ࡱ > / 1 ! " # $ % & ' ( ) * + , - . '` A bjbjLULU 2 .? .? o % G G G 8 G | PM U ` " ` ` ` b b b 7 9 9 9 9 9 9 $ H h ] | b b | | ] ` ` n | ` ` | 7 + ` T @5 G 9 ( 0 ɘ $ n m b k N r w b b b ] ] U X b b b | | | | t x $[ t x Strengthening the Reporting of Genetic Risk Prediction Studies (GRIPS): Explanation and Elaboration
A. Cecile J.W. Janssens1,*, John P.A. Ioannidis2,3,4,5,6, Sara Bedrosian7, Paolo Boffetta8,9, Siobhan M. Dolan10, Nicole Dowling7, Isabel Fortier11, Andrew N. Freedman12, Jeremy M. Grimshaw13,14, Jeffrey Gulcher15, Marta Gwinn7, Mark A. Hlatky16, Holly Janes17, Peter Kraft12, Stephanie Melillo7, Christopher J. ODonnell19,20, Michael J. Pencina21, David Ransohoff22, Sheri D. Schully12, Daniela Seminara12, Deborah M. Winn12, Caroline F. Wright23, Cornelia M. van Duijn1, Julian Little24, Muin J. Khoury7
1 Department of Epidemiology, Erasmus University Medical Center, Rotterdam, The Netherlands.
2 Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece.
3 Biomedical Research Institute, Foundation for Research and Technology, Ioannina, Greece.
4 Department of Medicine, Tufts University School of Medicine, Boston MA, USA.
5 Center for Genetic Epidemiology and Modeling and Tufts CTSI, Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston MA, USA.
6 Department of Epidemiology, Harvard School of Public Health, Boston MA, USA.
7 Office of Public Health Genomics, Centers for Disease Control and Prevention, Atlanta GA, USA.
8 The Tisch Cancer Institute, Mount Sinai School of Medicine, New York NY, USA.
9 International Prevention Research Institute, Lyon, France.
10 Department of Obstetrics & Gynecology and Womens Health, Albert Einstein College of Medicine / Montefiore Medical Center, Bronx NY, USA.
11 Public Population Project in Genomics (P3G), Montreal, Quebec, Canada.
12 Division of Cancer Control and Population Sciences, National Cancer Institute, National Institutes of Health, Bethesda MD, USA.
13 Clinical Epidemiology Program, Ottawa Hospital Research Institute, Ottawa ON, Canada;
14 Department of Medicine, University of Ottawa, Ottawa ON, Canada.
15 deCODE Genetics, Reykjavik, Iceland.
16 Department of Health Research and Policy, Stanford University, Palo Alto CA, USA.
17 Fred Hutchinson Cancer Research Center, Vaccine and Infectious Disease Institute and Division of Public Health Sciences, Seattle WA, USA.
18 Department of Epidemiology, Harvard School of Public Health, Boston MA, USA.
19 National Heart, Lung and Blood Institute (NHLBI) and the NHLBI's Framingham Heart Study, Framingham MA, USA;
20 Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Boston MA, USA.
21 Department of Biostatistics, Boston University, Boston MA, USA; Harvard Clinical Research Institute, Boston MA, USA.
22 University of North Carolina at Chapel Hill School of Medicine, Chapel Hill NC, USA.
23 PHG Foundation, Cambridge, United Kingdom.
24 Department of Epidemiology and Community Medicine, University of Ottawa, Ottawa ON, Canada.
* Corresponding author
A. Cecile J.W. Janssens, Erasmus University Medical Center, Department of Epidemiology, PO Box 2040, 3000 CA Rotterdam, the Netherlands. Email: a.janssens@erasmusmc.nl; Telephone: +31-10-7044214; Fax: +31-10-7044657.
Running head: GRIPS statement: Explanation & ElaborationSummary Points
The rapid and continuing progress in gene discovery for complex diseases is fuelling interest in the potential application of genetic risk models for clinical and public health practice.
The number of studies assessing the predictive ability is steadily increasing, but they vary widely in completeness of reporting and apparent quality.
Transparent reporting of the strengths and weaknesses of these studies is important to facilitate the accumulation of evidence on genetic risk prediction.
A multidisciplinary workshop sponsored by the Human Genome Epidemiology Network developed a checklist of 25 items recommended for strengthening the reporting of Genetic RIsk Prediction Studies (GRIPS), building on the principles established by prior reporting guidelines.
These recommendations aim to enhance the transparency, quality and completeness of study reporting, and thereby to improve the synthesis and application of information from multiple studies that might differ in design, conduct or analysis.
Introduction
The advent of genome-wide association studies has accelerated the discovery of novel genetic markers, in particular single nucleotide polymorphisms (SNPs) that are associated with risk for common complex diseases. Technological developments in large-scale genomic studies, such as whole genome sequencing, will facilitate the discovery of novel of common SNPs, as well as of rare variants, copy number variations, deletions/insertions, structural variations (e.g., inversions), and epigenetic effects that influence the regulation of gene expression. These developments are fuelling interest in the translation of this basic knowledge to health care practice. Knowledge about genetic risk factors may be used to target diagnostic, preventive and therapeutic interventions for complex disorders based on a persons genetic risk, or to complement existing risk models based on classical non-genetic factors such as the Framingham risk score for cardiovascular disease. Implementation of genetic risk prediction in health care requires a series of studies that encompass all phases of translational research ADDIN REFMGR.CITE Khoury2007KHOURY2007The continuum of translation research in genomic medicine: how can we accelerate the appropriate integration of human genome discoveries into health care and disease prevention?JournalKHOURY2007The continuum of translation research in genomic medicine: how can we accelerate the appropriate integration of human genome discoveries into health care and disease prevention?Khoury,M.J.Gwinn,M.Yoon,P.W.Dowling,N.Moore,C.A.Bradley,L.2007/10Delivery of Health CareDiseaseFeedbackGeneticGenetic ResearchGenomeGenome,HumanGenomicsGuidelinesHealthHumanHumansKnowledgePopulationPreventive MedicinePublic HealthResearchNot in File665674Genet.Med.910National Office of Public Health Genomics Centers for Disease Control and Prevention, Atlanta, Georgia 30341, USA. mkhoury@cdc.govPM:18073579Genet.Med.1Hlatky2009HLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationJournalHLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationHlatky,M.A.Greenland,P.Arnett,D.K.Ballantyne,C.M.Criqui,M.H.Elkind,M.S.Go,A.S.Harrell,F.E.,Jr.Hong,Y.Howard,B.V.Howard,V.J.Hsue,P.Y.Kramer,C.M.McConnell,J.P.Normand,S.L.O'Donnell,C.J.Smith,S.C.,Jr.Wilson,P.W.2009/5/5American Heart AssociationanalysisAssociationBiologicalBiological Markerscardiovascular diseaseCardiovascular DiseasesClinicaldiagnosisDiseaseEvaluation Studies as TopicHumansmethodsPopulationPrognosisResearchResearch DesignRiskRisk AssessmentSensitivity and SpecificitystandardsStatisticalUnited StatesNot in File24082416Circulation11917PM:19364974Circulation1[1,2], starting with a comprehensive evaluation of genetic risk prediction.
Genetic risk prediction studies typically concern the development and/or evaluation of models for the prediction of a particular health outcome, but there is considerable variation in their design, conduct and analysis. Genetic risk models most frequently predict risk of disease, but they are also being investigated for the prediction of prognostic outcome, treatment response or treatment side effects. Risk prediction models are used in research and clinical settings to classify individuals into homogeneous groups e.g., for randomization in clinical trials and for targeting preventive or therapeutic interventions. The main study designs are cohort, cross-sectional or case-control. The genetic risk factors often are SNPs, but other variants such as insertions/deletions, haplotypes and copy number variations can be included as well. The risk models are based on genetic variants only, or include both genetic and non-genetic risk factors. Risk prediction models are statistical algorithms, which can be simple genetic risk scores (e.g., risk allele counts), or be based on regression analyses (e.g., weighted risk scores or predicted risks) or on more complex analytic approaches such as support vector machine learning or classification trees. Papers on genetic risk prediction vary as to whether they present the development of a risk model only, the validation of one or more risk models only, or both development and validation of a risk model ADDIN REFMGR.CITE Janssens2009JANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJournalJANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJanssens,A.C.J.W.Van Duijn,C.M.2009/2/18ClinicalDiseaseepidemiologyHealthNetherlandsPublic HealthUniversitiesNot in File20Genome Med.12Department of Epidemiology, Erasmus University Medical Center Rotterdam, 3000 CA Rotterdam, The Netherlands. a.janssens@erasmusmc.nlPM:19341491Genome Med.1[3]. Lastly, studies vary in the measures used to assess model performance. So far, assessments have nearly always included measures of discrimination, but hardly any considered calibration ADDIN REFMGR.CITE Janssens2009JANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJournalJANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJanssens,A.C.J.W.Van Duijn,C.M.2009/2/18ClinicalDiseaseepidemiologyHealthNetherlandsPublic HealthUniversitiesNot in File20Genome Med.12Department of Epidemiology, Erasmus University Medical Center Rotterdam, 3000 CA Rotterdam, The Netherlands. a.janssens@erasmusmc.nlPM:19341491Genome Med.1[3]. Recent studies have additionally assessed measures of reclassification, despite debate on the appropriate use and interpretation of these measures ADDIN REFMGR.CITE Pencina2008PENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondJournalPENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondPencina,M.J.D'Agostino,R.B.,Sr.D'Agostino,R.B.,Jr.Vasan,R.S.2008/1/30AlgorithmsArea Under CurveAssociationBostoncardiovascular diseaseCardiovascular DiseasesClassificationDiseaseepidemiologyetiologyHumansMathematicsModelsModels,StatisticalPaperResearchResearch SupportRiskRisk AssessmentRisk FactorsRoc CurveSensitivity and SpecificityStatisticsstatistics & numerical dataUnited StatesUniversitiesNot in File157172Stat.Med.272Department of Mathematics and Statistics, Framingham Heart Study, Boston University, Boston, MA 02215, USA. mpencina@bu.eduPM:17569110Statistics in MedicineStat.Med.1Mihaescu2010MIHAESCU2010Improvement of risk prediction by genomic profiling: reclassification measures versus the area under the receiver operating characteristic curveJournalMIHAESCU2010Improvement of risk prediction by genomic profiling: reclassification measures versus the area under the receiver operating characteristic curveMihaescu,R.van Zitteren,M.van Hoek,M.Sijbrands,E.J.Uitterlinden,A.G.Witteman,J.C.Hofman,A.Hunink,M.G.Van Duijn,C.M.Janssens,A.C.2010/8/1Body Mass IndexClinicalDNA FingerprintingepidemiologyHumansMetagenomicsNetherlandsProspective StudiesRiskRisk FactorsRoc Curvestatistics & numerical dataUniversitiesNot in File353361Am J Epidemiol1723Department of Epidemiology, Erasmus MC, University Medical Center, Rotterdam, the NetherlandsPM:20562194Am J Epidemiol1[4,5].
So far most genetic prediction studies have shown that the predictive performance of genetic risk models is poor, with some exceptions such as those for age-related macular degeneration, hypertriglyceridemia and Crohns disease ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1Weersma2009WEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortJournalWEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortWeersma,R.K.Stokkers,P.C.van Bodegraven,A.A.van Hogezand,R.A.Verspaget,H.W.de Jong,D.J.van der Woude,C.J.Oldenburg,B.Linskens,R.K.Festen,E.A.van der,Steege G.Hommes,D.W.Crusius,J.B.Wijmenga,C.Nolte,I.M.Dijkstra,G.2009/3AdultAge of OnsetAllelesanalysisAssociationColitis,UlcerativeCrohn DiseaseDiseaseDisease SusceptibilityepidemiologyFemaleGastroenterologyGene Expression RegulationGenesGeneticGenetic Predisposition to DiseasegeneticsGenotypeHospitalsHumanHumansMaleMedicalmethodsMolecularMolecular BiologyMulticenter StudiesNetherlandsNod2 Signaling Adaptor ProteinOdds RatioPatientsPhenotypePolymorphismPolymorphism,GeneticReceptorsReceptors,InterleukinRegression AnalysisResearchResearch SupportRiskRisk AssessmentSingle NucleotideUniversitiesNot in File388395Gut583Department of Gastroenterology and Hepatology, University Medical Center Groningen and University of Groningen, P.O. Box 30001, 9700 RB Groningen, The Netherlands. R.K.Weersma@int.umcg.nlPM:18824555Gut1Wang2008WANG2008Polygenic determinants of severe hypertriglyceridemiaJournalWANG2008Polygenic determinants of severe hypertriglyceridemiaWang,J.Ban,M.R.Zou,G.Y.Cao,H.Lin,T.Kennedy,B.A.Anand,S.Yusuf,S.Huff,M.W.Pollex,R.L.Hegele,R.A.2008/7/1AllelesAssociationCanadaDiseaseFastingGenesGeneticGenotypehadLondonMutationObesityOdds RatioOntarioPatientsResearchUniversitiesNot in File28942899Hum.Mol.Genet.1718Vascular Biology Research Group and Robarts Research Institute and Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada N6A 5K8PM:18596051Human Molecular GeneticsHum.Mol.Genet.1[68]. While the poor performance is most likely due to the low number of variants that have been definitely linked to a phenotype to date, many publications lack sufficient details to judge methodological or analytic aspects. Information that is often missing includes details in the description of how the study was designed and conducted (e.g., how genetic variants were selected, how risk models or genetic risk scores were constructed and how risk categories were chosen), or how the results should be interpreted. An appropriate assessment of the studys strengths and weaknesses is not possible without this information. With increasing numbers of discovered genetic markers that can be used in future genetic risk prediction studies, it is crucial to enhance the quality of the reporting of these studies, since valid interpretation could be compromised by the lack of reporting of key information. There is ample evidence that prediction research often suffers from poor design and biases, and these might have an impact also on the results of the studies and on models of disease outcomes based on these studies ADDIN REFMGR.CITE Kyzas2007KYZAS2007AQuality of reporting of cancer prognostic marker studies: association with reported prognostic effectJournalKYZAS2007AQuality of reporting of cancer prognostic marker studies: association with reported prognostic effectKyzas,P.A.Denaxa-Kyza,D.Ioannidis,J.P.2007/2/7AssociationClinicalConfounding Factors (Epidemiology)diagnosisepidemiologyGreecehadHumansMedicineMedlineMeta-Analysis as TopicmethodsMolecularMolecular EpidemiologyNeoplasmsOdds RatioPredictive Value of TestsPrognosisResearch DesignRiskRisk AssessmentstandardsStatisticalTimeUniversitiesNot in File236243J.Natl.Cancer Inst.993Clinical and Molecular Epidemiology Unit, Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina 45110, GreecePM:17284718JNCI Journal of the National Cancer InstituteJ.Natl.Cancer Inst.1Kyzas2005KYZAS2005Selective reporting biases in cancer prognostic factor studiesJournalKYZAS2005Selective reporting biases in cancer prognostic factor studiesKyzas,P.A.Loizou,K.T.Ioannidis,J.P.2005/7/20AgedanalysisAssociationBiologicalCarcinoma,Squamous CellchemistryConfidence IntervalsEpidemiologic Research DesignepidemiologyFemaleGreeceHead and Neck NeoplasmsHumansLymphatic MetastasisMaleMedicineMeta-AnalysismethodsMiddle AgedMortalityOdds RatiopathologyPatientsPredictive Value of TestsPrognosisProtein p53Publication BiasResearch DesignRiskRisk AssessmentRisk FactorsStatisticalSurvival RateTumor Markers,BiologicalTumor Suppressor Protein p53United StatesUniversitiesNot in File10431055J.Natl.Cancer Inst.9714Department of Hygiene and Epidemiology, School of Medicine, University of Ioannina, Ioannina, GreecePM:16030302JNCI Journal of the National Cancer InstituteJ.Natl.Cancer Inst.1McShane2005MCSHANE2005REporting recommendations for tumor MARKer prognostic studies (REMARK)JournalMCSHANE2005REporting recommendations for tumor MARKer prognostic studies (REMARK)McShane,L.M.Altman,D.G.Sauerbrei,W.Taube,S.E.Gion,M.Clark,G.M.2005/8analysisBiologicalBiomedical ResearchdiagnosisGuidelinesHumansInformation DisseminationmethodsNeoplasmsResearchResearch DesignResearch SupportstandardsStatisticalTumor Markers,BiologicalNot in File416422Nat.Clin.Pract.Urol.28National Cancer Institute, Biometric Research Branch, Bethesda, MD 20892-7434, USA. Lm5h@nih.govPM:16482653Nat.Clin.Pract.Urol.1[911]. Although most prognostic studies published to date claim significant results ADDIN REFMGR.CITE Kyzas2007KYZAS2007Almost all articles on cancer prognostic markers report statistically significant resultsJournalKYZAS2007Almost all articles on cancer prognostic markers report statistically significant resultsKyzas,P.A.Denaxa-Kyza,D.Ioannidis,J.P.2007/11BiologicalDatabasesepidemiologyGreecehadHumansJournalism,MedicalMedicineMeta-AnalysismetabolismMortalityNeoplasmsPaperPeriodicals as TopicPrognosisPublication BiasRiskRisk FactorsstandardsStatisticaltrendsTumor Markers,BiologicalUniversitiesNot in File25592579Eur.J.Cancer4317Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, GreecePM:17981458Eur.J.Cancer1Tzoulaki2009TZOULAKI2009Assessment of claims of improved prediction beyond the Framingham risk scoreJournalTZOULAKI2009Assessment of claims of improved prediction beyond the Framingham risk scoreTzoulaki,I.Liberopoulos,G.Ioannidis,J.P.2009/12/2analysisClinicalCoronary DiseaseEnglandepidemiologyhadHealthHealth Status IndicatorsHumansKnowledgeLondonMedicinemethodsModels,StatisticalPopulationPrognosisPublic HealthReproducibility of ResultsRiskRisk AssessmentRisk FactorsNot in File23452352JAMA30221Department of Epidemiology and Public Health, Imperial College of Medicine, London, EnglandPM:19952321JAMA: The Journal of the American Medical AssociationJAMA1[12,13], very few translate to clinically useful applications, in part because study findings resulted from chance, methodological biases or the inclusion of risk factors that had not been previously replicated. Just as for observational epidemiological studies ADDIN REFMGR.CITE von Elm2004VONELM2004The scandal of poor epidemiological researchJournalVONELM2004The scandal of poor epidemiological researchvon Elm,ErikEgger,Matthias2004/10/16ResearchNot in File868869BMJ3297471http://www.bmj.comBritish Medical JournalBMJ1[14], poor reporting complicates the use of the specific study for research, clinical, or public health purposes and the deficiencies also hamper the synthesis of evidence across studies.
