Self-monitoring of blood glucose in type 2 diabetes: a new look at published trials

Abstract

To the Editor: Published randomised controlled trials (RCTs) of self-monitoring of blood glucose (SMBG) in type 2 diabetes have not reached a consistent conclusion with regard to the benefit of using SMBG, especially in non-insulin-treated patients. As a consequence, there is worldwide variability in the use of glucose test strips and policies on reimbursement [1]. We analysed 18 relevant RCTs [2–19], some of which had not been included in previous meta-analyses [20–25]. Why do these trials not provide a reliable answer? The main reason appears to be conceptual. SMBG per se is a diagnostic procedure and not an intervention. Clinical trials, though, compare different types of intervention vs outcome. Hence, the 18 RCTs did not simply compare the same treatment protocol plus/minus SMBG but compared intervention strategies, which differed in many aspects, i.e. SMBG-guided disease management strategies. Each of the RCTs studied a different strategy for SMBG-guided disease management. As a consequence, the clinical outcomes of the different trials can hardly be combined in a metaanalysis. This renders safe conclusions virtually impossible. An SMBG-guided disease management strategy should include some type of algorithm enabling the patient to translate SMBG readings into changes of diet, daily exercise and/or of dose of glucose-lowering medication. Next there should be some guidance for patients during their first months of using SMBG, and, finally, decisions on medication and advice from nurses or doctors should also make use of documented SMBG results. For this advice, some rules or an algorithm on how to translate blood glucose profiles into medical advice should exist in the context of a RCT. We have analysed the RCTs for the SMBG-based disease management strategy employed. Unfortunately, information on this crucial point is quite brief in most reports. Only seven of the 18 study protocols included an algorithm or some guideline to teach patients how to respond to elevated SMBG readings, whether they were fasting or postprandial (Table 1). Reinforcement of patient education by guidance during the study ranged from one visit in 6 months to initially weekly meetings, to continuous support via the Internet. Six trials did not report a procedure for consideration of documented SMBG data for making medical treatment decisions by doctors or nurses; a further two trials apparently evaluated fasting but not postprandial values (Table 1). We conclude that most RCTs of SMBG in type 2 diabetes do not provide a sufficiently detailed description of the SMBG-guided disease management strategy used, even though it is this intervention strategy that is analysed in the trial. As judged from the information provided, there is Diabetologia (2008) 51:686–688 DOI 10.1007/s00125-008-0946-7