Songfeng Wang

University of South Carolina, Columbia, SC, United States

Are you Songfeng Wang?

Claim your profile

Publications (6)24.93 Total impact

  • Songfeng Wang, Jiajia Zhang, Wenbin Lu
    [Show abstract] [Hide abstract]
    ABSTRACT: The Cox proportional hazards (PH) model with time-dependent covariates (referred to as the extended PH model) has been widely used in medical and health related studies to investigate the effects of time-varying risk factors on survival. Theories and practices regarding model estimation and fitting have been well developed for the extended PH model. However, little has been done regarding sample size calculations in survival studies involving a time-varying risk factor. A novel sample size formula based on the extended PH model is proposed by investigating the asymptotic distributions of the weighted log-rank statistics under the null and local alternative hypotheses. The derived sample size formula is an extension of the sample size formula for the standard Cox PH model. The performance of the proposed formula is evaluated by extensive simulations, and examples based on real data are given to illustrate the applications of the proposed methods.
    Computational Statistics & Data Analysis 01/2014; 74:217–227. · 1.30 Impact Factor
  • Chao Cai, Songfeng Wang, Wenbin Lu, Jiajia Zhang
    [Show abstract] [Hide abstract]
    ABSTRACT: Due to advances in medical research, more and more diseases can be cured nowadays, which largely increases the need for an easy-to-use software in calculating sample size of clinical trials with cure fractions. Current available sample size software, such as PROC POWER in SAS, Survival Analysis module in PASS, powerSurvEpi package in R are all based on the standard proportional hazards (PH) model which is not appropriate to design a clinical trial with cure fractions. Instead of the standard PH model, the PH mixture cure model is an important tool in handling the survival data with possible cure fractions. However, there are no tools available that can help design a trial with cure fractions. Therefore, we develop an R package NPHMC to determine the sample size needed for such study design.
    Computer methods and programs in biomedicine 10/2013; · 1.56 Impact Factor
  • Source
    Songfeng Wang, Jiajia Zhang, Andrew B Lawson
    [Show abstract] [Hide abstract]
    ABSTRACT: In the United States, prostate cancer is the third most common cause of death from cancer in males of all ages, and the most common cause of death from cancer in males over age 75. It has been recognized that the incidence of the prostate cancer is high in African Americans, and its occurrence and progression may be impacted by geographical factors. In order to investigate the spatial effects and racial disparities for prostate cancer in Louisiana, in this article we propose a normal mixture accelerated failure time spatial model, which does not require the proportional hazards assumption and allows the multi-model distribution to be modeled. The proposed model is estimated with a Bayesian approach and it can be easily implemented in WinBUGS. Extensive simulations show that the proposed model provides decent flexibility for a variety of parametric error distributions. The proposed method is applied to 2000-2007 Louisiana prostate cancer data set from the Surveillance, Epidemiology and End Results Program. The results reveal the possible spatial pattern and racial disparities for prostate cancer in Louisiana.
    Statistical Methods in Medical Research 11/2012; · 2.36 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This study evaluated the relation between apolipoprotein B (apoB) decrease and coronary heart disease, stroke, and cardiovascular disease risk. Bayesian random-effects meta-analysis was used to evaluate the association of mean absolute apoB decrease (milligrams per deciliter) with relative risk of coronary heart disease (nonfatal myocardial infarction and coronary heart disease death), stroke (nonfatal stroke and fatal stroke), or cardiovascular disease (coronary heart disease, stroke, and coronary revascularization). Analysis included 25 trials (n = 131,134): 12 on statin, 4 on fibrate, 5 on niacin, 2 on simvastatin-ezetimibe, 1 on ileal bypass surgery, and 1 on aggressive versus standard low-density lipoprotein (LDL) cholesterol and blood pressure targets. Combining the 25 trials, each 10-mg/dl decrease in apoB was associated with a 9% decrease in coronary heart disease, no decrease in stroke, and a 6% decrease in major cardiovascular disease risk. Non-high-density lipoprotein (non-HDL) cholesterol decrease modestly outperformed apoB decrease for