Reporting guidelines have been published for various research designs ADDIN REFMGR.CITE Simera2010SIMERA2010A catalogue of reporting guidelines for health researchJournalSIMERA2010A catalogue of reporting guidelines for health researchSimera,I.Moher,D.Hoey,J.Schulz,K.F.Altman,D.G.2010/1EducationGuidelinesHealthPaperResearchResearch SupportUniversitiesNot in File3553Eur.J.Clin.Invest401University of Oxford, Oxford, UKPM:20055895Eur.J.Clin.Invest1[15] and these contain many items that are also relevant to genetic risk prediction studies. In particular, the guidelines for genetic association studies (STREGA) have relevant items on the assessment of genetic variants, and the guidelines for observational studies (STROBE) have relevant items about the reporting of study design. The guidelines for diagnostic studies (STARD) and those for tumor marker prognostic studies (REMARK) include relevant items about test evaluation, and the REMARK guidelines include relevant items about risk prediction ADDIN REFMGR.CITE Von Elm2007VONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesJournalVONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesVon Elm,E.Altman,D.G.Egger,M.Pocock,S.J.Gotzsche,P.C.Vandenbroucke,J.P.2007/10/16Biomedical ResearchCase-Control StudiesCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologymethodsObservationPreventive MedicinePublishingResearchstandardsUniversitiesNot in Filee296PLoS.Med.410Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland. strobe@ispm.unibe.chPM:17941714PLoS.Med.1Little2009LITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementJournalLITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementLittle,J.Higgins,J.P.Ioannidis,J.P.Moher,D.Gagnon,F.von Elm,E.Khoury,M.J.Cohen,B.Davey-Smith,G.Grimshaw,J.Scheet,P.Gwinn,M.Williamson,R.E.Zou,G.Y.Hutchings,K.Johnson,C.Y.Tait,V.Wiens,M.Golding,J.Van Duijn,C.McLaughlin,J.Paterson,A.Wells,G.Fortier,I.Freedman,M.Zecevic,M.King,R.Infante-Rivard,C.Stewart,A.Birkett,N.2009/2/3analysisAssociationCanadaDiseaseepidemiologyGenesGeneticGenetic Predisposition to DiseasegeneticsGenomeGenomicsGuidelines as TopicHealthHumanHumansMedicinemethodsOntarioPeriodicals as TopicPopulationPublic HealthResearchResearch SupportstandardsStatisticalUniversitiesNot in Filee22PLoS.Med.62Canada Research Chair in Human Genome Epidemiology, University of Ottawa, Ottawa, Ontario, Canada. jlittle@uottawa.caPM:19192942PLoS.Med.1Bossuyt2003BOSSUYT2003MTowards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiativeJournalBOSSUYT2003MTowards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiativeBossuyt,P.M.Reitsma,J.B.Bruns,D.E.Gatsonis,C.A.Glasziou,P.P.Irwig,L.M.Lijmer,J.G.Moher,D.Rennie,D.de Vet,H.C.2003/1/4AlgorithmsBias (Epidemiology)ClinicalClinical Trials as TopicConsensusDiagnostic Techniques and ProceduresepidemiologyGuidelinesGuidelines as TopicMedicalmethodsNetherlandsPatientsPublishingResearchResearch DesignResearch SupportstandardsUniversitiesNot in File4144BMJ3267379Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, University of Amsterdam, PO Box 22700, 1100 DE Amsterdam, Netherlands. stard@amc.uva.nlPM:12511463British Medical JournalBMJ1McShane2005MCSHANE2005AReporting recommendations for tumor marker prognostic studiesJournalMCSHANE2005AReporting recommendations for tumor marker prognostic studiesMcShane,L.M.Altman,D.G.Sauerbrei,W.Taube,S.E.Gion,M.Clark,G.M.2005/12/20analysisBiologicalBiomedical ResearchdiagnosisHumansInformation DisseminationNeoplasmsResearchResearch DesignResearch SupportstandardsTumor Markers,BiologicalNot in File90679072J.Clin.Oncol.2336Biometric Research Branch, National Cancer Institute, Bethesda, MD, USAPM:16172462J.Clin.Oncol.1[1619]. However, none of these guidelines are fully suited to genetic risk prediction studies, an emerging field of investigations with specific methodological issues that need to be addressed, such as the handling of large numbers of genetic variants (from 10s to 10000s), which come with greater challenges and flexibility on how these can be dealt with in the analyses.
The main goal of this paper is to propose and justify a set of guiding principles for reporting results of Genetic RIsk Prediction Studies (GRIPS). To minimize confusion in the field, these recommendations build on prior reporting guidelines whenever possible. The intended audience for the reporting guideline is broad and includes epidemiologists, geneticists, statisticians, clinician scientists and laboratory-based investigators who undertake genetic risk prediction studies, as well as journal editors and reviewers who have to appraise the design, conduct and analysis of such studies. In addition, it includes 'users' of such studies who wish to understand the basic premise, design, and limitations of genetic prediction studies in order to interpret the results for their potential application in health care. These guidelines are also intended to ensure that essential data from genetic risk prediction studies are presented, which will facilitate information synthesis as part of systematic reviews and meta-analyses.
Finally, it is important to emphasize that these recommendations are guidelines only for how to report research; the recommendations do not prescribe how to perform genetic risk prediction studies. Nevertheless, we suggest that increased transparency of reporting might have a favorable effect on the quality of research, and thereby improve the translation into practice, as has been the case for the adoption of the CONSORT checklist in the reporting of randomized controlled trials ADDIN REFMGR.CITE Plint2006PLINT2006Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic reviewJournalPLINT2006Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic reviewPlint,A.C.Moher,D.Morrison,A.Schulz,K.Altman,D.G.Hill,C.Gaboury,I.2006/9/4AdoptionCanadaEditorial PolicieshadHumansMedlineOntarioPeriodicals as TopicPublishingQuality ControlRandomized Controlled Trials as TopicRiskstandardsUniversitiesNot in File263267Med.J.Aust.1855Department of Pediatrics, University of Ottawa, Ottawa, Ontario, Canada. plint@cheo.on.caPM:16948622Medical Journal of AustraliaMed.J.Aust.1[20].
Development of the GRIPS Statement
The GRIPS Statement was developed by a multidisciplinary panel of 25 risk prediction researchers, epidemiologists, geneticists, methodologists, statisticians and journal editors, seven of whom were also part of the STREGA initiative ADDIN REFMGR.CITE Little2009LITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementJournalLITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementLittle,J.Higgins,J.P.Ioannidis,J.P.Moher,D.Gagnon,F.von Elm,E.Khoury,M.J.Cohen,B.Davey-Smith,G.Grimshaw,J.Scheet,P.Gwinn,M.Williamson,R.E.Zou,G.Y.Hutchings,K.Johnson,C.Y.Tait,V.Wiens,M.Golding,J.Van Duijn,C.McLaughlin,J.Paterson,A.Wells,G.Fortier,I.Freedman,M.Zecevic,M.King,R.Infante-Rivard,C.Stewart,A.Birkett,N.2009/2/3analysisAssociationCanadaDiseaseepidemiologyGenesGeneticGenetic Predisposition to DiseasegeneticsGenomeGenomicsGuidelines as TopicHealthHumanHumansMedicinemethodsOntarioPeriodicals as TopicPopulationPublic HealthResearchResearch SupportstandardsStatisticalUniversitiesNot in Filee22PLoS.Med.62Canada Research Chair in Human Genome Epidemiology, University of Ottawa, Ottawa, Ontario, Canada. jlittle@uottawa.caPM:19192942PLoS.Med.1[17]. They attended a two-day meeting in Atlanta, GA, USA, in December 2009 sponsored by the Centers for Disease Control and Prevention on behalf of the Human Genome Epidemiology Network (HuGENet) ADDIN REFMGR.CITE Khoury1998KHOURY1998AThe Human Genome Epidemiology NetworkJournalKHOURY1998AThe Human Genome Epidemiology NetworkKhoury,M.J.Dorman,J.S.1998/7/1Computer Communication NetworksDatabases as TopicepidemiologyGenomeGenome,HumanHumanHuman Genome ProjectHumansInformation ServicesNot in File13Am J Epidemiol1481PM:9663396American Journal of EpidemiologyAm J Epidemiol1[21]. Participants discussed a draft version of the checklist that was prepared and distributed prior to the meeting. This draft version was developed based on existing reporting guidelines, namely STREGA ADDIN REFMGR.CITE Little2009LITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementJournalLITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementLittle,J.Higgins,J.P.Ioannidis,J.P.Moher,D.Gagnon,F.von Elm,E.Khoury,M.J.Cohen,B.Davey-Smith,G.Grimshaw,J.Scheet,P.Gwinn,M.Williamson,R.E.Zou,G.Y.Hutchings,K.Johnson,C.Y.Tait,V.Wiens,M.Golding,J.Van Duijn,C.McLaughlin,J.Paterson,A.Wells,G.Fortier,I.Freedman,M.Zecevic,M.King,R.Infante-Rivard,C.Stewart,A.Birkett,N.2009/2/3analysisAssociationCanadaDiseaseepidemiologyGenesGeneticGenetic Predisposition to DiseasegeneticsGenomeGenomicsGuidelines as TopicHealthHumanHumansMedicinemethodsOntarioPeriodicals as TopicPopulationPublic HealthResearchResearch SupportstandardsStatisticalUniversitiesNot in Filee22PLoS.Med.62Canada Research Chair in Human Genome Epidemiology, University of Ottawa, Ottawa, Ontario, Canada. jlittle@uottawa.caPM:19192942PLoS.Med.1[17], REMARK ADDIN REFMGR.CITE McShane2005MCSHANE2005AReporting recommendations for tumor marker prognostic studiesJournalMCSHANE2005AReporting recommendations for tumor marker prognostic studiesMcShane,L.M.Altman,D.G.Sauerbrei,W.Taube,S.E.Gion,M.Clark,G.M.2005/12/20analysisBiologicalBiomedical ResearchdiagnosisHumansInformation DisseminationNeoplasmsResearchResearch DesignResearch SupportstandardsTumor Markers,BiologicalNot in File90679072J.Clin.Oncol.2336Biometric Research Branch, National Cancer Institute, Bethesda, MD, USAPM:16172462J.Clin.Oncol.1[19], and STARD ADDIN REFMGR.CITE Bossuyt2003BOSSUYT2003MTowards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiativeJournalBOSSUYT2003MTowards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiativeBossuyt,P.M.Reitsma,J.B.Bruns,D.E.Gatsonis,C.A.Glasziou,P.P.Irwig,L.M.Lijmer,J.G.Moher,D.Rennie,D.de Vet,H.C.2003/1/4AlgorithmsBias (Epidemiology)ClinicalClinical Trials as TopicConsensusDiagnostic Techniques and ProceduresepidemiologyGuidelinesGuidelines as TopicMedicalmethodsNetherlandsPatientsPublishingResearchResearch DesignResearch SupportstandardsUniversitiesNot in File4144BMJ3267379Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, University of Amsterdam, PO Box 22700, 1100 DE Amsterdam, Netherlands. stard@amc.uva.nlPM:12511463British Medical JournalBMJ1[18]. These were selected from all available guidelines (see www.equator-network.org) because of their focus on observational study designs and genetic factors (STREGA), prediction models (REMARK), and test evaluation (REMARK and STARD). Methodological issues pertinent to risk prediction studies were addressed in presentations during the meeting. Workshop participants revised the initial recommendations both during the meeting and in extensive electronic correspondence after the meeting. To harmonize our recommendations for genetic risk prediction studies with previous guidelines, we chose the same wording and explanations for the items wherever possible. Finally, we tried to maintain consistency with previous guidelines for the evaluation of risk prediction studies of cardiovascular diseases and cancer ADDIN REFMGR.CITE Hlatky2009HLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationJournalHLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationHlatky,M.A.Greenland,P.Arnett,D.K.Ballantyne,C.M.Criqui,M.H.Elkind,M.S.Go,A.S.Harrell,F.E.,Jr.Hong,Y.Howard,B.V.Howard,V.J.Hsue,P.Y.Kramer,C.M.McConnell,J.P.Normand,S.L.O'Donnell,C.J.Smith,S.C.,Jr.Wilson,P.W.2009/5/5American Heart AssociationanalysisAssociationBiologicalBiological Markerscardiovascular diseaseCardiovascular DiseasesClinicaldiagnosisDiseaseEvaluation Studies as TopicHumansmethodsPopulationPrognosisResearchResearch DesignRiskRisk AssessmentSensitivity and SpecificitystandardsStatisticalUnited StatesNot in File24082416Circulation11917PM:19364974Circulation1Freedman2005FREEDMAN2005Cancer risk prediction models: a workshop on development, evaluation, and applicationJournalFREEDMAN2005Cancer risk prediction models: a workshop on development, evaluation, and applicationFreedman,A.N.Seminara,D.Gail,M.H.Hartge,P.Colditz,G.A.Ballard-Barbash,R.Pfeiffer,R.M.2005/5/18BreastBreast NeoplasmsColorectal Neoplasms,Hereditary NonpolyposisCommunicationConfidence IntervalsCoronary DiseaseCost-Benefit AnalysisDecision MakingepidemiologyetiologyEvaluation Studies as TopicFemaleGeneticGenetic Predisposition to DiseasegeneticsHumansMass ScreeningmethodsModelsModels,StatisticalMutationNeoplasmsOvarian NeoplasmsPopulationPredictive Value of Testsprevention & controlReproducibility of ResultsResearchResearch SupportRiskRisk AssessmentRisk FactorsScienceStatisticaltrendsNot in File715723J.Natl.Cancer Inst.9710Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD 20892-7344, USA. Andrew_Freedman@nih.govPM:15900041JNCI Journal of the National Cancer InstituteJ.Natl.Cancer Inst.1[2,22]. The final version of the checklist is presented in Table 1.
Scope of the GRIPS Statement
The GRIPS Statement is intended to maximize the transparency, quality and completeness of reporting on research methodology and findings in a particular study. Researchers can use the statement to inform their choice of study design and analyses, but the guidelines do not support or oppose the choice of any particular study design or method. For example, the guidelines recommend that the study population should be described, but do not specify which population is preferred in a particular study.
Items presented in the checklist are relevant for a wide array of observational risk prediction studies, because the checklist focuses on the main aspects in the design and analysis of risk prediction studies. GRIPS does not address randomized trials that may be performed to test risk models, nor does it specifically address decision analyses, cost-effectiveness analyses, assessment of health care needs or assessment of barriers to health care implementation ADDIN REFMGR.CITE Khoury2010KHOURY2010The emergence of translational epidemiology: from scientific discovery to population health impactJournalKHOURY2010The emergence of translational epidemiology: from scientific discovery to population health impactKhoury,M.J.Gwinn,M.Ioannidis,J.P.2010/9/1Diffusion of InnovationDiseaseepidemiologyGenomicsHealthHumanHumansKnowledgeMeta-Analysismethodsorganization & administrationPopulationPublic HealthRandomized Controlled TrialsResearchRiskRisk factorRoleTranslational ResearchNot in File517524Am.J.Epidemiol.1725Office of Public Health Genomics, Centers for Disease Control and Prevention, Atlanta, Georgia 30333, USA. muk1@cdc.govPM:20688899Am.J.Epidemiol.1[23]. Once the performance of a risk model has been established, these next steps towards implementation require further evaluation ADDIN REFMGR.CITE Moons2009MOONS2009Prognosis and prognostic research: application and impact of prognostic models in clinical practiceJournalMOONS2009Prognosis and prognostic research: application and impact of prognostic models in clinical practiceMoons,K.G.Altman,D.G.Vergouwe,Y.Royston,P.2009ClinicalClinical MedicineHealthMedicalModelsModels,BiologicalNetherlandsPrognosisResearchResearch DesignResearch SupportSciencestatistics & numerical dataUniversitiesNot in Fileb606BMJ338Julius Centre for Health Sciences and Primary Care, University Medical Centre Utrecht, Utrecht, Netherlands. k.g.m.moons@umcutrecht.nlPM:19502216British Medical JournalBMJ1Von Elm2007VONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesJournalVONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesVon Elm,E.Altman,D.G.Egger,M.Pocock,S.J.Gotzsche,P.C.Vandenbroucke,J.P.2007/10/16Biomedical ResearchCase-Control StudiesCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologymethodsObservationPreventive MedicinePublishingResearchstandardsUniversitiesNot in Filee296PLoS.Med.410Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland. strobe@ispm.unibe.chPM:17941714PLoS.Med.1[24,25]. For the reporting of these studies, which go beyond the assessment of genetic risk models as such, additional requirements apply. However, proper documentation of genetic predictive research according to GRIPS might facilitate the translation of research findings into clinical and public health practice.
How to use this paper
This paper illustrates and elaborates on the items of the GRIPS Statement that are published in several journals. We modeled this Explanation and Elaboration document along the lines of those developed for other reporting guidelines ADDIN REFMGR.CITE Altman2001ALTMAN2001The revised CONSORT statement for reporting randomized trials: explanation and elaborationJournalALTMAN2001The revised CONSORT statement for reporting randomized trials: explanation and elaborationAltman,D.G.Schulz,K.F.Moher,D.Egger,M.Davidoff,F.Elbourne,D.Gotzsche,P.C.Lang,T.2001/4/17AlgorithmsConsensusHealthHumansMedicalMedicinePublishingQuality ControlRandomized Controlled Trials as TopicResearchResearch SupportSciencestandardsStatisticsNot in File663694Ann.Intern.Med.1348ICRF Medical Statistics Group, Centre for Statistics in Medicine, Institute of Health Sciences, Old Road, Headington, Oxford OX3 7LF, United KingdomPM:11304107Ann.Intern.Med.1Bossuyt2003BOSSUYT2003JThe STARD statement for reporting studies of diagnostic accuracy: explanation and elaborationJournalBOSSUYT2003JThe STARD statement for reporting studies of diagnostic accuracy: explanation and elaborationBossuyt,P.M.Reitsma,J.B.Bruns,D.E.Gatsonis,C.A.Glasziou,P.P.Irwig,L.M.Moher,D.Rennie,D.de Vet,H.C.Lijmer,J.G.2003/1/7AlgorithmsBias (Epidemiology)ClinicalClinical Trials as TopicDiagnostic Techniques and ProceduresepidemiologyHealthMedicalNetherlandsPublishingReference StandardsReproducibility of ResultsResearch DesignSensitivity and SpecificitystandardsStatistics as TopicUniversitiesNot in FileW112Ann.Intern.Med.1381Department of Clinical Epidemiology and Biostatistics, Academic Medical Center, University of Amsterdam, PO Box 22700, 1100 DE Amsterdam, The Netherlands. p.m.bossuyt@amc.uva.nlPM:12513067Ann.Intern.Med.1Liberati2009LIBERATI2009CThe PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaborationJournalLIBERATI2009CThe PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaborationLiberati,A.Altman,D.G.Tetzlaff,J.Mulrow,C.Gotzsche,P.C.Ioannidis,J.P.Clarke,M.Devereaux,P.J.Kleijnen,J.Moher,D.2009/7/21ConsensusEvidence-Based PracticeHealthHumansItalyMeta-AnalysisMeta-Analysis as TopicPublishingQuality ControlResearchResearch SupportReview Literature as TopicSafetystandardsTerminology as TopicNot in Filee1000100PLoS.Med.67Universita di Modena e Reggio Emilia, Modena, Italy. alesslib@mailbase.itPM:19621070PLoS.Med.1Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1[2629]. The GRIPS Statement consist of 25 items grouped by article sections (title and abstract, introduction, methods, results and discussion). The discussion of each item in this paper follows a standardized format. First, we illustrate each item with one or more published examples of what we consider to be transparent reporting, drawn from the genetic risk prediction studies referenced in Table 2. Table or figure numbers in the examples refer to the tables and figures in the present manuscript, not the original article. Second, for each item, we explain in detail the rationale for its inclusion in the checklist. And third, we present details about each item that need to be addressed to ensure transparent reporting.
