The potential for automation to reduce time required to undertake the key stages of systematic reviews has seen a growth in software applications in recent years. A recent systematic review found no unified information extraction framework for automating data extraction in systematic reviews. Whilst a proliferation of risk of bias and “process management” tools exist, data extraction from text and tables in systematic review has received less attention. We reviewed available tools and evaluative studies to determine progress in automation in this area. Systematic searches of Medline, Embase, and Scopus were undertaken until May 2018. A bespoke search strategy was developed; search terms included automation, machine learning, and data mining. Studies developing automated tools or methods for data extraction from published primary studies were included. We excluded tools used for “process management”, risk of bias assessment, or solely for use on abstracts. Searches were cross-checked with the SRToolbox (York Health Economics Consortium). After deduplication, 3563 records were screened by two independent reviewers and 23 were marked for full paper screening. Six studies were included. Studies presented methods or proof-of-concept on the extraction of information from text in full papers. Bespoke programs were used to detect and classify elements within a paper, to identify key information, annotate, and extract the information. Similar techniques were applied to detection, classification, decomposition, and extraction of data from tables, a key source of trial data. One “user ready” (others were at the theoretical stage) tool was identified. ExaCT identifies key phrases containing data such as interventions and reported outcomes for the reviewer to check. However, extraction was limited, and development appears to have ceased. Our research demonstrates promising automation technologies which have the potential to assist rather than displace systematic reviewers, but challenges remain and established tools have yet to emerge.

Background: The recommended place for treatment of diabetic ketoacidosis in children is a paediatric High Dependency Unit. This facility is not available in all areas where children with type 1 diabetes mellitus are cared for.Aims: This study investigates the safety and efficacy of diabetic ketoacidosis management in a community general hospital without a paediatric high dependency unit.Methods: Data from children with diabetic ketoacidosis were collected from all diabetes related admissions in Dr Gray's Hospital, Elgin from 2001 to 2010. Observations were compared with safety indicators (pH, bicarbonate, glucose, electrolytes and cerebral oedema) and were reviewed for the recovery to normal values (pH, bicarbonate), without abnormal fluctuation (electrolytes, glucose) and without neurological complications (cerebral oedema).Results: The 114 patients generated 251 diabetes-related admissions, 118 for diabetic ketoacidosis treatment of whom 99 patients were treated with intravenous fluids and insulin. The mean time to recover to a pH of at least 7.30 was 655 minutes (120-1410 min). There were 79 (4.37% of 1808) glucose readings dropping more than 5.0 mmol/l per hour. There were six hypoglycaemic events (2.3-2.9 mmol/l) and in one case potassium dropped to 2.2 mmol/l. There was no case which developed into cerebral oedema.Conclusion: Treatment of diabetic ketoacidosis in a community general hospital managed with a protocol for fluids, insulin and strict monitoring has shown to be effective in achieving recovery and to safely avoid complications.

BACKGROUND Irritable bowel syndrome (IBS) is common, and causes pain, bloating and diarrhoea and/or constipation. It is a troublesome condition that reduces the quality of life but causes no permanent damage. Inflammatory bowel disease (IBD) comprises mainly ulcerative colitis (UC) and Crohn's disease (CD). Both cause serious complications and may lead to sections of the bowel having to be removed, although this is more common with CD. The presenting symptoms of IBS and IBD can be similar. Distinguishing them on clinical signs and symptoms can be difficult. Until recently, colonoscopy was often required to rule out IBD. In younger people, > 60% of colonoscopies showed no abnormality. Faecal calprotectin (FC) is a protein released by the white blood cells, neutrophils, found in inflamed areas of the bowel in IBD. Determining the level of FC in stool samples may help distinguish IBS from IBD. OBJECTIVE To review the value of FC for distinguishing between IBD and non-IBD. DATA SOURCES Sources included MEDLINE, EMBASE, The Cochrane Library, Web of Science, websites of journals and the European Crohn's and Colitis Organisation (conference abstracts 2012 and 2013), and contact