Self monitoring of blood glucose in non-insulin-treated Type 2 diabetes

Transcription

1 Self monitoring of blood glucose in non-insulin-treated Type 2 diabetes A report prepared by an NHS Diabetes Working Group Supporting, Improving, Caring

2

3 Foreword This report was commissioned by Dr Rowan Hillson MBE, National Clinical Director for Diabetes in England at the Department of Health. Dr Hillson said People with diabetes for whom self monitoring of blood glucose helps to improve health and well-being should be helped to use this technology in the best possible way for them. People with diabetes for whom it is not helpful should not be asked to do finger prick tests. A working group was set up with all stakeholders represented. The request came against uncertainty about the effectiveness of blood glucose monitoring in individuals with non-insulin-treated Type 2 diabetes and a background of rising costs of SMBG at a time of financial stringency. Methods In order to inform the Working Group s discussions, a systematic review was commissioned by the HTA Programme, from the Aberdeen Health Technology Assessment group. The working group met on two occasions and had correspondence on the draft report. Membership Professor Simon Heller (Chair) Eleanor Kennedy Francesca Arundel Kath Barnard Sheila Burston Andrew Farmer Adrian Griffin Professor Martin Gulliford Julia Lawton Maurice O Kane Mark Samuels Jay Solanki Grace Sweeney Bridget Turner Professor Norman Waugh Doris-Ann Williams Funding NHS Diabetes 3

4 Executive Summary The aim of diabetes care is to improve the health and well being of people with diabetes. 1) Self monitoring of blood glucose is an essential component of management in insulin treated individuals with diabetes, helping patients to achieve tight blood glucose control and to identify low blood glucose values before the development of severe hypoglycaemia. The situation in non-insulin-treated Type 2 diabetes is unclear. There is increasing concern that health service managers and GPs are using recently published evidence to prevent even individuals who find blood glucose monitoring useful from checking their blood glucose whenever they feel they need to. 2) NHS Diabetes has formed a working group to prepare a report examining the practice of self monitoring of blood glucose (SMBG) in people with non-insulin-treated Type 2 diabetes. The remit of the group was to: 1. provide recommendations regarding the place of SMBG in the management of non-insulintreated Type 2 diabetes in England 2. summarise existing guidelines regarding SMBG in the management of non-insulin-treated Type 2 diabetes 3. commission a systematic review of SMBG in non-insulin-treated Type 2 diabetes 4. identify gaps in the evidence and make recommendations for future research 3) The systematic review compromised a review of previous systematic reviews, and systematic review of new trials, with a meta-analysis of all old and new trials plus a review of qualitative and economic studies. Trials comparing simple SMBG with no SMBG, found a reduction in HbA1c of 0.21%. Trials of enhanced SMBG (with education, feedback, etc) showed a bigger reduction in HbA1c, of 0.52% compared to no SMBG. The qualitative studies showed that: SMBG could motivate and improve diabetes self-management in some people SMBG can cause adverse psychological effects including depression and selfchastisement There was a lack of education in how to interpret and use the data from monitoring There was a lack of interest in the results from health care professionals Failure to act on the results was common. 4) The data from clinical trials show that in patients treated by lifestyle modification or oral agents, SMBG with appropriate education and clear objectives leads to improvement in blood glucose control. The improvements in blood glucose control when SMBG is used without education and clear objectives are so small as to be of doubtful clinical benefit. There is also evidence that some psychosocial outcomes deteriorate. Pricking one s finger is painful and some people find it distressing. There is a strong case for stopping blood glucose monitoring activity in people who derive no benefit from it, particularly where it is damaging their quality of life. 5) SMBG should be available to people receiving sulphonylurea treatment because of the risk of hypoglycaemia with this treatment. 6) Some newly diagnosed individuals and others with diabetes of longer duration but not taking insulin clearly benefit from measuring their blood glucose. In those individuals who find modern technology essential to managing their diabetes, arbitrary withdrawal of treatment should not occur. It would limit patient choice and may lead to a deterioration in blood glucose control. 4

