von Elm E.,University of Lausanne |
Altman D.G.,Center for Statistics in Medicine |
Egger M.,University of Bern |
Egger M.,University of Cape Town |
And 3 more authors.
International Journal of Surgery | Year: 2014
Much biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalisability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case-control, and cross-sectional studies. We convened a 2-day workshop in September 2004, with methodologists, researchers, and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE Statement) that relate to the title, abstract, introduction, methods, results, and discussion sections of articles. 18 items are common to all three study designs and four are specific for cohort, case-control, or cross-sectional studies. A detailed Explanation and Elaboration document is published separately and is freely available on the Web sites of PLoS Medicine, Annals of Internal Medicine, and Epidemiology. We hope that the STROBE Statement will contribute to improving the quality of reporting of observational studies. © 2014 The Authors.
Hanney M.,King's College London |
Prasher V.,Birmingham Community Healthcare National Health Service Trust |
Williams N.,Center for Statistics in Medicine |
Jones E.L.,King's College London |
And 9 more authors.
The Lancet | Year: 2012
Background: Prevalence of Alzheimer's disease in people with Down's syndrome is very high, and many such individuals who are older than 40 years have pathological changes characteristic of Alzheimer's disease. Evidence to support treatment with Alzheimer's drugs is inadequate, although memantine is beneficial in transgenic mice. We aimed to assess safety and efficacy of memantine on cognition and function in individuals with Down's syndrome. Methods: In our prospective randomised double-blind trial, we enrolled adults (>40 years) with karyotypic or clinically diagnosed Down's syndrome, with and without dementia, at four learning disability centres in the UK and Norway. We randomly allocated participants (1:1) to receive memantine or placebo for 52 weeks by use of a computer-generated sequence and a minimisation algorithm to ensure balanced allocation for five prognostic factors (sex, dementia, age group, total Down's syndrome attention, memory, and executive function scales [DAMES] score, and centre). The primary outcome was change in cognition and function, measured with DAMES scores and the adaptive behaviour scale (ABS) parts I and II. We analysed differences in DAMES and ABS scores between groups with analyses of covariance or quantile regression in all patients who completed the 52 week assessment and had available follow-up data. This study is registered, number ISRCTN47562898. Findings: We randomly allocated 88 patients to receive memantine (72  had DAMES data and 75  had ABS data at 52 weeks) and 85 to receive placebo (74  and 73 ). Both groups declined in cognition and function but rates did not differ between groups for any outcomes. After adjustment for baseline score, there were non-significant differences between groups of -4·1 (95 CI -13·1 to 4·8) in DAMES scores, -8·5 (-20·1 to 3·1) in ABS I scores, and 2·0 (-7·2 to 11·3) in ABS II scores, all in favour of controls. 10 (11) of 88 participants in the memantine group and six (7) of 85 controls had serious adverse events (p=0·33). Five participants in the memantine group and four controls died from serious adverse events (p=0·77). Interpretation: There is a striking absence of evidence about pharmacological treatment of cognitive impairment and dementia in people older than 40 years with Down's syndrome. Despite promising indications, memantine is not an effective treatment. Therapies that are effective for Alzheimer's disease are not necessarily effective in this group of patients. Funding: Lundbeck. © 2012 Elsevier Ltd.
Reyes-Sandoval A.,University of Oxford |
Berthoud T.,University of Oxford |
Alder N.,Center for Statistics in Medicine |
Siani L.,Okairos Inc. |
And 5 more authors.
Infection and Immunity | Year: 2010
Protection against liver-stage malaria relies on the induction of high frequencies of antigen-specific CD8+ T cells. We have previously reported high protective levels against mouse malaria, albeit short-lived, by a single vaccination with adenoviral vectors coding for a liver-stage antigen (ME.TRAP). Here, we report that prime-boost regimens using modified vaccinia virus Ankara (MVA) and adenoviral vectors encoding ME.TRAP can enhance both short- and long-term sterile protection against malaria. Protection persisted for at least 6 months when simian adenoviruses AdCh63 and AdC9 were used as priming vectors. Kinetic analysis showed that the MVA boost made the adenoviral-primed T cells markedly more polyfunctional, with the number of gamma interferon (INF-γ), tumor necrosis factor alpha (TNF-α), and interleukin-2 (IL-2) triple-positive and INF-γ and TNF-α double-positive cells increasing over time, while INF-γ single-positive cells declined with time. However, IFN-γ production prevailed as the main immune correlate of protection, while neither an increase of polyfunctionality nor a high integrated mean fluorescence intensity (iMFI) correlated with protection. These data highlight the ability of optimized viral vector prime-boost regimens to generate more protective and sustained CD8+ T-cell responses, and our results encourage a more nuanced assessment of the importance of inducing polyfunctional CD8+ T cells by vaccination. Copyright © 2010, American Society for Microbiology. All Rights Reserved.