with experts. REVIEW METHODS Systematic review and economic modelling. Review Manager (RevMan) version 5.2 (The Cochrane Collaboration, The Nordic Cochrane Centre, Copenhagen, Denmark) was used for most analysis, with statistical analyses done in Stata version 12 (StataCorp LP, College Station, TX, USA). Forest plots and receiver operating characteristic curves were produced. Quality Assessment of Diagnostic Accuracy Studies was used for quality assessment. Economic modelling was done in Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA, USA). LIMITATIONS Studies were often small, most used only one calprotectin cut-off level, and nearly all came from secondary care populations. RESULTS Twenty-eight studies provided data for 2 × 2 tables and were included in meta-analyses, with seven in the most important comparison in adults (IBS vs. IBD) and eight in the key comparison in paediatrics (IBD vs. non-IBD). Most studies used laboratory enzyme-linked immunosorbent assay (ELISA) tests. For distinguishing between IBD and IBS in adults, these gave pooled sensitivity of 93% and specificity of 94% at FC cut-off level of 50 µg/g. Sensitivities at that cut-off ranged from 83% to 100%, and specificities from 60% to 100%. For distinguishing between IBD and non-IBD in paediatric populations with ELISA tests, sensitivities ranged from 95% to 100% at cut-off of 50 µg/g and specificities of 44-93%. Few studies used point-of-care testing but that seemed as reliable as ELISA, though perhaps less specific. The evidence did not provide any grounds for preferring one test over others on clinical effectiveness grounds. FC testing in primary care could reduce the need for referral and colonoscopies. Any quality-adjusted life-year gains are likely to be small because of the low prevalence of IBD and the high sensitivities of all of the tests, resulting in few false negatives with IBD. However, considerable savings could accrue. Areas of uncertainty include the optimum management of people with borderline results (50-150 µg/g), most of whom do not have IBD. Repeat testing may be appropriate before referral. CONCLUSIONS Faecal calprotectin can be a highly sensitive way of detecting IBD, although there are inevitably trade-offs between sensitivity and specificity, with some false positives (IBS with positive calprotectin) if a low calprotectin cut-off is used. In most cases, a negative calprotectin rules out IBD, thereby sparing most people with IBS from having to have invasive investigations, such as colonoscopy. STUDY REGISTRATION PROSPERO CRD 42012003287. FUNDING The National Institute for Health Research Health Technology Assessment programme.

Ford J, Cummins E, Sharma P, et al. Systematic review of the clinical effectiveness and cost-effectiveness, and economic evaluation, of denosumab for the treatment of bone metastases from solid tumours. Southampton (UK): NIHR Journals Library; 2013 Jul.(Health Technology Assessment, No. 17.29.)

Results: non-small cell lung cancer-Systematic review of the clinical effectiveness and cost-effectiveness, and economic evaluation, of denosumab for the treatment of bone metastases from solid tumours-NCBI Bookshelf

BACKGROUNDDenosumab offers an alternative, or additional, treatment for the prevention of skeletal-related events (SREs) in patients with bone metastases from solid tumours.OBJECTIVESThe aim of this review was to assess the clinical effectiveness and cost-effectiveness of denosumab, within its licensed indication, for the prevention of SREs in patients with bone metastases from solid tumours.DATA SOURCESDatabases searched were MEDLINE (1948 to April 2011), EMBASE (1980 to March 2011), The Cochrane Library (all sections; Issue 1, 2011) and Web of Science with Conference Proceedings (1970 to May 2011).REVIEW METHODSOnly randomised controlled trials (RCTs) assessing denosumab, bisphosphonates (BPs) or best supportive care (BSC) in patients with bone metastases were included. Systematic reviews and observational studies were used for safety and quality-of-life assessments. Study quality was assessed using the Cochrane risk of bias tool. Studies suitable for meta-analysis were synthesised using network meta-analysis (NMA). A systematic review was conducted for cost, quality-of-life and cost-effectiveness studies. The results of this informed the cost-utility modelling. This principally estimated the cost-effectiveness of denosumab relative to zoledronic acid for when BPs are currently recommended and relative to BSC when BPs are not recommended or are contraindicated.RESULTSA literature search identified 39 studies (eight suitable for NMA). Denosumab was effective in delaying time to first SRE and reducing the risk of multiple SREs compared with zoledronic acid. Generally speaking, denosumab was similar to