evaluate the effectiveness of a structured individual-based lifestyle education (SILE) program to reduce the hemoglobin A1c (HbA§ssub§1c§esub§) level in type 2 diabetes patients delivered by registered dietitians in primary care clinical settings. Methods. This was a 6-month prospective cluster randomized controlled trial in a primary care setting with randomization at the practice level. Twenty general practitioners in 20 clinics in Kanagawa prefecture, Japan, were involved. 193 adults (51% men, mean age 61.3 years) with type 2 diabetes and HbA§ssub§1c§esub§ ≥6.5% who received treatment in medical clinics were the participants. A SILE program was implemented through 4 sessions with trained registered dietitians during the 6-month study period. Results were compared with those of a control group who received usual care. The primary endpoint was the change in HbA§ssub§ 1c§esub§ levels at 6 months from baseline. Secondary endpoints were the changes at 6 months from baseline in fasting plasma glucose, lipid profile, blood pressure, BMI, energy, and nutrient intakes (whole day and each meal). Intention-to-treat analysis was conducted. Mixed-effects linear models were used to examine the effects of the treatment. Results: The mean change at 6 months from baseline in HbA§ssub§1c§esub§ was a 0.7% decrease in the intervention group (n = 100) and a 0.2% decrease in the control group (n = 93) (difference -0.5%, 95%CI: -0.2% to -0.8%, p = 0.004). After adjusting for baseline values and other factors, the difference was still significant (p = 0.003 ∼ 0.011). The intervention group had a significantly greater decrease in mean energy intake at dinner compared with the control group and a greater increase in mean vegetable intake for the whole day, breakfast, and lunch as shown in crude and adjusted models. A tendency toward improvement was observed in the other secondary endpoints but the improvement was not statistically significant. These results were confirmed by several sensitivity analyses. Conclusions: The SILE program that was provided in primary care settings for patients with type 2 diabetes resulted in greater improvement in HbA§ssub§1c§esub§ levels than usual diabetes care and education. Trial registration. © 2013 Adachi et al.; licensee BioMed Central Ltd.

Saeki H.,Fujifilm Co. | Tango T.,Center for Medical Statistics
Statistics in Medicine | Year: 2011

The efficacy of diagnostic procedures is generally evaluated on the basis of the results from multiple raters. However, there are few adequate methods of performing non-inferiority tests with confidence intervals to compare the accuracies (sensitivities or specificities) when multiple raters are considered. We propose new statistical methods for comparing the accuracies of two diagnostic procedures in a non-inferiority trial, on the basis of the results from multiple independent raters who are also independent of the study centers. We consider a study design in which each patient is subjected to two diagnostic procedures and all images are read by all raters. By assuming a multinomial distribution for matched-pair categorical data arising from the study design, we derive a score-based full menu, that is, a non-inferiority test, confidence interval and sample size formula, for inference of the difference in correlated proportions between the two diagnostic procedures. We conduct Monte Carlo simulation studies to examine the validity of the proposed methods, which showed that the proposed test has a size closer to the nominal significance level than a Wald-type test and that the proposed confidence interval has better empirical coverage probability than a Wald-type confidence interval. We illustrate the proposed methods with data from a study of diagnostic procedures for the diagnosis of oesophageal carcinoma infiltrating the tracheobronchial tree. © 2011 John Wiley & Sons, Ltd.

Tango T.,Center for Medical Statistics
Biostatistics | Year: 2016

For the analysis of longitudinal or repeated measures data, generalized linear mixed-effects models provide a flexible and powerful tool to deal with heterogeneity among subject response profiles. However, the typical statistical design adopted in usual randomized controlled trials is an analysis of covariance type analysis using a pre-defined pair of "pre-post" data, in which pre-(baseline) data are used as a covariate for adjustment together with other covariates. Then, the major design issue is to calculate the sample size or the number of subjects allocated to each treatment group. In this paper, we propose a new repeated measures design and sample size calculations combined with generalized linear mixed-effects models that depend not only on the number of subjects but on the number of repeated measures before and after randomization per subject used for the analysis. The main advantages of the proposed design combined with the generalized linear mixed-effects models are (1) it can easily handle missing data by applying the likelihood-based ignorable analyses under the missing at random assumption and (2) it may lead to a reduction in sample size, compared with the simple pre-post design. The proposed designs and the sample size calculations are illustrated with real data arising from randomized controlled trials. © 2015 The Author 2015. Published by Oxford University Press. All rights reserved.

