To the Editor: We thank Søholm et al [1] for their interest in our study comparing the effect of nocturnal hypoglycaemia on quality of life and mood in people with type 1 diabetes [2].

First, to address the question posed by Søholm and colleagues [1], ‘Does nocturnal hypoglycaemia really improve quality of life?’, our data did not show a general effect of nocturnal hypoglycaemia in our cohort, but rather an effect of asymptomatic nocturnal hypoglycaemia only, which was limited to the subgroup of participants with self-reported severely impaired hypoglycaemia awareness (hypoglycaemia unawareness) [2]. The same individuals, who are prone to severe hypoglycaemia, reported lower mean scores on the EuroQol-5D visual analogue scale (EQ-5D VAS) as compared with those with normal awareness, as would be expected.

As Søholm et al state in their letter [1], quality of life comprises a range of different elements in a person’s life, which can make evaluating quality of life a complex challenge. The EQ-5D (including the EQ-5D VAS) is a generic tool to assess health-related quality of life (HRQoL) [3]. The strength of the EQ-5D VAS is that it is a simple and quick examination, which can be applied repetitively, thereby making it possible to assess the impact of nocturnal hypoglycaemia on day-to-day HRQoL. Our study demonstrates that the EQ-5D VAS is useful for the detection of day-to-day changes in HRQoL in type 1 diabetes.

We agree with Søholm et al [1] that there are many other aspects of quality of life that need attention and we support the need for a more comprehensive examination of quality of life in relation to hypoglycaemia.

A major challenge when evaluating changes or differences in quality-of-life measurements is to determine the minimally important difference (MID), i.e. the smallest difference in scores that is not only statistically significant, but also clinically significant. We thank Søholm et al [1] for pointing out that the reference used in our paper was not correct. However, several other studies have reported the MID for the EQ-5D VAS to range between 6.9–10.8 [4,5,6]. These studies include people with chronic diseases other than diabetes (chronic obstructive pulmonary disease, stroke, cancer) and even though the EQ-5D and the EQ-5D VAS has been used widely in diabetes [7], we acknowledge that the clinical importance of changes and differences in EQ-5D VAS in diabetes remains unclear and needs to be properly established. However, as the level of MID is approximately the same throughout the different studies [4,5,6], the MID in diabetes is likely to be in the same range, which supports the suggestion that the 10.3-point difference in EQ-5D VAS between days with or without preceding asymptomatic nocturnal hypoglycaemia observed in our study in participants with hypoglycaemia unawareness is clinically significant [2]. This is further supported by studies assessing the impact of complications of diabetes on EQ-5D VAS [8, 9], which have shown that severe complications, such as blindness, amputation and cardiovascular disease, lead to similar differences in EQ-5D VAS of 3.6–11.2 points.

We agree with Søholm et al [1] that ‘as we focus on CGM-detected, asymptomatic hypoglycaemia in clinical care, we must understand its relevance and impact on self-management, health and quality of life’; hence, further evaluation of the impact of hypoglycaemia on other aspects of quality of life in people with type 1 diabetes is needed, and we hope that our results can inspire further research in the subject. We are aware of the ongoing Hypoglycaemia–MEasurement, ThResholds and ImpaCtS (Hypo-METRICS) study (ClinicalTrails.gov identifier: NTC04304963) and look forward to seeing potentially important outcomes from it.