zoledronic acid for quality of life, pain, overall survival and safety. The NMA demonstrated that denosumab was more effective in delaying SREs than placebo, but was limited by numerous uncertainties. Cost-utility modelling results for denosumab relative to zoledronic acid were driven by the availability of the patient access scheme (PAS) for denosumab. Without this, denosumab was not estimated to be cost-effective compared with zoledronic acid. With it, the cost-effectiveness ranged between dominance for breast and prostate cancer, to between £5400 and £15,300 per quality-adjusted life-year (QALY) for other solid tumours (OSTs) including non-small cell lung cancer (NSCLC) and £12,700 per QALY for NSCLC. Owing to small patient gains estimated, the cost-effectiveness of denosumab was very sensitive to the zoledronic acid price. Denosumab was not estimated to be cost-effective compared with BSC.LIMITATIONSOnly subgroup data were available for denosumab for NSCLC, and OSTs excluding NSCLC. The NMA was subject to numerous uncertainties. Owing to small patient gains estimated, the cost-effectiveness of denosumab was very sensitive to the zoledronic acid price.CONCLUSIONDenosumab, compared with zoledronic acid and placebo, is effective in delaying SREs, but is similar with regard to quality of life and pain. Cost-effectiveness showed that without the PAS denosumab was not estimated to be cost-effective relative to either zoledronic acid or BSC. With the PAS, denosumab was estimated to be cost-effective relative to zoledronic acid but not BSC.STUDY REGISTRATIONPROSPERO number CRD42011001418.FUNDINGThe National Institute for Health Research Health Technology Assessment programme.

BACKGROUND:Cystic fibrosis (CF) is an inherited disease that leads to damage to lungs, pancreas and other organs. Most people with CF die prematurely from lung disease, but survival has improved markedly over the decades and it is estimated that children born with CF now will live to an average age of 50 years. CF-related diabetes (CFRD) is due to damage to the pancreas, which, over time, loses its capacity to produce sufficient insulin. CFRD is becoming more common owing to the improved survival of people with CF. OBJECTIVES:The initial aim was to review the methods for screening for CFRD, which can be symptomless but still be causing harm. As the aim of screening and early detection is to allow earlier treatment, a second aim was to assess the effectiveness of treatments. However, during the review it became clear that there were problems with how CFRD is defined, uncertainty about when hyperglycaemia should be treated. DATA SOURCES:Details of relevant studies were obtained from the usual bibliometric databases - MEDLINE (1950-2008), EMBASE (1980-2008), The Cochrane Library (all sections), Web of Science (1970-2008). Websites of relevant bodies were searched for guidelines and reports. Conference abstracts were searched. Expert co-authors identified key papers. REVIEW METHODS:Systematic reviews of treatments and screening tests. Screening studies were data extracted if they provided sufficient data to construct 2 × 2 tables. Other screening studies were described in narrative manner. The background to CF and CFRD were described in a narrative manner, as was Chapter 2 on problems with defining CFRD. A model was constructed for cost-effectiveness analysis, but was not used because of lack of data. RESULTS:Diabetes is usually defined based on the level of blood glucose (BG) at which the risk of retinopathy occurs. For CFRD, it would be better to define it on the level at which the risk of lung disease (pulmonopathy) rises. There seems little place for treatments other than insulin, but the best insulin regimen remains to be confirmed. The best screening test may be by continuous glucose monitoring systems but further evidence is required. Screening may need to detect BG levels of > 8 mmol/l because that may be the level above which pulmonopathy starts in people with CF. LIMITATIONS:The evidence base for treatment is disappointing with few large randomised controlled trials. The key question is when treatment should start, perhaps at the post-prandial hyperglycaemia stage. Research is needed. Until that is done, we cannot be sure what we are screening for, and, therefore, which screening strategy should be used. CONCLUSIONS:The definition of CFRD should probably be based on pulmonopathy risk, rather than using the classical definition of diabetes. That implies that we should be screening for a wider range of hyperglycaemia than in other forms of diabetes, perhaps to detect BG excursions of > 8 mmol/l. Insulin treatment may need to start at lower levels than formerly accepted. FUNDING:The National Institute for Health Research Health Technology Assessment programme.