prediction of coronary heart disease (Bayes factor [BF] 1.45) and cardiovascular disease (BF 2.07) risk decrease; apoB decrease added to non-HDL cholesterol plus LDL cholesterol decrease slightly improved cardiovascular disease risk prediction (1.13) but did not improve coronary heart disease risk prediction (BF 1.03) and worsened stroke risk prediction (BF 0.83). In the 12 statin trials, apoB and non-HDL cholesterol decreases similarly predicted cardiovascular disease risk; apoB improved coronary heart disease prediction when added to non-HDL cholesterol/LDL cholesterol decrease (BF 3.33) but did not improve stroke risk prediction when added to non-HDL cholesterol/LDL cholesterol decrease (BF 1.06). In conclusion, across all drug classes, apoB decreases did not consistently improve risk prediction over LDL cholesterol and non-HDL cholesterol decreases. For statins, apoB decreases added information to LDL cholesterol and non-HDL cholesterol decreases for predicting coronary heart disease but not stroke or overall cardiovascular disease risk decrease.
    The American journal of cardiology 08/2012; · 3.58 Impact Factor
  • Songfeng Wang, Jiajia Zhang, Wenbin Lu
    [Show abstract] [Hide abstract]
    ABSTRACT: In clinical trials with time-to-event endpoints, it is not uncommon to see a significant proportion of patients being cured (or long-term survivors), such as trials for the non-Hodgkins lymphoma disease. The popularly used sample size formula derived under the proportional hazards (PH) model may not be proper to design a survival trial with a cure fraction, because the PH model assumption may be violated. To account for a cure fraction, the PH cure model is widely used in practice, where a PH model is used for survival times of uncured patients and a logistic distribution is used for the probability of patients being cured. In this paper, we develop a sample size formula on the basis of the PH cure model by investigating the asymptotic distributions of the standard weighted log-rank statistics under the null and local alternative hypotheses. The derived sample size formula under the PH cure model is more flexible because it can be used to test the differences in the short-term survival and/or cure fraction. Furthermore, we also investigate as numerical examples the impacts of accrual methods and durations of accrual and follow-up periods on sample size calculation. The results show that ignoring the cure rate in sample size calculation can lead to either underpowered or overpowered studies. We evaluate the performance of the proposed formula by simulation studies and provide an example to illustrate its application with the use of data from a melanoma trial. Copyright © 2012 John Wiley & Sons, Ltd.
    Statistics in Medicine 07/2012; · 2.04 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To determine the relationship between non-high-density lipoprotein cholesterol (HDL-C) lowering and coronary heart disease (CHD) risk reduction for various lipid-modifying therapies. Non-HDL-C is the second lipid target of therapy after low-density lipoprotein cholesterol (LDL-C). Randomized placebo or active-controlled trials were evaluated. The effect of mean non-HDL-C reduction on the relative risk of nonfatal myocardial infarction and CHD death was estimated using Bayesian random-effects meta-analysis models adjusted for study duration. Cochrane's Q was used to test for heterogeneity. Inclusion criteria were met by 14 statin (n = 100,827), 7 fibrate (n = 21,647), and 6 niacin (n = 4,445) trials, and 1 trial each of a bile acid sequestrant (n = 3,806), diet (n = 458), and ileal bypass surgery (n = 838). For statins, each 1% decrease in non-HDL-C resulted in an estimated 4.5-year CHD relative risk of 0.99 (95% Bayesian confidence interval: 0.98 to 1.00). The fibrate model did not differ from the statin model (Bayes factor K = 0.49) with no evidence of heterogeneity. The niacin model was moderately different from the statin model (K = 7.43), with heterogeneity among the trials (Q = 11.8, 5 df; p = 0.038). The only niacin monotherapy trial (n = 3,908) had a 1:1 relationship between non-HDL-C and risk reduction. No consistent relationships were apparent for the 5 small trials of niacin in combination. The 95% confidence intervals for the single trials of diet, bile acid sequestrants, and surgery also included the 1:1 relationship. Non-HDL-C is an important target of therapy for CHD prevention. Most lipid-modifying drugs used as monotherapy have an approximately 1:1 relationship between percent non-HDL-C lowering and CHD reduction.
    Journal of the American College of Cardiology 02/2009; 53(4):316-22. · 14.09 Impact Factor