Frequently, papers about genetic risk prediction are conducted using data from multiple populations. Many studies have combined multiple datasets to develop the risk model, for example by obtaining controls and cases from different populations ADDIN REFMGR.CITE Sparso2009SPARSO2009Combined analysis of 19 common validated type 2 diabetes susceptibility gene variants shows moderate discriminative value and no evidence of gene-gene interactionJournalSPARSO2009Combined analysis of 19 common validated type 2 diabetes susceptibility gene variants shows moderate discriminative value and no evidence of gene-gene interactionSparso,T.Grarup,N.Andreasen,C.Albrechtsen,A.Holmkvist,J.Andersen,G.Jorgensen,T.Borch-Johnsen,K.Sandbaek,A.Lauritzen,T.Madsbad,S.Hansen,T.Pedersen,O.2009/7AdultAgedAllelesanalysisCase-Control StudiesClinicalDenmarkDiabetes Mellitus,Type 2epidemiologyFemaleGenetic Predisposition to DiseaseGenetic VariationgeneticsHumansMalemethodsMiddle AgedModels,GeneticOdds RatioPatientsPenetrancePopulationReproducibility of ResultsResearchResearch SupportRiskRisk FactorsRoc CurveNot in File13081314Diabetologia527Steno Diabetes Center, Niels Steensens Vej 1, Gentofte, Denmark. tspr@steno.dkPM:19404609Diabetologia1Salinas2009SALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalityJournalSALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalitySalinas,C.A.Koopmeiners,J.S.Kwon,E.M.FitzGerald,L.Lin,D.W.Ostrander,E.A.Feng,Z.Stanford,J.L.2009/3/1AdultAgedCase-Control StudiesChromosomes,Human,Pair 17Chromosomes,Human,Pair 8ClinicalDiseaseFamilyGeneticGenetic Predisposition to DiseasegeneticsGenotypehadhistoryHumansLogistic ModelsMalemethodsMiddle AgedModelsMortalityOdds RatioPolymorphismPolymorphism,Single NucleotidePopulationPredictive Value of TestsPrognosisProportional Hazards ModelsProstatic NeoplasmsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideNot in File363372Prostate694Fred Hutchinson Cancer Research Center, Seattle, Washington 98109-1024, USAPM:19058137Prostate1Lauenborg2009LAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesJournalLAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesLauenborg,J.Grarup,N.Damm,P.Borch-Johnsen,K.Jorgensen,T.Pedersen,O.Hansen,T.2009/1/1AllelesanalysisAssociationBody Mass IndexDiabetes MellitusGeneticGenotypehistorymethodsOdds RatioPrevalenceResearchResearch DesignRiskWomenNot in File145150J Clin Endocrinol Metab941http://jcem.endojournals.org/cgi/content/abstract/94/1/145Journal of Clinical Endocrinology MetabolismJ Clin Endocrinol Metab1Weersma2009WEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortJournalWEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortWeersma,R.K.Stokkers,P.C.van Bodegraven,A.A.van Hogezand,R.A.Verspaget,H.W.de Jong,D.J.van der Woude,C.J.Oldenburg,B.Linskens,R.K.Festen,E.A.van der,Steege G.Hommes,D.W.Crusius,J.B.Wijmenga,C.Nolte,I.M.Dijkstra,G.2009/3AdultAge of OnsetAllelesanalysisAssociationColitis,UlcerativeCrohn DiseaseDiseaseDisease SusceptibilityepidemiologyFemaleGastroenterologyGene Expression RegulationGenesGeneticGenetic Predisposition to DiseasegeneticsGenotypeHospitalsHumanHumansMaleMedicalmethodsMolecularMolecular BiologyMulticenter StudiesNetherlandsNod2 Signaling Adaptor ProteinOdds RatioPatientsPhenotypePolymorphismPolymorphism,GeneticReceptorsReceptors,InterleukinRegression AnalysisResearchResearch SupportRiskRisk AssessmentSingle NucleotideUniversitiesNot in File388395Gut583Department of Gastroenterology and Hepatology, University Medical Center Groningen and University of Groningen, P.O. Box 30001, 9700 RB Groningen, The Netherlands. R.K.Weersma@int.umcg.nlPM:18824555Gut1[7,3032], or have derived risk models in multiple populations ADDIN REFMGR.CITE Lyssenko2008LYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesJournalLYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesLyssenko,V.Jonsson,A.Almgren,P.Pulizzi,N.Isomaa,B.Tuomi,T.Berglund,G.Altshuler,D.Nilsson,P.Groop,L.2008/11/20Body Mass IndexClinicalDiabetes MellitusDiseaseDnaFamilyGenesGenetichadhistoryInsulinmethodsPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSciencesecretionSingle NucleotideSmokingSwedenTimeUniversitiesNot in File22202232N.Engl.J.Med.35921Department of Clinical Sciences, Lund University, Malmo, Sweden. valeri.lyssenko@med.lu.sePM:19020324New England Journal of MedicineN.Engl.J.Med.1[33]. Studies may also use one or more populations to validate the model in independent samples. Readers need to be able to assess the similarities and differences among these populations in terms of the design of the study, selection of participants, data collection and analyses. Differences in the study designs and population characteristics that might impact the validity and generalizability of the findings should be reported. These may include ascertainment of participants, distributions of age, sex and ethnicity as well as the prevalence of risk factors, disease and co-morbidities ADDIN REFMGR.CITE Janssens2009JANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJournalJANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJanssens,A.C.J.W.Van Duijn,C.M.2009/2/18ClinicalDiseaseepidemiologyHealthNetherlandsPublic HealthUniversitiesNot in File20Genome Med.12Department of Epidemiology, Erasmus University Medical Center Rotterdam, 3000 CA Rotterdam, The Netherlands. a.janssens@erasmusmc.nlPM:19341491Genome Med.1[3]. Authors should describe any efforts made to harmonize the assessment methods, if these were different. The essential items that should be reported for each population are marked in Table 1.
Finally, genetic risk models may also be applied to predict other clinically relevant outcomes such as prognosis, treatment response and side effects of treatment. To improve the readability of the paper, the paper focuses on prediction of disease risk, but the items also apply to other health outcomes as well.
The GRIPS Checklist
For each checklist item shown in Table 1, this section provides examples of appropriate reporting from actual scientific articles of genetic risk models for diseases and health conditions, as well as an explanation of the importance and need for the item and helpful guidance about details that constitute transparent reporting.
TITLE and ABSTRACT
Item 1: (a) Identify the article as a study of risk prediction using genetic factors. (b) Use recommended keywords in the abstract: genetic or genomic, risk, prediction.
Examples. (Title) Combining information from common type 2 diabetes risk polymorphisms improves disease prediction. ADDIN REFMGR.CITE Weedon2006WEEDON2006Combining information from common type 2 diabetes risk polymorphisms improves disease predictionJournalWEEDON2006Combining information from common type 2 diabetes risk polymorphisms improves disease predictionWeedon,M.N.McCarthy,M.I.Hitman,G.Walker,M.Groves,C.J.Zeggini,E.Rayner,N.W.Shields,B.Owen,K.R.Hattersley,A.T.Frayling,T.M.2006/10/3AllelesCase-Control StudiesDiabetes MellitusDiseasehadmethodsOdds RatioRiskRoleTimePolymorphismNot in Filee374PLOS Med310PM:17020404PLOS MedicinePLOS Med1[34]
(Title) Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variables. ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1[6]
(Abstract) Recent studies have evaluated whether incorporating nontraditional risk factors improves coronary heart disease (CHD) prediction models. This 19862001 US study aggregated the contribution of multiple single nucleotide polymorphisms into a genetic risk score (GRS) and assessed whether the GRS plus traditional risk factors predict CHD better than traditional risk factors alone. ADDIN REFMGR.CITE Morrison2007MORRISON2007Prediction of coronary heart disease risk using a genetic risk score: the Atherosclerosis Risk in Communities StudyJournalMORRISON2007Prediction of coronary heart disease risk using a genetic risk score: the Atherosclerosis Risk in Communities StudyMorrison,A.C.Bare,L.A.Chambless,L.E.Ellis,S.G.Malloy,M.Kane,J.P.Pankow,J.S.Devlin,J.J.Willerson,J.T.Boerwinkle,E.2007/7/1African Continental Ancestry GroupAtherosclerosisBlacksbloodBlood PressureCase-Control StudiesCholesterolCoronary DiseaseDiseaseepidemiologyetiologyEuropean Continental Ancestry GroupFemaleFollow-Up StudiesGeneticgeneticsHealthHumanHumansMaleMiddle AgedModelsPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsProspective StudiesResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveScienceSingle NucleotideTexasUnited StatesUniversitiesWhitesNot in File2835Am J Epidemiol1661Human Genetics Center and Division of Epidemiology, University of Texas Health Science Center at Houston, Houston, TX 77030, USAPM:17443022American Journal of EpidemiologyAm J Epidemiol1[35]
(Abstract) The degree to which currently known genetic variants can improve the prediction of CHD risk beyond conventional risk factors in this disorder was investigated. ADDIN REFMGR.CITE van der Net2009VANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiaJournalVANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiavan der Net,J.B.Janssens,A.C.J.W.Defesche,J.C.Kastelein,J.J.Sijbrands,E.J.G.Steyerberg,E.W.2009/2/1Age of OnsetAgedcomplicationsCoronary DiseaseDiseaseFemaleGenetic Predisposition to DiseaseGenetic VariationgeneticsGenotypehadHealthHumansHyperlipoproteinemia Type IIMaleMiddle AgedNetherlandsPatientsPolymorphism,GeneticProportional Hazards ModelsPublic HealthRiskRisk FactorsNot in File375380Am.J.Cardiol.1033Department of Public Health, Erasmus MC-University Medical Center Rotterdam, Rotterdam, The NetherlandsPM:19166692Am.J.Cardiol.1[36]
Explanation. Public bibliographic databases have become an essential tool in knowledge synthesis and dissemination and a key source for identifying studies. To date, there is no single strategy that retrieves all or most papers on genetic risk prediction in these databases. Table 2 shows that the 24 studies of genetic risk prediction cited in this paper have used 17 different terms in their titles and one study made no reference to genetic factors at all ADDIN REFMGR.CITE Wu2007WU2007Projecting individualized probabilities of developing bladder cancer in white individualsJournalWU2007Projecting individualized probabilities of developing bladder cancer in white individualsWu,X.Lin,J.Grossman,H.B.Huang,M.Gu,J.Etzel,C.J.Amos,C.I.Dinney,C.P.Spitz,M.R.2007/11/1Case-Control StudiesClinicalEpidemiologic FactorsepidemiologyEuropean Continental Ancestry GroupFemaleGeneticHumansMalemethodsMiddle AgedModelsModels,StatisticalPatientsPopulationProbabilityResearchResearch SupportRiskRisk AssessmentRisk factorRisk FactorsRoc CurveTexasUniversitiesUrinary Bladder NeoplasmsNot in File49744981J.Clin.Oncol.2531Department of Epidemiology, The University of Texas M.D. Anderson Cancer Center, Houston, TX 77030, USA. xwu@mdanderson.orgPM:17971596J.Clin.Oncol.1[37]. PubMed Clinical Queries has implemented standardized search strategies for retrieving clinical prediction guides ADDIN REFMGR.CITE Wong2003WONG2003Developing optimal search strategies for detecting sound clinical prediction studies in MEDLINEJournalWONG2003Developing optimal search strategies for detecting sound clinical prediction studies in MEDLINEWong,S.S.Wilczynski,N.L.Haynes,R.B.Ramkissoonsingh,R.2003CanadaCausalityClinicaldiagnosishadHealthHumansInformation Storage and RetrievalMedical Subject HeadingsMedlinemethodsOntarioPrognosisPublishingRecallResearchResearch SupportRisk AssessmentSensitivity and SpecificityUniversitiesNot in File728732AMIA.Annu.Symp.Proc.McMaster University, Hamilton, Ontario, CanadaPM:14728269AMIA.Annu.Symp.Proc.1[38] and prognosis studies in general ADDIN REFMGR.CITE Wilczynski2004WILCZYNSKI2004ADeveloping optimal search strategies for detecting clinically sound prognostic studies in MEDLINE: an analytic surveyJournalWILCZYNSKI2004ADeveloping optimal search strategies for detecting clinically sound prognostic studies in MEDLINE: an analytic surveyWilczynski,N.L.Haynes,R.B.2004/6/9CanadaCase ReportClinicalepidemiologyHealthInformation Storage and RetrievalMedicalMedical Subject HeadingsMedlinemethodsOntarioPaperPrognosisPublishingResearchResearch SupportSensitivity and SpecificitytherapyTimeUniversitiesNot in File23BMC.Med.2Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, L8N 3Z5, Canada. wilczyn@mcmaster.ca <wilczyn@mcmaster.ca>PM:15189561BMC.Med.1[39], but these are inefficient strategies to retrieve genetic risk prediction studies. The broad versions of both types of PubMed Clinical Queries were able to ascertain most of the listed papers, but at the same time many other studies not related to this topic (Table 2). To facilitate identification and indexing, authors are encouraged to exploit all three opportunities, namely title, abstract and Medical Subject Headings (MeSH terms), to help ensure the capture of the article in the clinical queries and routine PubMed searches.
In the abstract, authors should explicitly describe their work as a study of genetic risk prediction by using the three keywords: genetic (or genomic), risk, and prediction. These words do not need to be mentioned in a specific combination or order. If the report focuses on genetic risk prediction as a main objective, authors are advised to mention the keywords in the title. The use of the keyword genetic or genomic is particularly important because a variety of genetic variants exists, such as chromosomes, SNPs, haplotypes or copy number variations. It will be difficult to retrieve all relevant studies if authors only use the specific terminology and not a broad descriptor like genetic variant. Table 2 shows that the combination of the keywords was by far more specific in identifying the prediction studies that are cited in this paper as compared with the PubMed Clinical Queries. The use of these keywords is also essential when risk prediction is not the main objective of a study, for example when prediction analysis is part of genome-wide association studies ADDIN REFMGR.CITE Aulchenko2009AULCHENKO2009Loci influencing lipid levels and coronary heart disease risk in 16 European population cohortsJournalAULCHENKO2009Loci influencing lipid levels and coronary heart disease risk in 16 European population cohortsAulchenko,Y.S.Ripatti,S.Lindqvist,I.Boomsma,D.Heid,I.M.Pramstaller,P.P.Penninx,B.W.Janssens,A.C.Wilson,J.F.Spector,T.Martin,N.G.Pedersen,N.L.Kyvik,K.O.Kaprio,J.Hofman,A.Freimer,N.B.Jarvelin,M.R.Gyllensten,U.Campbell,H.Rudan,I.Johansson,A.Marroni,F.Hayward,C.Vitart,V.Jonasson,I.Pattaro,C.Wright,A.Hastie,N.Pichler,I.Hicks,A.A.Falchi,M.Willemsen,G.Hottenga,J.J.de Geus,E.J.Montgomery,G.W.Whitfield,J.Magnusson,P.Saharinen,J.Perola,M.Silander,K.Isaacs,A.Sijbrands,E.J.Uitterlinden,A.G.Witteman,J.C.Oostra,B.A.Elliott,P.Ruokonen,A.Sabatti,C.Gieger,C.Meitinger,T.Kronenberg,F.Doring,A.Wichmann,H.E.Smit,J.H.McCarthy,M.I.Van Duijn,C.M.Peltonen,L.2009/1AdolescentAdultAgedAged,80 and overbloodBody Mass IndexCholesterol,HDLCohort StudiesCoronary DiseaseDiseaseepidemiologyEuropeEuropean Continental Ancestry GroupFemaleGenetic Predisposition to DiseaseGenetic ScreeninggeneticsGenome,HumanHumansLipidsMaleMetabolic Networks and PathwaysMiddle AgedNetherlandsPhenotypePolymorphism,Single NucleotideQuantitative Trait LociRiskRisk FactorsSex CharacteristicsUniversitiesNot in File4755Nat.Genet411[1] Department of Epidemiology and Biostatistics, Erasmus University Medical Center, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. [2] These authors contributed equally to this workPM:19060911Nat.Genet1[40]. To ensure that these articles are identifiable, authors should mention the prediction analysis in the abstract as well.
MeSH terms are another opportunity to identify an article as a study of genetic risk prediction, but this is often not under control of the author. The articles listed in Table 2 have been given a variety of MeSH terms and no single term or combination of terms would have retrieved all papers. To facilitate future synthesis of studies, we recommend that studies on this topic at least use the MeSH terms genetic predisposition to disease, risk assessment and predictive value of tests. These three terms are analogous to the keywords genetic, risk and prediction. Each MeSH term alone retrieved 18 of the articles listed in Table 2, and over 50,000 other articles (results not shown). The exact combination of the three MeSH terms did not retrieve any of these studies, but also only a little over 100 other papers in total. Consequently, assigning the three MeSH terms to genetic risk prediction studies potentially allows for a very specific search strategy to retrieve future articles.
INTRODUCTION
Item 2: Explain the scientific background and rationale for the prediction study.
Example. Knowledge about genetic and epidemiologic associations with the leading cause of blindness among the elderly, age-related macular degeneration, has grown exponentially in recent years. Several genetic variants with strong and consistent
associations with AMD have recently been identified. We also know that in addition to age, ethnicity, and family history, there are modifiable factors: smoking, nutritional
antioxidants and omega-3 fatty acid intake, and overall and abdominal adiposity. However, it remains unknown whether all these genetic and environmental factors act independently or jointly and to what extent they as a group can predict the occurrence of age-related macular degeneration (AMD) or progression to advanced AMD from early and intermediate stages. Such information might be useful for screening those at high risk due to a positive family history or having signs of early or intermediate disease, among whom some progress to advanced stages of AMD with visual loss. Early detection could reduce the growing societal burden due to AMD by targeting and emphasizing modifiable habits earlier in life and recommending more frequent surveillance for those highly susceptible to the disease. ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1[6]
Explanation. The background should inform the reader what is already known on the topic, and what gaps in knowledge justify conducting the present study. Relevant background information should include, but is not limited to, the following two topics:
First, what is known about the role of genetic factors in the outcome of interest, and in particular about the genetic variants that are being considered for inclusion in the prediction model? Such information could include a summary of how many genetic variants have been discovered and possibly what is the range of their observed effect sizes.
Second, the introduction should inform what alternative models for risk prediction are available or have been investigated for the outcome of interest, including models that are based on fewer genetic variants, the same variants, non-genetic risk factors or a combination of genetic and non-genetic factors. The assessment of the performance of these risk models can provide a reference value for the evaluation of the risk model under study ADDIN REFMGR.CITE Tzoulaki2009TZOULAKI2009Assessment of claims of improved prediction beyond the Framingham risk scoreJournalTZOULAKI2009Assessment of claims of improved prediction beyond the Framingham risk scoreTzoulaki,I.Liberopoulos,G.Ioannidis,J.P.2009/12/2analysisClinicalCoronary DiseaseEnglandepidemiologyhadHealthHealth Status IndicatorsHumansKnowledgeLondonMedicinemethodsModels,StatisticalPopulationPrognosisPublic HealthReproducibility of ResultsRiskRisk AssessmentRisk FactorsNot in File23452352JAMA30221Department of Epidemiology and Public Health, Imperial College of Medicine, London, EnglandPM:19952321JAMA: The Journal of the American Medical AssociationJAMA1Janssens2008JANSSENS2008AGenome-based prediction of common diseases: advances and prospectsJournalJANSSENS2008AGenome-based prediction of common diseases: advances and prospectsJanssens,A.C.J.W.Van Duijn,C.M.2008DiseaseNot in FileR166R173Hum Mol Genet17Hum Mol Genet1[13,41]. A comparison with earlier studies is most informative when essential information about the comparability of the studies is provided. Such information may include details about the setting (see below) and the age, sex and ethnicity of the population investigated.
For some topics, summarizing this information systematically would require formal systematic reviews of extensive bodies of literature and hundreds of pages, far beyond the typical short introduction of most research papers. Therefore, we recommend that the authors should be concise in reviewing the status of current risk research on the topic of interest and how the current study proposes to build on this existing evidence.