BACKGROUND The prevalence of type 2 diabetes mellitus (T2DM) is increasing in the UK and worldwide. Before the onset of T2DM, there are two conditions characterised by blood glucose levels that are above normal but below the threshold for diabetes. If screening for T2DM in introduced, many people with impaired glucose tolerance (IGT) will be found and it is necessary to consider how they should be treated. The number would depend on what screening test was used and what cut-offs were chosen. OBJECTIVE To review the clinical effectiveness and cost-effectiveness of non-pharmacological interventions, including diet and physical activity, for the prevention of T2DM in people with intermediate hyperglycaemia. DATA SOURCES Electronic databases, MEDLINE (1996-2011), EMBASE (1980-2011) and all sections of The Cochrane Library, were searched for systematic reviews, randomised controlled trials (RCTs) and other relevant literature on the effectiveness of diet and/or physical activity in preventing, or delaying, progression to T2DM.The databases were also searched for studies on the cost-effectiveness of interventions. REVIEW METHODS The review of clinical effectiveness was based mainly on RCTs, which were critically appraised. Subjects were people with intermediate hyperglycaemia, mainly with IGT. Interventions could be diet alone, physical activity alone, or the combination. For cost-effectiveness analysis, we updated the Sheffield economic model of T2DM. Modelling based on RCTs may not reflect what happens in routine care so we created a 'real-life' modelling scenario wherein people would try lifestyle change but switch to metformin after 1 year if they failed. RESULTS Nine RCTs compared lifestyle interventions (predominantly dietary and physical activity advice, with regular reinforcement and frequent follow-up) with standard care. The primary outcome was progression to diabetes. In most trials, progression was reduced, by over half in some trials. The best effects were seen in participants who adhered best to the lifestyle changes; a scenario of a trial of lifestyle change but a switch to metformin after 1 year in those who did not adhere sufficiently appeared to be the most cost-effective option. LIMITATIONS Participants in the RCTs were volunteers and their results may have been better than in general populations. Even among the volunteers, many did not adhere. Some studies were not long enough to show whether the interventions reduced cardiovascular mortality as well as diabetes. The main problem is that we know what people should do to reduce progression, but not how to persuade most to do it. CONCLUSION In people with IGT, dietary change to ensure weight loss, coupled with physical activity, is clinically effective and cost-effective in reducing progression to diabetes. FUNDING The National Institute for Health Research Health Technology Assessment programme.

This paper presents a summary of the evidence review group (ERG) report into the clinical effectiveness and cost-effectiveness of liraglutide in the treatment of type 2 diabetes mellitus, based upon the manufacturer's submission to the National Institute for Health and Clinical Excellence (NICE) as part of the single technology appraisal process. The manufacturer proposed the use of liraglutide as a second or third drug in patients with type 2 diabetes whose glycaemic control was unsatisfactory with metformin, with or without a second oral glucose-lowering drug. The submission included six manufacturer-sponsored trials that compared the efficacy of liraglutide against other glucose-lowering agents. Not all of the trials were relevant to the decision problem. The most relevant were Liraglutide Effects and Actions in Diabetes 5 (LEAD-5) (liraglutide used as part of triple therapy and compared against insulin glargine) and LEAD-6 [liraglutide in triple therapy compared against another glucagon-like peptide-1 agonist, exenatide]. Five of the six trials were published in full and one was then unpublished. Two doses of liraglutide, 1.2 and 1.8 mg, were used in some trials, but in the two comparisons in triple therapy, against glargine and exenatide, only the 1.8-mg dose was used. Liraglutide in both doses was found to be clinically effective in lowering blood glucose concentration [glycated haemoglobin (HbA1c)], reducing weight (unlike other glucose-lowering agents, such as sulphonylureas, glitazones and insulins, which cause weight gain) and also reducing systolic blood pressure (SBP). Hypoglycaemia was uncommon. The ERG carried out meta-analyses comparing the 1.2- and 1.8-mg doses of liraglutide, which suggested that there was no difference in control of diabetes, and only a slight difference in weight loss, insufficient to justify the extra cost. The cost-effectiveness analysis was carried out using the Center for Outcomes Research model. The health benefit was reported as quality-adjusted life-years (QALYs). The manufacturer estimated the cost-effectiveness to be £ 15,130 per QALY for liraglutide 1.8 mg compared with glargine, £ 10,054 per QALY for liraglutide 1.8 mg compared with exenatide, £ 10,465 per QALY for liraglutide 1.8 mg compared with sitagliptin, and £ 9851 per QALY for liraglutide 1.2 mg compared with sitagliptin. The ERG conducted additional sensitivity analyses and concluded that the factors that carried most weight were: in the comparison with glargine, the direct utility effects of body mass index (BMI) changes and SBP, with some additional contribution from HbA1c, in the comparison with exenatide, HbA1c, with some additional effects from cholesterol and triglycerides in the comparison with sitagliptin, HbA1c and direct utility effects of BMI changes. The European Medicines Agency has approved liraglutide in dual therapy with other oral glucose-lowering agents. NICE guidance recommends the use of liraglutide 1.2 mg in triple therapy when glycaemic control remains or becomes inadequate with a combination of two oral glucose-lowering drugs. The use of liraglutide 1.2 mg in a dual therapy is indicated only in patients who are intolerant of, or have contraindications to, three oral glucose-lowering drugs. The use of liraglutide 1.8 mg was not approved by NICE. The ERG recommends research into the (currently unlicensed) use of liraglutide in combination with long-acting insulin.

