Determining relative importance of variables in developing and validating predictive models

BMC Medical Research Methodology

Volumn 9, Issue 1, 2009, Pages

  Beyene, Joseph ^a,b   Atenafu, Eshetu G ^a   Hamid, Jemila S ^a   To, Teresa ^a,b   Sung, Lillian ^a,b  

^a HOSPITAL FOR SICK CHILDREN   (Canada)

^b UNIVERSITY OF TORONTO   (Canada)

Author keywords

[No Author keywords available]

Indexed keywords

ACCURACY; ACUTE LYMPHOBLASTIC LEUKEMIA; ARTICLE; BOOTSTRAPPING; HUMAN; PREDICTOR VARIABLE; PROSTATE CANCER; REPRODUCIBILITY; ROC CURVE; STATISTICAL MODEL; TUMOR LYSIS SYNDROME; VALIDATION PROCESS; CHILD; COMPARATIVE STUDY; FEMALE; MALE; METHODOLOGY; PREDICTION AND FORECASTING; RECEIVER OPERATING CHARACTERISTIC; REGRESSION ANALYSIS; RISK ASSESSMENT; RISK FACTOR; STATISTICS; VALIDATION STUDY;

CHILD; FEMALE; HUMANS; MALE; MODELS, STATISTICAL; PRECURSOR CELL LYMPHOBLASTIC LEUKEMIA-LYMPHOMA; PREDICTIVE VALUE OF TESTS; REGRESSION ANALYSIS; REPRODUCIBILITY OF RESULTS; RISK ASSESSMENT; RISK FACTORS; ROC CURVE; TUMOR LYSIS SYNDROME;

EID: 70350338235     PISSN: None     EISSN: 14712288     Source Type: Journal    
DOI: 10.1186/1471-2288-9-64     Document Type: Article

References (34)

