Meta-analysis of individual patient data from randomised trials: a review of methods used in practice

Background: Meta-analyses based on individual patient data (IPD) are regarded as the gold standard for systematic reviews. However, the methods used for analysing and presenting results from IPD meta-analyses have received little discussion. Methods We review 44 IPD meta-analyses published during the years 1999–2001. We summarize whether they obtained all the data they sought, what types of approaches were used in the analysis, including assumptions of common or random effects, and how they examined the effects of covariates. Results: Twenty-four out of 44 analyses focused on time-to-event outcomes, and most analyses (28) estimated treatment effects within each trial and then combined the results assuming a common treatment effect across trials. Three analyses failed to stratify by trial, analysing the data is if they came from a single mega-trial. Only nine analyses used random effects methods. Covariate-treatment interactions were generally investigated by subgrouping patients. Seven of the meta-analyses included data from less than 80% of the randomized patients sought, but did not address the resulting potential biases. Conclusions: Although IPD meta-analyses have many advantages in assessing the effects of health care, there are several aspects that could be further developed to make fuller use of the potential of these time-consuming projects. In particular, IPD could be used to more fully investigate the influence of covariates on heterogeneity of treatment effects, both within and between trials. The impact of heterogeneity, or use of random effects, are seldom discussed. There is thus considerable scope for enhancing the methods of analysis and presentation of IPD meta-analysis.

Design and baseline characteristics of the Food4Me study: a web-based randomised controlled trial of personalised nutrition in seven European countries

Improving lifestyle behaviours has considerable potential for reducing the global burden of non-communicable diseases, promoting better health across the life-course and increasing well-being. However, realising this potential will require the development, testing and implementation of much more effective behaviour change interventions than are used conventionally. Therefore, the aim of this study was to conduct a multi-centre, web-based, proof-of-principle study of personalised nutrition (PN) to determine whether providing more personalised dietary advice leads to greater improvements in eating patterns and health outcomes compared to conventional population-based advice. A total of 5,562 volunteers were screened across seven European countries; the first 1,607 participants who fulfilled the inclusion criteria were recruited into the trial. Participants were randomly assigned to one of the following intervention groups for a 6-month period: Level 0-control group-receiving conventional, non-PN advice; Level 1-receiving PN advice based on dietary intake data alone; Level 2-receiving PN advice based on dietary intake and phenotypic data; and Level 3-receiving PN advice based on dietary intake, phenotypic and genotypic data. A total of 1,607 participants had a mean age of 39.8 years (ranging from 18 to 79 years). Of these participants, 60.9 % were women and 96.7 % were from white-European background. The mean BMI for all randomised participants was 25.5 kg m(-2), and 44.8 % of the participants had a BMI ≥ 25.0 kg m(-2). Food4Me is the first large multi-centre RCT of web-based PN. The main outcomes from the Food4Me study will be submitted for publication during 2015.

Dietary fat modifications and blood pressure in subjects with the metabolic syndrome in the LIPGENE dietary intervention study

Hypertension is a key feature of the metabolic syndrome. Lifestyle and dietary changes may affect blood pressure (BP), but the knowledge of the effects of dietary fat modification in subjects with the metabolic syndrome is limited. The objective of the present study was to investigate the effect of an isoenergetic change in the quantity and quality of dietary fat on BP in subjects with the metabolic syndrome. In a 12-week European multi-centre, parallel, randomised controlled dietary intervention trial (LIPGENE), 486 subjects were assigned to one of the four diets distinct in fat quantity and quality: two high-fat diets rich in saturated fat or monounsaturated fat and two low-fat, high-complex carbohydrate diets with or without 1·2 g/d of very long-chain n-3 PUFA supplementation. There were no overall differences in systolic BP (SBP), diastolic BP or pulse pressure (PP) between the dietary groups after the intervention. The high-fat diet rich in saturated fat had minor unfavourable effects on SBP and PP in males.