BACKGROUNDScreening for gestational diabetes has long been a controversial topic. A previous Health Technology Assessment (HTA) report reviewed literature on screening for gestational diabetes mellitus (GDM) and assessed the case for screening against the criteria set by the National Screening Committee.OBJECTIVETo update a previous HTA report which reviewed the literature on screening for GDM by examining evidence that has emerged since that last report, including the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS), the Maternal and Fetal Medicine Units Network (MFMUN) trial and the Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) study. To review data on recent trends in maternal age at birth and on the prevalence of overweight and obesity and the effect on prevalence of GDM.DATA SOURCESA systematic review and meta-analysis of the literature was carried out. The bibliographic databases used were MEDLINE (1996 to January 2009), EMBASE (1996 to December 2009), the Cochrane Library 2008 issue 4, the Centre for Reviews and Dissemination database and the Web of Science.REVIEW METHODSFor the review of treatment with oral drugs versus insulin, a full systematic review and meta-analysis was carried out. The results of the ACHOIS, MFMUN and HAPO studies were summarised and their implications discussed. Findings of a selection of other recent studies, relevant to the continuum issue, were summarised. Some recent screening studies were reviewed, including a particular focus on studies of screening earlier in pregnancy.RESULTSThe HAPO results showed a linear relationship between plasma glucose and adverse outcomes - there is a continuum of risk with no clear threshold which could divide women into those with gestational diabetes and those without. There was good evidence from trials and the meta-analysis that women who fail to control hyperglycaemic in pregnancy on lifestyle measures alone can be safely and effectively be treated with oral agents, metformin or glibenclamide, rather than going directly to insulin. Evidence showed few differences in results between glibenclamide and insulin and metformin and insulin. The exceptions were that there was less maternal hypoglycaemia with glibenclamide, but less neonatal hypoglycaemia and lower birthweight with insulin, and there was less maternal weight gain with metformin. The ACHOIS and MFMUN trials showed reductions in perinatal complications among infants born to mothers who were provided with more intensive dietary advice, blood glucose monitoring and insulin when required. The HAPO study demonstrated adverse outcomes over a much wider range of blood glucose (BG) than the traditional definition of GDM. In the HAPO study, no one measure of BG came out as being clearly the best, although fasting plasma glucose (FPG) was as good as any, and had advantages of being more convenient than an oral glucose tolerance test (OGTT), but correlations between fasting and post-load levels were quite poor. Two screening strategies dominated; (1) selection by the American Diabetes Association criteria followed by the 75-g OGTT [incremental cost-effectiveness ratio (ICER) 3678 pounds], and (2) selection by high-risk ethnicity followed by the 75-g OGTT (ICER 21,739 pounds). Studies indicated that costs are about 1833 pounds higher for pregnancies complicated by gestational diabetes, suggesting that prevention would be worthwhile.LIMITATIONSNot all of the HAPO results have been published, and none of the reviewed economic studies resolved the most difficult issue - at what level of BG does intervention become cost-effective?CONCLUSIONSThe evidence base has improved since the last HTA review in 2002. There is now good evidence for treatment of oral drugs instead of insulin and it looks increasingly as if FPG could be the test of choice. However some key uncertainties remain to be resolved, which can be done by further analysis of the already collected HAPO data and by using the UK model used in developing the NICE guidelines to assess the cost-effectiveness of intervention in each of the seven HAPO categories.

