In our article, published in the March edition of this journal, we state that ‘it may be better to focus on other risk factors, such as smoking, high blood pressure, and lipid profile disturbances, than to aim for increasingly lower therapeutic values for HbA1c.’1 The validity of our conclusions was confirmed by a recently published large retrospective study.2 Although the design was different, it emphasised the absence of benefit of strict glycaemic control in patients with longer diabetes duration. In fact, this study even showed an increased mortality in patients with HbA1c under 7.5% who underwent treatment intensification with insulin.
In his comment to our article, Searle points out that there are no baseline differences in these risk factors between the survivors and the deceased in our study.3 Although this observation is correct, we respectfully disagree that it contradicts our statement. Absence of differences in baseline characteristics, for example smoking, does not mean that smoking is not an independent risk factor for mortality. To answer the question whether smoking, blood pressure, and cholesterol levels are related to mortality, Cox regression analyses, including correction for confounders, are an option in order to better interpret a (possible) effect of, in this case, HbA1c on mortality. For example, in the same study cohort, we studied the relationship between mortality and lipid profile in different age groups.4 In this study, higher cholesterol levels did relate to mortality.
We agree with Searle that the benefits of interventions, as studied in randomised controlled trials, do not necessarily translate to improvements in daily practice. Many trials include a selected population and are, therefore, not representative of the general population. However, our results more or less confirm the results of these trials, like the UKPDS, that we discussed in our article.