1. Statistics in medical journals: development in the 1980s. DG Altman, Stat Med 1991 10 1897 913 10.1002/sim.4780101206 1805317
2. FE Harell, Regression modeling strategies: with application to linear models, logistic regression and survival analysis Springer-Verlag, New York 2001
3. Logistic regression models in medical research are poorly presented. R Bender U Grouven, BMJ 1996 313 628 8806274
4. What do you mean by validating a prognostic model? DG Altman P Royston, Stat Med 2000 19 4 453 73 10.1002/(SICI)1097-0258(20000229)19:4<453::AID- SIM350>3.0.CO;2-5 10694730
5. Features at presentation predict children with acute lymphoblastic leukemia at low risk for tumor lysis syndrome. TH Truong J Beyene J Hitzler O Abla AM Maloney S Weitzman L Sung, Cancer 2007 110 8 1832 9 10.1002/cncr.22990 17724682
6. A predictive model for detection of tumor lysis syndrome during AML induction therapy. AR Mato BE Riccio L Qin DF Heitjan M Carroll A Loren DL Porter A Perl E Stadtmauer D Tsai A Gewirtz SM Luger, Leukemia and Lymphoma 2006 47 5 877 883 10.1080/10428190500404662 16753873
7. Tumor classification by partial least squares using microarray gene expression data. DV Nguyen DM Rocke, Bioinformatics 2002 18 1 39 50 10.1093/bioinformatics/18.1.39 11836210
8. Predictive models for breast cancer susceptibility from single nucleotide polymorphisms. J Listgarten S Damaraju B Poulin L Cook J Dufour A Driga J Mackey D Wishart R Greiner B Zanke, Clinical Cancer Research 2004 10 2725 2737 10.1158/1078-0432.CCR-1115-03 15102677
9. Integrated modeling of clinical and gene expression information for personalized prediction of disease out comes. J Pittman E Huang H Dressman CF Horng SH Cheng MH Tsou CM Chen A Bild ES Iversen AT Huang JR Nevins M West, PNAS 2004 101 22 8431 36 10.1073/pnas.0401736101 15152076
10. A prediction model for lung cancer diagnosis that integrates genomic and clinical features. J Beane P Sebastiani TH Whitfield K Steiling YM Dumas ME Lenburg A Spira, Cancer Prevention Research 2008 1 56 64 10.1158/1940-6207.CAPR- 08-0011 19138936
11. Towards integrated clinic-genomic models for personalized medicine: combining gene expression signatures and clinical factors in breast cancer outcomes prediction. JR Nevins ES Huang H Dressman J Pittman AT Huang M West, Human Molecular Genetics 2003 12 2 153 7 10.1093/hmg/ddg287 12928487
12. Evaluation of logistic regression reporting in current obstetrics and gynecology literature. RT Mikolajczyk A Disilvesto J Zhang, Obstetrics and Gynecology 2008 111 2 Pt 1 413 9 18238980
13. An appraisal of multivariable logistic models in pulmonary and critical care literature. M Moss AD Wellman GA Cotsonis, Chest 2003 123 923 928 10.1378/chest.123.3.923 12628895
14. The risk of determining risk with multivariable models. J Concato AR Feinstein TR Holford, Ann Intern Med 1993 118 201 210 8417638
15. Statistical reviewing policies of medical journals: caveat lector? SN Goodman DG Altman SL George, J Gen Intern Med 1998 13 753 756 10.1046/j.1525-1497.1998.00227.x 9824521
16. Predictive accuracy and explained variation. M Schemper, Stat Med 2003 22 2299 2308 10.1002/sim.1486 12854094
17. Internal Validation of Predictive Models: Efficiency of some procedures for logistic regression analysis. EW Steyerberg FE Harrell Jr GJ Borsboom MJ Eijkemans Y Vergouwe JD Habbema, Journal of Clinical Epidemiology 2001 54 8 774 781 10.1016/S0895-4356(01)00341-9 11470385
18. Methods of variable selection in regression modeling. PA Murtagugh, Communications in Statistics - Simulation and Computation 1998 27 711 734 10.1080/03610919808813505
19. Backward, forward and stepwise automated subset selection algorithms: Frequency of obtaining authentic and noise variables. S Derksen HJ Keselman, British Journal of Mathematical and Statistical Psychology 1992 45 265 282
20. Can the strengths of AIC and BIC be shared? A conflict between model indentification and regression estimation. Y Yang, Biometrica 2005 92 4 937 950 10.1093/biomet/92.4.937 (Pubitemid 41681244)
21. Statistical Modeling: The two cultures. L Breiman, Statistical Science 2001 16 3 199 231 10.1214/ss/1009213726
22. DW Hosmer S Lemeshow, Applied Logistic Regression Wiley, New York Second 2000
23. AC Davison D Hinkley, Bootstrap Methods and their Applications Cambridge: Cambridge Series in Statistical and Probabilistic Mathematics 1997
24. Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. J Carpenter J Bithell, Stat Med 2000 19 1141 1164 10.1002/(SICI)1097-0258(20000515)19:9<1141::AID-SIM479>3.0.CO;2-F 10797513
25. A bootstrap resampling procedure for model building: Applicatioon to the Cox regression model. W Sauerbrei M Schumacher, Statistics in Medicine 1992 11 2093 2109 10.1002/sim.4780111607 1293671
26. Bootstrap methods for developing predictive models. PC Austin JV Tu, The American Statistician 2004 58 2 131 7 10.1198/0003130043277
27. Automated variable selection methods for logistic regression produced unstable models for predicting acute myocardial infarction mortality. PC Austin JV Tu, Journal of Clinical Epidemiology 2004 57 1138 1146 10.1016/j.jclinepi. 2004.04.003 15567629
28. Selection of subsets of regression variables. AJ Miller, Journal of the Royal Statistical Society [Ser A] 1984 147 389 425 10.2307/2981576
29. The analysis and selection of variables in linear regression. RR Hocking, Biometrics 1976 32 1 49 10.2307/2529336
30. Bayesian Model Selection in Social Research. AE Raftery, Sociological Methodology 1995 25 111 195 10.2307/271063
31. Gleason score as predictor of clinicopathologic findings and biochemical (PSA) progression following radical prostatectomy. MS Guimarates MM Quintal LR Meirelles LA Maqna U Ferreira A Billis, Int Braz J Urol 2008 34 1 23 29 18341718
32. How well does the Gleason score predict prostate cancer death? A 20-year follow up of a population based cohort in Sweden. O Andrén K Fall L Franzén SO Andersson JE Johansson MA Rubin, J Urol 2006 175 4 1337 40 10.1016/S0022-5347(05)00734-2 16515993
33. Use of Percent Free Prostate-Specific Antigen as a Predictor of the Pathological Features of Clinically Localized Prostate Cancer. J Morote G Encabo IM Torres, Eur Urol 2000 38 225 229 10.1159/000020283 10895016
34. Preoperative serum prostate specific antigen does not reflect biochemical failure rates after radical prostatectomy in men with large volume cancers. M Noguchi TA Stamey J McNeal CM Yemoto, J Urol 2000 164 5 1596 1600 10.1016/S0022-5347(05)67036-X 11025712