OBJECTIVES:To examine whether or not self-monitoring of blood glucose (SMBG) is worthwhile, in terms of glycaemic control, hypoglycaemia, quality of life (QoL) and cost per quality-adjusted life-year (QALY), in people with type 2 diabetes (T2DM) who were not treated with insulin or who were on basal insulin in combination with oral agents. DATA SOURCES:Literature searched included systematic reviews published since 1996, and a systematic review and meta-analyses of randomised controlled trials (RCTs) identified from the reviews, and from searches for more recent trials, along with review of qualitative and economic studies. Search strategies were limited to the English language and to articles published since 1996, and included: databases searched from 1996 to April 2009 - The Cochrane Library, MEDLINE, EMBASE, PsycINFO, Web of Science - limited to meeting abstracts; and websites. REVIEW METHODS:The intervention was self-testing of blood glucose with a meter and test strips. Studies included adult patients with T2DM on any oral treatment or combination of regimens, including lifestyle, oral agents or once-daily basal insulin. Existing systematic reviews of SMBG were summarised and results compared. Evidence synthesis of all of the studies meeting the inclusion criteria was carried out using a narrative review. Data were analysed by outcome and subgroups. HbA1c data from RCTs were summarised using a meta-analysis. Heterogeneity was calculated using the chi-squared and I2 methods. The following analyses were carried out: SMBG compared to self-monitoring of urine glucose, SMBG versus no SMBG, more intensive SMBG versus less intensive SMBG, and more intensive SMBG versus no SMBG. Available qualitative data gained from in-depth interview studies, repeated interviews, and questionnaire and survey data were summarised. RESULTS:The review identified 30 RCTs, although few were of high quality. Ten trials comparing SMBG with no SMBG showed statistically significant reduction in HbA1C of 0.21%, which may not be considered clinically significant. A similar, though not statistically significant difference, was shown where SMBG with education was compared to SMBG without education or feedback. RCTs showed no consistent effect on hypoglycaemic episodes and no impact on medication changes. Review of cost-effectiveness studies showed that costs of SMBG per annum vary considerably (10-259 pounds). Although some studies assert that SMBG may lead to savings in health-care costs which may offset the costs of testing, the best analysis to date (DiGEM - Diabetes Glycaemic Education and Monitoring) concluded that SMBG was not cost-effective. Qualitative studies revealed that there was a lack of education in how to interpret and use the data from SMBG, and that failure to act on the results was common. CONCLUSIONS:The evidence suggested that SMBG is of limited clinical effectiveness in improving glycaemic control in people with T2DM on oral agents, or diet alone, and is therefore unlikely to be cost-effective. SMBG may lead to improved glycaemic control only in the context of appropriate education - both for patients and health-care professionals - on how to respond to the data, in terms of lifestyle and treatment adjustment. Also, SMBG may be more effective if patients are able to self-adjust drug treatment. Further research is required on the type of education and feedback that are most helpful, characteristics of patients benefiting most from SMBG, optimal timing and frequency of SMBG, and the circumstances under which SMBG causes anxiety and/or depression.