Human microbiota in health and disease

Each human body plays host to a microbial population which is both numerically vast (at around 1014 microbial cells) and phenomenally diverse (over 1,000 species). The majority of the microbial species in the gut have not been cultured but the application of culture-independent approaches for high throughput diversity and functionality analysis has allowed characterisation of the diverse microbial phylotypes present in health and disease. Studies in monozygotic twins, showing that these retain highly similar microbiota decades after birth and initial colonisation, are strongly indicative that diversity of the microbiome is host-specific and affected by the genotype. Microbial diversity in the human body is reflected in both richness and evenness. Diversity increases steeply from birth reaching its highest point in early adulthood, before declining in older age. However, in healthy subjects there appears to be a core of microbial phylotypes which remains relatively stable over time. Studies of individuals from diverse geopraphies suggest that clusters of intestinal bacterial groups tend to occur together, constituting ‘enterotypes’. So variation in intestinal microbiota is stratified rather than continuous and there may be a limited number of host/microbial states which respond differently to environmental influences. Exploration of enterotypes and functional groups may provide biomarkers for disease and insights into the potential for new treatments based on manipulation of the microbiome. In health, the microbiota interact with host defences and exist in harmonious homeostasis which can then be disturbed by invading organisms or when ‘carpet bombing’ by antibiotics occurs. In a portion of individuals with infections, the disease will resolve itself without the need for antibiotics and microbial homeostasis with the host’s defences is restored. The administration of probiotics (live microorganisms which when administered in adequate amounts confer a health benefit on the host) represents an artificial way to enhance or stimulate these natural processes. The study of innate mechanisms of antimicrobial defence on the skin, including the production of numerous antimicrobial peptides (AMPs), has shown an important role for skin commensal organisms. These organisms may produce AMPs, and also amplify the innate immune responses to pathogens by activating signalling pathways and processing host produced AMPs. Research continues into how to enhance and manipulate the role of commensal organisms on the skin. The challenges of skin infection (including diseases caused by multiply resistant organisms) and infestations remain considerable. The potential to re-colonise the skin to replace or reduce pathogens, and exploring the relationship between microbiota elsewhere and skin diseases are among a growing list of research targets. Lactobacillus species are among the best known ‘beneficial’ bacterial members of the human microbiota. Of the approximately 120 species known, about 15 are known to occur in the human vagina. These organisms have multiple properties, including the production of lactic acid, hydrogen peroxide and bacteriocins, which render the vagina inhospitable to potential pathogens. Depletion of the of the normal Lactobacillus population and overgrowth of vaginal anaerobes, accompanied by the loss of normal vaginal acidity can lead to bacterial vaginosis – the commonest cause of abnormal vaginal discharge in women. Some vaginal anaerobes are associated with the formation of vaginal biofilms which serve to act as a reservoir of organisms which persists after standard antibiotic therapy of bacterial vaginosis and may help to account for the characteristically high relapse rate in the condition. Administration of Lactobacillus species both vaginally and orally have shown beneficial effects in the treatment of bacterial vaginosis and such treatments have an excellent overall safety record. Candida albicans is a frequent coloniser of human skin and mucosal