Item 3: Specify the study objectives and state the specific model(s) that is/are investigated. State if the study concerns the development of the model(s), the validation effort of the model(s), or both.
Examples. We examined subjects in two large Scandinavian prospective studies with a median follow-up period of 23.5 years to determine whether these genetic variants alone or in combination with clinical risk factors might predict the future development of type 2 diabetes and whether these variants were associated with changes in insulin secretion or action over time. ADDIN REFMGR.CITE Lyssenko2008LYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesJournalLYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesLyssenko,V.Jonsson,A.Almgren,P.Pulizzi,N.Isomaa,B.Tuomi,T.Berglund,G.Altshuler,D.Nilsson,P.Groop,L.2008/11/20Body Mass IndexClinicalDiabetes MellitusDiseaseDnaFamilyGenesGenetichadhistoryInsulinmethodsPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSciencesecretionSingle NucleotideSmokingSwedenTimeUniversitiesNot in File22202232N.Engl.J.Med.35921Department of Clinical Sciences, Lund University, Malmo, Sweden. valeri.lyssenko@med.lu.sePM:19020324New England Journal of MedicineN.Engl.J.Med.1[33]
The present study was designed to evaluate whether the findings of Zheng et al. could be replicated in a population-based sample of American Caucasian men and to evaluate how the combination of SNP genotypes and family history function in prediction models for prostate cancer risk and for prostate cancer-specific mortality. ADDIN REFMGR.CITE Salinas2009SALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalityJournalSALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalitySalinas,C.A.Koopmeiners,J.S.Kwon,E.M.FitzGerald,L.Lin,D.W.Ostrander,E.A.Feng,Z.Stanford,J.L.2009/3/1AdultAgedCase-Control StudiesChromosomes,Human,Pair 17Chromosomes,Human,Pair 8ClinicalDiseaseFamilyGeneticGenetic Predisposition to DiseasegeneticsGenotypehadhistoryHumansLogistic ModelsMalemethodsMiddle AgedModelsMortalityOdds RatioPolymorphismPolymorphism,Single NucleotidePopulationPredictive Value of TestsPrognosisProportional Hazards ModelsProstatic NeoplasmsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideNot in File363372Prostate694Fred Hutchinson Cancer Research Center, Seattle, Washington 98109-1024, USAPM:19058137Prostate1[31]
Explanation. Objectives refer to the specific research questions that are investigated in the study. For genetic risk prediction studies, the objectives should specify which models are investigated for the prediction of which outcome in which population and setting. Furthermore, authors should state whether the report concerns the development of a novel risk model (and if so, whether some sort of internal or external validation is performed) or about a replication or validation of an earlier model. Finally, any planned subgroup and interaction analyses should be specified, including a priori hypotheses or a statement that subgroup and interaction effects were explored without any hypothesis.
METHODS
Item 4: Specify the key elements of the study design and describe the setting, locations and relevant dates, including periods of recruitment, follow-up and data collection.
Examples. The Rotterdam Study is a prospective, population-based, cohort study among 7,983 inhabitants of a Rotterdam suburb, designed to investigate determinants of chronic diseases. Participants were aged 55 years and older. Baseline examinations took place from 1990 until 1993. Follow-up examinations were performed in 19931994, 19971999, and 20022004. Between these exams, continuous surveillance on major disease outcomes was conducted. Information on vital status was obtained from
municipal health authorities. ADDIN REFMGR.CITE van Hoek2008VANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyJournalVANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyvan Hoek,M.Dehghan,A.Witteman,J.C.Van Duijn,C.M.Uitterlinden,A.G.Oostra,B.A.Hofman,A.Sijbrands,E.J.Janssens,A.C.2008/11ADAM ProteinsAgedCation Transport ProteinsClinicalCyclin-Dependent Kinase 5Cyclin-Dependent Kinase Inhibitor p15Cyclin-Dependent Kinase Inhibitor p16Diabetes Mellitus,Type 2ethnologyEuropean Continental Ancestry GroupGenetic Predisposition to DiseaseGenetic TestinggeneticsGenome,HumanGenome-Wide Association StudyHumanHumansMembrane ProteinsmethodsMiddle AgedNeoplasm ProteinsNetherlandsPolymorphism,GeneticProspective StudiesProteinsResearch DesignRiskRisk FactorsRNA-Binding ProteinsTCF Transcription FactorsUniversitiesNot in File31223128Diabetes5711Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, the NetherlandsPM:18694974Diabetes1[42]
A cohort of 2,576 men and 2,636 women from a general population (aged 3065 years at inclusion) participated in the DESIR longitudinal study and were clinically and biologically evaluated at inclusion, at 3-, 6-, and 9-year visits. ADDIN REFMGR.CITE Vaxillaire2008VAXILLAIRE2008Impact of common type 2 diabetes risk polymorphisms in the DESIR prospective studyJournalVAXILLAIRE2008Impact of common type 2 diabetes risk polymorphisms in the DESIR prospective studyVaxillaire,M.Veslot,J.Dina,C.Proenca,C.Cauchi,S.Charpentier,G.Tichet,J.Fumeron,F.Marre,M.Meyre,D.Balkau,B.Froguel,P.2008/1AdultAgedAllelesCohort StudiesDiabetes Mellitus,Type 2DiseaseEnvironmental ExposureepidemiologyFastingFemaleFollow-Up StudiesFranceGenesGeneticgeneticsGenotypeGlucokinaseGlucose IntoleranceHumansHyperglycemiaIncidenceInsulinInsulin ResistanceMalemethodsMiddle AgedMulticenter StudiesPolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePopulationPrevalenceProportional Hazards ModelsProspective StudiesResearchResearch DesignResearch SupportRiskRisk FactorsSyndromeUniversitiesNot in File244254Diabetes571UMR8090 and Institute of Biology, Lille 2 University, CNRS and Pasteur Institute, Lille, France. martine.vaxillaire@good.ibl.frPM:17977958Diabetes1[43]
Explanation. Key elements about the study design include whether the analyses were performed in: a cohort study, which follows a group of individuals over time to identify incident cases of disease; a cross sectional study, which examines prevalent disease in a defined population; or a case-control study, which compares individuals with the trait of interest to those without ADDIN REFMGR.CITE Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1Von Elm2007VONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesJournalVONELM2007EThe Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studiesVon Elm,E.Altman,D.G.Egger,M.Pocock,S.J.Gotzsche,P.C.Vandenbroucke,J.P.2007/10/16Biomedical ResearchCase-Control StudiesCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologymethodsObservationPreventive MedicinePublishingResearchstandardsUniversitiesNot in Filee296PLoS.Med.410Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland. strobe@ispm.unibe.chPM:17941714PLoS.Med.1Little2009LITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementJournalLITTLE2009CSTrengthening the REporting of Genetic Association Studies (STREGA): an extension of the STROBE statementLittle,J.Higgins,J.P.Ioannidis,J.P.Moher,D.Gagnon,F.von Elm,E.Khoury,M.J.Cohen,B.Davey-Smith,G.Grimshaw,J.Scheet,P.Gwinn,M.Williamson,R.E.Zou,G.Y.Hutchings,K.Johnson,C.Y.Tait,V.Wiens,M.Golding,J.Van Duijn,C.McLaughlin,J.Paterson,A.Wells,G.Fortier,I.Freedman,M.Zecevic,M.King,R.Infante-Rivard,C.Stewart,A.Birkett,N.2009/2/3analysisAssociationCanadaDiseaseepidemiologyGenesGeneticGenetic Predisposition to DiseasegeneticsGenomeGenomicsGuidelines as TopicHealthHumanHumansMedicinemethodsOntarioPeriodicals as TopicPopulationPublic HealthResearchResearch SupportstandardsStatisticalUniversitiesNot in Filee22PLoS.Med.62Canada Research Chair in Human Genome Epidemiology, University of Ottawa, Ottawa, Ontario, Canada. jlittle@uottawa.caPM:19192942PLoS.Med.1[17,29,44]. Setting refers to how participants were recruited, for example through hospitals, outpatient clinics, screening centers or registries, and location refers to the country, region and cities, if relevant. Stating the dates of data-collection rather than the duration of the follow-up helps to place the study in historical context and is particularly important in the context of changes in diagnostic methods (e.g., imaging and use of biomarkers), and changes in the assessment of genotype and other risk factors.
Researchers should also state whether the data were de novo collected specifically for the purpose stated in the introduction, or whether the analyses were conducted using previously collected data ADDIN REFMGR.CITE Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1[29]. The secondary use of existing data is not necessarily less credible, but a statement might help to explain limitations in the study, including, but not limited to, relevant data not being assessed or the presence of peculiar population characteristics.
Item 5: Describe eligibility criteria for participants, and sources and methods of selection of participants.
Examples. (Eligibility criteria) The diagnosis of diabetes in case subjects was based on either current treatment with diabetes-specific medication or laboratory evidence of hyperglycemia if treated with diet alone. Patients with confirmed diagnosis of monogenic diabetes and those treated with regular insulin therapy within 1 year of diagnosis were excluded. Case subjects in this study had an age at diagnosis between 35 and 70 years, inclusive. Control subjects had not been diagnosed with diabetes at the time of recruitment or subsequently and were excluded if there was evidence of hyperglycemia during recruitment (fasting glucose >7.0 mmol/l, A1C >6.4%) or if they were >80 years old. ADDIN REFMGR.CITE Lango2008LANGO2008Assessing the combined impact of 18 common genetic variants of modest effect sizes on type 2 diabetes riskJournalLANGO2008Assessing the combined impact of 18 common genetic variants of modest effect sizes on type 2 diabetes riskLango,H.Palmer,C.N.Morris,A.D.Zeggini,E.Hattersley,A.T.McCarthy,M.I.Frayling,T.M.Weedon,M.N.2008/6/30AllelesAssociationClinicalDiseaseGeneticgeneticshadhigherMedicalmethodsOdds RatioPopulationResearchResearch DesignRiskScienceNot in File31293135Diabetes5711Genetics of Complex Traits, Institute of Biomedical and Clinical Science, Peninsula Medical School, Magdalen Road, Exeter, UKPM:18591388Diabetes1[45]
(Sources and methods of selection) The study population consisted of 283 women with previous gestational diabetes mellitus who were admitted to the Department of Obstetrics, Copenhagen University Hospital, Rigshospitalet, Denmark, during 19781996 and who had participated in a follow-up study during 20002002. ADDIN REFMGR.CITE Lauenborg2009LAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesJournalLAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesLauenborg,J.Grarup,N.Damm,P.Borch-Johnsen,K.Jorgensen,T.Pedersen,O.Hansen,T.2009/1/1AllelesanalysisAssociationBody Mass IndexDiabetes MellitusGeneticGenotypehistorymethodsOdds RatioPrevalenceResearchResearch DesignRiskWomenNot in File145150J Clin Endocrinol Metab941http://jcem.endojournals.org/cgi/content/abstract/94/1/145Journal of Clinical Endocrinology MetabolismJ Clin Endocrinol Metab1[32]
Explanation. The predictive performance of a risk model might vary with the population in which the test is applied, and is preferably assessed by testing a random sample of individuals from the population at risk of the disease or outcome. The eligibility criteria, source and methods of selection of the study participants thus inform readers about the assumed target population for testing as well as about the representativeness of the study population. Knowledge of the selection criteria is essential in appraising the validity and generalizability of the study results. Eligibility criteria may be presented as inclusion and exclusion criteria, specifying characteristics such as age, sex, ancestry, ethnicity and/or geographical region, and, for case-control studies, diagnosis and comorbidity. The source refers to the populations from which the participants were selected and to the methods of selectionwhether participants were, for example, randomly invited, referred or self-selected. The diagnostic criteria should be clearly described, including references to standards, if applicable.
For cohort and cross-sectional studies, the population base from which participants were invited (e.g., from a general population, specific region or hospital) should be specified. Depending on the aim of the cohort, typical eligibility criteria may include age, sex, ethnicity, specific risk factors, and for cohorts of patients, diagnosis, disease duration or stage, and comorbidity ADDIN REFMGR.CITE Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1[29].
For case-control studies, one should specify the (diagnostic) criteria that were used to select cases, and the criteria for selecting the controls. The extent to which controls were screened for absence of symptoms related to the disease or outcome under study should be described. Description of the criteria should enable understanding of the spectrum of disease involved. Case-control studies sometimes compare very severe cases with very healthy controls, particularly if the data were previously collected primarily for gene discovery ADDIN REFMGR.CITE Maller2006MALLER2006Common variation in three genes, including a noncoding variant in CFH, strongly influences risk of age-related macular degenerationJournalMALLER2006Common variation in three genes, including a noncoding variant in CFH, strongly influences risk of age-related macular degenerationMaller,J.George,S.Purcell,S.Fagerness,J.Altshuler,D.Daly,M.J.Seddon,J.M.2006/9AllelesanalysisBostonCase-Control StudiesDiseaseGenetic ResearchgeneticsGenotypeHumanMassachusettsRecurrenceResearchRiskSiblingsNot in File10551059Nat.Genet.389[1] Center for Human Genetic Research, Massachusetts General Hospital, 185 Cambridge St., Boston, Massachusetts 02114, USA. [2] Program in Medical and Population Genetics, Broad Institute of Harvard and MIT, 7 Cambridge Center, Cambridge, Massachusetts 02142, USAPM:16936732Nature GeneticsNat.Genet.1Wang2008WANG2008Polygenic determinants of severe hypertriglyceridemiaJournalWANG2008Polygenic determinants of severe hypertriglyceridemiaWang,J.Ban,M.R.Zou,G.Y.Cao,H.Lin,T.Kennedy,B.A.Anand,S.Yusuf,S.Huff,M.W.Pollex,R.L.Hegele,R.A.2008/7/1AllelesAssociationCanadaDiseaseFastingGenesGeneticGenotypehadLondonMutationObesityOdds RatioOntarioPatientsResearchUniversitiesNot in File28942899Hum.Mol.Genet.1718Vascular Biology Research Group and Robarts Research Institute and Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada N6A 5K8PM:18596051Human Molecular GeneticsHum.Mol.Genet.1[8,46]. Such stringent selection of participants is an effective strategy for gene discovery, but predictive performance might be overestimated compared with assessment in unselected populations where controls might have early symptoms or risk factors of disease. Furthermore, for case-control studies, it is important to specify whether cases and controls were matched and how, as overmatching might affect the predictive power of that factor in the sample relative to its predictive power in an unmatched population.
Item 6: Clearly define all participant characteristics, risk factors and outcomes. Clearly define genetic variants using a widely-used nomenclature system.
Examples. (Predictors) We selected six SNPs from six loci on the basis of their association with levels of LDL or HDL cholesterol in at least one previous study. These six SNPs were, for association with LDL cholesterol, APOB (apolipoprotein B, rs693), PCSK9 (proprotein convertase subtilisin/kexin type 9, rs11591147), and LDLR (low-density lipoprotein receptor, rs688); and for association with HDL cholesterol, CETP (cholesteryl ester transfer protein, rs1800775), LIPC (hepatic lipase, rs1800588), and LPL (lipoprotein lipase, rs328). ADDIN REFMGR.CITE Kathiresan2008KATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsJournalKATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsKathiresan,S.Melander,O.Anevski,D.Guiducci,C.Burtt,N.P.Roos,C.Hirschhorn,J.N.Berglund,G.Hedblad,B.Groop,L.Altshuler,D.M.Newton-Cheh,C.Orho-Melander,M.2008/3/20AffectAllelesAssociationbloodcardiovascular diseaseCardiovascular DiseasesCholesterolCholesterol,HDLCholesterol,LDLClassificationClinicalCoronary DiseaseDietDiseaseFemalegeneticsGenotypehadHumansMaleMassachusettsmethodsMiddle AgedModelsMortalityMultivariate AnalysisMyocardial InfarctionPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideStrokeTimeNot in File12401249N.Engl.J.Med.35812Cardiovascular Disease Prevention Center, Cardiology Division, Massachusetts General Hospital, MA 02114, USA. skathiresan@partners.orgPM:18354102New England Journal of MedicineN.Engl.J.Med.1[47]
(Predictors) Another example is provision of the information in tabular form (See Table 3) ADDIN REFMGR.CITE Meigs2008MEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesJournalMEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesMeigs,J.B.Shrader,P.Sullivan,L.M.McAteer,J.B.Fox,C.S.Dupuis,J.Manning,A.K.Florez,J.C.Wilson,P.W.D'Agostino,R.B.,Sr.Cupples,L.A.2008/11/20AllelesbloodBlood PressureBody Mass IndexBostonCholesterolClinicalDiabetes MellitusFamilyFastingGeneticGenotypehistoryKnowledgeMassachusettsmethodsOdds RatioPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideStatisticsNot in File22082219N.Engl.J.Med.35921General Medicine Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA. jmeigs@partners.orgPM:19020323New England Journal of MedicineN.Engl.J.Med.1[48].
(Predictors) We defined a positive self reported family history of diabetes as a report that one or both parents had diabetes; this definition is more than 56% sensitive and 97% specific for confirmed parental diabetes. [] We considered diabetes to be present in a parent when medication was prescribed to control the diabetes or when the casual plasma glucose level was 11.1 mmol per liter or higher or 200.0 mg per deciliter or higher at any examination. ADDIN REFMGR.CITE Meigs2008MEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesJournalMEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesMeigs,J.B.Shrader,P.Sullivan,L.M.McAteer,J.B.Fox,C.S.Dupuis,J.Manning,A.K.Florez,J.C.Wilson,P.W.D'Agostino,R.B.,Sr.Cupples,L.A.2008/11/20AllelesbloodBlood PressureBody Mass IndexBostonCholesterolClinicalDiabetes MellitusFamilyFastingGeneticGenotypehistoryKnowledgeMassachusettsmethodsOdds RatioPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideStatisticsNot in File22082219N.Engl.J.Med.35921General Medicine Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA. jmeigs@partners.orgPM:19020323New England Journal of MedicineN.Engl.J.Med.1[48]
(Outcomes) The prespecified composite end point of cardiovascular events was defined as myocardial infarction, ischemic stroke, and death from coronary heart disease. Myocardial infarction was defined on the basis of codes 410 and I21 in the International Classification of Diseases, 9th Revision and 10th Revision (ICD-9 and ICD-10), respectively. Ischemic stroke was defined on the basis of codes 434 or 436 (ICD-9) and I63 or I64 (ICD-10). ADDIN REFMGR.CITE Kathiresan2008KATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsJournalKATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsKathiresan,S.Melander,O.Anevski,D.Guiducci,C.Burtt,N.P.Roos,C.Hirschhorn,J.N.Berglund,G.Hedblad,B.Groop,L.Altshuler,D.M.Newton-Cheh,C.Orho-Melander,M.2008/3/20AffectAllelesAssociationbloodcardiovascular diseaseCardiovascular DiseasesCholesterolCholesterol,HDLCholesterol,LDLClassificationClinicalCoronary DiseaseDietDiseaseFemalegeneticsGenotypehadHumansMaleMassachusettsmethodsMiddle AgedModelsMortalityMultivariate AnalysisMyocardial InfarctionPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideStrokeTimeNot in File12401249N.Engl.J.Med.35812Cardiovascular Disease Prevention Center, Cardiology Division, Massachusetts General Hospital, MA 02114, USA. skathiresan@partners.orgPM:18354102New England Journal of MedicineN.Engl.J.Med.1[47]
Explanation. All participant characteristics, genetic and non-genetic risk factors, and outcomes that are considered and used in the analyses, should be defined and described unambiguously. Disease outcomes should be defined by reference to established diagnostic criteria or justification of study-specific criteria, if such are employed. Both the selection of genetic and non-genetic risk factors should be clarified. Authors should specify whether all known risk factors are included, and, if not, why some are excluded. Genetic variants should be described using widely-used nomenclature ADDIN REFMGR.CITE Wain2002WAIN2002Guidelines for human gene nomenclatureJournalWAIN2002Guidelines for human gene nomenclatureWain,H.M.Bruford,E.A.Lovering,R.C.Lush,M.J.Wright,M.W.Povey,S.2002/4GenesGuidelinesGuidelines as TopicHumanHumansLaboratoriesLondonResearchResearch SupportTerminology as TopicUniversitiesNot in File464470Genomics794HUGO Gene Nomenclature Committee, The Galton Laboratory, Department of Biology, University College London, Wolfson House, 4, Stephenson Way, London, NW1 2HE, UKPM:11944974Genomics1[49]. For example, SNPs could be presented with rs numbers with allusion to the pertinent reference database and build (e.g., HapMap release 27) ADDIN REFMGR.CITE Sherry2001SHERRY2001dbSNP: the NCBI database of genetic variationJournalSHERRY2001dbSNP: the NCBI database of genetic variationSherry,S.T.Ward,M.H.Kholodov,M.Baker,J.Phan,L.Smigielski,E.M.Sirotkin,K.2001/1/1AnimalsAssociationBiotechnologyDatabases,FactualGeneticGenetic VariationgeneticsGenomeHealthHumanHuman Genome ProjectHumansInformation ServicesInternetMedicineNational Institutes of Health (U.S.)National Library of Medicine (U.S.)Polymorphism,Single NucleotideResearchResearch SupportUnited StatesNot in File308311Nucleic Acids Res.291National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health, Bethesda, MD, 20894, USA. sherry@ncbi.nlm.nih.govPM:11125122Nucleic Acids ResearchNucleic Acids Res.1[50]. When proxies (surrogate markers) are considered, the correlation with the intended variant should be quantified, for example in terms of R2 along with the population used to derive the correlation. When variants are obtained by imputation, the imputation method and reference database should be described along with an estimate of the quality of the imputation.