BACKGROUND The National Institute for Health and Clinical Excellence (NICE) was reviewing its previous guidance on continuous subcutaneous insulin infusion (CSII). The review provided an assessment of evidence which had been published since the previous NICE appraisal (TA 151) in 2007. OBJECTIVES To examine the clinical effectiveness and cost-effectiveness of using CSII to treat diabetes. To update the previous assessment report by reviewing evidence that has emerged since the last appraisal, and to take account of developments in alternative therapies, in particular the long-acting analogue insulins, which cause fewer problems with hypoglycaemia. DATA SOURCES A systematic review of the literature and an economic evaluation were carried out. The bibliographic databases used were MEDLINE and EMBASE, 2002 to June 2007. The Cochrane Library (all sections), the Science Citation Index (for meeting abstracts only) and the website of the 2007 American Diabetes Association were also searched. REVIEW METHODS The primary focus for type 1 diabetes mellitus (T1DM) was the comparison of CSII with multiple daily injection (MDI), based on the newer insulin analogues, but trials of neutral protamine Hagedorn (NPH)-based MDI that had been published since the last assessment were identified and described in brief. For type 2 diabetes mellitus (T2DM), all trials of MDI versus CSII were included, whether the long-acting insulin was analogue or not, because there was no evidence that analogue-based MDI was better than NPH-based MDI. Trials that were shorter than 12 weeks were excluded. Information on the patients' perspectives was obtained from four sources: the submission from the pump users group--Insulin Pump Therapy (INPUT); interviews with parents of young children who were members of INPUT; some recent studies; and from a summary of findings from the previous assessment report. Economic modelling used the Center for Outcomes Research (CORE) model, through an arrangement with the NICE and the pump manufacturers, whose submission also used the CORE model. RESULTS The 74 studies used for analysis included eight randomised controlled trials (RCTs) of CSII versus analogue-based MDI in either T1DM or T2DM, eight new (since the last NICE appraisal) RCTs of CSII versus NPH-based MDI in T1DM, 48 observational studies of CSII, six studies of CSII in pregnancy, and four systematic reviews. The following benefits of CSII were highlighted: better control of blood glucose levels, as reflected by glycated haemoglobin (HbA1c) levels, with the size of improvement depending on the level before starting CSII; reduction in swings in blood glucose levels, and in problems due to the dawn phenomenon; fewer problems with hypoglycaemic episodes; reduction in insulin dose per day, thereby partly off-setting the cost of CSII; improved quality of life, including a reduction in the chronic fear of severe hypoglycaemia; more flexibility of lifestyle--no need to eat at fixed intervals, more freedom of lifestyle and easier participation in social and physical activity; and benefits for the patients' family. The submission from INPUT emphasised the quality of life gains from CSII, as well as improved control and fewer hypoglycaemic episodes. Also, there was a marked discrepancy between the improvement in social quality of life reported by successful pump users, and the lack of convincing health-related quality of life gains reported in the trials. With regard to economic evaluation, the main cost of CSII is for consumables, such as tubing and cannulas, and is about 1800-2000 pounds per year. The cost of the pump, assuming 4-year life, adds another 430-720 pounds per annum. The extra cost compared with analogue-based MDI averages 1700 pounds. Most studies, assuming a reduction in HbA1c level of 1.2%, found CSII to be cost-effective. LIMITATIONS The most important weakness of the evidence was the very small number of randomised trials of CSII against the most modern forms of MDI, using analogue insulins. CONCLUSIONS Based on the totality of evidence, using observational studies to supplement the limited data from randomised trials against best MDI, CSII provides some advantages over MDI in T1DM for both children and adults. However, there was no evidence that CSII is better than analogue-based MDI in T2DM or in pregnancy. Further trials with larger numbers and longer durations comparing CSII and optimised MDI in adults, adolescents and children are needed. In addition, there should be a trial of CSII versus MDI with similar provision of structured education in both arms. A trial is also needed for pregnant women with pre-existing diabetes, to investigate using CSII to the best effect.