membranes, and is a normal part of the microbiota in the mouth, gut and vagina. Nevertheless Candida albicans is the most common fungal pathogen worldwide and is a leading cause of serious and often fatal nosocomial infections. What turns this organism from a commensal to a pathogen is a combination of increasing virulence in the organism and predisposing host factors that compromise immunity. There has been considerable research into the use of probiotic Lactobacillus spp. in vaginal candidiasis. Studies in reconstituted human epithelium and monolayer cell cultures have shown that L. rhamnosus GG can protect mucosa from damage caused by Candida albicans, and enhance the immune responses of mucosal surfaces. Such findings offer the promise that the use of such probiotic bacteria could provide new options for antifungal therapy. Studies of changes of the human intestinal microbiota in health and disease are complicated by its size and diversity. The Alimentary Pharmabiotic Centre in Cork (Republic of Ireland) has the mission to ‘mine microbes for mankind’ and its work illustrates the potential benefits of understanding the gut microbiota. Work undertaken at the centre includes: mapping changes in the microbiota with age; studies of the interaction between the microbiota and the gut; potential interactions between the gut microbiota and the central nervous system; the potential for probiotics to act as anti-infectives including through the production of bacteriocins; and the characterisation of interactions between gut microbiota and bile acids which have important roles as signalling molecules and in immunity. The important disease entity where the role of the gut microbiota appears to be central is the Irritable Bowel Syndrome (IBS). IBS patients show evidence of immune activation, impaired gut barrier function and abnormal gut microbiota. Studies with probiotics have shown that these organisms can exert anti-inflammatory effects in inflammatory bowel disease and may strengthen the gut barrier in IBS of the diarrhoea-predominant type. Formal randomised trials of probiotics in IBS show mixed results with limited benefit for some but not all. Studies confirm that administered probiotics can survive and temporarily colonise the gut. They can also stimulate the numbers of other lactic acid bacilli in the gut, and reduce the numbers of pathogens. However consuming live organisms is not the only way to influence gut microbiota. Dietary prebiotics are selectively fermented ingredients that can change the composition and/or activity of the gastrointestinal microbiota in beneficial ways. Dietary components that reach the colon, and are available to influence the microbiota include poorly digestible carbohydrates, such as non-starch polysaccharides, resistant starch, non-digestible oligosaccharides (NDOs) and polyphenols. Mixtures of probiotic and prebiotic ingredients that can selectively stimulate growth or activity of health promoting bacteria have been termed ‘synbiotics’. All of these approaches can influence gut microbial ecology, mainly to increase bifidobacteria and lactobacilli, but metagenomic approaches may reveal wider effects. Characterising how these changes produce physiological benefits may enable broader use of these tactics in health and disease in the future. The current status of probiotic products commercially available worldwide is less than ideal. Prevalent problems include misidentification of ingredient organisms and poor viability of probiotic microorganisms leading to inadequate shelf life. On occasions these problems mean that some commercially available products cannot be considered to meet the definition of a probiotic product. Given the potential benefits of manipulating the human microbiota for beneficial effects, there is a clear need for improved regulation of probiotics. The potential importance of the human microbiota cannot be overstated. ‘We feed our microbes, they talk to us and we benefit. We just have to understand and then exploit this.’ (Willem de Vos).

