In his counterpoint article, Davidson (1) argues that self-monitoring of blood glucose (SMBG) in type 2 diabetic subjects not using insulin is a waste of money. However, as the discussions accompanying and following publication of a new meta-analysis by Welschen et al. (2) demonstrate, the evidence is far from conclusive either for or against use of SMBG in this patient group.
As pointed out by Ipp et al. (3), many of the trials of SMBG conducted thus far have been underpowered to detect a significant impact and therefore individually cannot reliably conclude that SMBG does or does not influence HbA1c (A1C). In an attempt to bring some clarity to the current situation, Welschen et al. performed a meta-analysis based upon pooling of more recent randomized trials with the conclusion that SMBG affords a modest but significant 0.39% reduction in A1C. According to Davidson, even if this effect is clinically relevant, it is likely to be outweighed by the cost of providing SMBG. To accurately answer that claim, we undertook a cost-effectiveness analysis of SMBG using a Markov state model of diabetes to assess the clinical impact and related cost when SMBG is provided to non–insulin-requiring patients within the German health care system. Assuming a modest improvement in A1C of 0.39%, the result was a slight increase in life expectancy (0.083 years) and reduced cost of complications (70% attributable to microvascular events). This finding is in line with the results of the U.K. Prospective Diabetes Study, in which a 1% reduction in A1C corresponded to a reduction in complications (4). In our analysis, the cost per life-year gained was ∼€31,000 and therefore, from a health insurance perspective, acceptable. Over a 10-year period and taking into consideration cost savings due to reduced complications, SMBG employed at a frequency of seven times/week would account for ∼6% of the total direct costs covered by health insurance.
While current evidence is not perfect, it supports, on both clinical and economic grounds, the use of SMBG in type 2 diabetic subjects not using insulin. Therefore, it would be premature to consider withdrawal of this treatment option. As noted by Ipp et al. (3), now is the time for industry to fund large multicenter trials with sufficient power to confirm the findings obtained by pooling small randomized controlled trials.
K.N., K.M.E.-A., and C.W. have received grant/research support from Hoffman-La Roche.