BACKGROUNDIn May 2008, the National Institute for Health and Clinical Excellence (NICE) issued an updated guideline [clinical guideline (CG) 66] for the management of all aspects of type 2 diabetes. This report aims to provide information on new drug developments to support a 'new drugs update' to the 2008 guideline.OBJECTIVETo review the newer agents available for blood glucose control in type 2 diabetes from four classes: the glucagon-like peptide-1 (GLP-1) analogue exenatide; dipeptidyl peptidase-4 (DPP-4) inhibitors sitagliptin and vildagliptin; the long-acting insulin analogues, glargine and detemir; and to review concerns about the safety of the thiazolidinediones.DATA SOURCESThe following databases were searched: MEDLINE (1990-April 2008), EMBASE (1990-April 2008), the Cochrane Library (all sections) Issue 2, 2008, and the Science Citation Index and ISI Proceedings (2000-April 2008). The websites of the American Diabetes Association, the European Association for the Study of Diabetes, the US Food and Drug Administration, the European Medicines Evaluation Agency and the Medicines and Healthcare Products Regulatory Agency were searched, as were manufacturers' websites.REVIEW METHODSData extraction was carried out by one person, and checked by a second. Studies were assessed for quality using standard methods for reviews of trials. Meta-analyses were carried out using the Cochrane Review Manager (RevMan) software. Inclusion and exclusion criteria were based on current standard clinical practice in the UK, as outlined in NICE CG 66. The outcomes for the GLP-1 analogues, DPP-4 inhibitors and the long-acting insulin analogues were: glycaemic control, reflected by glycated haemoglobin (HbA1c) level, hypoglycaemic episodes, changes in weight, adverse events, quality of life and costs. Modelling of the cost-effectiveness of the various regimes used the United Kingdom Prospective Diabetes Study (UKPDS) Outcomes Model.RESULTSExenatide improved glycaemic control by around 1%, and had the added benefit of weight loss. The gliptins were effective in improving glycaemic control, reducing HbA1c level by about 0.8%. Glargine and detemir were equivalent to Neutral Protamine Hagedorn (NPH) (and to each other) in terms of glycaemic control but had modest advantages in terms of hypoglycaemia, especially nocturnal. Detemir, used only once daily, appeared to cause slightly less weight gain than glargine. The glitazones appeared to have similar effectiveness in controlling hyperglycaemia. Both can cause heart failure and fractures, but rosiglitazone appears to slightly increase the risk of cardiovascular events whereas pioglitazone reduces it. Eight trials examined the benefits of adding pioglitazone to an insulin regimen; in our meta-analysis, the mean reduction in HbA1c level was 0.54% [95% confidence interval (CI) -0.70 to -0.38] and hypoglycaemia was marginally more frequent in the pioglitazone arms [relative risk (RR) 1.27, 95% CI 0.99 to 1.63]. In most studies, those on pioglitazone gained more weight than those who were not. In terms of annual drug acquisition costs among the non-insulin regimes for a representative patient with a body mass index of around 30 kg/m2, the gliptins were the cheapest of the new drugs, with costs of between 386 pounds and 460 pounds. The glitazone costs were similar, with total annual costs for pioglitazone and for rosiglitazone of around 437 pounds and 482 pounds, respectively. Exenatide was more expensive, with an annual cost of around 830 pounds. Regimens containing insulin fell between the gliptins and exenatide in terms of their direct costs, with a NPH-based regimen having an annual cost of around 468 pounds for the representative patient, whereas the glargine and detemir regimens were more expensive, at around 634 pounds and 716 pounds, respectively. Comparisons of sitagliptin and rosiglitazone, and of vidagliptin and pioglitazone slowed clinical equivalence in terms of quality-adjusted life-years (QALYs), but the gliptins were marginally less costly. Exenatide, when compared with glargine, appeared to be cost-effective. Comparing glargine with NPH showed an additional anticipated cost of around 1800 pounds. Within the comparison of detemir and NPH, the overall treatment costs for detemir were slightly higher, at between 2700 pounds and 2600 pounds.LIMITATIONSThe UKPDS Outcomes Model does not directly address aspects of the treatments under consideration, for example the direct utility effects from weight loss or weight gain, severe hypoglycaemic events and the fear of severe hypoglycaemic events. Also, small differences in QALYs among the drugs lead to fluctuations in incremental cost-effectiveness ratios.CONCLUSIONSExenatide, the gliptins and detemir were all clinically effective. The long-acting insulin analogues glargine and detemir appeared to have only slight clinical advantages over NPH, but had much higher costs and did not appear to be cost-effective as first-line insulins for type 2 diabetes. Neither did exenatide appear to be cost-effective compared with NPH but, when used as third drug after failure of dual oral combination therapy, exenatide appeared cost-effective relative to glargine in this analysis. The gliptins are similar to the glitazones in glycaemic control and costs, and appeared to have fewer long-term side effects. Therefore, it appears, as supported by recent NICE guidelines, that NPH should be the preferred first-line insulin for the treatment of type 2 diabetes. More economic analysis is required to establish when it becomes cost-effective to switch from NPH to a long-acting analogue. Also, long-term follow-up studies of exenatide and the gliptins, and data on combined insulin and exenatide treatment, would be useful.

As adding metformin to insulin therapy has been advocated in type 1 diabetes, we conducted a systematic review of published clinical trials and clinical trial databases to assess the effects on HbA1c, weight, insulin-dose requirement and adverse effects.