Neeser et al. (1) challenge my argument that self-monitoring of blood glucose (SMBG) in type 2 diabetic patients not taking insulin is not beneficial for lowering glycemia and therefore is a waste of (a lot of) money (2).
In a meta-analysis of six randomized controlled trials, Welschen et al. (3) found a significant reduction of 0.39% in HbA1c (A1C) levels in type 2 diabetic patients not taking insulin who performed SMBG. Welschen et al. did point out that the reduction was only significant in two of the trials and that the conclusion that SMBG was beneficial “should be interpreted with caution, as the methodological quality of the trials … was limited in four of the six included studies.” A large number of nonrandomized studies were also negative (2, 3). Be that as it may, Neeser et al. (1), using a Markov model on data from the German health care system, state that a 0.39% reduction of A1C levels resulted in a 30-day (0.083 years) increase of life expectancy and “reduced cost of complications (70% attributable to microvascular events).” The cost per life-year gained was ∼€31,000 (or $36,400) and “therefore, from a health insurance perspective, acceptable.” They conclude that it is “premature to consider withdrawal of this treatment option” and suggest that industry should fund large multicenter trials to determine whether SMBG is helpful in this situation.
Several points can be made in response. Although we are not given the costs of SMBG in non–insulin-requiring patients in the German health care system, I would emphasize that in the U.S., a conservative estimate of the cost of SMBG in these patients is nearly $1.5 billion/year (2). This is a tremendous amount of money for an activity for which there is little (to be charitable) or no evidence for a beneficial outcome. If this were a drug, it certainly would not have received Food and Drug Administration approval. In a sense, therefore, SMBG in patients not taking insulin represents a very expensive “off-label” use. Of course, calls for larger studies are usually appropriate, but realistically speaking, why would industry fund studies that have an excellent chance of showing what a number of smaller studies have already shown, especially since they are making a lot of money in that market already? And, if a larger study were negative, wouldn’t there be cries to do even larger ones? After all, one can really never prove a negative. There is always the possibility that another slight twist or an even larger study could be positive. If it takes a very large number of subjects to show a significant positive result, the clinical benefit must be difficult to uncover. At some point, one has to conclude that enough is enough and we have to accept the results at hand. In the meantime, large amounts of money are being diverted from better uses in our health care system.