5 7) Training programmes for professionals need to be modified to address this important area and incorporate it within the local diabetes care package. In the absence of clear evidence for general benefit, training should include the need to support patients who find SMBG important in managing their diabetes successfully and stopping monitoring in those who do not find it helpful. 8) Blood glucose monitoring equipment supplied directly by retail pharmacists is often wasted and ineffective without guidance on how to integrate the information into ongoing care. To help people with diabetes, written information agreed by healthcare professionals, industry and pharmacists should be developed, providing simple educational principles of how to use the technology to aid self-management. 9) Future research should include a detailed exploration of the factors which determine success in incorporating glucose monitoring into diabetes self-management. This will inform the development of more effective interventions, evaluated by the most appropriate study designs to inform improved clinical practice and better targeting of resources. 10) Savings can be made by stopping blood glucose monitoring among those who derive no benefit and it is important that the resources generated by such activity are used locally to support other aspects of diabetes care, particularly education of both staff and patients. Recommendations for SMBG in people with non-insulin-treated diabetes 1. SMBG with appropriate structured education should be available to people receiving sulphonylurea treatment to identify hypoglycaemic episodes. 2. In keeping with the recommendations contained within NICE Clinical Guideline CG 87, SMBG should only be provided routinely to people with Type 2 diabetes not treated with insulin or sulphonylureas where there is an agreed purpose or goal to testing. 3. SMBG should be used only within a care package, accompanied by structured education which should include clear instructions as to the place of monitoring and how results can be used to reinforce lifestyle change, adjust therapy or alert healthcare professionals. This should include regular review to identify and support those who find it useful while discouraging those who gain no clinical benefit from continuing to test. 4. Individuals with non-insulin treated diabetes who are motivated by SMBG activity and use the information to maximise the effect of lifestyle and medication should be encouraged to continue to monitor. 5. Staff training in the use of SMBG to support changes in lifestyle and self-adjustment of medications is required 6. Savings from a reduction in SMBG in individuals with non-insulin treated diabetes should be used to provide both structured education and training of professionals. 7. Future research should focus on how to identify those who will gain most from SMBG and establish how they integrate it successfully into their approach to self-management. Introduction The introduction, in the late 1970s of blood glucose monitoring (self management of blood glucose SMBG) to enable self-management of diabetes is now recognised as a revolution in the management of type 1 diabetes. Yet, 30 years later the application of SMBG to type 2 diabetes continues to provoke considerable controversy. Almost all of those involved in diabetes care would agree that technology which allows insulin treated individuals to achieve tight blood glucose control while identifying and treating hypoglycaemia is an essential component of diabetes treatment. Yet the NHS spends millions of pounds in providing monitoring strips to people with diabetes; many either ignore the process altogether while others record the results diligently with no idea of what they mean or what to do with them. Whilst some individuals clearly benefit from testing, recent 5

6 evidence suggests that monitoring impairs quality of life and increases symptoms of anxiety and depression in many individuals. The suggestion that blood glucose monitoring may be a waste of time and money is difficult to accept, particularly by patients who derive considerable value from the exercise. There is increasing concern that health service managers and GPs are using recently published evidence to prevent even individuals who find blood glucose monitoring useful from checking their blood glucose whenever they feel they need to. These issues have prompted NHS Diabetes to commission a working group to prepare a report examining the practice of self monitoring of blood glucose (SMBG) in Type 2 diabetes. The remit of the group was to: 1. commission a systematic review of SMBG in Type 2 diabetes 2. identify gaps in the evidence and make recommendations for future research 3. summarise existing guidelines regarding SMBG in the management of Type 2 diabetes 4. provide recommendations regarding the place of SMBG in the management of Type 2 diabetes in the UK Brief Historical Review and Background Self monitoring of blood glucose rapidly became established in the management of diabetes in the early 1980s following the demonstration that patients with Type 1 diabetes could use mains operated glucose meters at home during pregnancy to guide adjustments in insulin dose. (1) The contrast between its precision even with the relatively crude instruments of the time when compared with testing for urinary glucose was immense. The ability to identify glucose values in real time and identify hypoglycaemia at an early stage were major advantages. Blood glucose monitoring rapidly superseded urine testing and became established as the monitoring method of choice in individuals with Type 1 diabetes. Technological advances with smaller, more accurate portable meters running on batteries, contributed to the advent of intensive insulin therapy with multiple injections of insulin adjusted according to the prevailing blood glucose. This paved the way for trials such as the DCCT which established the benefits of tight blood glucose control in reducing the risk of tissue complications.(2) The high frequency of severe hypoglycaemia made it essential that individuals pursuing this level of blood glucose control had access to technology through the NHS which could reliably identify low glucose values. Similar arguments have been advanced to support the provision of blood glucose testing for patients with insulin treated Type 2 diabetes. They also run significant risks of severe hypoglycaemia, particularly when pursuing tight glucose targets although episodes are less frequent than for those with Type 1 diabetes. Thus there is little stated disagreement that patients who are attempting to control their blood glucose tightly using insulin should be allowed/encouraged to undertake SMBG. It is important to recognise that patients treated with sulphonylureas are also at risk of severe hypoglycaemia. A recent study indicated that around 7% of individuals taking sulphonylureas with a mean HbA1c of 7.5% experienced at least one severe episode over 9-12 months a rate comparable to patients with Type 2 diabetes recently started on insulin therapy.(3) There is therefore a strong case that SMBG should also be provided to individuals taking sulphonylureas. Despite almost universal agreement that SMBG should be available to individuals with diabetes who are treated with insulin, the evidence to support these recommendations is surprisingly scanty. A recent systematic review found no overall improvement in blood glucose control in trials involving individuals with Type 1 diabetes or significant reductions in hypoglycaemia.(4) Perhaps this reflects the lack of interest in undertaking formal RCTs in an area where the advantages of such technology are thought to be obvious. Nevertheless the failure of RCTs to demonstrate any major benefits of these new technologies in terms of blood glucose control or 6

