Case Study : Intensive glucose lowering therapy and serious cardiovascular outcomes

Question 1

Study Population (ACCORD–glycemic trial)

Answer 1

Patients with T2 diabetes who were at risk for CVD, from 77 clinical centers across the USA and Canada

Question 2

Study Design

Answer 2

RCT (8 different Tx groups including BP and lipid trials) FU planned for 4-8 years (5.6 mean years) but actually ended intensive therapy arm after 3.5 years

Question 3

Exposure assessment

Answer 3

Randomized 1:1 intensive glucose lowering or standard therapy (open label)

Question 4

Sample Size

Answer 4

5,128 intensive therapy and 5123 standard therapy

Question 5

Primary outcome

Answer 5

CVD composite: 1st occurrence of nonfatal MI, nonfatal stroke, CV death (all adjudicated by central committee blinded to Tx group)

Question 6

Outcome assessment

Answer 6

Assessed by investigators at scheduled visits.
-Schedule for glycemic control: Intensive therapy–every month for 4 months, then every 2 months & standard therapy–every 4 months

Question 7

Association

Answer 7

No primary endpoint (HR = 0.9), but death from any cause HR= 1.22 and death from CV cause HR = 1.35

Question 8

Potential Biases

Answer 8

Covariates well balanced at baseline but at the end of the study, more weight gain and hypoglycemia among the intensive therapy arm

Question 9

ACORD Compared to an RCT and an LST

Answer 9

ACCORD had randomization, large sample size, similar inclusion criteria as RCT, lengthy questionnaire like RCT, a mix of hard points, it required patient visits like RCT, physicians and site monitoring status unknown, unknown whether they would continue follow up if treatment was discontinued, and finally, this is somewhat naturalistic like LST but not RCT.