UK Food Standards Agency cis-monounsaturated fatty acid workshop report

The UK Food Standards Agency convened a group of expert scientists to review current research investigating the optimal dietary intake for n-9 cis-monounsaturated fatty acids (MUFA). The aim was to review the mechanisms underlying the reported beneficial effects of MUFA on CHD risk, and to establish priorities for future research. The issue of optimal MUFA intake is contingent upon optimal total fat intake; however, there is no consensus of opinion on what the optimal total fat intake should be. Thus, it was recommended that a large multi-centre study should look at the effects on CHD risk of MUFA replacement of saturated fatty acids in relation to varying total fat intakes; this study should be of sufficient size to take account of genetic variation, sex, physical activity and stage of life factors, as well as being of sufficient duration to account for adaptation to diets. Recommendations for studies investigating the mechanistic effects of MUFA were also made. Methods of manipulating the food chain to increase MUFA at the expense of saturated fatty acids were also discussed.

Effective elements of cognitive behaviour therapy for psychosis: results of a novel type of subgroup analysis based on principal stratification

Background. Meta-analyses show that cognitive behaviour therapy for psychosis (CBT-P) improves distressing positive symptoms. However, it is a complex intervention involving a range of techniques. No previous study has assessed the delivery of the different elements of treatment and their effect on outcome. Our aim was to assess the differential effect of type of treatment delivered on the effectiveness of CBT-P, using novel statistical methodology. Method. The Psychological Prevention of Relapse in Psychosis (PRP) trial was a multi-centre randomized controlled trial (RCT) that compared CBT-P with treatment as usual (TAU). Therapy was manualized, and detailed evaluations of therapy delivery and client engagement were made. Follow-up assessments were made at 12 and 24 months. In a planned analysis, we applied principal stratification (involving structural equation modelling with finite mixtures) to estimate intention-to-treat (ITT) effects for subgroups of participants, defined by qualitative and quantitative differences in receipt of therapy, while maintaining the constraints of randomization. Results. Consistent delivery of full therapy, including specific cognitive and behavioural techniques, was associated with clinically and statistically significant increases in months in remission, and decreases in psychotic and affective symptoms. Delivery of partial therapy involving engagement and assessment was not effective. Conclusions. Our analyses suggest that CBT-P is of significant benefit on multiple outcomes to patients able to engage in the full range of therapy procedures. The novel statistical methods illustrated in this report have general application to the evaluation of heterogeneity in the effects of treatment.

Impact of geographical region on urinary metabolomic and plasma fatty acid profiles in subjects with the metabolic syndrome across Europe: the LIPGENE study

The application of metabolomics in multi-centre studies is increasing. The aim of the present study was to assess the effects of geographical location on the metabolic profiles of individuals with the metabolic syndrome. Blood and urine samples were collected from 219 adults from seven European centres participating in the LIPGENE project (Diet, genomics and the metabolic syndrome: an integrated nutrition, agro-food, social and economic analysis). Nutrient intakes, BMI, waist:hip ratio, blood pressure, and plasma glucose, insulin and blood lipid levels were assessed. Plasma fatty acid levels and urine were assessed using a metabolomic technique. The separation of three European geographical groups (NW, northwest; NE, northeast; SW, southwest) was identified using partial least-squares discriminant analysis models for urine (R 2 X: 0•33, Q 2: 0•39) and plasma fatty acid (R 2 X: 0•32, Q 2: 0•60) data. The NW group was characterised by higher levels of urinary hippurate and N-methylnicotinate. The NE group was characterised by higher levels of urinary creatine and citrate and plasma EPA (20 : 5 n-3). The SW group was characterised by higher levels of urinary trimethylamine oxide and lower levels of plasma EPA. The indicators of metabolic health appeared to be consistent across the groups. The SW group had higher intakes of total fat and MUFA compared with both the NW and NE groups (P≤ 0•001). The NE group had higher intakes of fibre and n-3 and n-6 fatty acids compared with both the NW and SW groups (all P< 0•001). It is likely that differences in dietary intakes contributed to the separation of the three groups. Evaluation of geographical factors including diet should be considered in the interpretation of metabolomic data from multi-centre studies.

Application of dried blood spots to determine vitamin D status in a large nutritional study with unsupervised sampling: the Food4Me project

An efficient and robust method to measure vitamin D (25-hydroxy vitamin D3 (25(OH)D3) and 25-hydroxy vitamin D2 in dried blood spots (DBS) has been developed and applied in the pan-European multi-centre, internet-based, personalised nutrition intervention study Food4Me. The method includes calibration with blood containing endogenous 25(OH)D3, spotted as DBS and corrected for haematocrit content. The methodology was validated following international standards. The performance characteristics did not reach those of the current gold standard liquid chromatography-MS/MS in plasma for all parameters, but were found to be very suitable for status-level determination under field conditions. DBS sample quality was very high, and 3778 measurements of 25(OH)D3 were obtained from 1465 participants. The study centre and the season within the study centre were very good predictors of 25(OH)D3 levels (P<0·001 for each case). Seasonal effects were modelled by fitting a sine function with a minimum 25(OH)D3 level on 20 January and a maximum on 21 July. The seasonal amplitude varied from centre to centre. The largest difference between winter and summer levels was found in Germany and the smallest in Poland. The model was cross-validated to determine the consistency of the predictions and the performance of the DBS method. The Pearson's correlation between the measured values and the predicted values was r 0·65, and the sd of their differences was 21·2 nmol/l. This includes the analytical variation and the biological variation within subjects. Overall, DBS obtained by unsupervised sampling of the participants at home was a viable methodology for obtaining vitamin D status information in a large nutritional study.