7 hypoglycaemia indicates the likely difficulty in demonstrating benefit among individuals who were not perceived to be at significant risk of hypoglycaemia or where the link between immediate knowledge of blood glucose is less obviously connected to changes in therapy. The case for blood glucose testing in those treated with lifestyle or oral agents which do not cause hypoglycaemia is largely focussed on the potential of self monitoring to alter behaviour favourably leading to improved metabolic control. It is possible that increased awareness of prevailing blood glucose levels together with the demonstration that certain activities (e.g. eating or exercise) can lower or raise them has clear potential educational value and might be an important motivating factor. This could encourage patients to sustain positive changes in diabetes self-management. However, these hypotheses while attractive, have not to date been supported by evidence from clinical trials. There are other situations where the use of blood glucose monitoring could be expected to be clinically useful (Table 1) although here too benefit is largely unproven. The uncertainty around the data presented above and the true place of SMBG in management is also reflected by the considerable geographical variation in the use of SMBG in Type 2 diabetes. If patient outcomes are similar between centres it suggests that SMBG is not an important determinant of blood glucose outcome. There is also a need to draw together the work which has been undertaken to measure the views of patients in terms of quality of life and other psychosocial measures. Faced with these difficulties, the working group reviewed national and international guidance including Diabetes UK (appendices 2 and 3), International Diabetes Federation (appendix 4), and NICE (appendix 5), and commissioned a systematic review to update and summarise the evidence surrounding SMBG in Type 2 diabetes. This review, undertaken by Norman Waugh and colleagues from the Aberdeen Health Technology Assessment Group University of Aberdeen is provided in full in Appendix 1 with a summary provided below. 7

8 Self monitoring of blood glucose in type 2 diabetes: systematic review. Background The prevalence of type 2 diabetes has been rising in the United Kingdom, and around 4% of the population now have the condition. Good control of blood glucose level is important in preventing or delaying the complications of type 2 diabetes, such as heart disease, peripheral vascular disease, visual loss and renal failure. However many people with type 2 diabetes do not have good control of their blood glucose. The usual method for monitoring glycaemic control is by measuring glycated haemoglobin, or HbA1c, which gives an average of the blood glucose over three months. If it is high, then control needs to be improved. NICE recommends that most people with type 2 diabetes should aim to keep HbA1c at 6.5% or under, though targets should be tailored to the individual. However, HbA1c does not tell patients what their blood glucose is doing on a day to day basis. Selfmonitoring by testing for urinary glucose is one way of checking when blood glucose is high, but is only a rough guide. A more accurate measure can be obtained by blood testing, which is done by pricking the skin to get a drop of blood, putting that blood on a testing strip, and reading the result with a small meter. This can be done at different times of day, before or after meals, or before or after physical activity. Meters are cheap (about 14), and the NHS requires manufacturers to provide them free of charge if needed, so the main cost is the test strips, at about 14 for a pack of 50. Main question Is self-monitoring of blood glucose worthwhile in people with type 2 diabetes not treated with insulin, or on only basal insulin in combination with oral agents, in terms of glycaemic control, hypoglycaemia, quality of life and other relevant outcomes, and cost per quality adjusted life year? Methods Review of systematic reviews published since 1996, and a systematic review and meta-analyses of randomised controlled trials identified from the reviews, and from searches for more recent trials. Review of qualitative and economic studies. Search strategy: Electronic databases: including The Cochrane Library [all sections] (Issue 2, 2009), MEDLINE (1996-April 2009), Embase (1996-April 2009), PsycINFO (1996 to April 2009), Web of Science limited to meeting abstracts (1996-April 2009) Web sites of EASD, ADA and Diabetes UK searched for meeting abstracts in April Websites of FDA, MHRA, SMBG International Working Group, Current Controlled Trials, ClinicalTrials.gov Contact with experts in the field Scrutiny of bibliographies of retrieved papers The searches were limited to English language and to articles published since 1996 (due to the number of recent good quality systematic reviews) and in order to reflect current meter technologies. The search strategy did not include limits for study design, as all types of studies were screened manually for potential inclusion. Systematic reviews We found 11 systematic reviews published in the last 10 years, most in the last few years. Most were of good quality. They contained from three to 13 RCTs out of a total of 20 RCTS. Their conclusions on glycaemic control varied, with some saying there was no benefit, others saying there was benefit, and some saying that there was no conclusive evidence of benefit. Much of the apparent disagreement may arise from the level of HbA1c which was considered to prove benefit, since the differences in meta-analysis were often of the order of 0.2%, which can be statistically 8

