Display options
Cite
Miao J, Sinha S, Wang S, et al. Analysis of Multivariate Disease Classification Data in the Presence of Partially Missing Disease Traits. J Biom Biostat. 2014;5doi: 10.4172/2155-6180.1000197.
Miao, J., Sinha, S., Wang, S., Diver, W. R., & Gapstur, S. M. (2014). Analysis of Multivariate Disease Classification Data in the Presence of Partially Missing Disease Traits. Journal of biometrics & biostatistics, 5. https://doi.org/10.4172/2155-6180.1000197
Miao, Jingang, et al. "Analysis of Multivariate Disease Classification Data in the Presence of Partially Missing Disease Traits." Journal of biometrics & biostatistics vol. 5 (2014). doi: https://doi.org/10.4172/2155-6180.1000197
Miao J, Sinha S, Wang S, Diver WR, Gapstur SM. Analysis of Multivariate Disease Classification Data in the Presence of Partially Missing Disease Traits. J Biom Biostat. 2014;5. doi: 10.4172/2155-6180.1000197. PMID: 25530913; PMCID: PMC4270282.
Copy Download .nbib Format: NLM
Share it on

J Biom Biostat. 2014;5. doi: 10.4172/2155-6180.1000197.

Analysis of Multivariate Disease Classification Data in the Presence of Partially Missing Disease Traits.

Journal of biometrics & biostatistics

Jingang Miao, Samiran Sinha, Suojin Wang, W Ryan Diver, Susan M Gapstur

Affiliations

  1. Department of Statistics, Texas A&M University, College Station, TX 77843, USA.
  2. Epidemiology Research Program, American Cancer Society, Inc. Atlanta, GA 30303, USA.

PMID: 25530913 PMCID: PMC4270282 DOI: 10.4172/2155-6180.1000197

Abstract

In modern cancer epidemiology, diseases are classified based on pathologic and molecular traits, and different combinations of these traits give rise to many disease subtypes. The effect of predictor variables can be measured by fitting a polytomous logistic model to such data. The differences (heterogeneity) among the relative risk parameters associated with subtypes are of great interest to better understand disease etiology. Due to the heterogeneity of the relative risk parameters, when a risk factor is changed, the prevalence of one subtype may change more than that of another subtype does. Estimation of the heterogeneity parameters is difficult when disease trait information is only partially observed and the number of disease subtypes is large. We consider a robust semiparametric approach based on the pseudo-conditional likelihood for estimating these heterogeneity parameters. Through simulation studies, we compare the robustness and efficiency of our approach with that of the maximum likelihood approach. The method is then applied to analyze the associations of weight gain with risk of breast cancer subtypes using data from the American Cancer Society Cancer Prevention Study II Nutrition Cohort.

Keywords: Dimension reduction; Disease classification; Estimating equations; Etiologic heterogeneity; Missing disease trait; Polytomous logistic model

References

  1. Rev Med Suisse. 2012 May 23;8(342):1101-4 - PubMed
  2. Biometrics. 2006 Dec;62(4):980-5 - PubMed
  3. Blood. 2008 Dec 15;112(13):5150-60 - PubMed
  4. Am J Epidemiol. 2001 Nov 15;154(10):954-62 - PubMed
  5. Biometrika. 2010 Sep;97(3):683-698 - PubMed
  6. Am J Epidemiol. 2012 Sep 15;176(6):512-8 - PubMed
  7. Cancer. 2012 Oct 1;118(19):4860-8 - PubMed
  8. Stat Med. 1997 Feb 28;16(4):385-95 - PubMed
  9. Stat Med. 2008 Sep 20;27(21):4238-53 - PubMed
  10. Cancer. 2002 Jan 15;94(2):500-11 - PubMed
  11. Natl Cancer Inst Monogr. 1985 May;67:49-52 - PubMed

Publication Types

Grant support