Graf A.C.,Center for Medical Statistics | Bauer P.,Center for Medical Statistics
Statistics in Medicine | Year: 2011

We calculate the maximum type 1 error rate of the pre-planned conventional fixed sample size test for comparing the means of independent normal distributions (with common known variance) which can be yielded when sample size and allocation rate to the treatment arms can be modified in an interim analysis. Thereby it is assumed that the experimenter fully exploits knowledge of the unblinded interim estimates of the treatment effects in order to maximize the conditional type 1 error rate. The 'worst-case' strategies require knowledge of the unknown common treatment effect under the null hypothesis. Although this is a rather hypothetical scenario it may be approached in practice when using a standard control treatment for which precise estimates are available from historical data. The maximum inflation of the type 1 error rate is substantially larger than derived by Proschan and Hunsberger (Biometrics 1995; 51:1315-1324) for design modifications applying balanced samples before and after the interim analysis. Corresponding upper limits for the maximum type 1 error rate are calculated for a number of situations arising from practical considerations (e.g. restricting the maximum sample size, not allowing sample size to decrease, allowing only increase in the sample size in the experimental treatment). The application is discussed for a motivating example. © 2011 John Wiley & Sons, Ltd..

Yamaoka K.,Teikyo University | Tango T.,Center for Medical Statistics
BMC Medicine | Year: 2012

Background: To evaluate the effect of lifestyle modifications on metabolic syndrome (MetS) as assessed by its resolution and improved values for its components.Methods: This was a systematic review and meta-analysis. Searches were performed of MEDLINE and the Cochrane Database from January 1966 to October 2011 to identify randomized controlled trials (RCTs) related to the study objective. The included studies were RCTs restricted to the English language, with a follow-up period of 6 months or more, which reported overall resolution of MetS or values of MetS components (fasting blood glucose, waist circumference, high-density lipoprotein (HDL), triglycerides, and systolic and diastolic blood pressure (SBP, DBP)). Two investigators independently assessed study eligibility. The effect sizes were the relative proportion of patients with resolved MetS and mean differences in MetS component values from baseline to 1-year follow-up in a lifestyle-modification intervention (LMI) group versus a control (conventional lifestyle education or no treatment) group. Meta-analyses were conducted using a random-effects model.Results: Eleven interventions in eight RCTs were used for the meta-analyses. The relative proportion of patients with resolved MetS in the intervention group was approximately 2.0 (95% CI 1.5 to 2.7) times greater in the intervention group compared with the control group (7 interventions, n = 2.839). LMI (5 interventions, n = 748) significantly reduced mean values for SBP by -6.4 mmHg (95% CI -9.7 to -3.2), DBP by -3.3 mmHg (95% CI -5.2 to -1.4), triglycerides by -12.0 mg/dl (95% CI -22.2 to -1.7), waist circumference by -2.7 cm (95% CI -4.6 to -0.9), and fasting blood glucose by -11.5 mg/dl (95% CI -22.4 to -0.6) (5 interventions), but reductions were not significant for HDL (1.3 mg/dl; 95% CI -0.6 to 3.1).Conclusions: The LMI was effective in resolving MetS and reducing the severity of related abnormalities (fasting blood glucose, waist circumference, SBP and DBP, and triglycerides) in subjects with MetS. © 2012 Yamaoka and Tango; licensee BioMed Central Ltd.

Loading Center for Medical Statistics collaborators
Loading Center for Medical Statistics collaborators