9 significant but not clinically important. There was some evidence that studies in which patients were given feedback in response to SMBG values and / or in which SMBG results were used to modify therapeutic regimens were more effective than those without feedback or use of SMBG for therapy modification. Effects also tended to be larger for patients with higher baseline HbA1c values. Randomised controlled trials (RCTs) We found 26 RCTs, ranging in size from under 30 to over 800 patients, and in duration from 12 weeks to 30 months. Only four trials scored highly on quality assessment. Components of the SMBG interventions were not well described in many cases. Half of the trials reported a reduction in HbA1c, and all that did find favourable results included an educational component and /or feedback. Ten trials compared simple SMBG with no SMBG, and found a reduction in HbA1c of 0.21%, which was statistically significant but of doubtful clinical significance. Four trials of enhanced SMBG (for example, with education, feedback, etc) showed a bigger reduction in HbA1c, of 0.52% compared to no SMBG. When SMBG enhanced with an educational or feedback component was compared to simple SMBG (5 trials), there was an HbA1c reduction of 0.2%, however, this did not quite reach statistical significance. Three RCTs showed no difference between SMBG and urine testing. Differences in the frequency of hypoglycaemic episodes were inconsistent. There was no difference in weight or BMI. There was no increase in medication changes with SMBG versus no SMBG, which may explain why HbA1c is not improved. Few studies examined quality of life, but the two best ones for this outcome (both from the UK, DIGEM and ESMON) reported a net adverse effect on anxiety and/or depression. Results from other studies were less clear-cut. Observational studies There were 36 relevant observational studies. These are more prone to bias, from confounding factors, and association does not necessarily mean cause. Eighteen showed no difference in HbA1c, 12 showed a reduction (but often very small) and some showed an increase in HbA1c on SMBG. This latter finding may be because SMBG was started because of poor glycaemic control. Qualitative studies The qualitative studies had some fairly consistent messages: There was a lack of education in how to interpret and use the data from SMBG In some patients, SMBG caused adverse psychological effects including depression and self-chastisement while others found it a useful tool for reassurance, assessing effects of behaviour and empowerment There was a lack of education in how to interpret and use the data from SMBG There was a lack of interest in the results from health care professionals Failure to act on the results was common The cost-effectiveness literature There was a mixture of studies, some just about costs, some looking at possible savings, others at cost-effectiveness. Some were funded by the manufacturers of testing strips and meters; these tended to be more favourable by making more generous assumptions on the effect on HbA1c. The cost of SMBG in people with type 2 diabetes in England is uncertain, but probably around 30 million a year of which at least half could be saved by adhering to previous guidelines and by applying the findings of DIGEM in the sulphonylurea only group. The reported costs per annum of SMBG vary amongst studies, from 10 to 259, the lowest being an estimate about 10 a year for infrequent testers on diet alone. Several studies asserted that SMBG can lead to savings which offset testing costs, and some estimated that SMBG could lead to savings from reduced costs in other health care. These studies tended to have more optimistic assumptions. However most of these studies failed to allow for the potentially negative impact of SMBG on aspects of quality of life. 9

