Advice
Clinical and technical evidence
Clinical and technical evidence
A literature search was carried out for this briefing in accordance with the interim process and methods statement for medtech innovation briefings. This briefing includes the most relevant or best available published evidence relating to the clinical effectiveness of the technology. Further information about how the evidence for this briefing was selected is available on request by contacting mibs@nice.org.uk.
Published evidence
There are 4 studies summarised in this briefing, including 1 randomised controlled trial, 2 retrospective studies and 1 prospective pilot study. In total, there were 611 people included in these studies. In addition to the studies presented in full, there is 1 feasibility study in 46 people (Bergenstal et al. 2012) and a pilot study in 14 people (Bashan et al. 2012). These included people with type 1 or type 2 diabetes and are not presented here in detail.
The clinical evidence and its strengths and limitations is summarised in the overall assessment of the evidence.
Overall assessment of the evidence
These studies were done in appropriate populations and reported relevant outcome measurements. Three of the studies were done in the UK and so are generalisable to NHS practice. Two of these provided longer term data over a period of more than 2 years. However, these UK‑based studies were single‑arm, meaning there is no comparable evidence to standard care in the UK. One study was a randomised controlled trial and used an appropriate comparator. However, it was limited by a 6‑month follow-up period and the study was done in the US. This study showed a reduction in HbA1c levels in the intervention group compared with the control, with no statistical difference in hypoglycaemic events. The company highlights that reducing HbA1c levels while not increasing the risk of hypoglycaemia is a key benefit of the technology. All studies were limited by the evidence being on the physical device rather than the app. Although the company states that the software is the same in both versions of the device, there is no evidence showing whether the app version alone would affect efficacy of the technology.
Bergenstal et al. (2019)
Intervention and comparator
Intervention: d‑Nav insulin guidance system (a handheld device that contains a glucose meter and software that adjusts insulin dose) and healthcare professional support.
Comparator: healthcare professional support alone.
Key outcomes
The study was done over a 6‑month period. At baseline, mean HbA1c was 8.7% (standard deviation [SD] 0.8%) in the intervention group and 8.5% (SD 0.8%) in the control group. The mean decrease in HbA1c after 6 months was 1% (SD 1%) in the intervention group and 0.3% (SD 0.9%) in the control group (p<0.0001). There was a similar frequency of hypoglycaemic events per month in each group (p=0.96).
Strengths and limitations
This multicentre randomised controlled study used an appropriate comparator, and baseline characteristics between groups appear similar, although no statistical comparisons were done. However, this study is limited by a relatively short (6‑month) follow-up period and the study was done in the US, which could limit generalisability to the NHS. This study was also done on the device version rather than the app for this technology.
Harper et al. (2018)
Study size, design and location
Retrospective cohort study of 246 people with type 2 diabetes who use insulin in the UK.
Intervention and comparator
Intervention: d‑Nav insulin guidance system (a handheld device that contains a glucose meter and software that adjusts insulin dose); no comparator.
Key outcomes
Patient data was collected for 2.8 years (plus or minus 0.9). There were 70.3% of people who went through periods when their total daily dose was reduced. HbA1c decreased from 9.4% (plus or minus 1.6%) at baseline to 7.1% (plus or minus 0.9%). The frequency of hypoglycaemic episodes (less than 54 mg/dl) was low at 0.5 (plus or minus 0.6) episodes per month.
Strengths and limitations
This study was done in the UK and included people who have used the service for more than 19.5 months who have had no gap in insulin dosage information greater than 3 months. However, the study was observational and had no comparator group. Because the data was collected retrospectively, there was also a lack of information on additional glucose‑lowering medications. This study was funded by the company and used the d‑Nav device rather than the app.
Donnelly et al. (2015)
Intervention and comparator
Intervention: d‑Nav insulin guidance system (a handheld device that contains a glucose meter and software that adjusts insulin dose); no comparator.
Key outcomes
There were 28 people who withdrew from the study, 19 of which were related to not being able to use the device correctly. Of the 94 people who completed the study, the mean (and standard deviation) HbA1c decreased from 77 mmol/mol (plus or minus 15 mmol/mol; 9.2% plus or minus 1.4%) at baseline to 62 mmol/mol (plus or minus 13 mmol/mol; 7.8% plus or minus 1.2%) at the 3‑to-5‑month clinic visit and to 59 mmol/mol (plus or minus 13 mmol/mol; 7.5% plus or minus 1.2%) at the 6‑to-12‑month clinic visit. The study reported a low frequency of minor hypoglycaemia (blood glucose less than or equal to 3.6 mmol/litre).
Harper et al. (2016)
Study size, design and location
Retrospective analysis of 62 people who had used the d-Nav technology for more than a year in the UK.
Intervention and comparator
Intervention: d‑Nav insulin guidance system (a handheld device that contains a glucose meter and software that adjusts insulin dose); no comparator.
Key outcomes
People were studied over an average of 2.1 years (plus or minus 0.5 years' standard deviation). After 9 months, stability in HbA1c level was achieved with an average HbA1c of 7.4% plus or minus 0.2% (57.4 mmol/mol plus or minus 1 mmol/mol). On average, insulin dose decrease events happened in 56.5% of people, with an average of 0.8 (plus or minus 0.4) events per year.
Strengths and limitations
Although this study was done in the UK, it is limited by being a retrospective study with no comparator group. The analysis was also limited to people who have used the service for more than 1 year and excluded those who had 2 or more episodes of severe hypoglycaemia in the past year. This study was also funded by the company and was on the d‑Nav device rather than the app.