Item 7: (a) Describe sources of data and details of methods of assessment (measurement) for each variable. (b) Give a detailed description of genotyping and other laboratory methods.
Examples. (Sources of data) Phenotyping was performed by the participating
gastroenterologist from each university medical center by reviewing a patients chart retrospectively. ADDIN REFMGR.CITE Weersma2009WEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortJournalWEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortWeersma,R.K.Stokkers,P.C.van Bodegraven,A.A.van Hogezand,R.A.Verspaget,H.W.de Jong,D.J.van der Woude,C.J.Oldenburg,B.Linskens,R.K.Festen,E.A.van der,Steege G.Hommes,D.W.Crusius,J.B.Wijmenga,C.Nolte,I.M.Dijkstra,G.2009/3AdultAge of OnsetAllelesanalysisAssociationColitis,UlcerativeCrohn DiseaseDiseaseDisease SusceptibilityepidemiologyFemaleGastroenterologyGene Expression RegulationGenesGeneticGenetic Predisposition to DiseasegeneticsGenotypeHospitalsHumanHumansMaleMedicalmethodsMolecularMolecular BiologyMulticenter StudiesNetherlandsNod2 Signaling Adaptor ProteinOdds RatioPatientsPhenotypePolymorphismPolymorphism,GeneticReceptorsReceptors,InterleukinRegression AnalysisResearchResearch SupportRiskRisk AssessmentSingle NucleotideUniversitiesNot in File388395Gut583Department of Gastroenterology and Hepatology, University Medical Center Groningen and University of Groningen, P.O. Box 30001, 9700 RB Groningen, The Netherlands. R.K.Weersma@int.umcg.nlPM:18824555Gut1[7]
(Sources of data) All clinical measurements were performed in practice by [the first author] (first measurement) and a nurse practitioner (second, third and fourth measurements with in-between periods of 3 months). ADDIN REFMGR.CITE Plat2009PLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectJournalPLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectPlat,A.W.Stoffers,H.E.J.H.Klungel,O.H.van Schayck,C.P.de Leeuw,P.W.Soomers,F.L.Schiffers,P.M.Kester,A.D.M.Kroon,A.A.2009PolymorphismRiskPopulationNot in File659667J Hum Hypertens23J Hum Hypertens1[51]
(Methods of assessment) Weight was measured in underwear to the nearest 0.1 kg on Soehnle electronic scales. We measured height in bare feet to the nearest 1 mm by using a stadiometer with the participant standing erect with head in the Frankfort plane. We calculated body mass index as weight (kilograms)/height (metres) squared. We measured waist circumference, taken as the smallest circumference at or below the costal margin, with participants unclothed in the standing position by using a fibreglass tape measure at 600 g tension. We measured systolic blood pressure and diastolic blood pressure twice in the sitting position after five minutes rest with the Hawksley random zero sphygmomanometer. We took the average of the two readings t o b e t h e m e a s u r e d b l o o d p r e s s u r e . W e t o o k v e n o u s b l o o d i n t h e f a s t i n g s t a t e o r a t l e a s t f i v e h o u r s a f t e r a l i g h t , f a t f r e e b r e a k f a s t , b e f o r e a t w o h o u r 7 5 g o r a l g l u c o s e t o l e r a n c e t e s t w a s d o n e . S e r u m f o r l i p i d a n a l y s e s w a s r e f r i g e r a t e d a t "4 C a n d a s s a y e d within 72 hours. We used a Cobas Fara centrifugal analyzer (Roche Diagnostics System, Nutley, NJ) to measure cholesterol and triglyceride concentrations. We measured high density lipoprotein cholesterol by precipitating non-high density lipoprotein cholesterol with dextran sulfate-magnesium chloride with the use of a centrifuge and measuring cholesterol in the supernatant fluid. We used the Friedewald formula to calculate low density lipoprotein cholesterol concentration. ADDIN REFMGR.CITE Talmud2010TALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyJournalTALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyTalmud,P.J.Hingorani,A.D.Cooper,J.A.Marmot,M.G.Brunner,E.J.Kumari,M.Kivimaki,M.Humphries,S.E.2010AllelesArea Under CurveBody Mass IndexCalibrationCholesterolCohort StudiesdiagnosisDiseaseFamilyFastingGeneticgeneticsGenotypeGlucose Tolerance TesthistoryLondonMedicalMedicineModelsOdds RatioPhenotypePolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideSmokingTriglyceridesUniversitiesWomenNot in Fileb4838BMJ340Centre of Cardiovascular Genetics, Department of Medicine, University College London, London WC1E 6JF. p.talmud@ucl.ac.ukPM:20075150British Medical JournalBMJ1[52]
(Outcomes) Women with gestational diabetes mellitus in the years 19781985 were diagnosed by a 3h, 50g oral glucose tolerance test (OGTT), whereas women with gestational diabetes mellitus in 19871996 were diagnosed by a 3h, 75g OGTT. ADDIN REFMGR.CITE Lauenborg2009LAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesJournalLAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesLauenborg,J.Grarup,N.Damm,P.Borch-Johnsen,K.Jorgensen,T.Pedersen,O.Hansen,T.2009/1/1AllelesanalysisAssociationBody Mass IndexDiabetes MellitusGeneticGenotypehistorymethodsOdds RatioPrevalenceResearchResearch DesignRiskWomenNot in File145150J Clin Endocrinol Metab941http://jcem.endojournals.org/cgi/content/abstract/94/1/145Journal of Clinical Endocrinology MetabolismJ Clin Endocrinol Metab1[32]
(Genotyping) Genotyping was performed with the use of matrix-assisted laser desorptionionization time of-flight mass spectrometry on a MassARRAY platform (Sequenom), as described previously. All SNPs were in HardyWeinberg equilibrium (P>0.001). The genotyping success rate was 96%. Using 15 samples analyzed in quadruplicate, we found the genotyping error rate to be <0.7%. ADDIN REFMGR.CITE Kathiresan2008KATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsJournalKATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsKathiresan,S.Melander,O.Anevski,D.Guiducci,C.Burtt,N.P.Roos,C.Hirschhorn,J.N.Berglund,G.Hedblad,B.Groop,L.Altshuler,D.M.Newton-Cheh,C.Orho-Melander,M.2008/3/20AffectAllelesAssociationbloodcardiovascular diseaseCardiovascular DiseasesCholesterolCholesterol,HDLCholesterol,LDLClassificationClinicalCoronary DiseaseDietDiseaseFemalegeneticsGenotypehadHumansMaleMassachusettsmethodsMiddle AgedModelsMortalityMultivariate AnalysisMyocardial InfarctionPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideStrokeTimeNot in File12401249N.Engl.J.Med.35812Cardiovascular Disease Prevention Center, Cardiology Division, Massachusetts General Hospital, MA 02114, USA. skathiresan@partners.orgPM:18354102New England Journal of MedicineN.Engl.J.Med.1[47]
Explanation. Apart from the selection and definitions of the variables, the sources and methods used for the assessment can impact the quality of the study. Important quality concerns are the potential for misclassification of risk factors and outcomes, as well as the accuracy of genotyping ADDIN REFMGR.CITE Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1[29]. Sources of data basically refer to who did the data collection and how. Were the data collected by research physicians or trained students? Were questionnaires completed in an interview or based on self-report, and was the genotyping performed in house or by a specialized laboratory? Methods of assessment refer to the specific techniques or questionnaires that were used. If methods have been published previously, provide a reference. The laboratory procedures used to measure biomarkers should be described in sufficient detail for others to be able to perform them and evaluate the generalizability of prediction models that include them. For less widely-used assessments, such as questionnaires and procedures that are developed by the researchers themselves, authors should report validity and reliability information about the quality of the assessment ADDIN REFMGR.CITE Pepe2008PEPE2008AGauging the performance of SNPs, biomarkers, and clinical factors for predicting risk of breast cancerJournalPEPE2008AGauging the performance of SNPs, biomarkers, and clinical factors for predicting risk of breast cancerPepe,M.S.Janes,H.E.2008/7/16Area Under CurveBiologicalBreastBreast NeoplasmsClinicalFemaleGeneticGenetic MarkersgeneticsHumansPolymorphism,Single NucleotidePredictive Value of TestsResearchResearch DesignResearch SupportRiskRisk AssessmentRisk FactorsRoc CurveTumor Markers,BiologicalNot in File978979J.Natl.Cancer Inst.10014PM:18612128JNCI Journal of the National Cancer InstituteJ.Natl.Cancer Inst.1[53]. When different assessments are used at baseline and follow-up (e.g., baseline assessments done by research physicians and follow-up assessments obtained from medical records of the general practitioner) these should be explained. When there is an arbitration process for outcomes (e.g., centralized team arbitrating on outcomes based on information contributed by local investigators in peripheral centers), this process should be specified.
Item 8: (a) Describe how genetic variants were handled in the analyses. (b) Explain how other quantitative variables were handled in the analyses. If applicable, describe which groupings were chosen, and why.
Examples. (Genetic variants) Using these 18 SNPs, we constructed a genotype score ranging from 0 to 36 on the basis of the number of risk alleles [see Table 3 for coding of the risk alleles]. ADDIN REFMGR.CITE Meigs2008MEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesJournalMEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesMeigs,J.B.Shrader,P.Sullivan,L.M.McAteer,J.B.Fox,C.S.Dupuis,J.Manning,A.K.Florez,J.C.Wilson,P.W.D'Agostino,R.B.,Sr.Cupples,L.A.2008/11/20AllelesbloodBlood PressureBody Mass IndexBostonCholesterolClinicalDiabetes MellitusFamilyFastingGeneticGenotypehistoryKnowledgeMassachusettsmethodsOdds RatioPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideStatisticsNot in File22082219N.Engl.J.Med.35921General Medicine Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA. jmeigs@partners.orgPM:19020323New England Journal of MedicineN.Engl.J.Med.1[48]
(Genetic variants) For the first analysis of the effects of the polymorphic DNA variants, we used additive genetic models. In addition, we tested dominant and recessive alternative models for the best fit []. Multivariate linear regression analyses were used to test correlations between genotype and phenotype. Non-normally distributed variables were logtransformed before analysis. The effect size of a genetic or clinical risk factor on the risk of type 2 diabetes was calculated from multivariate regression analysis, with adjustment for age and sex, with the use of Nagelkerke R square. We estimated the predictive value of a combination of risk alleles (each person could have 0, 1, or 2 of them, for a total of 22) in 11 genes, which significantly predicted the risk of diabetes by defining subjects with more than 12 risk alleles (about 20%) as being at high risk and those with fewer than 8 risk alleles (about 20%) as being at low risk. ADDIN REFMGR.CITE Lyssenko2008LYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesJournalLYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesLyssenko,V.Jonsson,A.Almgren,P.Pulizzi,N.Isomaa,B.Tuomi,T.Berglund,G.Altshuler,D.Nilsson,P.Groop,L.2008/11/20Body Mass IndexClinicalDiabetes MellitusDiseaseDnaFamilyGenesGenetichadhistoryInsulinmethodsPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSciencesecretionSingle NucleotideSmokingSwedenTimeUniversitiesNot in File22202232N.Engl.J.Med.35921Department of Clinical Sciences, Lund University, Malmo, Sweden. valeri.lyssenko@med.lu.sePM:19020324New England Journal of MedicineN.Engl.J.Med.1[33]
(Other variables) Multivariate unconditional logistic regression analysis was performed to evaluate the relationships between prevalence or progression of AMD and all the genotypes plus various risk factors, controlling for age (70 years or older versus younger than 70), sex, and education (high school or less versus more than high school), cigarette smoking (never, past, or current), and body mass index (BMI), which was calculated as the weight in kilograms divided by the square of the height in meters (<25, 2529.9, and 30+). ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1[6]
Explanation. There are many approaches to data analysis of genetic variants; thus, specification and clarification of this handling is particularly relevant. Genetic variants may be entered in regression analysis separately as dominant or recessive effects e.g., ADDIN REFMGR.CITE Podgoreanu2006PODGOREANU2006Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after cardiac surgeryJournalPODGOREANU2006Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after cardiac surgeryPodgoreanu,M.V.White,W.D.Morris,R.W.Mathew,J.P.Stafford-Smith,M.Welsby,I.J.Grocott,H.P.Milano,C.A.Newman,M.F.Schwinn,D.A.Perioperative Genetics and Safety Outcomes Study (PEGASUS) Investigative Team,2006/7/4analysisClinicalIntercellular Adhesion Molecule-1Interleukin-6methodsMyocardial InfarctionOdds RatioPatientsRiskRisk FactorssurgeryNot in FileI275Circulation1141_supplhttp://circ.ahajournals.org/cgi/content/abstract/114/1_suppl/I-275Circulation1Humphries2007HUMPHRIES2007Candidate gene genotypes, along with conventional risk factor assessment, improve estimation of coronary heart disease risk in healthy UK menJournalHUMPHRIES2007Candidate gene genotypes, along with conventional risk factor assessment, improve estimation of coronary heart disease risk in healthy UK menHumphries,S.E.Cooper,J.A.Talmud,P.J.Miller,G.J.2007/1AlgorithmsanalysisbloodBlood PressureCholesterolCoronary DiseasediagnosisDiseaseFalse Positive ReactionsGenesGeneticgeneticsGenotypeGreat BritainHumansInterleukin-6Lipoprotein LipaseLondonMaleMedicalmethodsMiddle AgedMIDDLE-AGED MENMultivariate AnalysisPolymorphism,Single NucleotidePredictive Value of TestsProspective StudiesReference ValuesResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSmokingUniversitiesNot in File816Clin.Chem.531Centre for Cardiovascular Genetics, Department of Medicine, British Heart Foundation Laboratories, Royal Free and University College Medical School, London, United Kingdom. rmhaseh@ucl.ac.ukPM:17130180Clinical ChemistryClin.Chem.1[54,55], per allele (additive or log-additive) effects ADDIN REFMGR.CITE Lauenborg2009LAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesJournalLAUENBORG2009Common Type 2 Diabetes Risk Gene Variants Associate with Gestational DiabetesLauenborg,J.Grarup,N.Damm,P.Borch-Johnsen,K.Jorgensen,T.Pedersen,O.Hansen,T.2009/1/1AllelesanalysisAssociationBody Mass IndexDiabetes MellitusGeneticGenotypehistorymethodsOdds RatioPrevalenceResearchResearch DesignRiskWomenNot in File145150J Clin Endocrinol Metab941http://jcem.endojournals.org/cgi/content/abstract/94/1/145Journal of Clinical Endocrinology MetabolismJ Clin Endocrinol Metab1[32], or genotype categories ADDIN REFMGR.CITE Paynter2009PAYNTER2009Cardiovascular disease risk prediction with and without knowledge of genetic variation at chromosome 9p21.3JournalPAYNTER2009Cardiovascular disease risk prediction with and without knowledge of genetic variation at chromosome 9p21.3Paynter,N.P.Chasman,D.I.Buring,J.E.Shiffman,D.Cook,N.R.Ridker,P.M.2009/1/20bloodBlood PressureBostoncardiovascular diseaseCardiovascular DiseasesCholesterolChromosomes,Human,Pair 9Chronic DiseaseClassificationCohort StudiesDiseaseepidemiologyFamilyFemaleFollow-Up StudiesGeneticGenetic VariationgeneticsGenomeGenotypehadHealthhistoryHumansKaplan-Meiers EstimateKnowledgeMiddle AgedMyocardial InfarctionPatientsPolymorphismPolymorphism,Single NucleotideProbabilityProportional Hazards ModelsProspective StudiesResearchResearch SupportRiskRisk factorRisk FactorsSmokingUnited StatesWomenNot in File6572Ann.Intern.Med.1502Brigham and Women's Hospital, 900 Commonwealth Avenue East, Boston, MA 02215, USA. npaynter@partners.orgPM:19153409Ann.Intern.Med.1van Hoek2008VANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyJournalVANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyvan Hoek,M.Dehghan,A.Witteman,J.C.Van Duijn,C.M.Uitterlinden,A.G.Oostra,B.A.Hofman,A.Sijbrands,E.J.Janssens,A.C.2008/11ADAM ProteinsAgedCation Transport ProteinsClinicalCyclin-Dependent Kinase 5Cyclin-Dependent Kinase Inhibitor p15Cyclin-Dependent Kinase Inhibitor p16Diabetes Mellitus,Type 2ethnologyEuropean Continental Ancestry GroupGenetic Predisposition to DiseaseGenetic TestinggeneticsGenome,HumanGenome-Wide Association StudyHumanHumansMembrane ProteinsmethodsMiddle AgedNeoplasm ProteinsNetherlandsPolymorphism,GeneticProspective StudiesProteinsResearch DesignRiskRisk FactorsRNA-Binding ProteinsTCF Transcription FactorsUniversitiesNot in File31223128Diabetes5711Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, the NetherlandsPM:18694974Diabetes1[42,56]. Any of these three approaches can be followed depending on what was the best fitting genetic model for each variant ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1Wang2008WANG2008Polygenic determinants of severe hypertriglyceridemiaJournalWANG2008Polygenic determinants of severe hypertriglyceridemiaWang,J.Ban,M.R.Zou,G.Y.Cao,H.Lin,T.Kennedy,B.A.Anand,S.Yusuf,S.Huff,M.W.Pollex,R.L.Hegele,R.A.2008/7/1AllelesAssociationCanadaDiseaseFastingGenesGeneticGenotypehadLondonMutationObesityOdds RatioOntarioPatientsResearchUniversitiesNot in File28942899Hum.Mol.Genet.1718Vascular Biology Research Group and Robarts Research Institute and Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada N6A 5K8PM:18596051Human Molecular GeneticsHum.Mol.Genet.1Weersma2009WEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortJournalWEERSMA2009Molecular prediction of disease risk and severity in a large Dutch Crohn's disease cohortWeersma,R.K.Stokkers,P.C.van Bodegraven,A.A.van Hogezand,R.A.Verspaget,H.W.de Jong,D.J.van der Woude,C.J.Oldenburg,B.Linskens,R.K.Festen,E.A.van der,Steege G.Hommes,D.W.Crusius,J.B.Wijmenga,C.Nolte,I.M.Dijkstra,G.2009/3AdultAge of OnsetAllelesanalysisAssociationColitis,UlcerativeCrohn DiseaseDiseaseDisease SusceptibilityepidemiologyFemaleGastroenterologyGene Expression RegulationGenesGeneticGenetic Predisposition to DiseasegeneticsGenotypeHospitalsHumanHumansMaleMedicalmethodsMolecularMolecular BiologyMulticenter StudiesNetherlandsNod2 Signaling Adaptor ProteinOdds RatioPatientsPhenotypePolymorphismPolymorphism,GeneticReceptorsReceptors,InterleukinRegression AnalysisResearchResearch SupportRiskRisk AssessmentSingle NucleotideUniversitiesNot in File388395Gut583Department of Gastroenterology and Hepatology, University Medical Center Groningen and University of Groningen, P.O. Box 30001, 9700 RB Groningen, The Netherlands. R.K.Weersma@int.umcg.nlPM:18824555Gut1[68]. Alternatively, genetic variants may be entered combined as risk scores ADDIN REFMGR.CITE Lyssenko2008LYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesJournalLYSSENKO2008Clinical risk factors, DNA variants, and the development of type 2 diabetesLyssenko,V.Jonsson,A.Almgren,P.Pulizzi,N.Isomaa,B.Tuomi,T.Berglund,G.Altshuler,D.Nilsson,P.Groop,L.2008/11/20Body Mass IndexClinicalDiabetes MellitusDiseaseDnaFamilyGenesGenetichadhistoryInsulinmethodsPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSciencesecretionSingle NucleotideSmokingSwedenTimeUniversitiesNot in File22202232N.Engl.J.Med.35921Department of Clinical Sciences, Lund University, Malmo, Sweden. valeri.lyssenko@med.lu.sePM:19020324New England Journal of MedicineN.Engl.J.Med.1Talmud2010TALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyJournalTALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyTalmud,P.J.Hingorani,A.D.Cooper,J.A.Marmot,M.G.Brunner,E.J.Kumari,M.Kivimaki,M.Humphries,S.E.2010AllelesArea Under CurveBody Mass IndexCalibrationCholesterolCohort StudiesdiagnosisDiseaseFamilyFastingGeneticgeneticsGenotypeGlucose Tolerance TesthistoryLondonMedicalMedicineModelsOdds RatioPhenotypePolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideSmokingTriglyceridesUniversitiesWomenNot in Fileb4838BMJ340Centre of Cardiovascular Genetics, Department of Medicine, University College London, London WC1E 6JF. p.talmud@ucl.ac.ukPM:20075150British Medical JournalBMJ1Kathiresan2008KATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsJournalKATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsKathiresan,S.Melander,O.Anevski,D.Guiducci,C.Burtt,N.P.Roos,C.Hirschhorn,J.N.Berglund,G.Hedblad,B.Groop,L.Altshuler,D.M.Newton-Cheh,C.Orho-Melander,M.2008/3/20AffectAllelesAssociationbloodcardiovascular diseaseCardiovascular DiseasesCholesterolCholesterol,HDLCholesterol,LDLClassificationClinicalCoronary DiseaseDietDiseaseFemalegeneticsGenotypehadHumansMaleMassachusettsmethodsMiddle AgedModelsMortalityMultivariate AnalysisMyocardial InfarctionPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideStrokeTimeNot in File12401249N.Engl.J.Med.35812Cardiovascular Disease Prevention Center, Cardiology Division, Massachusetts General Hospital, MA 02114, USA. skathiresan@partners.orgPM:18354102New England Journal of MedicineN.Engl.J.Med.1[33,47,52]. Risk scores often simply sum the number of risk alleles or genotypes (unweighted), or sum their beta-coefficients from regression analyses (weighted). When using risk scores, authors should explain which of the alleles or genotypes is considered as the risk variant, as this is not necessarily the less common (minor) variant (see Table 3). The description of the coding of the genetic variants should enable other researchers to replicate the analyses for validation or updating of the risk model.
