gms | German Medical Science

MAINZ//2011: 56. GMDS-Jahrestagung und 6. DGEpi-Jahrestagung

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V.
Deutsche Gesellschaft für Epidemiologie e. V.

26. - 29.09.2011 in Mainz

Artikel

Artikel empfehlen

Adjustment for baseline values when outcome changes are of interest – a comparison of recommendations for randomized controlled trials and genetic association studies

Meeting Abstract

Suche in Medline nach

  • André Scherag - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Essen
  • Sonali Pechlivanis - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Essen
  • Carolin Pütter - Institut für Medizinische Informatik, Biometrie und Epidemiologie, Essen
  • Susanne Moebus - Institut für Medizinische Informatik, Biometrie und Epidemiologie (for the Heinz Nixdorf Recall Study Investigators), Essen
  • Karl-Heinz Jöckel - Institut für Medizinische Informatik, Biometrie und Epidemiologie (for the Heinz Nixdorf Recall Study Investigators), Essen

Mainz//2011. 56. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie (gmds), 6. Jahrestagung der Deutschen Gesellschaft für Epidemiologie (DGEpi). Mainz, 26.-29.09.2011. Düsseldorf: German Medical Science GMS Publishing House; 2011. Doc11gmds062

doi: 10.3205/11gmds062, urn:nbn:de:0183-11gmds0625

Veröffentlicht: 20. September 2011

© 2011 Scherag et al.
Dieser Artikel ist ein Open Access-Artikel und steht unter den Creative Commons Lizenzbedingungen (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.de). Er darf vervielfältigt, verbreitet und öffentlich zugänglich gemacht werden, vorausgesetzt dass Autor und Quelle genannt werden.

Gliederung

Text

Objective: In observational studies like cohort studies [1] and more recently in genetic association studies [2] it has been claimed that adjusting for baseline values leads to biased effect size estimates. However, in randomized controlled trials analysts are strongly encouraged to adjustment for baseline values when analysing outcome changes depending on treatment (e.g., [3], [4]).

Methods: Using the theory of “Directed Acyclic Graphs” and simulations we review several possible relationships between genotype, baseline values and baseline changes, characterize the impact of adjustment for baseline values on the estimated effect and describe the impact on type I and type II error levels. Finally, we analyse established genetic variants associated to obesity and their relationship to intraindividual 5-year changes of body mass index in the Heinz Nixdorf Recall study ([5]; n=3,995).

Results/conclusions: We demonstrate that both adjusting and non-adjusting for baseline values may lead to biased effect size estimates in genetic association studies. In a genome-wide association study setting, where effect sizes are known to be distorted anyway (e.g., [6]), it may be worthwhile to utilize the superior statistical efficiency of an adjusted analysis. Independent replication samples should be used to characterize the effect and to perform sensitivity analyses (e.g., [7]).

Gliederung

References

1.
Glymour MM, Weuve J, Berkman LF, Kawachi I, Robins JM. When is baseline adjustment useful in analyses of change? An example with education and cognitive change. Am J Epidemiol. 2005;162(3):267-78.
2.
McArdle PF, Whitcomb BW. Improper adjustment for baseline in genetic association studies of change in phenotype. Hum Hered. 2009;67(3):176-82.
3.
Committee for Proprietary Medicinal Products (CPMP). Points to consider on adjustment for baseline covariates. Stat Med. 2004;23(5):701-9.
4.
Senn S. Change from baseline and analysis of covariance revisited. Stat Med. 2006;25(24):4334-44.
5.
Pechlivanis S, Scherag A, Mühleisen TW, Möhlenkamp S, Horsthemke B, Boes T, Bröcker-Preuss M, Mann K, Erbel R, Jöckel KH, Nöthen MM, Moebus S. Heinz Nixdorf Recall Study Group. Coronary artery calcification and its relationship to validated genetic variants for diabetes mellitus assessed in the Heinz Nixdorf recall cohort. Arterioscler Thromb Vasc Biol. 2010;30(9):1867-72.
6.
Kraft P. Curses--winner's and otherwise--in genetic epidemiology. Epidemiology. 2008;19(5):649-51, discussion 657-658.
7.
Greenland S. Basic methods for sensitivity analysis of biases. Int J Epidemiol. 1996;25(6):1107-16.