Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Assessment of intermediate (process) outcomes in the BARI 2D clinical trial

Liu, Linxi (2016) Assessment of intermediate (process) outcomes in the BARI 2D clinical trial. Master's Thesis, University of Pittsburgh. (Unpublished)

Submitted Version

Download (649kB)


BARI 2D is a 2-by-2 factorial clinical trial identifying optimal therapies for patients with documented stable coronary artery disease and Type 2 Diabetes. Previous results of the BARI 2D study indicated that there was a non-significant association between the therapies and 5-year survival outcomes. Our study hypothesizes that several intermediate variables may exist whose trajectories of improving or worsening in response to treatments act against each other leading to the previous non-significant associations. Before randomization, the participants were stratified to two types of revascularization - PCI or CABG. Then each participant was randomized to receive one cardiac therapy (prompt revascularization plus intensive medicine or intensive medicine alone), and also one glycemic therapy (insulin-sensitization or insulin-provision). Five intermediate outcomes BMI, SBP, HDL, LDL and Hba1c were studied. Linear regression was conducted to assess the difference of each intermediate outcome between therapies at Year 1 and at Year 3. Also conducted was a comparison of the trajectories over time by therapies using longitudinal repeated measures mixed models. In 2368 patients (mean age 62.4 years), the improvement of the intermediate outcomes was generally notable over the first year, and then slowly diminished over time. At both Year 1 and 3, insulin-provision resulted in higher BMI and Hba1c, and lower HDL than insulin-sensitization, irrespective of the assigned cardiac therapy and revascularization stratum. Longitudinally, insulin-provision resulted in higher Hba1c, irrespective of the assigned cardiac therapy and revascularization stratum; and an interaction effect of the cardiac and glycemic therapies on BMI was found in CABG stratum. These results suggested that insulin-sensitization therapy is superior to insulin-provision therapy generally. In CABG stratum, the effect of cardiac therapy on BMI depends on the assignment of glycemic therapy. The public health significance of this study is that, though the cancelling-out hypotheses for these five intermediate variables may be overly optimistic, it involves a potentially illuminating perspective to explain the mechanisms through which the BARI 2D treatment therapies affect multiple intermediate outcomes. This in turn could also help inform and enhance the targeted adjuvant therapies, thus resulting in improved survival outcomes by better focusing on controlling the harmful intermediate variables.


Social Networking:
Share |


Item Type: University of Pittsburgh ETD
Status: Unpublished
CreatorsEmailPitt UsernameORCID
Liu, Linxilil92@pitt.eduLIL92
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Thesis AdvisorBoudreau, RobertBoudreauR@edc.pitt.eduROB21
Committee MemberBrooks, Mariambrooks@pitt.eduMBROOKS
Committee MemberMarroquin, Oscarmarroquinoc@upmc.eduOSM3
Date: 29 June 2016
Date Type: Publication
Defense Date: 6 May 2016
Approval Date: 29 June 2016
Submission Date: 12 May 2016
Access Restriction: 1 year -- Restrict access to University of Pittsburgh for a period of 1 year.
Number of Pages: 47
Institution: University of Pittsburgh
Schools and Programs: School of Public Health > Epidemiology
Degree: MS - Master of Science
Thesis Type: Master's Thesis
Refereed: Yes
Uncontrolled Keywords: Coronary Artery Disease; Type 2 Diabetes; Intermediate; Risk Factors; BMI; HDL; LDL; SBP; Hba1c
Date Deposited: 29 Jun 2016 18:16
Last Modified: 01 May 2017 05:15


Monthly Views for the past 3 years

Plum Analytics

Actions (login required)

View Item View Item