Quantitative variables can be handled as continuous or be categorized. Transformations may be required when the relationships between the variables and the outcome are not linear, and these should be specified. Frequently, quantitative variables are categorized before inclusion in the analyses. A well-known example is body mass index, which is categorized as underweight, normal weight, overweight and obese. The rationale and thresholds used for categorization should be explained, particularly when they deviate from commonly used cut-offs based on clinical or epidemiological studies.
Item 9: Specify the procedure and data used for the derivation of the risk model. Specify which candidate variables were initially examined or considered for inclusion in models. Include details of any variable selection procedures and other model-building issues. Specify the horizon of risk prediction (e.g., 5-year risk).
Examples. (Model derivation) We constructed multivariable proportional-hazards models to examine the association between the genotype score and the time to the first cardiovascular event, excluding subjects who had had a previous myocardial infarction or ischemic stroke. We first confirmed that the proportional-hazards assumption was met. The hazard ratio for the genotype score as a continuous measure was estimated in a model adjusting for all 14 available baseline covariates. Cumulative incidence curves were constructed according to the genotype score with the use of Cox regression analysis. ADDIN REFMGR.CITE Kathiresan2008KATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsJournalKATHIRESAN2008Polymorphisms associated with cholesterol and risk of cardiovascular eventsKathiresan,S.Melander,O.Anevski,D.Guiducci,C.Burtt,N.P.Roos,C.Hirschhorn,J.N.Berglund,G.Hedblad,B.Groop,L.Altshuler,D.M.Newton-Cheh,C.Orho-Melander,M.2008/3/20AffectAllelesAssociationbloodcardiovascular diseaseCardiovascular DiseasesCholesterolCholesterol,HDLCholesterol,LDLClassificationClinicalCoronary DiseaseDietDiseaseFemalegeneticsGenotypehadHumansMaleMassachusettsmethodsMiddle AgedModelsMortalityMultivariate AnalysisMyocardial InfarctionPolymorphismPolymorphism,Single NucleotideProportional Hazards ModelsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideStrokeTimeNot in File12401249N.Engl.J.Med.35812Cardiovascular Disease Prevention Center, Cardiology Division, Massachusetts General Hospital, MA 02114, USA. skathiresan@partners.orgPM:18354102New England Journal of MedicineN.Engl.J.Med.1[47]
(Variable selection) Twenty-three candidate genes involved in the pathogenesis of inflammation and myocardial ischemia-reperfusion injury were selected a priori based on previous transcription profiling in humans and animal models, pathway analysis, a review of linkage and association studies reported in the literature, and expert opinion. Forty-eight SNPs were subsequently selected in these process-specific candidate genes, based on literature review, genomic context, and predictive analyses with an emphasis on functionally important variants. ADDIN REFMGR.CITE Podgoreanu2006PODGOREANU2006Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after cardiac surgeryJournalPODGOREANU2006Inflammatory gene polymorphisms and risk of postoperative myocardial infarction after cardiac surgeryPodgoreanu,M.V.White,W.D.Morris,R.W.Mathew,J.P.Stafford-Smith,M.Welsby,I.J.Grocott,H.P.Milano,C.A.Newman,M.F.Schwinn,D.A.Perioperative Genetics and Safety Outcomes Study (PEGASUS) Investigative Team,2006/7/4analysisClinicalIntercellular Adhesion Molecule-1Interleukin-6methodsMyocardial InfarctionOdds RatioPatientsRiskRisk FactorssurgeryNot in FileI275Circulation1141_supplhttp://circ.ahajournals.org/cgi/content/abstract/114/1_suppl/I-275Circulation1[54]
(Model building issues) Both univariate and multivariate odds ratios (ORs) were calculated with a binary-logistic regression model to evaluate the relationship between polymorphisms and prevalent CVD. For that purpose, dummy variables were created using the homozygous wild-type genotype as reference category. Age and gender, both demographic variables, were incorporated in both the univariate as well as in the multivariate linear regression analyses Adjustment for potential confounders was performed by incorporating smoking, alcohol, diabetes mellitus, waist circumference, serum creatinine, mean systolic and diastolic blood pressure, microalbuminuria and dyslipidaemia into these models. To avoid collinearity, waist circumference was used instead of waist-to-hip ratio or body mass index and condensed measures such as diabetes and dyslipidaemia were used, as defined earlier. ADDIN REFMGR.CITE Plat2009PLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectJournalPLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectPlat,A.W.Stoffers,H.E.J.H.Klungel,O.H.van Schayck,C.P.de Leeuw,P.W.Soomers,F.L.Schiffers,P.M.Kester,A.D.M.Kroon,A.A.2009PolymorphismRiskPopulationNot in File659667J Hum Hypertens23J Hum Hypertens1[51]
Explanation. Because of the potential for flexibility in the derivation of the risk model, authors need to clarify why and how they constructed the model as they did and which data they used. This clarification includes a specification of the variables, defined in item 6, that were initially considered and which procedures were followed for a final selection (e.g., backward deletion or forward inclusion, and the criteria for deletion and inclusion), if applicable. Clarification also includes a specification of the study participants included in the analysis, if different from the total study population, transformations of the variables, the choice of statistical model (e.g., logistic or Cox proportional hazards models), and the handling of interaction effects between predictors in the model (see also item 13). The specification also concerns the rationale for constructing separate models for subgroups, e.g., for different ethnic groups, or including the stratification variable as a variable or interaction effect in a model for the total population.
Authors should also specify and explain the horizon of the risk prediction, when appropriate (e.g., in cohort studies, whether the model predicts, for instance, 5-year or lifetime risk). When more complicated risk prediction models are developed using statistical learning methods such as regularized regression or support vector machines, these should be explained and specified in sufficient detail that others can implement these models in other data sets. For some more complex black box models (such as random forests) this may require making a software implementation of the final model available. The description of the data used should include whether a selection of the population was used for the derivation of the model, how this subpopulation was selected, and how censored data were handled in cohort studies.
Some studies aim only to validate and further apply an already existing model. In this case, it should simply be stated that a previous model was used with appropriate reference to the previous study or studies that developed the model along with a succinct description of its features.
Item 10: Specify the procedure and data used for the validation of the risk model.
Example. The internal validity of the prediction models was assessed using bootstrapping techniques. A total of 100 random bootstrap samples were drawn with replacement from the [total] group of 1,337 patients. The discriminative accuracy of the 100 prediction models as fit on these bootstrap samples was determined for each bootstrap sample and for the original group (n =1,337). This comparison gives an impression of how overoptimistic the model is, i.e., how much the performance of the model would deteriorate when applied to a new group of similar patients. ADDIN REFMGR.CITE van der Net2009VANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiaJournalVANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiavan der Net,J.B.Janssens,A.C.J.W.Defesche,J.C.Kastelein,J.J.Sijbrands,E.J.G.Steyerberg,E.W.2009/2/1Age of OnsetAgedcomplicationsCoronary DiseaseDiseaseFemaleGenetic Predisposition to DiseaseGenetic VariationgeneticsGenotypehadHealthHumansHyperlipoproteinemia Type IIMaleMiddle AgedNetherlandsPatientsPolymorphism,GeneticProportional Hazards ModelsPublic HealthRiskRisk FactorsNot in File375380Am.J.Cardiol.1033Department of Public Health, Erasmus MC-University Medical Center Rotterdam, Rotterdam, The NetherlandsPM:19166692Am.J.Cardiol.1[36]
Evaluation of model predictive performance using the same dataset used for fitting the model usually leads to a biased assessment. To obtain an unbiased assessment of discriminatory power of the multivariate regression models, a tenfold cross-validation was used in the ROC analysis and in the IDI analysis. Tenfold crossvalidation randomly divides the data into ten (roughly) equal subsets and repeatedly uses any nine subsets for model fitting and the remaining subset as validation until each of the ten subsets has been used exactly once as validation data. ADDIN REFMGR.CITE Lin2009LIN2009Risk prediction of prevalent diabetes in a Swiss population using a weighted genetic score--the CoLaus StudyJournalLIN2009Risk prediction of prevalent diabetes in a Swiss population using a weighted genetic score--the CoLaus StudyLin,X.Song,K.Lim,N.Yuan,X.Johnson,T.Abderrahmani,A.Vollenweider,P.Stirnadel,H.Sundseth,S.S.Lai,E.Burns,D.K.Middleton,L.T.Roses,A.D.Matthews,P.M.Waeber,G.Cardon,L.Waterworth,D.M.Mooser,V.2009/4AdultAgedAllelesBlood PressureBody Mass IndexClinicalDiabetes MellitusDiabetes Mellitus,Type 2DiseaseepidemiologyEuropean Continental Ancestry GroupFamilyFemaleGene FrequencyGenesGeneticGenetic MarkersGenetic Predisposition to DiseasegeneticsGoalshistoryHumansMalemethodsMiddle AgedPolymorphismPolymorphism,Single NucleotidePopulationPrevalenceResearchResearch SupportRiskSingle Nucleotidestatistics & numerical dataSwitzerlandNot in File600608Diabetologia524Discovery Analytics, GlaxoSmithKline, Collegeville, PA, USAPM:19139842Diabetologia1[57]
Explanation. Assessment of the risk model in the same population as that from which the model was derived generally leads to more positive conclusions than when the evaluation is conducted in an independent population ADDIN REFMGR.CITE Ioannidis2008IOANNIDIS2008BWhy most discovered true associations are inflatedJournalIOANNIDIS2008BWhy most discovered true associations are inflatedIoannidis,J.P.2008/9AssociationClinicalClinical TrialsClinical Trials as TopicData Interpretation,StatisticalepidemiologyGreeceHumansLinkage DisequilibriumMedicinemethodsModels,StatisticalMolecular EpidemiologySensitivity and SpecificityStatisticalUniversitiesNot in File640648Epidemiology195Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece. jioannid@cc.uoi.grPM:18633328Epidemiology1[58]. Therefore, validation of the risk model, reassessing the performance of the model in another dataset, is an essential part of model evaluation ADDIN REFMGR.CITE Bleeker2003BLEEKER2003External validation is necessary in prediction research: a clinical exampleJournalBLEEKER2003External validation is necessary in prediction research: a clinical exampleBleeker,S.E.Moll,H.A.Steyerberg,E.W.Donders,A.R.Derksen-Lubsen,G.Grobbee,D.E.Moons,K.G.2003/9Bacterial InfectionsChild,PreschoolClinicalcomplicationsConfidence IntervalsetiologyFever of Unknown OriginHumansInfantInfectionLikelihood FunctionsMedicalmethodsModelsNetherlandsPatientsPredictive Value of TestsResearchResearch SupportNot in File826832J.Clin.Epidemiol.569Erasmus Medical Center/Sophia Children's Hospital Department of Pediatrics, Room Sp 1545 Dr Molewaterplein 60, 3015 GJ Rotterdam, The Netherlands. sebleeker@yahoo.comPM:14505766J.Clin.Epidemiol.1[59], especially when models are developed with the specific intention to apply them in health care. There are two main types of validation: internal validation in the same population or external validation in an independent sample. Internal validation is useful to prevent optimistic assessments, but it does not inform about the performance of the model in other samples of the same population ADDIN REFMGR.CITE Altman2009ALTMAN2009Prognosis and prognostic research: validating a prognostic modelJournalALTMAN2009Prognosis and prognostic research: validating a prognostic modelAltman,D.G.Vergouwe,Y.Royston,P.Moons,K.G.2009Acute DiseaseChildColorectal SurgerycomplicationsCoughHumansMedicineModels,BiologicalMortalityPositive-Pressure RespirationPredictive Value of TestsPrognosisReproducibility of ResultsResearchResearch SupportRisk FactorsStatisticsThoracic Surgical ProceduresUniversitiesNot in Fileb605BMJ338Centre for Statistics in Medicine, University of Oxford, Oxford OX2 6UD. doug.altman@csm.ox.ac.ukPM:19477892British Medical JournalBMJ1[60]. Moreover, many methods of standard internal validation, such as cross-validation, can still give inflated estimates of classification accuracy, even if properly performed. Authors should report whether they performed (internal or external) validation, and describe the procedure of the validation process. For example, for internal validation, authors should describe what part of the population was used to derive the risk model and what part was used for the validation, and whether they, for example, used cross validation and bootstrapping techniques ADDIN REFMGR.CITE Altman2009ALTMAN2009Prognosis and prognostic research: validating a prognostic modelJournalALTMAN2009Prognosis and prognostic research: validating a prognostic modelAltman,D.G.Vergouwe,Y.Royston,P.Moons,K.G.2009Acute DiseaseChildColorectal SurgerycomplicationsCoughHumansMedicineModels,BiologicalMortalityPositive-Pressure RespirationPredictive Value of TestsPrognosisReproducibility of ResultsResearchResearch SupportRisk FactorsStatisticsThoracic Surgical ProceduresUniversitiesNot in Fileb605BMJ338Centre for Statistics in Medicine, University of Oxford, Oxford OX2 6UD. doug.altman@csm.ox.ac.ukPM:19477892British Medical JournalBMJ1[60]. For external validation, they should describe the populations that are used for the validation, particularly the comparability with the population that was used to derive the risk model. If the model is already validated elsewhere in previous research, this should also be stated. So far, none of the genetic risk prediction studies had performed an external validation of the risk model ADDIN REFMGR.CITE Janssens2009JANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJournalJANSSENS2009AGenome-based prediction of common diseases: methodological considerations for future researchJanssens,A.C.J.W.Van Duijn,C.M.2009/2/18ClinicalDiseaseepidemiologyHealthNetherlandsPublic HealthUniversitiesNot in File20Genome Med.12Department of Epidemiology, Erasmus University Medical Center Rotterdam, 3000 CA Rotterdam, The Netherlands. a.janssens@erasmusmc.nlPM:19341491Genome Med.1[3].
Item 11: Specify how missing data were handled.
Examples. Variables with missing values were hypertension (1%), smoking (10%), BMI (14%), plasma HDL cholesterol (19%), plasma LDL cholesterol (20%), and plasma triglycerides (16%). We applied a multiple imputation method (aregImpute function of the R statistical package; version 2.5.1; www.r-project.org) to impute these missing values in our Cox proportional hazards models because imputation decreases bias in the hazard ratios that may occur when patients with incomplete information are excluded from the analysis. In a secondary analysis, we used the full data set (n = 2,145) and multiple imputation to impute both missing values for conventional risk factors and missing genotype data. This analysis gave discriminative accuracies for the 3 prediction models virtually identical to the analysis without imputation of missing genotype data []. ADDIN REFMGR.CITE van der Net2009VANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiaJournalVANDERNET2009AUsefulness of genetic polymorphisms and conventional risk factors to predict coronary heart disease in patients with familial hypercholesterolemiavan der Net,J.B.Janssens,A.C.J.W.Defesche,J.C.Kastelein,J.J.Sijbrands,E.J.G.Steyerberg,E.W.2009/2/1Age of OnsetAgedcomplicationsCoronary DiseaseDiseaseFemaleGenetic Predisposition to DiseaseGenetic VariationgeneticsGenotypehadHealthHumansHyperlipoproteinemia Type IIMaleMiddle AgedNetherlandsPatientsPolymorphism,GeneticProportional Hazards ModelsPublic HealthRiskRisk FactorsNot in File375380Am.J.Cardiol.1033Department of Public Health, Erasmus MC-University Medical Center Rotterdam, Rotterdam, The NetherlandsPM:19166692Am.J.Cardiol.1[36]
Explanation. Missing data are inevitable in observational studies. Authors should specify the percentage of missing values in their data, indicate whether there are theoretical or empirical grounds that missingness could be non-random, and specify how missing data were handled in the analyses. Authors should specify the methods used to deal with the missing data (e.g., complete case analysis, imputation, reweighting) and the assumptions that underlie this choice. Assumptions may include the distribution of the data and whether data were missing completely at random, or related to other variables, including the outcome of the study ADDIN REFMGR.CITE Royston2009ROYSTON2009Prognosis and prognostic research: Developing a prognostic modelJournalROYSTON2009Prognosis and prognostic research: Developing a prognostic modelRoyston,P.Moons,K.G.Altman,D.G.Vergouwe,Y.2009Biomedical ResearchClinicalClinical TrialsData CollectionHumansKidney NeoplasmsLondonMortalityMultivariate AnalysisPrognosisRegression AnalysisResearchResearch SupportRisk AssessmentRoc CurveNot in Fileb604BMJ338MRC Clinical Trials Unit, London NW1 2DA. pr@ctu.mrc.ac.ukPM:19336487British Medical JournalBMJ1[61].
