Clinical Edge

Summaries of Must-Read Clinical Literature, Guidelines, and FDA Actions

Real-World Effectiveness vs RCT Efficacy in T2D

Diabetes Care; ePub 2017 Aug 11; Carls, et al

In patients with type 2 diabetes (T2D), poor medication adherence is primarily why real-world (RW) effectiveness is significantly less than randomized controlled trials (RCT) efficacy, according to a recent study. The study compared the change in HbA1c of RW patients with T2D 12 months after starting a glucagon-like peptide-1 receptor agonist (GLP-1RA) or dipeptidyl peptidase-4 inhibitor (DPP4) with published findings from RCTs evaluating these drugs. Researchers found:

  • RW patients initiating a GLP-1RA (n=221) or a DPP4 (n=652) had smaller reductions in HbA1c than reported in RCTs. The reductions in the RW vs RCTs were: -0.52 vs 1.30% for GLP-1s and -0.51 vs. -0.68% for DPP4s.
  • Baseline HbA1c additional medications and adherence were significant explanatory factory in the RW HbA1c change.
  • Poor medication adherence accounted for ∼three-fourths of the gap between RW and expected RCT results.


Carls GS, Tuttle E, Tan RD, et al. Understanding the gap between efficacy in randomized controlled trials and effectiveness in real-world use of GLP-1RA and DPP4 therapies in patients with type 2 diabetes. [Published online ahead of print August 11, 2017]. Diabetes Care. doi:10.2337/dc16-2725.


We know that randomized trials give us the most accurate assessment of the effectiveness of a medication. However, it has become increasingly recognized over the last 10 years that randomized trials may not be the best reflection of how a medication works in a real-world setting. The population of patients chosen for a randomized trial may be different in key ways from the patients we take care of day-to-day. The fact that patients took the initiative to sign up for the randomized trial differentiates them from those individuals who did not sign up. These patients may be more motivated than others to take their mediations as prescribed and to persist in taking their medication even if they have some side effects. It is notable that in randomized trials, the GLP-1s were much more potent than DPP-4s, with A1c reductions of -1.30 vs -0.68, while in the real-world the differences between the 2 medication classes were a lot less at -0.52 vs. -0.51. Perhaps this is due to the increased rate of GI side effects with GLP-1s compared to DPP-4s, or an effect of one being an injection and the other an oral medication. Information like this is essential in order to both understand how medications work in the real world and to inform us about medication classes where the gap between RCT and RW use is great, so that we can better enhance compliance with taking medications. —Neil Skolnik, MD