10 The cost-effectiveness analyses vary in their assumptions, with those funded by industry producing lower ICERs. The best analysis to date is that from the DIGEM trial (funded by the UK HTA Programme) which after taking into account all costs, gains and disutilities, concluded that SMBG was not cost-effective. Conclusion The current evidence suggests that SMBG is of limited clinical effectiveness in improving glycaemic control in people with type 2 diabetes or oral agents, or diet alone, and is therefore unlikely to be cost-effective. There were insufficient data for those on a single basal insulin to reach any conclusion. No data are available on the possible benefits of SMBG in selected patient subgroups. SMBG can only be expected to lead to improved glycaemic control in the context of appropriate education both for patients and health care professionals on how to respond to the readings, in terms of lifestyle and treatment adjustment. It may be more effective if patients are able to selfadjust drug treatment. In the authors opinion, at a time when funds are scarce, the case for investment in blood glucose monitoring in type 2 diabetes, not treated with insulin, is not proven. Further research is required on the type of education and feedback that are most helpful, characteristics of patients benefitting most from SMBG, optimal timing and frequency of SMBG, and the circumstances under which SMBG causes anxiety and/or depression. Commentary on findings The findings of the short report are consistent with other recent reviews and indicate that SMBG alone in individuals with Type 2 diabetes treated by diet or oral agents overall leads to minimal improvements in blood glucose control of the order of 0.2% HbA1c. This degree of improvement is of doubtful clinical relevance. When taken together with the evidence from some recent trials that SMBG is associated with worsening of psychosocial measures such as depression it is unsurprising that some have concluded that SMBG is ineffective in this large group of patients. This has led to the argument that the NHS could save resources by simply refusing to reimburse the costs of meters and strips for those on oral agents. To access the full copy of the HTA report visit Limitations of systematic reviews in hiding specific benefits of SMBG By design, systematic reviews combine data from all participants in randomised controlled trials and estimate the average effect of an intervention. Thus, although the mean improvement in blood glucose control may be modest, the overall figure includes the results from individuals whose blood glucose control improves markedly. Such individuals may be using SMBG to alter and maintain changes in lifestyle which results in improved blood glucose control (effects which are hidden among the overall result). The qualitative studies have confirmed the ways SMBG can benefit some individuals with diabetes. These include: Establishing the relationship between their physical symptoms and their blood sugar. Providing reassurance Assessing the effects of specific behaviours Promotion of adherence to self management Empowering them to take more control over their health care and ability to contribute to physician s evaluation of their status However, only a few individuals with Type 2 diabetes currently use SMBG in these ways to manage their diabetes successfully. Clinical experience suggests that a decision to withhold blood glucose monitoring technology from patients to whom monitoring has become an essential component of their self-management strategy might be damaging. Such a policy could risk both alienating them from their professional carers as well as leading to a deterioration in their metabolic control and a consequent increased risk of complications. 10

