Antonio Gasparrini: Open access: a researcher's perspective

  • View
    319

  • Download
    1

Embed Size (px)

Transcript

  • 1.Open access: a researchers perspectiveAntonio GasparriniLondon School of Hygiene and Tropical Medicine, London, UK Open Research and DataOpen Access Week 22 October 2012 - Birbeck College, London

2. My backgroundGraduated in biology in Italy, then 4 years working asepidemiologist in a cancer research center in FlorenceMSc + postgraduate school (still in Italy) + PhD (in UK) inmedical statistics,Worked at LSHTM in the last 5 years, mainly in statisticalmethodology and software development 3. My MRC fellowshipAwarded a Research Methodology fellowship from MRC (Dec2011 Nov 2014)Project developed on my previous researchSuccess of this project critical for next funding applicationNeed to comply with the MRC regulations on open accessMy budget for open access costs: 6000 in total 4. OutlineSome points:My perspective: as scientist and junior academicPublishing: steps and costsMy publications as a case studyOpen research: beyond publications 5. The scientists perspectiveI favour a system which:guarantees high-quality researchallows the independent assessment of research ndingsensures the dissemination of the such ndings 6. The academics perspectiveI favour a system which:covers the costs of my researchdelivers a fast and eective peer-review processprovides tools for disseminating my work 7. Publishing a research paper: stepsLiterature reviewDrafting the manuscriptChoice of the journal and submissionReview and acceptanceCopyright agreementOpen access feePublicationActors: the researcher, the institution, the research community,the funder, the journals/publishersAn ecient and fair system? 8. A rst articlePublished online in Statistics in Medicine (2012):The choice of the journalCopyright transferredOpen access fee 2250Impact factor 1.99 9. Submitted and published versions StatisticsResearch ArticleResearch Article in Medicine Received 9 August 2011, Accepted 11 May 2012 Published online in Wiley Online LibraryReceived XXXX (wileyonlinelibrary.com) DOI: 10.1002/sim.5471(www.interscience.wiley.com) DOI: 10.1002/sim.0000Multivariate meta-analysis for non-linear andMultivariate meta-analysis forother multi-parameter associations non-linear and other multi-parameter associationsA. Gasparrinia , B. Armstrongb , M. G. Kenwarda A. Gasparrini,a * B. Armstrongb and M. G. KenwardaIn this paper we formalize the application of multivariate meta-analysis and meta-regression to synthesize estimatesof multi-parameter associations obtained from different studies. This modelling approach extends the standardIn this paper, we formalize the application of multivariate meta-analysis and meta-regression to synthesizetwo-stage analysis used to combine results across different sub-groups or populations. The most straightforwardestimates of multi-parameter associations obtained from different studies. This modelling approach extends the standard two-stage analysis used to combine results across different sub-groups or populations. The mostapplication is for the meta-analysis of non-linear relationships, described for example by regression coefcientsstraightforward application is for the meta-analysis of non-linear relationships, described for example byof splines or other functions, but the methodology easily generalizes to any setting where complex associationsregression coefcients of splines or other functions, but the methodology easily generalizes to any settingare described by multiple correlated parameters. The modelling framework of multivariate meta-analysis iswhere complex associations are described by multiple correlated parameters. The modelling framework ofimplemented in the package mvmeta within the statistical environment R. As an illustrative example, we propose multivariate meta-analysis is implemented in the package mvmeta within the statistical environment R. Asa two-stage analysis for investigating the non-linear exposure-response relationship between temperature and non-an illustrative example, we propose a two-stage analysis for investigating the non-linear exposureresponseaccidental mortality using time series data from multiple cities. Multivariate meta-analysis represents a useful relationship between temperature and non-accidental mortality using time-series data from multiple cities.analytical tool for studying complex associations through a two-stage procedure. Copyright c 2011 John Wiley & Multivariate meta-analysis represents a useful analytical tool for studying complex associations through a two-stage procedure. Copyright 2012 John Wiley & Sons, Ltd.Sons, Ltd. Keywords: meta-analysis; multivariate analysis; multivariate meta-analysis; non-linear; splinesKeywords: meta-analysis; multivariate analysis; multivariate meta-analysis; non-linear; splines 1. Introduction1. Introduction Meta-analysis is a standard, well-grounded statistical procedure for combining the evidence from