Contrary to the widely-held belief, there is no proof that non-insulin-dependent patients with Type 2 diabetes benefit from glucose self-monitoring. This is the conclusion of the final report of the Institute for Quality and Efficiency in Health Care (IQWiG).
The current data are quantitatively and qualitatively inadequate. The few trials that are suitable for investigating these questions have not included or have insufficiently reported many outcomes important to patients. Owing to their short duration, it is also not possible to draw any conclusions on the long-term benefit of glucose self-monitoring.
Besides drug therapy, lifestyle, especially diet and exercise, also plays an important role in the treatment of Type 2 diabetes. Many experts assume that blood glucose self-monitoring helps patients in adapting their lifestyle, because the measured values enable them to see the direct effect of diet and physical activity and then to take suitable measures. The result should be that their blood glucose is better controlled and acute and long-term complications are reduced — at least this is the assumption.
In order to examine whether these assumptions can be scientifically proven, IQWiG searched for comparative trials with and without self-monitoring. Self-monitoring could also be a component of a complex education and treatment program, such as are often offered to patients with diabetes. These trials were included if the participants in the treatment and control groups received the same treatment regimen — the only difference being that one group was with self-monitoring and the other without.
All trials included in the assessment additionally investigated the impact of blood glucose self-monitoring on the HbA1c value. The joint analysis revealed that blood glucose self-monitoring actually does assist in lowering blood glucose. However, the difference was marginal compared to the group that did not carry out self-monitoring. It was inside the range that is acceptable within the context of drug approval for describing a new drug as “not inferior” compared to existing drugs. No health advantage from this difference can therefore be anticipated.
Overall, IQWiG and its external experts found 6 randomized controlled trials that were suitable for investigating the impact of medical interventions on the course of the disease. In all the included trials, education was a component in the therapy strategy. All 6 trials investigated the benefit of blood glucose self-monitoring; no suitable clinical comparisons were identified on urine glucose self-monitoring. The duration of the included trials was between 6 and 12 months, in other words, none of them were designed to investigate the long-term benefit of self-monitoring.
However, data on criteria that were important for the patient-relevant benefit were not even collected in these trials. This applies in particular to concomitant and late complications caused by diabetes, such as sight loss or cardiac disease. Other outcomes, such as quality of life and patient satisfaction, were in fact investigated in a few trials but inadequately reported, so that the results cannot be accepted as reliable. Yet even the few available data did not display any advantage for self-monitoring.
According to IQWiG and its external experts, therefore, the quality of trials on glucose self-monitoring is still inadequate overall. What is lacking are trials of longer duration that enable the long-term effects of glucose self-monitoring to be evaluated. Even the Canadian Agency for Drugs and Technologies in Health (CADTH) has complained in its most recent assessment of blood glucose self-monitoring that there is a lack of long-term trials.
Institute for Quality and Efficiency in Health Care, January 2010