Biography

Hormuzd A. Katki received a B.S. in Math from the University of Chicago then an M.S. in Statistics from Carnegie-Mellon University. He joined NCI in 1999 as a Staff Scientist. He received a Ph.D in Biostatistics from Johns Hopkins University in 2006 and received the Margaret Merrell Award for research by a Biostatistics doctoral student.

Research Interests

Models to estimate absolute risks of diseases or of carrying mutations

Such models can be used to improve etiologic studies, make risk predictions for clinical use, and form the basis for intervention program planning and evaluation. (1) I am working on a model for absolute risk of cervical neoplasia endpoints based on a history of HPV tests, pap smears, biopsies, and other risk factors. This model integrates all risk factors into a single number that could form the basis for decisions about clinical management. (2) I have improved models that predict who in a family carries inherited deleterious mutations (such as in BRCA1/2), based on family history of disease. I have accounted for errors in reported family history, medical interventions that alter disease risks, and properly accounting for multiple diseases. (3) Absolute risk models developed on external data could improve control of confounding factors via prognosis scores and propensity scores. I will apply these ideas to the long-term follow-up of the HPV vaccine trial at NCI.

Survival analysis for cohorts with missing covariate information

Analyses of multiple exposures within established cohorts are a ubiquitous epidemiologic study design. To save resources while retaining statistical efficiency, the exposures are measured on most disease cases and only a well-chosen sample of the controls. Analyzing such studies as case-control studies ignores the information in the controls missing exposure measurements. Such studies are better analyzed as two-phase designs which extract information from all cohort members (subsets of this design include the case-cohort and nested case-control designs). Our methods estimate survival curves and attributable risks (rather than mere relative risks), realize impressive information gains by using the entire cohort, and permit optimal sampling schemes for controls to have exposure measurements. My R package NestedCohort provides software for these methods.

Improving inference from p-values by using Bayes Factors

The limitations of p-values as measures of evidence are exposed in high-dimensional bioinformatic data and by the quest to discover ever-smaller effects in epidemiology. The Bayes Factor (and the related likelihood ratio) have better theoretical justifications as evidential measures and help subject-matter experts better understand the data so they can apply their outside knowledge to draw comprehensive conclusions. However, the use of Bayes Factors necessitates a paradigm-shift for how science is done. I proposed False Report Bayes Factors to bridge this gap and allow researchers to gain extra insight from p-values and from Bayes Factors. I hope this work points to a future of better interpretation of medical findings by scientists, clinicians, and the lay public.

Keywords

Absolute risk, attributable risk, BRCA1/2, BRCAPRO, BayesMendel, HPV, growth curves, two-phase design, case-cohort, nested case-control