Item 12: Specify all measures used for the evaluation of the risk model including, but not limited to, measures of model fit and predictive ability.
Examples. We calculated odds ratios and 95% confidence intervals associated with each additional risk allele for each SNP individually and in the genotype score. Using C statistics , we evaluated the discriminatory capability of the models with the genotype score as compared with the models without the genotype score. We also evaluated risk reclassification with the use of the genotype score, according to the method developed by Pencina et al. for determining net reclassification improvement. We assessed model calibration using the HosmerLemeshow chi-square test. We used categories of genotype score to calculate likelihood ratios and posterior probabilities of diabetes. Statistical analyses were performed with the use of SAS software, version 8 (SAS Institute). A two-tailed P value of less than 0.05 was considered to indicate statistical significance. ADDIN REFMGR.CITE Meigs2008MEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesJournalMEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesMeigs,J.B.Shrader,P.Sullivan,L.M.McAteer,J.B.Fox,C.S.Dupuis,J.Manning,A.K.Florez,J.C.Wilson,P.W.D'Agostino,R.B.,Sr.Cupples,L.A.2008/11/20AllelesbloodBlood PressureBody Mass IndexBostonCholesterolClinicalDiabetes MellitusFamilyFastingGeneticGenotypehistoryKnowledgeMassachusettsmethodsOdds RatioPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideStatisticsNot in File22082219N.Engl.J.Med.35921General Medicine Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA. jmeigs@partners.orgPM:19020323New England Journal of MedicineN.Engl.J.Med.1[48]
Our primary measure of discrimination was the Harrell c-index, a generalization of the area under the receiver-operating characteristic curve that allows for censored data. The c-index assesses the ability of the risk score to rank women who develop incident cardiovascular disease higher than women who do not. We assessed general calibration across deciles of predicted risk by using the HosmerLemeshow goodness-of-fit test to compare the average predicted risk with the KaplanMeier risk estimate within each decile and considered a chi-square value of 20 or higher (P < 0.01) to be poor calibration. We assessed risk reclassification by sorting the predicted 10-year risk for each model into 4 categories (<5%, 5% to <10%, 10% to <20%, and e"2 0 % ) . W e t h e n c o m p a r e d t h e a s s i g n e d c a t e g o r i e s f o r a p a i r o f m o d e l s . F o r e a c h p a i r , w e c a l c u l a t e d t h e p r o p o r t i o n o f p a r t i c i p a n t s w h o w e r e r e c l a s s i f i e d b y t h e c o m p a r i s o n m o d e l v e r s u s t h e r e f e r e n c e m o d e l ; w e c o n s i d e r e d r e c l a s s i f i c a t i o n t o b e c o r r e c t i f t h e KaplanMeier risk estimate for the reclassified group was closer to the comparison category than the reference. We computed the HosmerLemeshow statistic for the reclassification tables, which assesses agreement between the KaplanMeier risk estimate and predicted risk within the reclassified categories. We also computed the Net Reclassification Improvement, which compares the shifts in reclassified categories by observed outcome, and the Integrated Discrimination Improvement, which directly compares the average difference in predicted risk for women who go on to develop cardiovascular disease with women who do not for the 2 models, on the women who were not censored before 8 years. ADDIN REFMGR.CITE Paynter2009PAYNTER2009Cardiovascular disease risk prediction with and without knowledge of genetic variation at chromosome 9p21.3JournalPAYNTER2009Cardiovascular disease risk prediction with and without knowledge of genetic variation at chromosome 9p21.3Paynter,N.P.Chasman,D.I.Buring,J.E.Shiffman,D.Cook,N.R.Ridker,P.M.2009/1/20bloodBlood PressureBostoncardiovascular diseaseCardiovascular DiseasesCholesterolChromosomes,Human,Pair 9Chronic DiseaseClassificationCohort StudiesDiseaseepidemiologyFamilyFemaleFollow-Up StudiesGeneticGenetic VariationgeneticsGenomeGenotypehadHealthhistoryHumansKaplan-Meiers EstimateKnowledgeMiddle AgedMyocardial InfarctionPatientsPolymorphismPolymorphism,Single NucleotideProbabilityProportional Hazards ModelsProspective StudiesResearchResearch SupportRiskRisk factorRisk FactorsSmokingUnited StatesWomenNot in File6572Ann.Intern.Med.1502Brigham and Women's Hospital, 900 Commonwealth Avenue East, Boston, MA 02215, USA. npaynter@partners.orgPM:19153409Ann.Intern.Med.1[56]
Explanation. A thorough assessment of a risk prediction model comprises many different aspects, but generally includes at least the following questions: (1) How well does the model fit the underlying data?; and (2) What is the predictive ability of the model? Several measures are available to answer each question, and the methods section should clearly describe which measures were used to answer which questions ADDIN REFMGR.CITE Pencina2008PENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondJournalPENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondPencina,M.J.D'Agostino,R.B.,Sr.D'Agostino,R.B.,Jr.Vasan,R.S.2008/1/30AlgorithmsArea Under CurveAssociationBostoncardiovascular diseaseCardiovascular DiseasesClassificationDiseaseepidemiologyetiologyHumansMathematicsModelsModels,StatisticalPaperResearchResearch SupportRiskRisk AssessmentRisk FactorsRoc CurveSensitivity and SpecificityStatisticsstatistics & numerical dataUnited StatesUniversitiesNot in File157172Stat.Med.272Department of Mathematics and Statistics, Framingham Heart Study, Boston University, Boston, MA 02215, USA. mpencina@bu.eduPM:17569110Statistics in MedicineStat.Med.1Janes2008JANES2008Assessing the value of risk predictions by using risk stratification tablesJournalJANES2008Assessing the value of risk predictions by using risk stratification tablesJanes,H.Pepe,M.S.Gu,W.2008/11/18CalibrationClassificationClinicalDiseaseMedicalmethodsModelsModels,StatisticalPopulationResearchRiskRisk AssessmentStatisticalNot in File751760Ann Intern.Med.14910Fred Hutchinson Cancer Research Center, Seattle, Washington 98109, USA. hjanes@scharp.orgPM:19017593Ann Intern.Med.1[4,62]. Measures of model fit (also referred to as calibration) include the Hosmer Lemeshow statistic, R2, log-likelihood and Akaike information criterion (AIC), and measures of predictive ability (also called discrimination measures) include the area under the receiver operating characteristic curve (AUC), discrimination slope and Brier score. These measures can be accompanied by figures and tables, including calibration plots (see in ADDIN REFMGR.CITE Altman2009ALTMAN2009Prognosis and prognostic research: validating a prognostic modelJournalALTMAN2009Prognosis and prognostic research: validating a prognostic modelAltman,D.G.Vergouwe,Y.Royston,P.Moons,K.G.2009Acute DiseaseChildColorectal SurgerycomplicationsCoughHumansMedicineModels,BiologicalMortalityPositive-Pressure RespirationPredictive Value of TestsPrognosisReproducibility of ResultsResearchResearch SupportRisk FactorsStatisticsThoracic Surgical ProceduresUniversitiesNot in Fileb605BMJ338Centre for Statistics in Medicine, University of Oxford, Oxford OX2 6UD. doug.altman@csm.ox.ac.ukPM:19477892British Medical JournalBMJ1[60]), risk distributions (see Figure 1), AUC plots (see Figure 2), discrimination plots (see in ADDIN REFMGR.CITE Steyerberg2010STEYERBERG2010Assessing the Performance of Prediction Models: A Framework for Traditional and Novel MeasuresJournalSTEYERBERG2010Assessing the Performance of Prediction Models: A Framework for Traditional and Novel MeasuresSteyerberg,E.W.Vickers,A.J.Cook,N.R.Gerds,T.Gonen,M.Obuchowski,N.Pencina,M.J.Kattan,M.W.2010/1Behavioral & Social SciencesClinicalmethodsModelsPatientsRoleStatisticsNot in File128138Epidemiology211Epidemiology1[63]) and predictiveness curves (see in ADDIN REFMGR.CITE Pepe2008PEPE2008Integrating the predictiveness of a marker with its performance as a classifierJournalPEPE2008Integrating the predictiveness of a marker with its performance as a classifierPepe,M.S.Feng,Z.Huang,Y.Longton,G.Prentice,R.Thompson,I.M.Zheng,Y.2008/2/1analysisBiologicalBiological MarkersbloodClassificationDiseaseEpidemiologic MethodsHumansLogistic ModelsMalemethodsModelsModels,TheoreticalPopulationPredictive Value of TestsProstate-Specific AntigenResearchResearch SupportRiskRisk AssessmentRisk factorRisk FactorsRoc CurveSensitivity and SpecificityStatisticalNot in File362368Am J Epidemiol.1673Fred Hutchinson Cancer Research Center, Seattle, WA 98109, USA. mspepe@u.washington.eduPM:17982157Am J Epidemiol.1[64]). The description of the methods used should clarify also what measures of uncertainty are employed (e.g., 95% confidence intervals) and specify any tests used to determine the significance of the findings. When p-values are reported, authors should indicate what p-value threshold they considered for statistical significance.
When two risk models are compared and one is an expanded version of the other, the assessment of the risk models includes the two questions for each model. Increases in AUC or in discrimination slope (called integrated discrimination improvement, IDI) provide simple ways to assess improvement of one model over the other [58]. Recent studies have also assessed whether the improvement of risk models also reclassifies people into different risk categories ADDIN REFMGR.CITE Hlatky2009HLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationJournalHLATKY2009Criteria for evaluation of novel markers of cardiovascular risk: a scientific statement from the American Heart AssociationHlatky,M.A.Greenland,P.Arnett,D.K.Ballantyne,C.M.Criqui,M.H.Elkind,M.S.Go,A.S.Harrell,F.E.,Jr.Hong,Y.Howard,B.V.Howard,V.J.Hsue,P.Y.Kramer,C.M.McConnell,J.P.Normand,S.L.O'Donnell,C.J.Smith,S.C.,Jr.Wilson,P.W.2009/5/5American Heart AssociationanalysisAssociationBiologicalBiological Markerscardiovascular diseaseCardiovascular DiseasesClinicaldiagnosisDiseaseEvaluation Studies as TopicHumansmethodsPopulationPrognosisResearchResearch DesignRiskRisk AssessmentSensitivity and SpecificitystandardsStatisticalUnited StatesNot in File24082416Circulation11917PM:19364974Circulation1Janssens2010JANSSENS2010Assessment of improved prediction beyond traditional risk factors: when does a difference make a difference?JournalJANSSENS2010Assessment of improved prediction beyond traditional risk factors: when does a difference make a difference?Janssens,A.C.J.W.Khoury,M.J.2010RiskRisk FactorsNot in File35Circ: Cardiovasc Genet3Circ: Cardiovasc Genet1[2,65]. These measures of reclassification, such as the percentage of total reclassification and net reclassification improvement ADDIN REFMGR.CITE Pencina2008PENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondJournalPENCINA2008Evaluating the added predictive ability of a new marker: from area under the ROC curve to reclassification and beyondPencina,M.J.D'Agostino,R.B.,Sr.D'Agostino,R.B.,Jr.Vasan,R.S.2008/1/30AlgorithmsArea Under CurveAssociationBostoncardiovascular diseaseCardiovascular DiseasesClassificationDiseaseepidemiologyetiologyHumansMathematicsModelsModels,StatisticalPaperResearchResearch SupportRiskRisk AssessmentRisk FactorsRoc CurveSensitivity and SpecificityStatisticsstatistics & numerical dataUnited StatesUniversitiesNot in File157172Stat.Med.272Department of Mathematics and Statistics, Framingham Heart Study, Boston University, Boston, MA 02215, USA. mpencina@bu.eduPM:17569110Statistics in MedicineStat.Med.1Cook2009COOK2009Advances in measuring the effect of individual predictors of cardiovascular risk: the role of reclassification measuresJournalCOOK2009Advances in measuring the effect of individual predictors of cardiovascular risk: the role of reclassification measuresCook,N.R.Ridker,P.M.2009/6/2Blood PressureBostoncardiovascular diseaseCardiovascular DiseasesClinicalDiseaseepidemiologyFemaleHumansMedicalmethodsMiddle AgedModelsProportional Hazards ModelsResearchResearch SupportRiskRisk AssessmentRisk FactorsRoleWomenNot in File795802Ann.Intern.Med.15011Donald W. Reynolds Center for Cardiovascular Research and the Center for Cardiovascular Disease Prevention, Brigham and Women's Hospital, Harvard Medical School, 900 Commonwealth Avenue East, Boston, MA 02215, USA. ncook@rics.bwh.harvard.eduPM:19487714Ann.Intern.Med.1[4,66], are calculated from a reclassification table (Table 7). When risk categories are used (e.g., for the calculation of reclassification measures), the rationale for the cut-off values should be presented with either appropriate reference to previous work showing that this is a standard choice, or appropriate justification for the choice of cut-offs made by the authors. When several different cut-off categorizations have been studied, all of them should be reported.
Item 13: Describe all subgroups, interactions and exploratory analyses that were examined
Examples. (Subgroups) [In introduction:] However, it remains unknown whether all these genetic and environmental factors act independently or jointly and to what extent they as a group can predict the occurrence of AMD or progression to advanced AMD from early and intermediate stages. Such information may be useful for screening those at high risk due to a positive family history or having signs of early or intermediate disease, among whom some progress to advanced stages of AMD with visual loss. [In Methods:] Individuals with advanced AMD were compared to the control group of persons with no AMD, and progressors were compared to nonprogressors
with regard to genotype and risk factor data. ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1[6]
(Interactions) Multiplicative interactions were tested for each pair of [all 6] SNPs by including both main effects and an interaction term (a product of two main effects) in a logistic regression model. ADDIN REFMGR.CITE Zheng2008ZHENG2008Cumulative association of five genetic variants with prostate cancerJournalZHENG2008Cumulative association of five genetic variants with prostate cancerZheng,S.L.Sun,J.Wiklund,F.Smith,S.Stattin,P.Li,G.Adami,H.O.Hsu,F.C.Zhu,Y.Balter,K.Kader,A.K.Turner,A.R.Liu,W.Bleecker,E.R.Meyers,D.A.Duggan,D.Carpten,J.D.Chang,B.L.Isaacs,W.B.Xu,J.Gronberg,H.2008/1/16analysisAssociationBostondiagnosisepidemiologyFamilyGeneticGenomicshadHealthhistoryHumanMassachusettsMedicalmethodsMultivariate AnalysisOdds RatioPolymorphismPopulationPublic HealthResearchScienceSingle NucleotideSwedenUniversitiesNot in File910919N.Engl.J.Med.358From the Center for Human Genomics (S.L.Z., J.S., S.S., G.L., F.-C.H., Y.Z., A.R.T., W.L., E.R.B., D.A.M., B.-L.C., J.X.) and the Departments of Biostatistical Sciences (F.-C.H.) and Urology (A.K.K.), Wake Forest University School of Medicine, Winston-Salem, NC; the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm (F.W., H.-O.A., K.B., H.G.); the Department of Urology, Umea University Hospital, Umea, Sweden (P.S.); the Department of Epidemiology, Harvard School of Public Health, Boston (H.-O.A.); Translational Genomics Research Institute, Phoenix, AZ (D.D., J.D.C.); and Johns Hopkins Medical Institutions, Baltimore (W.B.I.). This article (10.1056/NEJMoa075819) was published at www.nejm.org on January 16, 2008. It will appear in the February 28 issue of the JournalPM:18199855New England Journal of MedicineN.Engl.J.Med.1[67]
Explanation. For the evaluation of the predictive performance there might be subgroups in which the risk model performs better than in the initial study population, and there might be genetic variants that jointly have a larger impact on disease risk. The large number of possible analyses that include subgroups or interactions, however, increases the likelihood of finding at least some statistically significant effect by chance ADDIN REFMGR.CITE Oxman1992OXMAN1992A consumer's guide to subgroup analysesJournalOXMAN1992A consumer's guide to subgroup analysesOxman,A.D.Guyatt,G.H.1992/1/1analysisCausalityDecision Support TechniquesGuidelinesHealthMeta-AnalysisMeta-Analysis as TopicOntarioPaperRandomized Controlled Trials as TopicResearchResearch DesignResearch SupportScienceStatisticalUniversitiesNot in File7884Ann Intern.Med.1161McMaster University Health Sciences Centre, Hamilton, OntarioPM:1530753Ann Intern.Med.1[68]. Authors should therefore not only clarify all additional subgroup analyses they performed, but also indicate whether these were planned based on a priori clinical or epidemiological evidence, or arose in an exploratory fashion. Similarly, authors should also explain whether interaction effects were considered and, if so, which ones and why, and how the selection in the final model was done (see item 9). These descriptions should include any methods used to prevent over interpretation of the results, e.g., methods that adjust the p-value thresholds to adjust for multiple testing. Planned analyses of subgroups and interactions should logically follow from the introduction (see item 3); exploratory analyses can be introduced in the methods.
RESULTS
Item 14: Report the numbers of individuals at each stage of the study. Give reasons for non-participation at each stage. Report the number of participants not genotyped, and reasons why they were not genotyped.
