NIHR DC Discover

Woman testing for blood glucose levels

NIHR Signal Self-monitoring of blood glucose provides no important benefit for most people with type 2 diabetes

Published on 23 October 2018

doi: 10.3310/signal-000663

Patients with type 2 diabetes who monitor their blood glucose themselves may see small, short-term improvements in blood sugar control. This is not enough to be clinically important or outweigh the costs and personal inconvenience of long-term self-testing.

Self-monitoring is a well-established strategy for type 1 diabetes and for people with type 2 who need insulin. The benefit for all people with type 2 is debatable. This review pooled 24 randomised controlled trials comparing self-monitoring with any control strategy for people not taking insulin.

Self-monitoring gave a 0.3 percentage point reduction in glycated haemoglobin (HbA1c) at six months. This is just below the 0.4% threshold for a meaningful clinical difference in this measure of average 3-month sugar control. People who had poorer blood glucose control at the start saw a greater benefit. However, there was no difference between the self-monitoring and control groups by 12 months.

The review supports current guideline recommendations that self-monitoring is not routinely used for people with type 2 diabetes controlled on diet or tablets.  

Share your views on the research.

Why was this study needed?

There are around 3.7 million people living with diabetes in the UK. Around 90% of those affected have type 2 diabetes.

Management of type 2 diabetes usually begins with lifestyle changes, followed by the addition of oral blood-glucose-lowering medications, progressing to additional drugs and insulin if needed. Blood glucose is monitored by measuring HbA1c every three to six months. The aim is to keep HbA1c ideally below 6.5% or 48mmol/mol.

Self-monitoring of blood glucose is important for patients treated with insulin. A 2012 Cochrane review identified 12 trials assessing benefit for people with type 2 diabetes who are not taking insulin. It found that monitoring gave only small improvements in blood glucose that were not maintained beyond six months.

This review took another look at the topic to establish whether there is now enough evidence to revise clinical practice.

What did this study do?

This systematic review identified 24 recent randomised controlled trials involving 5,454 people with type 2 diabetes. Average HbA1c at baseline was 8.1% (65-66mmol/mol).

The majority of trials compared self-monitoring of blood glucose with no intervention, two compared with urine glucose monitoring and two compared with once-weekly monitoring. Twelve studies took ≥seven glucose measures per week, others took daily measures, and some took multiple daily measures on a few days per week.

The reviewers excluded trials of continuous self-monitoring; where monitoring was part of a more complex intervention (such as education); and where patients were taking several daily insulin injections (basal bolus). Four studies included people on less intensive insulin.

The evidence was judged as low to moderate quality, mainly because participants and assessors were aware of the treatment they were allocated to and study results were variable.

What did it find?

  • At three months, HbA1c was 0.31% lower (95% confidence interval [CI]: -0.57 to -0.05) in the self-monitoring compared with the control group. This is just below the 0.4% threshold considered to be a meaningful clinical difference. The meta-analysis included 11 studies (2,558 patients) with highly variable results giving less confidence in the pooled estimate.  
  • A similar difference was maintained at six months: HbA1c 0.34% lower (95% CI -0.52 to -0.17) in the self-monitoring group. This meta-analysis included 19 studies (4,338 patients), again with mixed results.
  • After 12 months, there was no difference between the self-monitoring and control groups (HbA1c -0.10%; 95% CI -0.28 to +0.08), based on 10 studies (2,427 patients).
  • Patients whose HbA1c was higher than 8% at baseline benefited more from self-monitoring. They showed 0.83% greater reduction in HbAc1 at three months (95% CI -1.55 to -0.11; four studies) and -0.48% at six months (95% CI -0.77 to -0.19; 11 studies). However, they too showed no benefit after 12 months (0.01%; 95% CI -0.10 to +0.12; four studies).

What does current guidance say on this issue?

The 2015 NICE guideline on management of type 2 diabetes recommends that self-monitoring should not be routinely offered unless the person:

  • is taking insulin,
  • has low blood glucose (hypoglycaemia),
  • is taking oral medication that may increase their risk of hypoglycaemia while driving or operating machinery,
  • is pregnant or planning to become pregnant.

Doctors may consider offering self-monitoring of blood glucose in the short-term for people starting treatment with steroids or to confirm suspected hypoglycaemia.

What are the implications?

This study finds an additional 12 trials compared with a 2012 Cochrane review and arrives at broadly the same conclusion. For people with type 2 diabetes who are not using insulin, any benefit from self-monitoring is small and doesn’t last beyond six months.