11 What about the majority who apparently get no benefit? Conversely, systematic reviews also show that there are many individuals who derive no benefit from SMBG in terms of blood glucose control (indeed, this may deteriorate) and in whom monitoring presumably fails to help them to manage their diabetes successfully. There are many potential reasons why SMBG may not contribute to successful diabetes selfmanagement, some of which may be amenable to additional professional input and support. Some of the qualitative studies confirm clinical experience that many patients merely record results with no real idea of both what the measurements signify or what they should do with them. There is frequently a gap in understanding between patient and professional. The former often believe that they are measuring blood glucose to allow the professional both to see how hard they are working and then give advice on alterations in therapy. Professionals on the other hand frequently take the view that the purpose of monitoring is largely to aid self-management and that they do not either need or wish to see these records. If professionals show little or no interest in the results of the patients hard work this may cause resentment and damage the therapeutic relationship. Some of these issues could be resolved by better mutual understanding and more constructive and informed consultations. NICE CG87 guidance (5) recommends in people using SMBG that there is a need to assess at least annually and in a structured way: self-monitoring skills the quality and appropriate frequency of testing the use made of the results obtained the impact on quality of life the continued benefit the equipment used Nevertheless, we need to accept that many individuals, who despite additional input find SMBG unhelpful, should not continue to monitor. This will both relieve such patients of a considerable burden as well as making significant savings on a fruitless exercise. Thus, just as it seems important to guarantee that those who derive benefit from such technology are not prevented from using it, it is also necessary not to continue monitoring or waste resources on those who derive no benefit and whom are harmed psychologically. Such individuals may gain value from urine testing or can be monitored by regular measurement of HbA1c alone. The place of SMBG within an overall educational approach and the problem of providing SMBG in isolation The systematic reviews support the belief that self monitoring is more effective when provided as part of a self-management education package, reporting overall improvements in HbA1c of around 0.5%, a clinically relevant improvement. This reinforces the importance of providing structured education packages for newly diagnosed individuals which include monitoring as one of the components. This might involve SMBG, although testing of urinary glucose, clearly cheaper, might be both acceptable and as effective, a hypothesis currently being tested in an ongoing trial. One firm recommendation is that SMBG should only be taught as part of an overall education package and not on its own. However, the most effective way of integrating SMBG into educational programmes is unclear and needs further research. The importance of not providing SMBG in isolation is also relevant to an important current issue in which blood glucose meters are frequently sold to the public by private pharmacists. It seems likely that since monitoring which results from such activities is largely provided without information on how to integrate the information into ongoing care, monitoring is often wasted and ineffective. The practice also creates further tension when primary care teams are requested to provide additional glucose strips. One constructive action might be to provide written information agreed by healthcare professionals, industry and pharmacists which 11

12 could be provided along with blood glucose equipment, providing simple educational principles of how to use the technology to aid self management. In the absence of any clear evidence as to the best way to integrate information into educational packages this might be based on a pragmatic approach which could include information on normal blood glucose levels, the effect of eating and other activities on blood glucose, values which indicate successful alterations in lifestyle, those alerting individuals of the need to contact their professional carers and practical suggestions on the frequency of monitoring. Integrating SMBG into diabetes care and educating professionals Monitoring strategies need to be included in care planning, a process which is currently being developed and piloted in diabetes management. Thus, professional and patient need to agree together the purpose of the exercise, particularly as part of goal setting, and its value should be regularly reviewed. In those who derive no value then SMBG could be stopped with metabolic control monitored in other ways, perhaps urine testing for whom this is acceptable together with regular measurement of HbA1c. It is important that regular HbA1c testing be universally available for people with Type 2 diabetes although the best way of using such information to enable more effective selfmanagement also needs to be established. Not only is SMBG used haphazardly but different professionals who contribute to the care of individuals with diabetes have different views on both the purpose and role of monitoring. Patients must receive a consistent message from all the professionals they come into contact with. It is important that local guidelines which usually include policies on the use of medication and care pathways develop an agreed approach to monitoring in the management of patients on diet or oral agents. Clinical experience and information from qualitative research suggest that many of the staff who have most contact with those using oral agents, particularly practice nurses and GPs, need a better understanding of the role of blood glucose monitoring for those on oral agents. Front-line staff need to know how self monitoring fits within the overall care package and have the knowledge and competency to discuss this with patients in the framework of an overall approach to the treatment of Type 2 diabetes in its early stages. This should include an understanding of the need to individualise therapy, supporting and encouraging those who find monitoring important to successful self-management and stopping it in those who derive no benefit, in terms of either blood glucose control or deteriorating psychosocial outcomes. These aspects of diabetes care should be incorporated into local training events to ensure that staff are competent and individuals with diabetes receive a consistent message. Research gaps There is no longer a place for trials which measure the effect of blood glucose monitoring in isolation and compared to usual practice. There are sufficient studies showing slight or no differences when SMBG is used on its own and a more relevant question is why some studies which integrate monitoring into an education package show improved blood glucose control while others do not. We remain unsure as to the best way in which SMBG should be used in this group of individuals. Some have suggested that the results of SMBG could be used to self-adjust oral medication but there is potential for harm as well as benefit. The development of the most appropriate approaches requires considerable further work in a complex field of research. There is an urgent need for more collaboration with social scientists to identify characteristics of those who do well (or badly) and in particular explore how those who find most value in self monitoring use knowledge of their blood glucose concentration to alter lifestyle or improve other areas of self management. This would help to bridge the gap of fundamentally different expectations between professionals and patients on the role and ownership of blood glucose 12

