Volume 6, Issue 1 (20 2010)
irje 2010, 6(1): 18-25 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Soltanian A, Faghihzadeh S, Gerami A, Mehdibarzi D, Jing Cheng J. Estimation of Treatment Effects in Crossover Clinical Trials with Noncompliance. irje 2010; 6 (1) :18-25
URL: http://irje.tums.ac.ir/article-1-89-en.html
URL: http://irje.tums.ac.ir/article-1-89-en.html
1- , s.faghihzadeh@modares.ac.ir
Abstract: (13324 Views)
Background & Objectives: In clinical trials some of participants do not take assignment treatment. Intention-to-treat (ITT) is one of the strategies to analyze of clinical trials with control. ITT estimation will be invalid and incorrect to show of treatment effects in case of existing non-compliance in participants. In this study we adjusted noncompliance effect to compare of active treatment and placebo.
Methods: To demonstrate efficiency of proposed model, a dataset of crossover clinical trial with 42 patients with knee osteoarthritis was used. To estimate the non-compliance’s effect adjusting at comparison of treatment effects, we use mean of compliance proportion at periods in sequences. The parameters were estimated by maximum likelihood method. ( could you ask authors to have a look at what they wrote and compare with Farsi version)
Results: The results show that baseline variables distributions like duration of disease, severity of disease, age and sex, were not significant (p>0.05). The standard error estimation of treatment effects ( ) based on adjusted model were less than standard model (0.09 and 0.12, respectively). In addition, likelihood ratio statistics based on adjusted model were less than standard model (1177.7 versus 1205.1).
Conclusion: Based on estimation of standard errors and likelihood ratio statistics at adjusted and standard models, we observe that adjusted model is more efficient than standard model.
Methods: To demonstrate efficiency of proposed model, a dataset of crossover clinical trial with 42 patients with knee osteoarthritis was used. To estimate the non-compliance’s effect adjusting at comparison of treatment effects, we use mean of compliance proportion at periods in sequences. The parameters were estimated by maximum likelihood method. ( could you ask authors to have a look at what they wrote and compare with Farsi version)
Results: The results show that baseline variables distributions like duration of disease, severity of disease, age and sex, were not significant (p>0.05). The standard error estimation of treatment effects ( ) based on adjusted model were less than standard model (0.09 and 0.12, respectively). In addition, likelihood ratio statistics based on adjusted model were less than standard model (1177.7 versus 1205.1).
Conclusion: Based on estimation of standard errors and likelihood ratio statistics at adjusted and standard models, we observe that adjusted model is more efficient than standard model.
Keywords: Crossover clinical trials, Intention-to-treat, Noncompliance, Maximum likelihood estimator
Type of Study: Research |
Subject:
General
Received: 2009/07/12 | Accepted: 2009/12/19 | Published: 2013/09/3
Received: 2009/07/12 | Accepted: 2009/12/19 | Published: 2013/09/3
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