Donepezil for the symptomatic treatment of patients with mild to moderate Alzheimer's disease: a meta-analysis of individual patient data from randomised controlled trials

Background: The objective was to evaluate the efficacy and tolerability of donepezil (5 and 10 mg/day) compared with placebo in alleviating manifestations of mild to moderate Alzheimer's disease (AD). Method: A systematic review of individual patient data from Phase II and III double-blind, randomised, placebo-controlled studies of up to 24 weeks and completed by 20 December 1999. The main outcome measures were the ADAS-cog, the CIBIC-plus, and reports of adverse events. Results: A total of 2376 patients from ten trials were randomised to either donepezil 5 mg/day (n = 821), 10 mg/day (n = 662) or placebo (n = 893). Cognitive performance was better in patients receiving donepezil than in patients receiving placebo. At 12 weeks the differences in ADAS-cog scores were 5 mg/day-placebo: - 2.1 [95% confidence interval (CI), - 2.6 to - 1.6; p < 0.001], 10 mg/day-placebo: - 2.5 ( - 3.1 to - 2.0; p < 0.001). The corresponding results at 24 weeks were - 2.0 ( - 2.7 to - 1.3; p < 0.001) and - 3.1 ( - 3.9 to - 2.4; p < 0.001). The difference between the 5 and 10 mg/day doses was significant at 24 weeks (p = 0.005). The odds ratios (OR) of improvement on the CIBIC-plus at 12 weeks were: 5 mg/day-placebo 1.8 (1.5 to 2.1; p < 0.001), 10 mg/day-placebo 1.9 (1.5 to 2.4; p < 0.001). The corresponding values at 24 weeks were 1.9 (1.5 to 2.4; p = 0.001) and 2.1 (1.6 to 2.8; p < 0.001). Donepezil was well tolerated; adverse events were cholinergic in nature and generally of mild severity and brief in duration. Conclusion: Donepezil (5 and 10 mg/day) provides meaningful benefits in alleviating deficits in cognitive and clinician-rated global function in AD patients relative to placebo. Increased improvements in cognition were indicated for the higher dose. Copyright © 2004 John Wiley & Sons, Ltd.

Integrated motivational interviewing and cognitive behavioural therapy for people with psychosis and comorbid substance misuse: randomised controlled trial

Objectives To evaluate the effectiveness of integrated motivational interviewing and cognitive behaviour therapy in addition to standard care for patients with psychosis and a co-morbid substance use problem. Design Two-centre, open, rater-blind randomised controlled trial Setting UK Secondary Care Participants 327 patients with clinical diagnoses of schizophrenia, schizophreniform or schizoaffective disorder and DSM-IV diagnoses of drug and/or alcohol dependence or abuse Interventions Participants were randomly allocated to integrated motivational interviewing and cognitive behaviour therapy or standard care. Therapy has two phases. Phase one – “motivation building” – concerns engaging the patient, then exploring and resolving ambivalence for change in substance use. Phase two –“Action” – supports and facilitates change using cognitive behavioural approaches. Up to 26 therapy sessions were delivered over one year. Main outcomes The primary outcome was death from any cause or admission to hospital in the 12 months after therapy. Secondary outcomes were frequency and amount of substance use (Timeline Followback), readiness to change, perceived negative consequences of use, psychotic symptom ratings, number and duration of relapses, global assessment of functioning and deliberate self harm, at 12 and 24 months, with additional Timeline Followback assessments at 6 and 18 months. Analysis was by intention-to-treat with robust treatment effect estimates. Results 327 participants were randomised. 326 (99.7%) were assessed on the primary outcome, 246 (75.2%) on main secondary outcomes at 24 months. Regarding the primary outcome, there was no beneficial treatment effect on hospital admissions/ death during follow-up, with 20.2% (33/163) of controls and 23.3% (38/163) of the therapy group deceased or admitted (adjusted odds-ratio 1.16; P= 0.579; 95% confidence interval 0.68 to 1.99). For secondary outcomes there was no treatment effect on frequency of substance use or perceived negative consequences, but a statistically significant effect of therapy on amount used per substance-using day (adjusted odds-ratios: (a) for main substance 1.50; P=0.016; 1.08 to 2.09, (b) all substances 1.48; P=0.017; 1.07 to 2.05). There was a statistically significant treatment effect on readiness to change use at 12 months (adjusted odds-ratio 2.05; P=0.004; 1.26 to 3.31), not maintained at 24 months. There were no treatment effects on assessed clinical outcomes. Conclusions Integrated motivational interviewing and cognitive behaviour therapy for people with psychosis and substance misuse does not improve outcome in terms of hospitalisation, symptom outcomes or functioning. It does result in a reduction in amount of substance use which is maintained over the year’s follow up. Trial registration Current Controlled Trials: ISRCTN14404480