Molyneux A.J.,University of Oxford |
Birks J.,Center for Statistics in Medicine |
Clarke A.,University of Oxford |
Sneade M.,University of Oxford |
Kerr R.S.C.,John Radcliffe Hospital
The Lancet | Year: 2015
Background Previous analyses of the International Subarachnoid Aneurysm Trial (ISAT) cohort have reported on the risks of recurrent subarachnoid haemorrhage and death or dependency for a minimum of 5 years and up to a maximum of 14 years after treatment of a ruptured intracranial aneurysm with either neurosurgical clipping or endovascular coiling. At 1 year there was a 7% absolute and a 24% relative risk reduction of death and dependency in the coiling group compared with the clipping group, but the medium-term results showed the increased need for re-treatment of the target aneurysm in the patients given coiling. We report the long-term follow-up of patients in this UK cohort. Methods In ISAT, patients were randomly allocated to either neurosurgical clipping or endovascular coiling after a subarachnoid haemorrhage, assuming treatment equipoise, between Sept 12, 1994, and May 1, 2002. We followed up 1644 patients in 22 UK neurosurgical centres for death and clinical outcomes for 10·0-18·5 years. We assessed dependency as self-reported modified Rankin scale score obtained through yearly questionnaires. Data for recurrent aneurysms and rebleeding events were collected from questionnaires and from hospital and general practitioner records. The Office for National Statistics supplied data on deaths. This study is registered, number ISRCTN49866681. Findings At 10 years, 674 (83%) of 809 patients allocated endovascular coiling and 657 (79%) of 835 patients allocated neurosurgical clipping were alive (odds ratio [OR] 1·35, 95% CI 1·06-1·73). Of 1003 individuals who returned a questionnaire at 10 years, 435 (82%) patients treated with endovascular coiling and 370 (78%) patients treated with neurosurgical clipping were independent (modified Rankin scale score 0-2; OR 1·25; 95% CI 0·92-1·71). Patients in the endovascular treatment group were more likely to be alive and independent at 10 years than were patients in the neurosurgery group (OR 1·34, 95% CI 1·07-1·67). 33 patients had a recurrent subarachnoid haemorrhage more than 1 year after their initial haemorrhage (17 from the target aneurysm). Interpretation Although rates of increased dependency alone did not differ between groups, the probability of death or dependency was significantly greater in the neurosurgical group than in the endovascular group. Rebleeding was more likely after endovascular coiling than after neurosurgical clipping, but the risk was small and the probability of disability-free survival was significantly greater in the endovascular group than in the neurosurgical group at 10 years. Funding UK Medical Research Council. © 2015 Elsevier Ltd.
Immunogenicity and tolerability of recombinant serogroup B meningococcal vaccine administered with or without routine infant vaccinations according to different immunization schedules: A randomized controlled trial
Gossger N.,University of Oxford |
Snape M.D.,University of Oxford |
Yu L.-M.,Center for Statistics in Medicine |
Finn A.,University of Bristol |
And 13 more authors.