There are some limitations to the evidence in terms of the diversity of the trials, patient groups and monitoring regimens. But overall it supports guideline recommendations that self-monitoring is not used routinely for people with type 2 diabetes unless there is a specific reason to do so.

Citation and Funding

Machry RV, Rados DV, Gregório GR, Rodrigues TC. Self-monitoring blood glucose improves glycemic control in type 2 diabetes without intensive treatment: A systematic review and meta-analysis. Diabetes Res Clin Pract. 2018;142:173-87.

Funded by Fundo de Incentivo a Pesquisa do Hospital de Clinicas de Porto Alegre, Brazil.  

Bibliography

Diabetes UK. Diabetes prevalence 2017. London: Diabetes UK; 2017.

NHS. Blood glucose test strips for patients with type 2 diabetes mellitus. PrescQIPP Bulletin. 2013;46.

NICE. Type 2 diabetes in adults: management. NG28. London: National Institute for Health and Care Excellence; 2015.

Malanda UL, Welschen LM, Riphagen II, et al. Self monitoring of blood glucose in patients with type 2 diabetes mellitus who are not using insulin. Cochrane Database of Syst Rev. 2012;(1):CD005060.

Why was this study needed?

There are around 3.7 million people living with diabetes in the UK. Around 90% of those affected have type 2 diabetes.

Management of type 2 diabetes usually begins with lifestyle changes, followed by the addition of oral blood-glucose-lowering medications, progressing to additional drugs and insulin if needed. Blood glucose is monitored by measuring HbA1c every three to six months. The aim is to keep HbA1c ideally below 6.5% or 48mmol/mol.

Self-monitoring of blood glucose is important for patients treated with insulin. A 2012 Cochrane review identified 12 trials assessing benefit for people with type 2 diabetes who are not taking insulin. It found that monitoring gave only small improvements in blood glucose that were not maintained beyond six months.

This review took another look at the topic to establish whether there is now enough evidence to revise clinical practice.

What did this study do?

This systematic review identified 24 recent randomised controlled trials involving 5,454 people with type 2 diabetes. Average HbA1c at baseline was 8.1% (65-66mmol/mol).

The majority of trials compared self-monitoring of blood glucose with no intervention, two compared with urine glucose monitoring and two compared with once-weekly monitoring. Twelve studies took ≥seven glucose measures per week, others took daily measures, and some took multiple daily measures on a few days per week.

The reviewers excluded trials of continuous self-monitoring; where monitoring was part of a more complex intervention (such as education); and where patients were taking several daily insulin injections (basal bolus). Four studies included people on less intensive insulin.

The evidence was judged as low to moderate quality, mainly because participants and assessors were aware of the treatment they were allocated to and study results were variable.

What did it find?

  • At three months, HbA1c was 0.31% lower (95% confidence interval [CI]: -0.57 to -0.05) in the self-monitoring compared with the control group. This is just below the 0.4% threshold considered to be a meaningful clinical difference. The meta-analysis included 11 studies (2,558 patients) with highly variable results giving less confidence in the pooled estimate.  
  • A similar difference was maintained at six months: HbA1c 0.34% lower (95% CI -0.52 to -0.17) in the self-monitoring group. This meta-analysis included 19 studies (4,338 patients), again with mixed results.
  • After 12 months, there was no difference between the self-monitoring and control groups (HbA1c -0.10%; 95% CI -0.28 to +0.08), based on 10 studies (2,427 patients).
  • Patients whose HbA1c was higher than 8% at baseline benefited more from self-monitoring. They showed 0.83% greater reduction in HbAc1 at three months (95% CI -1.55 to -0.11; four studies) and -0.48% at six months (95% CI -0.77 to -0.19; 11 studies). However, they too showed no benefit after 12 months (0.01%; 95% CI -0.10 to +0.12; four studies).

What does current guidance say on this issue?

The 2015 NICE guideline on management of type 2 diabetes recommends that self-monitoring should not be routinely offered unless the person:

  • is taking insulin,
  • has low blood glucose (hypoglycaemia),
  • is taking oral medication that may increase their risk of hypoglycaemia while driving or operating machinery,
  • is pregnant or planning to become pregnant.

Doctors may consider offering self-monitoring of blood glucose in the short-term for people starting treatment with steroids or to confirm suspected hypoglycaemia.