Casein-derived lactotripeptides reduce systolic and diastolic blood pressure in a meta-analysis of randomised clinical trials

There is an urgent need to treat individuals with high blood pressure (BP) with effective dietary strategies. Previous studies suggest a small, but significant decrease in BP after lactotripeptides (LTP) ingestion, although the data are inconsistent. The study aim was to perform a comprehensive meta-analysis of data from all relevant randomised controlled trials (RCT). Medline, Cochrane library, EMBASE and Web of Science were searched until May 2014. Eligibility criteria were RCT that examined the effects of LTP on BP in adults, with systolic BP (SBP) and diastolic BP (DBP) as outcome measures. Thirty RCT met the inclusion criteria, which resulted in 33 sets of data. The pooled treatment effect for SBP was −2.95 mmHg (95% CI: −4.17, −1.73; p < 0.001), and for DBP was −1.51 mmHg (95% CI: −2.21, −0.80; p < 0.001). Sub-group analyses revealed that reduction of BP in Japanese studies was significantly greater, compared with European studies (p = 0.002 for SBP and p < 0.001 for DBP). The 24-h ambulatory BP (AMBP) response to LTP supplementation was statistically non-significant (p = 0.101 for SBP and p = 0.166 for DBP). Both publication bias and “small-study effect” were identified, which shifted the treatment effect towards less significant SBP and non-significant DBP reduction after LTP consumption. LTP may be effective in BP reduction, especially in Japanese individuals; however sub-group, meta-regression analyses and statistically significant publication biases suggest inconsistencies.

Diet rich in high glucoraphanin broccoli reduces plasma LDL cholesterol: evidence from randomised controlled trials

Cruciferous-rich diets have been associated with reduction in plasma LDL-cholesterol (LDL-C), which may be due to the action of isothiocyanates derived from glucosinolates that accumulate in these vegetables. This study tests the hypothesis that a diet rich in high glucoraphanin (HG) broccoli will reduce plasma LDL-C. METHODS AND RESULTS: One hundred and thirty volunteers were recruited to two independent double-blind, randomly allocated parallel dietary intervention studies, and were assigned to consume either 400 g standard broccoli or 400 g HG broccoli per week for 12 weeks. Plasma lipids were quantified before and after the intervention. In study 1 (37 volunteers), the HG broccoli diet reduced plasma LDL-C by 7.1% (95% CI: -1.8%, -12.3%, p = 0.011), whereas standard broccoli reduced LDL-C by 1.8% (95% CI +3.9%, -7.5%, ns). In study 2 (93 volunteers), the HG broccoli diet resulted in a reduction of 5.1% (95% CI: -2.1%, -8.1%, p = 0.001), whereas standard broccoli reduced LDL-C by 2.5% (95% CI: +0.8%, -5.7%, ns). When data from the two studies were combined the reduction in LDL-C by the HG broccoli was significantly greater than standard broccoli (p = 0.031). CONCLUSION: Evidence from two independent human studies indicates that consumption of high glucoraphanin broccoli significantly reduces plasma LDL-C