Examples. Among 3648 identified subjects with prostate cancer, 3161 (87%) agreed to participate. DNA samples from blood, tumornodemetastasis (TNM) stage, Gleason grade (as determined by biopsy), and levels of prostate-specific antigen (PSA) at diagnosis were available for 2893 subjects (92%). ADDIN REFMGR.CITE Zheng2008ZHENG2008Cumulative association of five genetic variants with prostate cancerJournalZHENG2008Cumulative association of five genetic variants with prostate cancerZheng,S.L.Sun,J.Wiklund,F.Smith,S.Stattin,P.Li,G.Adami,H.O.Hsu,F.C.Zhu,Y.Balter,K.Kader,A.K.Turner,A.R.Liu,W.Bleecker,E.R.Meyers,D.A.Duggan,D.Carpten,J.D.Chang,B.L.Isaacs,W.B.Xu,J.Gronberg,H.2008/1/16analysisAssociationBostondiagnosisepidemiologyFamilyGeneticGenomicshadHealthhistoryHumanMassachusettsMedicalmethodsMultivariate AnalysisOdds RatioPolymorphismPopulationPublic HealthResearchScienceSingle NucleotideSwedenUniversitiesNot in File910919N.Engl.J.Med.358From the Center for Human Genomics (S.L.Z., J.S., S.S., G.L., F.-C.H., Y.Z., A.R.T., W.L., E.R.B., D.A.M., B.-L.C., J.X.) and the Departments of Biostatistical Sciences (F.-C.H.) and Urology (A.K.K.), Wake Forest University School of Medicine, Winston-Salem, NC; the Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm (F.W., H.-O.A., K.B., H.G.); the Department of Urology, Umea University Hospital, Umea, Sweden (P.S.); the Department of Epidemiology, Harvard School of Public Health, Boston (H.-O.A.); Translational Genomics Research Institute, Phoenix, AZ (D.D., J.D.C.); and Johns Hopkins Medical Institutions, Baltimore (W.B.I.). This article (10.1056/NEJMoa075819) was published at www.nejm.org on January 16, 2008. It will appear in the February 28 issue of the JournalPM:18199855New England Journal of MedicineN.Engl.J.Med.1[67]
[In methods:] In short, the Rotterdam Study is a prospective, population based, cohort study among 7,983 inhabitants of a Rotterdam suburb, designed to investigate determinants of chronic diseases [In Results:] A total of 6,544 participants were successfully genotyped for at least one polymorphism. Complete genotype information on all polymorphisms was present in 5,297 subjects (of whom 490 were incident cases and 545 were prevalent cases). ADDIN REFMGR.CITE van Hoek2008VANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyJournalVANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyvan Hoek,M.Dehghan,A.Witteman,J.C.Van Duijn,C.M.Uitterlinden,A.G.Oostra,B.A.Hofman,A.Sijbrands,E.J.Janssens,A.C.2008/11ADAM ProteinsAgedCation Transport ProteinsClinicalCyclin-Dependent Kinase 5Cyclin-Dependent Kinase Inhibitor p15Cyclin-Dependent Kinase Inhibitor p16Diabetes Mellitus,Type 2ethnologyEuropean Continental Ancestry GroupGenetic Predisposition to DiseaseGenetic TestinggeneticsGenome,HumanGenome-Wide Association StudyHumanHumansMembrane ProteinsmethodsMiddle AgedNeoplasm ProteinsNetherlandsPolymorphism,GeneticProspective StudiesProteinsResearch DesignRiskRisk FactorsRNA-Binding ProteinsTCF Transcription FactorsUniversitiesNot in File31223128Diabetes5711Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, the NetherlandsPM:18694974Diabetes1[42]
Explanation. The study report should clearly present the number of participants that were eligible for the study and how many were included in the final analyses. The authors should report the main reasons for non-participation, so that the reader can judge the extent to which the population available for the analyses is a representative selection of those who were eligible. Any evidence for missingness not completely at random should be presented ADDIN REFMGR.CITE Little1987LITTLE1987Statistical analysis with missing dataBook, WholeLITTLE1987Statistical analysis with missing dataLittle,R.J.A.Rubin,D.B.1987Not in FileNew YorkJohn Wiley & Sonsdefinities missing at random etc.2[69]. A flowchart can help clarify complex datasets, and is particularly useful for follow-up studies. A flowchart presents the exact numbers and the structure of the study (see example in ADDIN REFMGR.CITE Vandenbroucke2007VANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationJournalVANDENBROUCKE2007AStrengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaborationVandenbroucke,J.P.von Elm,E.Altman,D.G.Gotzsche,P.C.Mulrow,C.D.Pocock,S.J.Poole,C.Schlesselman,J.J.Egger,M.2007/10/16Case-Control StudiesClinicalCohort StudiesCross-Sectional StudiesEpidemiologic Research DesignepidemiologyGuidelines as TopicMedicalmethodsNetherlandsObservationPublishingResearchResearch SupportstandardsUniversitiesNot in Filee297PLoS.Med.410Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The NetherlandsPM:17941715PLoS.Med.1[29]). When a flowchart of the study has been previously published and the flow of participants is the same, a reference to the earlier publication can save space. For cohort studies, descriptive information about the follow-up time, e.g., in terms of the range, median and interquartile range of follow-up duration, should be provided.
Frequently, studies do not have complete genotype information for all participants for many reasons, including budget issues, unavailability of DNA material and genotyping quality issues. Because some reasons might impact the validity of the study, the number of participants that were not genotyped and the reasons should be reported. An example is survivor bias, which might occur when genotyping is performed on DNA obtained in one of the follow-up assessments of a cohort study (see example ADDIN REFMGR.CITE Talmud2010TALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyJournalTALMUD2010Utility of genetic and non-genetic risk factors in prediction of type 2 diabetes: Whitehall II prospective cohort studyTalmud,P.J.Hingorani,A.D.Cooper,J.A.Marmot,M.G.Brunner,E.J.Kumari,M.Kivimaki,M.Humphries,S.E.2010AllelesArea Under CurveBody Mass IndexCalibrationCholesterolCohort StudiesdiagnosisDiseaseFamilyFastingGeneticgeneticsGenotypeGlucose Tolerance TesthistoryLondonMedicalMedicineModelsOdds RatioPhenotypePolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideSmokingTriglyceridesUniversitiesWomenNot in Fileb4838BMJ340Centre of Cardiovascular Genetics, Department of Medicine, University College London, London WC1E 6JF. p.talmud@ucl.ac.ukPM:20075150British Medical JournalBMJ1[52]).
Item 15: Report demographic and clinical characteristics of the study population, including risk factors used in the risk modeling.
Examples. The mean age of cases was similar to that of controls, 59.9 and 59.6 years, respectively. In comparison with controls, a higher proportion of cases had a first-degree family history of prostate cancer (see Table 4). The majority of cases had serum PSA values of 4.09.9 ng/ml at diagnosis, localized stage disease and Gleason
scores of 5 or 6; most were treated with radical prostatectomy. ADDIN REFMGR.CITE Salinas2009SALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalityJournalSALINAS2009Clinical utility of five genetic variants for predicting prostate cancer risk and mortalitySalinas,C.A.Koopmeiners,J.S.Kwon,E.M.FitzGerald,L.Lin,D.W.Ostrander,E.A.Feng,Z.Stanford,J.L.2009/3/1AdultAgedCase-Control StudiesChromosomes,Human,Pair 17Chromosomes,Human,Pair 8ClinicalDiseaseFamilyGeneticGenetic Predisposition to DiseasegeneticsGenotypehadhistoryHumansLogistic ModelsMalemethodsMiddle AgedModelsMortalityOdds RatioPolymorphismPolymorphism,Single NucleotidePopulationPredictive Value of TestsPrognosisProportional Hazards ModelsProstatic NeoplasmsResearchResearch SupportRiskRisk factorRisk FactorsRoc CurveSingle NucleotideNot in File363372Prostate694Fred Hutchinson Cancer Research Center, Seattle, Washington 98109-1024, USAPM:19058137Prostate1[31]
Explanation. The authors should describe their populations in as much detail as is needed for the readers to judge the generalizability of the results. This description should include relevant demographic information, such as age, sex and ethnicity, and information on other risk factors and relevant pathology, e.g., early disease characteristics and comorbidity. Continuous variables are preferably described by means and standard deviations, and when their distributions are skewed, by medians and inter-quartile ranges. Variables that have a small number of response categories are preferably presented as percentages and numbers. This descriptive information is preferably presented separately for those people with and without the outcome of interest.
Item 16: Report unadjusted associations between the variables in the risk model(s) and the outcome. Report adjusted estimates and their precision from the full risk model(s) for each variable.
Examples. Table 5 displays the unadjusted association between demographic, environmental, and genetic variables and incident advanced AMD as well as the sample sizes within the groups. All factors except gender were related to progression. Baseline macular status was strongly related to progression. Both modifiable factors (smoking and BMI) and genetic variants were also associated with worsening of macular disease over time. The antioxidant/mineral treatment group had a lower rate of progression. Table 6 displays the multivariate adjusted ORs for incident advanced AMD and shows that, after adjustment for genotypes, older age, smoking, and higher BMI were related to a higher rate of progression. Baseline grade was a strong predictor of incident advanced AMD, and antioxidantmineral treatment was protective. The two CFH variants each independently increased risk of progression about two- to threefold, with similar increased risk for C3, comparing the homozygous risk and nonrisk genotypes. Variants in the two complement genes C2 and CFB reduced risk, although the association with CFB was not significant for progression to incident advanced AMD. ADDIN REFMGR.CITE Seddon2009SEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesJournalSEDDON2009Prediction model for prevalence and incidence of advanced age-related macular degeneration based on genetic, demographic, and environmental variablesSeddon,J.M.Reynolds,R.Maller,J.Fagerness,J.A.Daly,M.J.Rosner,B.2009/5AgedAlgorithmsAllelesanalysisAtrophyBody Mass IndexBostonComplement C2Complement C3Complement Factor BComplement Factor HDiseaseDnaEducationEnvironmentepidemiologyEpidemiology,MolecularFemaleGenesGeneticgeneticsGenotypeHumanHumansIncidenceMacular DegenerationMaleMassachusettsMedicalmethodsModelsModels,GeneticMulticenter StudiesOdds RatioPhenotypePolymorphismPolymorphism,GeneticPolymorphism,Single NucleotidePrevalenceProspective StudiesProteinsResearchResearch SupportRiskRisk FactorsRoc CurveSmokingSpectrometry,Mass,Matrix-Assisted Laser Desorption-IonizationStatisticsUniversitiesNot in File20442053Invest Ophthalmol.Vis.Sci.505Ophthalmic Epidemiology and Genetics Service, Tufts University School of Medicine and Tufts Medical Center, Boston, Massachusetts 02111, USA. jseddon@tuftsmedicalcenter.orgPM:19117936Invest Ophthalmol.Vis.Sci.1[6]
Explanation. To understand which risk factors have contributed to the distribution in risk predictions, authors should report model estimates for each, e.g., regression coefficients, such as odds ratios or hazard ratios, and confidence intervals from each full model considered for all risk factors included. Adjusted estimates should be presented next to the unadjusted estimates, so that readers are able to judge the extent to which the findings change by the inclusion of other risk factors in the model. This is particularly relevant for models that combine genetic and non-genetic risk factors, because non-genetic risk factors can be intermediate factors in the biological pathway ADDIN REFMGR.CITE Janssens2008JANSSENS2008AGenome-based prediction of common diseases: advances and prospectsJournalJANSSENS2008AGenome-based prediction of common diseases: advances and prospectsJanssens,A.C.J.W.Van Duijn,C.M.2008DiseaseNot in FileR166R173Hum Mol Genet17Hum Mol Genet1[41] and many non-genetic risk factors have complex correlation patterns ADDIN REFMGR.CITE Ioannidis2009IOANNIDIS2009AResearching genetic versus nongenetic determinants of disease: a comparison and proposed unificationJournalIOANNIDIS2009AResearching genetic versus nongenetic determinants of disease: a comparison and proposed unificationIoannidis,J.P.Loy,E.Y.Poulton,R.Chia,K.S.2009/11/18ClinicalDiseaseepidemiologyGeneticGenotypeGreeceMedicineMolecularMolecular EpidemiologyResearchResearch DesignstandardsUniversitiesNot in File7ps8Sci.Transl.Med.17Clinical and Molecular Epidemiology Unit, Department of Hygiene and Epidemiology, University of Ioannina School of Medicine, Ioannina, Greece. jioannid@cc.uoi.grPM:20368180Sci.Transl.Med.1Smith2007SMITH2007Clustered environments and randomized genes: a fundamental distinction between conventional and genetic epidemiologyJournalSMITH2007Clustered environments and randomized genes: a fundamental distinction between conventional and genetic epidemiologySmith,G.D.Lawlor,D.A.Harbord,R.Timpson,N.Day,I.Ebrahim,S.2007/12AgedAssociationCluster AnalysisConfounding Factors (Epidemiology)Cross-Sectional StudiesDiseaseEnvironmentEnvironmental ExposureEpidemiologic MethodsepidemiologyetiologyFemaleGenesGeneticGenetic Predisposition to DiseaseGenetic VariationgeneticsGenotypeGreat BritainHealth BehaviorHeart DiseasesHumansLife StyleMedicinemethodsMiddle AgedMulticenter StudiesPhenotypephysiopathologyPolymorphism,Single NucleotideReproducibility of ResultsResearchResearch SupportRiskRisk AssessmentRisk factorRisk FactorsSocioeconomic FactorsUniversitiesNot in Filee352PLoS.Med.412Department of Social Medicine, University of Bristol, Bristol, United Kingdom. george.davey-smith@bristol.ac.ukPM:18076282PLoS.Med.1[70,71]. Note that several studies have presented adjusted effect sizes for genetic variants (e.g., ADDIN REFMGR.CITE van Hoek2008VANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyJournalVANHOEK2008Predicting type 2 diabetes based on polymorphisms from genome-wide association studies: a population-based studyvan Hoek,M.Dehghan,A.Witteman,J.C.Van Duijn,C.M.Uitterlinden,A.G.Oostra,B.A.Hofman,A.Sijbrands,E.J.Janssens,A.C.2008/11ADAM ProteinsAgedCation Transport ProteinsClinicalCyclin-Dependent Kinase 5Cyclin-Dependent Kinase Inhibitor p15Cyclin-Dependent Kinase Inhibitor p16Diabetes Mellitus,Type 2ethnologyEuropean Continental Ancestry GroupGenetic Predisposition to DiseaseGenetic TestinggeneticsGenome,HumanGenome-Wide Association StudyHumanHumansMembrane ProteinsmethodsMiddle AgedNeoplasm ProteinsNetherlandsPolymorphism,GeneticProspective StudiesProteinsResearch DesignRiskRisk FactorsRNA-Binding ProteinsTCF Transcription FactorsUniversitiesNot in File31223128Diabetes5711Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, the NetherlandsPM:18694974Diabetes1Meigs2008MEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesJournalMEIGS2008Genotype score in addition to common risk factors for prediction of type 2 diabetesMeigs,J.B.Shrader,P.Sullivan,L.M.McAteer,J.B.Fox,C.S.Dupuis,J.Manning,A.K.Florez,J.C.Wilson,P.W.D'Agostino,R.B.,Sr.Cupples,L.A.2008/11/20AllelesbloodBlood PressureBody Mass IndexBostonCholesterolClinicalDiabetes MellitusFamilyFastingGeneticGenotypehistoryKnowledgeMassachusettsmethodsOdds RatioPolymorphismResearchResearch SupportRiskRisk factorRisk FactorsSingle NucleotideStatisticsNot in File22082219N.Engl.J.Med.35921General Medicine Division, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA. jmeigs@partners.orgPM:19020323New England Journal of MedicineN.Engl.J.Med.1Plat2009PLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectJournalPLAT2009The contribution of six polymorphisms to cardiovascular risk in a Dutch high-risk primary care population: the HIPPOCRATES projectPlat,A.W.Stoffers,H.E.J.H.Klungel,O.H.van Schayck,C.P.de Leeuw,P.W.Soomers,F.L.Schiffers,P.M.Kester,A.D.M.Kroon,A.A.2009PolymorphismRiskPopulationNot in File659667J Hum Hypertens23J Hum Hypertens1[42,48,51]) that were adjusted only for non-genetic risk factors. This is not the same as effect sizes for genetic variants from the full model, where coefficients are additionally adjusted for the other genetic variants as well. When regression methods were used for the prediction of risks, the intercept of the full model should be reported to facilitate future replication and validation of the risk model (see Table 6). For complex models where exhaustive specification of parameter estimates is not feasible, authors should provide software implementations of the risk prediction algorithm (see item 24).
Item 17: Report distributions of predicted risks and/or risk scores.
Examples. The distribution of predicted risks or risk scores is best presented in a figure, see Figure 1 ADDIN REFMGR.CITE Lango2008LANGO2008Assessing the combined impact of 18 common genetic variants of modest effect sizes on type 2 diabetes riskJournalLANGO2008Assessing the combined impact of 18 common genetic variants of modest effect sizes on type 2 diabetes riskLango,H.Palmer,C.N.Morris,A.D.Zeggini,E.Hattersley,A.T.McCarthy,M.I.Frayling,T.M.Weedon,M.N.2008/6/30AllelesAssociationClinicalDiseaseGeneticgeneticshadhigherMedicalmethodsOdds RatioPopulationResearchResearch DesignRiskScienceNot in File31293135Diabetes5711Genetics of Complex Traits, Institute of Biomedical and Clinical Science, Peninsula Medical School, Magdalen Road, Exeter, UKPM:18591388Diabetes1[45].
Explanation. Distributions of predicted risks inform the reader about the spread of risks in the population, as well as the frequencies at the higher and lower ends of the distribution. Preferably the report should present separate distributions for participants with and those without the outcome of interest, as this illustrates the discriminative accuracy of the risk model. The more the two distributions disperse, the higher the AUC. Authors should label the highest and lowest category by their actual range at least once. For example, Figure 1 shows that the lowest category is labeled 10-11 risk alleles, rather than 0-11, which informs readers that none of the participants had 0 to 9 risk alleles.
Item 18: Report measures of model fit and predictive ability, and any other performance measures, if pertinent.
Examples. We also evaluated whether genetic risk factors would further increase the risk imposed by an increase in the BMI or a decrease in the disposition index. There was a stepwise increase in diabetes risk with an increasing number of risk alleles and increasing quartiles of BMI (Fig. [not shown]) or a disposition index above or below the median. Therefore, carriers of more than 12 risk alleles who were in the highest quartile of BMI (263 of 826 subjects vs. 45 of 874 s u b j e c t s ) o r w h o h a d a l o w d i s p o s i t i o n i n d e x ( 5 8 o f 1 5 3 s u b j e c t s v s . 1 7 o f 1 6 8 s u b j e c t s ) h a d a n o d d s r a t i o f o r t y p e 2 d i a b e t e s o f 8 . 0 ( 9 5 % C I , 5 . 7 1 t o 1 1 . 1 9 ; P = 9 . 1 1 0 "3 4 ) a n d 5 . 8 ( 9 5 % C I , 3 . 1 8 t o 1 0 . 6 1 , P = 1 . 1 1 0 "8 ) , r e s p e c t i v e l y ( F i g . [ n o t s h o w n ] ) . T h e C s t a t i s t i c s h a d m i n i m a l y e t s i g n i f i c a n t i m p r o v e m e n t a f t e r t h e a d d i t i o n o f d a t a f r o m t h e g e n o t y p e d D N A v a r i a n t s t o t h e c l i n i c a l m o d e l ( f r o m 0 . 7 4 t o 0 . 7 5 , P = 1 . 0 1 0 "4 ) ( S u p p l e m e n t a r y T a b l e [ n o t s h o w n ] ) . & w e a l s o r e c l a s s i f i e d s u b j e c t s i n t o t h r e e r i s k c a t e g o r i e s ( 0 t o d"1 0 % , > 1 0 t o d"2 0 % , a n d > 2 0 % ) u s i n g t h e n e t - r e c l a s s i f i c a t i o n - i m p r o v e m e n t m e t h o d ( S u p p l e m e n t a r y T a b l e [ n o t s h o w n ] ) . B y a d d i n g g e n e t i c f a c t o r s t o c l i n i c a l f a c t o r s , w e c o u l d r e c l a s s i f y 9 % o f t h e M P P s u b j e c t s ( P = 2 . 5 1 0 "5 ) a n d 2 0 % o f t h e B o t n i a s u bjects (P = 0.05) to a higher risk category. Also, the use of the integrated-discrimination- improvement method, which did not require predefined risk categories, significantly improved the prediction of future diabetes in both the MPP subjects (P = 3.710 "1 4 ) a n d t h e B o t n i a s u b j e c t s ( P = 0 . 0 0 1 ) . A D D I N R E F M G R . C I T E <