What are the implications?

This study finds an additional 12 trials compared with a 2012 Cochrane review and arrives at broadly the same conclusion. For people with type 2 diabetes who are not using insulin, any benefit from self-monitoring is small and doesn’t last beyond six months.

There are some limitations to the evidence in terms of the diversity of the trials, patient groups and monitoring regimens. But overall it supports guideline recommendations that self-monitoring is not used routinely for people with type 2 diabetes unless there is a specific reason to do so.

Citation and Funding

Machry RV, Rados DV, Gregório GR, Rodrigues TC. Self-monitoring blood glucose improves glycemic control in type 2 diabetes without intensive treatment: A systematic review and meta-analysis. Diabetes Res Clin Pract. 2018;142:173-87.

Funded by Fundo de Incentivo a Pesquisa do Hospital de Clinicas de Porto Alegre, Brazil.  

Bibliography

Diabetes UK. Diabetes prevalence 2017. London: Diabetes UK; 2017.

NHS. Blood glucose test strips for patients with type 2 diabetes mellitus. PrescQIPP Bulletin. 2013;46.

NICE. Type 2 diabetes in adults: management. NG28. London: National Institute for Health and Care Excellence; 2015.

Malanda UL, Welschen LM, Riphagen II, et al. Self monitoring of blood glucose in patients with type 2 diabetes mellitus who are not using insulin. Cochrane Database of Syst Rev. 2012;(1):CD005060.

Self-Monitoring Blood Glucose improves glycemic control in Type 2 Diabetes without intensive treatment: a systematic review and meta-analysis

Published on 2 June 2018

Vaz Machry, R.,Varvaki Rados, D.,Ribeiro de Gregorio, G.,Costa Rodrigues, T.

Diabetes Res Clin Pract , 2018

AIMS: Systematic review and meta-analysis to evaluate the effect of Self-Monitoring of Blood Glucose (SMBG) on glycemic control in patients with type 2 Diabetes (T2D). METHODS: We searched the Medline, Embase, Cochrane Central, and ClinicalTrials.gov databases up to 20 July 2017. We also performed a manual search of abstracts from recent meetings of the American Diabetes Association and the European Association for the Study of Diabetes. STUDY SELECTION: randomized controlled trials (RCTs) conducted in patients with T2D comparing any kind of SMBG to a control group. Two independent reviewers assessed the eligibility of references. Influence of SMBG in glycated hemoglobin (HbA1c) was aggregated as weighted mean difference accessed by direct random effect meta-analyses at 12, 24 weeks and 1 year. Sub-analyses were made to assess the effects of previous glycemic control and number of tests performed. RESULTS: SMBG was associated with a reduction of HbA1c at 12 weeks (-0.31%; 95% CI: -0.57 to -0.05) and 24 weeks (-0.34%; 95%CI: -0.52 to -0.17), but no difference was found for 1 year. Subgroup analysis including studies with baseline HbA1c greater than 8% showed a higher reduction of HbA1c: -0.83% (95% CI: -1.55 to -0.11) at 12 weeks, and -0.48% (95% CI: -0.77 to -0.19) at 24 weeks, with no difference for 1 year nor for the stratification for number the tests. CONCLUSION: SMBG seems to lead to a slightly better glycemic control in the short term in patients with T2D. Patients decompensated at baseline appear to have the greatest benefit. PROSPERO register: CRD42016033558.

Expert commentary

Recently, there has been controversy about the value of self-monitoring of blood glucose by people with type 2 diabetes who are not treated with insulin.

The systematic review of 24 studies including 5,454 people with type 2 diabetes shows a small, short-term reduction in HbA1c (a measure of glucose control) which is unlikely to be of clinical relevance.

There is little in this paper to advocate a general recommendation for routine self-monitoring by people with type 2 diabetes, although some individuals may still benefit.

Richard Holt, Professor in Diabetes and Endocrinology, Honorary Consultant Physician, University of Southampton

Expert commentary

Combining similar trials and pooling large amounts of data gives us new insights into clinical interventions, like self-testing for diabetes.

This analysis of 24 similar trials looking at self-monitoring of blood glucose and glycaemic control, points to possible early benefits especially for patients who have poor glycaemic control at the start of the trials.

There are significant variations in the regularity of blood glucose monitoring and use of oral agents which need to be taken into consideration in the interpretation of the data.

Dr Elizabeth Robertson, Director of Research, Diabetes UK