13 measurements. Information from such research is essential to designing more effective ways of integrating monitoring into structured educational packages. There is little evidence to guide either patient or professional as to the optimum frequency of monitoring and we currently have little information as to how frequently those who use SMBG successfully in this group undertake monitoring. It is also unclear whether blood glucose should be monitored regularly or in bursts, perhaps at critical periods, i.e. if blood glucose control begins to deteriorate, in an attempt to improve motivation or identify those who need additional therapy. Economics It is clear that considerable savings could be made by stopping monitoring in people who do not derive benefit. In a time of financial stringency, the cost pressures are becoming irresistible. However, those who are motivated by SMBG activity and use the information to maximise the effect of lifestyle and medication must be allowed to continue to monitor. There is no place for a one size fits all approach; such a policy would deprive some individuals of what they regard as a cornerstone of treatment. An important incentive which should be considered by local commissioners would be the ability to hypothecate the resource currently allocated to blood glucose monitoring and permit any savings to be allocated to other aspects of community diabetes care. This might be best diverted to provide structured diabetes education for both patient and professional. Paradoxically, such an approach might in time, lead to an increase in the amount of monitoring but hopefully undertaken more efficiently and with more tangible benefit. Conclusion The data from clinical trials show increasingly clearly that SMBG in patients treated by lifestyle modification or oral agents leads to overall changes in blood glucose control which are minor and of little clinical relevance unless associated with education and clear objectives. There is also evidence that some psychosocial outcomes deteriorate. There is a strong case for stopping blood glucose monitoring activity in perhaps the majority of patients in this group who derive no benefit from the activity and particularly where it is damaging their quality of life. However, we believe that some newly diagnosed individuals clearly benefit from measuring their blood glucose. In those individuals who find modern technology essential to managing their diabetes, arbitrary withdrawal of treatment would limit patient choice and may lead to a deterioration in blood glucose control. Training programmes for professionals need to be modified to address this important area and incorporate it within the local diabetes care package. In the absence of clear evidence for general benefit, training should include the need to support patients who find SMBG important in managing their diabetes successfully and stopping monitoring in those who derive no benefit. Future research should include a detailed exploration of the factors which determine success in incorporating SMBG into diabetes selfmanagement. This will inform the development of more effective interventions, evaluated by the most appropriate study designs to inform improved clinical practice and better targeting of resources. Finally although savings can be made by stopping blood glucose monitoring among those who derive no benefit, it is important that the resources generated by such activity are used locally to support other aspects of diabetes care, particularly education of both staff and patients. Recommendations for SMBG in people with non-insulin-treated diabetes 1. SMBG with appropriate structured education should be available to people receiving sulphonylurea treatment to identify hypoglycaemic episodes. 2. In keeping with the recommendations contained within NICE Clinical Guideline CG87, SMBG should only be provided routinely to people with Type 2 diabetes not treated with insulin or sulphonylureas where there is an agreed purpose or goal to testing. 13

14 3. SMBG should be used only within a care package, accompanied by structured education which should include clear instructions as to the place of monitoring and how results can be used to reinforce lifestyle change, adjust therapy or alert healthcare professionals. This should include regular review to identify and support those who find it useful while discouraging those who gain no clinical benefit from continuing to test. 4. Individuals with non-insulin treated diabetes who are motivated by SMBG activity and use the information to maximise the effect of lifestyle and medication should be encouraged to continue to monitor. 5. Staff training in the use of SMBG to support changes in lifestyle and self-adjustment of medications is required 6. Savings from a reduction in SMBG in individuals with non-insulin treated diabetes should be used to provide both structured education and training of professionals. 7. Future research should focus on how to identify those who will gain most from SMBG and establish how they integrate it successfully into their approach to self-management. Acknowledgements We acknowledge the assistance of Christine Clar, Kath Barnard, Pamela Royle of the Aberdeen Health Technology Assessment Group and thank the Health Technology Assessment Programme for commissioning the systematic review References 1. Sonksen PH, Judd SL, Lowy C. Home monitoring of blood-glucose. Method for improving diabetic control. Lancet. 1978;1: The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329: UK Hypoglycaemia Study Group. Risk of hypoglycaemia in types 1 and 2 diabetes: effects of treatment modalities and their duration. Diabetologia. 2007;50: Coster S, Gulliford MC, Seed PT, Powrie JK, Swaminathan R. Monitoring blood glucose control in diabetes mellitus: a systematic review. Health Technol Assess. 2000;4:i-iv, NICE. NICE Clinical Guideline 87. Type 2 Diabetes: the management of type 2 diabetes. London: National Institute of Clinical Excellence; "Clar C, Barnard K, Cummins E, Royle P, Waugh N. NIHR Health Technology Assessment programme. Self monitoring of blood glucose in type 2 diabetes: systematic review. HTA ref. 09/19/01. In press (March 2010). Project abstract available from See full reference list in STA report Appendix Table 1 Situations where SMBG can provide clinically useful information Intensifying or changing treatment regimes to improve blood glucose control Driving if treated with sulphonylureas Intercurrent illness to guide the appropriate action Regular and/or intensive physical activity such as swimming, scuba diving, gardening if on oral agents with the potential to cause hypoglycaemia Optimising blood glucose control prior to conception and during pregnancy, including gestational diabetes Newly diagnosed to give information and understanding about how food/activity interact and effect on diabetes management; If regular HbA1c monitoring is not possible or convenient; Reassurance at specific times, To assist improved blood glucose control during in-patient stay 14