Selected Publications

• Katki HA, Blackford A, Chen S, and Parmigiani G. Multiple Diseases in Carrier Probability Estimation: Accounting for Surviving All Cancers Other than Breast and Ovary in BRCAPRO. Statistics in Medicine, 2008; 27(22): 4532-4548.
• Katki HA. Invited Commentary: Evidence-based Evaluation of p-values and Bayes Factors. American Journal of Epidemiology,, 2008; 168(4): 384-388.
• Katki HA, Mark SD. Survival Analysis for Cohorts with Missing Covariate Information. R News, 2008; 8(1): 14-19.
• Katki HA. Invited Discussion of "Estimates of human immunodeciency virus prevalence and proportion diagnosed based on Bayesian multiparameter synthesis of surveillance data". Journal of the Royal Statistical Society, Series A, 2008; 171(3): 575-576.
• Katki HA, Gail MH, Greene MH. Keynote Comment - Breast-cancer risk in BRCA-Mutation-Negative Women from BRCA-Mutation-Positive Families, Lancet Oncology, 2007; 8(12): 1042-1043.
• Katki HA. Incorporating Medical Interventions into Carrier Probability Estimation for Genetic Counseling. BMC Medical Genetics, 2007; Mar 22, 8:13
• Mark SD and Katki HA. Specifying and Implementing Nonparametric and Semiparametric Survival Estimators in Two-Stage (sampled) Cohort Studies with Missing Case Data. Journal of the American Statistical Association , 2006; 101(474):460-471
• Katki HA, Effect of Misreported Family History on Mendelian Mutation Prediction Models. Biometrics , 2006; 62(2):478-487
• Katki HA, Engels EA, and Rosenberg PS. Assessing Uncertainty in Reference Intervals via Tolerance Intervals: Application to a Mixed Model Describing HIV Infection. Statistics in Medicine, 2005; 24(20):3185-3198.
• Chen S, Wang W, Broman KW, Katki HA, and Parmigiani G. BayesMendel: an R environment for Mendelian Risk Prediction. Statistical Applications in Genetics and Molecular Biology , 2004; 3(1) Article 21.
• Rosenberg PS, Katki H, Swanson CA, Brown LM, Wacholder S, Hoover RN. Quantifying epidemiologic risk factors using non-parametric regression: model selection remains the greatest challenge. Statistics in Medicine , 2003; 22(21):3369-3381.
• Engels EA, Katki HA, Nielsen NM, Winther JF, Hjalgrim H, Gjerris F, Rosenberg PS, Frisch M, Cancer incidence in Denmark following exposure to poliovirus vaccine contaminated with simian virus 40. Journal of the National Cancer Institute, 2003; 95:532-539.
• Gail, M.H. Katki, H.A. Re: All-cause mortality in randomized trials of cancer screening. Journal of the National Cancer Institute, 2002; 94(11):862-866.
• Mark, S.D., Katki, H. Influence Function Based Variance Estimation and Missing Data Issues in Case-Cohort Studies. Lifetime Data Analysis, 2001; 7:331-344.
• Engels, E.A., Rosenberg, P.S., Katki, H., Goedert, J.J., Biggar, R.J. Trends in Human Immunodeficiency Virus Type 1 (HIV) Viral Load Levels among HIV-infected Children with Hemophilia. Journal of Infectious Diseases, 2001; 184(3):364-368.
• Mark, S.D., Qiao, Y.-L. Dawsey, S.M., Wu, Y.-P., Katki, H., Gunter, E., Fraumeni, Jr., J.F., Blot, W.J., Dong, Z.-W., Taylor, P.R. Prospective Study of Serum Selenium Levels and Incident Esophageal and Gastric Cancers. Journal of the National Cancer Institute, 2000; 92:21:1753-1763.
• Katki, H., Weiss, G.H., Keifer, J.E., Taitelbaum, H., Spencer, R.G.S. Optimization of Magnetization Transfer Experiments to Measure First-Order Rate Constants and Spin-Lattice Relaxation Times. NMR in Biomedicine, 1996; 9:135-139.

Collaborators

DCEG Collaborators

• Philip Castle, Ph.D.; Nilanjan Chatterjee, Ph.D.; Anil Chaturvedi, Ph.D.; Mitchell Gail, M.D, Ph.D; Mark Greene, M.D; Barry Graubard, Ph.D; Allan Hildesheim, Ph.D.; Aimee Kreimer, Ph.D.; Philip Rosenberg, Ph.D; Mark Schiffman, M.D.; Rachael Stolzenberg-Solomon, Ph.D.; Sophia Wang, Ph.D.; Sholom Wacholder, Ph.D; Regina Ziegler, Ph.D.

Other NCI Collaborators

• Diane Solomon, M.D.

Other Scientific Collaborators

• Andrew Bergen, Ph.D., Stanford Research International, CA
• Sining Chen, Ph.D, Johns Hopkins University, Baltimore, MD
• Giovanni Parmigiani, Ph.D, Johns Hopkins University, Baltimore, MD
• Chris Sanders, Ph.D., MedCo Health, MD
• Ravi Varadhan, Ph.D., Johns Hopkins University, Baltimore, MD
• Jon Wakefield, Ph.D., University of Washington, WA