JAMA - Journal of the American Medical Association | Year: 2012
Context: In the absence of an effective vaccine, serogroup B Neisseria meningitidis (MenB) remains a major cause of invasive disease in early childhood in developed countries. Objective: To determine the immunogenicity and reactogenicity of a multicomponent MenB vaccine (4CMenB) and routine infant vaccines when given either concomitantly or separately. Design, Setting, and Participants: Phase 2b, multicenter, open-label, parallel-group, randomized controlled study of 1885 infants enrolled at age 2 months from August 2008 to July 2010 in Europe. Intervention: Participants were randomized 2:2:1:1 to receive (1) 4CMenB at 2, 4, and 6 months with routine vaccines (7-valent pneumococcal and combined diphtheria, tetanus, acellular pertussis, inactivated polio, hepatitis B, Haemophilus influenzae type b vaccines); (2) 4CMenB at 2, 4, and 6 months and routine vaccines at 3, 5, and 7 months; (3) 4CMenB with routine vaccines at 2, 3, and 4 months; or (4) routine vaccines alone at 2, 3, and 4 months. Main Outcome Measures: Percentage of participants with human complement serum bactericidal activity (hSBA) titer of 1:5 or greater against 3 MenB strains specific for vaccine antigens (NZ98/254, 44/76-SL, and 5/99). Results: After three 4CMenB vaccinations, 99% or more of infants developed hSBA titers of 1:5 or greater against strains 44/76-SL and 5/99. For NZ98/254, this proportion was 79% (95% CI, 75.2%-82.4%) for vaccination at 2, 4, and 6 months with routine vaccines, 86.1% (95% CI, 82.9%-89.0%) for vaccination at 2, 4, and 6 months without routine vaccines, and 81.7% (95% CI, 76.6%-86.2%) for vaccination at 2, 3, and 4 months with routine vaccines. Responses to routine vaccines given with 4CMenB were noninferior to routine vaccines alone for all antigens, except for the responses to pertactin and serotype 6B pneumococcal polysaccharide. Fever was seen following 26% (158/602) to 41% (247/607) of 4CMenB doses when administered alone, compared with 23% (69/304) to 36% (109/306) after routine vaccines given alone and 51% (306/605) to 61% (380/624) after 4CMenB and routine vaccines administered together. Conclusion: A 4CMenB vaccine is immunogenic against reference strains when administered with routine vaccines at 2, 4, and 6 or at 2, 3, and 4 months of age, producing minimal interference with the response to routine infant vaccinations. Trial Registration: clinicaltrials.gov Identifier: NCT00721396. ©2012 American Medical Association. All rights reserved.
Copson E.,University of Southampton |
Eccles B.,University of Southampton |
Maishman T.,University of Southampton |
Gerty S.,University of Southampton |
And 9 more authors.
Journal of the National Cancer Institute | Year: 2013
Background Breast cancer at a young age is associated with poor prognosis. The Prospective Study of Outcomes in Sporadic and Hereditary Breast Cancer (POSH) was designed to investigate factors affecting prognosis in this patient group. Methods Between 2000 and 2008, 2956 patients aged 40 years or younger were recruited to a UK multicenter prospective observational cohort study (POSH). Details of tumor pathology, disease stage, treatment received, and outcome were recorded. Overall survival (OS) and distant disease-free interval (DDFI) were assessed using Kaplan-Meier curves. All statistical tests were two-sided. Results Median age of patients was 36 years. Median tumor diameter was 22mm, and 50% of patients had positive lymph nodes; 59% of tumors were grade 3, 33.7% were estrogen receptor (ER) negative, and 24% were human epidermal growth factor receptor 2 (HER2) positive. Five-year OS was higher for patients with ER-positive than ER-negative tumors (85.0%, 95% confidence interval [CI] = 83.2% to 86.7% vs 75.7%, 95% CI = 72.8% to 78.4%; P <. 001), but by eight years, survival was almost equal. The eight-year OS of patients with ER-positive tumors was similar to that of patients with ER-negative tumors in both HER2-positive and HER2-negative subgroups. The flexible parametric survival model for OS shows that the risk of death increases steadily over time for patients with ER-positive tumors in contrast to patients with ER-negative tumors, where risk of death peaked at two years. Conclusions These results confirm the increased frequency of ER-negative tumors and early relapse in young patients and also demonstrate the equally poor longer-term outlook of young patients who have ER-positive tumors with HER2-negative or -positive disease. © 2013 © Crown copyright 2013.
Simera I.,Center for Statistics in Medicine
Journal of the Pakistan Medical Association | Year: 2013
Substantial evidence demonstrates widespread shortcomings in the reporting of health research publications. Reporting guidelines represent an effective tool to help improve the completeness and transparency of published papers that are much needed for their future use. Examples of key reporting guidelines (CONSORT, STROBE, COREQ, ENTREQ, PRISMA, STARD, and SQUIRE) are introduced here together with other resources supporting the writing of high quality research publications that are provided by the EQUATOR Network (www.equator-network.org).
Dechartres A.,French Institute of Health and Medical Research |
Dechartres A.,Center dEpidemiologie Clinique |
Dechartres A.,University of Paris Descartes |
Altman D.G.,Center for Statistics in Medicine |
And 11 more authors.
