Most clinical trials assessing drugs for diabetes do not evaluate the medication’s impact on quality of life, the risk of developing complications or other outcomes that patients regard as important, argues a commentary in The Lancet this week.

Diabetes trials usually focus on the medication’s effect on blood sugar levels, a surrogate for the risk of microvascular and macrovascular complications. Using surrogate outcomes allows “smaller, shorter and cheaper trials” and the drugs reach the market more quickly.

However, these “apparent benefits are … a mirage and the apparent savings represent false economy," write Victor Montori and Gunjan Gandhi from the Mayo Clinic in the US and Gordon Guyatt of McMaster University in Canada.

"Any savings are quickly overwhelmed by costs associated with potentially ineffective, or even harmful (yet heavily advertised), expensive therapies and the incremental costs of treating the harms these interventions might cause," they say. "Patients and society may end up paying dearly for drugs that cause more harm than good."

Most studies use glycated haemoglobin (HbA1c) as a surrogate. This assumes that the lower the HbA1c, the lower the risk of complications from diabetes. However, the authors comment that HbA1c is less valid as a surrogate marker “when patients have a constellation of metabolic abnormalities, when the most common complications are macrovascular, and when the treatments have multiple poorly understood effects”.

These issues are common in type 2 diabetes. The studies suggesting that rosiglitazone (GlaxoSmithKline's Avandia) may increase cardiovascular risk further reduced the credibility of HbA1c “as an adequate surrogate”, the authors add.

Measuring effect

The commentary notes that only one in five randomised trials in diabetes published in leading medical journals is powered to measure the drugs’ effect on outcomes that are important to patients. Only 14% of randomised diabetes studies on clinical trial registries assess patient-important outcomes as primary endpoints.

As a result, doctors do not have the high-quality evidence needed to answer fundamental questions, such as whether an asymptomatic patient can expect a better prognosis if they follow a regimen that will reduce HbA1c by 0.5%.

The authors argue that the medical community should insist trials ascertain a potential therapy’s effect on outcomes that are significant to patients, concluding: "This policy will prevent the premature dissemination of therapies that ultimately prove harmful, facilitate patients' participation in decision-making, and speed the day when we can confidently offer treatments that will provide long-term benefits to patients with diabetes."

Montori V, Gandhi G and Guyatt G. Patient-important outcomes in diabetes - time for consensus. Lancet 2007;370:1104-6