A randomised controlled trial of positive memory training for the treatment of depression within schizophrenia

Abstract Background: Depression is highly prevalent within individuals diagnosed with schizophrenia, and is associated with an increased risk of suicide. There are no current evidence based treatments for low mood within this group. The specific targeting of co-morbid conditions within complex mental health problems lends itself to the development of short-term structured interventions which are relatively easy to disseminate within health services. A brief cognitive intervention based on a competitive memory theory of depression, is being evaluated in terms of its effectiveness in reducing depression within this group. Methods/Design: This is a single blind, intention-to-treat, multi-site, randomized controlled trial comparing Positive Memory Training plus Treatment as Usual with Treatment as Usual alone. Participants will be recruited from two NHS Trusts in Southern England. In order to be eligible, participants must have a DSM-V diagnosis of schizophrenia or schizo-affective disorder and exhibit at least a mild level of depression. Following baseline assessment eligible participants will be randomly allocated to either the Positive Memory Training plus Treatment as Usual group or the Treatment as Usual group. Outcome will be assessed at the end of treatment (3-months) and at 6-month and 9-month post randomization by assessors blind to group allocation. The primary outcome will be levels of depression and secondary outcomes will be severity of psychotic symptoms and cost-effectiveness. Semi-structured interviews will be conducted with all participants who are allocated to the treatment group so as to explore the acceptability of the intervention. Discussion: Cognitive behaviour therapy is recommended for individuals diagnosed with schizophrenia. However, the number of sessions and length of training required to deliver this intervention has caused a limit in availability. The current trial will evaluate a short-term structured protocol which targets a co-morbid condition often considered of primary importance by service users. If successful the intervention will be an important addition to current initiatives aimed at increasing access to psychological therapies for people diagnosed with severe mental health problems. Trial registration: Current Controlled Trials. ISRCTN99485756. Registered 13 March 2014.

An investigation of the shortcomings of the CONSORT 2010 statement for the reporting of group sequential randomised controlled trials: a methodological systematic review

Background It can be argued that adaptive designs are underused in clinical research. We have explored concerns related to inadequate reporting of such trials, which may influence their uptake. Through a careful examination of the literature, we evaluated the standards of reporting of group sequential (GS) randomised controlled trials, one form of a confirmatory adaptive design. Methods We undertook a systematic review, by searching Ovid MEDLINE from the 1st January 2001 to 23rd September 2014, supplemented with trials from an audit study. We included parallel group, confirmatory, GS trials that were prospectively designed using a Frequentist approach. Eligible trials were examined for compliance in their reporting against the CONSORT 2010 checklist. In addition, as part of our evaluation, we developed a supplementary checklist to explicitly capture group sequential specific reporting aspects, and investigated how these are currently being reported. Results Of the 284 screened trials, 68(24%) were eligible. Most trials were published in “high impact” peer-reviewed journals. Examination of trials established that 46(68%) were stopped early, predominantly either for futility or efficacy. Suboptimal reporting compliance was found in general items relating to: access to full trials protocols; methods to generate randomisation list(s); details of randomisation concealment, and its implementation. Benchmarking against the supplementary checklist, GS aspects were largely inadequately reported. Only 3(7%) trials which stopped early reported use of statistical bias correction. Moreover, 52(76%) trials failed to disclose methods used to minimise the risk of operational bias, due to the knowledge or leakage of interim results. Occurrence of changes to trial methods and outcomes could not be determined in most trials, due to inaccessible protocols and amendments. Discussion and Conclusions There are issues with the reporting of GS trials, particularly those specific to the conduct of interim analyses. Suboptimal reporting of bias correction methods could potentially imply most GS trials stopping early are giving biased results of treatment effects. As a result, research consumers may question credibility of findings to change practice when trials are stopped early. These issues could be alleviated through a CONSORT extension. Assurance of scientific rigour through transparent adequate reporting is paramount to the credibility of findings from adaptive trials. Our systematic literature search was restricted to one database due to resource constraints.