15 Appendix 1: Self monitoring of blood glucose in type 2 diabetes: systematic review. A short report commissioned by the NIHR HTA Programme on behalf of a Department of Health Working Group. Produced by: Aberdeen Health Technology Assessment Group Authors: Dr Christine Clar, Researcher in Systematic Reviews, Berlin Dr Katharine Barnard, Health Psychologist, University of Southampton Dr Ewen Cummins, Health Economist, McMaster Development Consultants, Glasgow Dr Pamela Royle, Research Fellow, University of Aberdeen Professor Norman Waugh, Professor of Public Health, University of Aberdeen Address for correspondence: Professor N.R.Waugh, Department of Public Health, Medical School Buildings Foresterhill, Aberdeen AB25 2ZD. Tel: n.r.waugh@abdn.ac.uk Date completed: 21sr December, 2009 Source of funding: This report was commissioned by the NIHR HTA Programme as project number HTA 09/19/01 Disclaimer: The views expressed in this report are those of the authors and not necessarily those of the NIHR HTA Programme. Any errors are the responsibility of the authors Declared competing interests of the authors: None Acknowledgements: We thank the following for commenting on a near final draft of this report, but absolve them from any errors in the final report, responsibility for which rests with the review team; Professor Simon Heller, Sheffield Dr Mark Houliston, general practitioner and lead clinician, Grampian Diabetes Managed Clinical Network. Dr Maurice O Kane, Londonderry We also thank the members of the Department of Health Working Group on self monitoring of blood glucose, whose discussions and comments on both this research review and the drafts of the Working Group Report were very useful. The members of the working group included Simon Heller (Chair) Eleanor Kennedy Andrew Farmer Adrian Griffin Martin Gulliford Julia Lawton Maurice O Kane Mark Samuels Grace Sweeney Bridget Turner 15

16 Contributions of authors: Christine Clar led the review of clinical effectiveness, commented on other sections, and edited the report after peer review and editorial comments Katharine Barnard led the review of qualitative studies and commented on other sections. Ewen Cummins led the review of economic studies. Pamela Royle did searches, quality assurance and editing, and commented on all sections. Norman Waugh wrote chapters 1 and 4, and did final editing. About the Aberdeen HTA group. The Aberdeen Health Technology Assessment Group is part of the Institute of Applied Health Sciences (IAHS), which is part of the College of Medicine and Life Sciences of the University of Aberdeen. The Institute of Applied Health Sciences is made up of discrete but methodologically related research groups. The HTA Group is drawn mainly from the Health Services Research Unit, Public Health, and the Health Economics Research Unit. The HTA Group produces independent health technology assessments reports (TARs) for the UK HTA Programme, which commissions TARs for NICE and other bodies, such as the National Screening Committee. It also carries out evidence reviews to support the NICE Single Technology Appraisal Programme. Particular interests include evaluation of nonpharmacological technologies, screening, and diabetes. Previous TARs from Aberdeen include; Inhaled insulin for diabetes (for NICE) Continuous subcutaneous insulin infusion (NICE) Screening for type 2 diabetes (NSC and DH) Non-pharmacological prevention of diabetes in those with impaired glucose tolerance (DH) Newer drugs for type 2 diabetes (NICE) We also do Cochrane reviews on diabetic topics. 16

17

18

19

20 Further copies of this publication can be ordered from Prontaprint, by ing or tel: , quoting DIABETES 129