inde-Meta-analysis is a standard, well-grounded statistical procedure for combining the evidence from independent studies pendent studies that address the same research hypothesis [1]. This methodology was developed origi-that address the same research hypothesis [1]. This methodology was developed originally for pooling the results fromnally for pooling the results from published observational or experimental studies, for which individualpublished observational or experimental studies, for which individual data were not available. Recently, meta-analysis has data were not available. Recently, meta-analysis has been described more broadly as a research synthesisbeen described more broadly as a research synthesis method, with the aim of estimating an average association across method, with the aim of estimating an average association across studies and to explore the degree andstudies and to explore the degree and sources of heterogeneity [2]. The analytical approach adopted in this context may be sources of heterogeneity [2]. The analytical approach adopted in this context may be described as adescribed as a two-stage hierarchical procedure: in the rst stage, study-specic estimates of the association of interest are two-stage hierarchical procedure: in the rst stage, study-specic estimates of the association of interestderived from individual data, controlling for individual-level covariates; in the second stage, these estimates are combined are derived from individual data, controlling for individual-level covariates; in the second stage, theseacross studies, optionally exploring the association with study-level predictors. The two-stage approach, a specic form estimates are combined across studies, optionally exploring the association with study-level predictors.of individual patient data (IPD) meta-analysis, has been shown to be a exible and computationally efcient method [3], The two-stage approach, a specic form of individual patient data (IPD) meta-analysis, has been shownand has been adopted in different contexts: to pool estimates from multiple randomized controlled trials [4]; to combineresults from survival models on time-to-event data in multi-centre cohorts [5]; and to synthesize associations from Poissonto be a exible and computationally efcient method [3] and has been adopted in different contexts: totime series models in multi-city analyses [6]. pool estimates from multiple randomized controlled trials [4]; to combine results from survival models The common approach to two-stage meta-analysis consists of summarizing the association in a single parameteron time-to-event data in multi-centre cohorts [5]; and to synthesize associations from Poisson time-seriesestimate from the rst stage, optionally controlling for individual-level confounders. This procedure allows standard meta-models in multi-city analyses [6].analytic techniques to be applied. However, complex associations, such as non-linear exposure-responses, are usuallyThe common approach to two-stage meta-analysis consists of summarizing the association in a singledescribed with functions dened by multiple parameters, and require more sophisticated meta-analytical approaches, parameter estimate from the rst stage, optionally controlling for individual-level confounders. Thiscapable of handling the multivariate nature of the summary estimates. Multivariate meta-analysis, a method originallyprocedure allows standard meta-analytic techniques to be applied. However, complex associations, such as non-linear exposureresponses, are usually described with functions dened by multiple parameters and require more sophisticated meta-analytical approaches capable of handling the multivariate nature of the summary estimates. Multivariate meta-analysis, a method originally developed to pool multiplea Department of Medical Statistics, London School of Hygiene and Tropical MedicinebDepartment of Social and Environmental Health Research, London School of Hygiene and Tropical MedicineCorrespondence to: Antonio Gasparrini, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK.E-mail: antonio.gasparrini@lshtm.ac.uk a Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, U.K. b Department of Social and Environmental Health Research, London School of Hygiene and Tropical Medicine, London, U.K.Contract/grant sponsor: Medical Research Council (UK), grants G0701030 and G1002296 *Correspondence to: Antonio Gasparrini, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, U.K. E-mail: antonio.gasparrini@lshtm.ac.ukStatist. Med. 2011, 00 118Copyright c 2011 John Wiley & Sons, Ltd.Prepared using simauth.cls [Version: 2010/03/10 v3.00] Copyright 2012 John Wiley & Sons, Ltd.Statist. Med. 2012 10. A second articleFirstly submitted to Biostatistics:Copyright transferredOpen access fee 2250Impact factor 2.145Rejected, re-submitted to BMC Med Res Method:Copyright retainedOpen access fee 1475 (1255 with LSHTM discount)Impact factor 2.67 11. A third articlePublished in Journal