JAMA - Journal of the American Medical Association | Year: 2014
IMPORTANCE: A persistent dilemma when performing meta-analyses is whether all available trials should be included in the meta-analysis. OBJECTIVES: To compare treatment outcomes estimated by meta-analysis of all trials and several alternative analytic strategies: single most precise trial (ie, trial with the narrowest confidence interval), meta-analysis restricted to the 25% largest trials, limit meta-analysis (a meta-analysis model adjusted for small-study effect), and meta-analysis restricted to trials at low overall risk of bias. DATA SOURCES: One hundred sixty-three meta-analyses published between 2008 and 2010 in high-impact-factor journals and between 2011 and 2013 in the Cochrane Database of Systematic Reviews: 92 (705 randomized clinical trials [RCTs]) with subjective outcomes and 71 (535 RCTs) with objective outcomes. DATA SYNTHESIS: For each meta-analysis, the difference in treatment outcomes between meta-analysis of all trials and each alternative strategy, expressed as a ratio of odds ratios (ROR), was assessed considering the dependency between strategies. A difference greater than 30% was considered substantial. RORs were combined by random-effects meta-analysis models to obtain an average difference across the sample. An ROR greater than 1 indicates larger treatment outcomes with meta-analysis of all trials. Subjective and objective outcomes were analyzed separately. RESULTS: Treatment outcomes were larger in the meta-analysis of all trials than in the single most precise trial (combined ROR, 1.13 [95%CI, 1.07-1.19]) for subjective outcomes and 1.03 (95%CI, 1.01-1.05) for objective outcomes). The difference in treatment outcomes between these strategies was substantial in 47 of 92 (51%) meta-analyses of subjective outcomes (meta-analysis of all trials showing larger outcomes in 40/47) and in 28 of 71 (39%) meta-analyses of objective outcomes (meta-analysis of all trials showing larger outcomes in 21/28). The combined ROR for subjective and objective outcomes was, respectively, 1.08 (95%CI, 1.04-1.13) and 1.03 (95%CI, 1.00-1.06) when comparing meta-analysis of all trials and meta-analysis of the 25% largest trials, 1.17 (95%CI, 1.11-1.22) and 1.13 (95%CI, 0.82-1.55) when comparing meta-analysis of all trials and limit meta-analysis, and 0.94 (95%CI, 0.86-1.04) and 1.03 (95%CI, 1.00-1.06) when comparing meta-analysis of all trials and meta-analysis restricted to trials at low risk of bias. CONCLUSIONS AND RELEVANCE: Estimation of treatment outcomes in meta-analyses differs depending on the strategy used. This instability in findings can result in major alterations in the conclusions derived from the analysis and underlines the need for systematic sensitivity analyses.
Dekkers O.M.,Leiden University |
Egger M.,University of Bern |
Altman D.G.,Center for Statistics in Medicine |
Vandenbroucke J.P.,Leiden University
Annals of Internal Medicine | Year: 2012
Case series are a commonly reported study design, but the label "case series" is used inconsistently and sometimes incorrectly. Mislabeling impairs the appropriate indexing and sorting of evidence. This article tries to clarify the concept of case series and proposes a way to distinguish them from cohort studies. In a cohort study, patients are sampled on the basis of exposure and are followed over time, and the occurrence of outcomes is assessed. A cohort study may include a comparison group, although this is not a necessary feature. A case series may be a study that samples patients with both a specific outcome and a specific exposure, or one that samples patients with a specific outcome and includes patients regardless of whether they have specific exposures. Whereas a cohort study, in principle, enables the calculation of an absolute risk or a rate for the outcome, such a calculation is not possible in a case series. © 2012 American College of Physicians.
Royston P.,MRC Clinical Trials Unit |
Altman D.G.,Center for Statistics in Medicine
Statistics in Medicine | Year: 2010
Logistic regression models are widely used in medicine for predicting patient outcome (prognosis) and constructing diagnostic tests (diagnosis). Multivariable logistic models yield an (approximately) continuous risk score, a transformation of which gives the estimated event probability for an individual. A key aspect of model performance is discrimination, that is, the model's ability to distinguish between patients who have (or will have) an event of interest and those who do not (or will not). Graphical aids are important in understanding a logistic model. The receiver-operating characteristic (ROC) curve is familiar, but not necessarily easy to interpret. We advocate a simple graphic that provides further insight into discrimination, namely a histogram or dot plot of the risk score in the outcome groups. The most popular performance measure for the logistic model is the c-index, numerically equivalent to the area under the ROC curve. We discuss the comparative merits of the c-index and the (standardized) mean difference in risk score between the outcome groups. The latter statistic, sometimes known generically as the effect size, has been computed in slightly different ways by several different authors, including Glass, Cohen and Hedges. An alternative measure is the overlap between the distributions in the outcome groups, defined as the area under the minimum of the two density functions. The larger the overlap, the weaker the discrimination. Under certain assumptions about the distribution of the risk score, the c-index, effect size and overlap are functionally related. We illustrate the ideas with simulated and real data sets. Copyright © 2010 John Wiley & Sons, Ltd.