Skip to content
View commentaries on this research

This is a plain English summary of an original research article

Telemedicine, such as text messaging or internet support systems used to communicate with patients, improves long-term blood sugar control in adults with type 1 or type 2 diabetes.

Telemedicine gave small reductions in HbA1c (a measure of overall diabetes control over 12 weeks) compared with usual care at all follow-up times. It was most effective in the short-term, reducing HbA1c by about 0.6% (6mmol/mol) by three months. There was less difference in the medium and long-term, around 0.3% (3mmol/mol) reduction up to one year or more. There was no effect on quality of life, mortality or risk of episodes of low blood sugar.

This review included 111 trials and 23,648 adults with type 1 or type 2 diabetes. A wide variety of interventions were categorised as telehealth. Texting or web portal ways of communicating were associated with slightly larger average effects. Few studies were performed in the UK.

Adding telemedicine to current UK practice for diabetes care may require additional funding and extra staff time so it is important to determine the interface that is most effective.

Why was this study needed?

In 2013, over 3.2 million adults were diagnosed with diabetes in England (6% prevalence) and Wales (6.7% prevalence). Around 90% of these diagnoses are for type 2 diabetes.

In 2012, management of type 2 diabetes and its complications cost the NHS around £8.8 billion a year. Due to the growing burden of type 2 diabetes, this is estimated to increase to £15.1 billion by 2035.

Telemedicine involves delivery of health services through interfaces such as text messaging or Web portals. In this review it was defined as some electronic form of provider-to-patient communication. It has been shown to have benefits in managing several other chronic diseases.

Previous reviews have shown that telemedicine could improve blood sugar control in diabetes, but studies have varied in the delivery formats assessed. This systematic review aimed to update this knowledge, examining the effect of different forms of telemedicine when added to usual care.

What did this study do?

The systematic review and meta-analysis identified 111 randomised controlled trials including 23,648 adults with type 1 or type 2 diabetes.

Telemedicine was delivered via telephone in most cases (59%), via clinical decision support systems in a third, and by Web portals or text messages in roughly 20% each. Nurses provided care in 37% of cases, doctors in 29% and non-specialised support staff in 23%.   Frequency of contact varied from daily to less than monthly.

The main outcome of interest was control of HbA1c. The authors aimed to identify which telemedicine formats were most effective.

Over a third of trials came from the United States with few UK-based. Common sources of bias were around patient allocation and outcome assessment.

What did it find?

  • Telemedicine gave modest reductions in HbA1c at all follow-up times. The mean difference (MD) compared to usual care was:
    • At 3 months or less: -0.57%, 95% confidence interval (CI) -0.74% to -0.40% (including 39 trials).
    • At 4-12 months: -0.28%, 95% CI -0.37% to -0.20% (including 87 trials).
    • At 12 months or more: -0.26%, 95% CI -0.46% to -0.06% (including 5 trials). Across all analyses there was significant variation in the results of the individual trials (heterogeneity), suggesting that the true size of the effects should be interpreted with some caution.
  • The researchers explored whether population characteristics or telemedicine delivery format could explain these differences. Text messaging and Web portals were associated with a greater effect than telephone-based systems (MD -0.28% [95% CI -0.52 to -0.05] for text and -0.35% [95% CI -0.56 to -0.14] for Web portal vs. telephone). Interventions where providers adjusted medication according to patient data gave greater improvements than trials without this component (MD -0.23%, 95% CI -0.42 to -0.05]). Asian trials also demonstrated greater effect than North American trials.
  • There was no evidence to suggest telemedicine enhanced quality of life, affected mortality or reduced the risk of low blood sugar (hypoglycaemic) episodes.

What does current guidance say on this issue?

NICE guidelines on the management of type 1 and type 2 diabetes provide recommendations around the frequency of monitoring of HbA1c, which is typically three to six monthly. For adults with type 1 diabetes, and most people with type 2 diabetes, the target HbA1c is 48 mmol/mol (6.5%) or lower. In type 2 diabetes, NICE recommend patients are involved in discussions around their individual target, taking into account adverse effects such as hypoglycaemia. Self-monitoring of blood sugar is recommended for type 1 diabetes, but only for type 2 in specific circumstances.

Both guidelines recommend telephone support, but neither mentions other forms of telemedicine.

What are the implications?

Earlier studies also support the idea that telemedicine could improve diabetes care. The UK government has plans to train NHS professionals to deliver healthcare using digital technology and to encourage the use of digital services.

There is need to establish which telemedicine interface or format is best, and whether the interventions improve outcomes at a reasonable cost. Texting and web platforms seem promising and potentially cheaper than telephone services as staff time could be minimised.

Technology moves quickly and the latest mobile apps will now need evaluation. These are already more popular than some of the technologies available when these trials were undertaken.

Citation and Funding

Faruque LI, Wiebe N, Ehteshami-Afshar A, et al. Effect of telemedicine on glycated hemoglobin in diabetes: a systematic review and meta-analysis of randomized trials. Canadian Medical Association Journal. 2017;

The work was funded by a team grant to the Interdisciplinary Chronic Disease Collaboration from Alberta-Innovates-Health Solutions; two Alberta Heritage Foundation for Medical Research Population Health Scholar Awards, the Roy and Vi Baay Chair in Kidney Research and Alberta Health with the Universities of Alberta and Calgary. The funding agencies did not have a role in the study conception, study analysis or writing of the manuscript.


NICE. Guidance on type 1 diabetes in adults: diagnosis and management. NG17. London: National Institute for Health and Care Excellence; 2015.

NICE. Type 2 diabetes: prevention in people at high risk. PH38. London: National Institute for Health and Care Excellence; 2012.

DH. New plans to expand the use of digital technology across the NHS. London: Department of Health; 2016.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article

The new HbA1c units (mmol/mol) have replaced the older unit (%) used in this Canadian study. An HbA1c of 6.5% is equivalent to 48mmol/mol and is the threshold sometimes used for diagnosing diabetes. For further values in the ranges used in this study:

  • 7.0% is equivalent to 53mmol/mol
  • 7.5% is equivalent to 58mmol/mol
  • 8.0% is equivalent to 64mmol/mol
  • 8.5% is equivalent to 69mmol/mol
  • 9.0% is equivalent to 75mmol/mol

The HbA1c test measures how much haemoglobin in the blood has become glycated (chemically bonded with glucose) and gives an estimate of overall glucose control over the preceding 12 weeks.

Back to top