Skip to main content
Explore URMC

URMC Logo

menu
URMC / Labs / Brent Johnson Lab / Research Projects

Research Projects and Interests

Adaptive Treatment Policies, Dynamic Treatment Regimes, Causal Inference

An ongoing research interest of mine is the statistical analysis of data from non-randomized studies. Within this genre, I actively research dynamic treatment regimes, semi-parametric efficiency, and locally optimal double robust estimators. My main application areas have been in infusion trials, HIV therapeutic trials, and recently application to tertiary treatments for patients diagnosed with ALS.

References:

  • Lu X, Johnson BA. (2015) Direct estimation of the mean outcome on treatment when treatment assignment and discontinuation compete. Biometrika. 102(4), 797-807.
  • Johnson BA, Ribaudo H, Gulick RM, Eron JJ. (2013) Modeling clinical endpoints as a function of time of switch to second-line ART with incomplete data on switching times. Biometrics. 69, 732-740.
  • Li L, Eron J, Ribaudo H, Gulick RM, Johnson BA. (2012) Evaluating the effect of early versus late ARV regimen change after failure on the initial regimen: results from the AIDS Clinical Trials Group Study A5095. Journal of the American Statistical Association. 107, 542-554.
  • Johnson BA. (2008) Treatment-competing events in dynamic regimes. Lifetime Data Analysis. 14, 196-215.
  • Johnson BA and Tsiatis AA. (2005) Semiparametric inference in observational duration-response studies, with duration possibly right-censored. Biometrika. 92, 605-618.
  • Johnson BA and Tsiatis AA. (2004) Estimating mean response as a function of treatment duration in an observational study, where treatment duration may be informatively censored. Biometrics. 60, 315-323.

The Analysis of Censored Marked Endpoints

I have started looking actively at the statistical analysis of marked endpoints with applications to therapeutic HIV trials. By marked endpoint in a clinical trial setting, I mean an endpoint that is measured when the event occurs, and not measured with the event is censored. Examples of such events include lifetime medical cost and the strain of the virus in vaccine efficacy studies.

References:

  • Johnson BA, Long Q, Huang Y, Chansky K, and Redman, M. (2015) Model selection and inference for censored lifetime medical expenditures. Biometrics. (In Press)
  • Johnson BA. (2015) On Huberized calibration regression for censored medical cost data. Statistics in Biosciences. 7, 367-378.

Variable Selection

I have investigated the theory and methods for regularized coefficient estimation in semi-parametric models with application to censored and missing data. I also proposed rank-based survival ensembles to complement the mboost package in R. Below are some references whose PDFs can be downloaded on my publications page as well as some software that can also be downloaded from the software page.

References:

  • Chung M, Long Q, and Johnson BA. (2013) A tutorial on rank-based coefficient estimation for censored data in small- and large-scale problems. Statistics and Computing. 23, 601-614.
  • Long Q, Chung M, Moreno C, Johnson BA. (2011) Risk prediction for cancer recurrence through regularized estimation with simultaneous adjustment for nonlinear clinical effects. Annals of Applied Statistics. (In press).
  • Johnson BA, Long Q, Chung M. (2011) On path restoration for censored outcomes. Biometrics. (In press).
  • Johnson BA, Long Q. (2011) Survival ensembles by the sum of pairwise differences. Annals of Applied Statistics. (In press).
  • Johnson BA. (2009) Rank-based estimation in the L1-regularized partly linear model for censored outcomes with application to integrated analyses of clinical predictors and gene expression data. Biostatistics. 10, 659-666.
  • Johnson BA. (2009) On lasso for censored data. Electronic Journal of Statistics. 3, 485-506.
  • Johnson BA, Lin DY, and Zeng D. (2008) Penalized estimating functions and variable selection in semiparametric regression models. Journal of the American Statistical Association. 103, 672-680.
  • Johnson BA and Peng L. (2008) Rank-based variable selection. Journal of Nonparametric Statistics. 20, 241-252.
  • Johnson BA. (2008) Variable selection in semiparametric linear regression with censored data. Journal of the Royal Statistical Society, Series B. 70, 351-370.

Treatment and Prevention of HIV and AIDS

Beginning in 2004, I began applying my interest in complex treatment strategies to HIV and AIDS research. At that time, I was a postdoctoral fellow and met Dr. Joseph Eron, Jr., Professor of Medicine and Director of the Clinical Core at UNC-CH, through a colleague in Biostatistics, Professor Michael Hudgens. Dr. Eron and I received an NIH grant award for our project in 2007. Briefly, we conceived of a novel strategy to estimate the causal effect of delayed switch from a failing antiretroviral regimen. Rather than conditioning the analysis only on those patients that failed, we estimate the combined effect of failing on the initial regimen and switching early or late to second-line regimen. Interestingly, in our analysis of the ACTG 5095 data, we found that our method detected mild clinical benefit, on average, to switching within 8 weeks of confirmed virologic failure of an efavirenz-containing regimen whereas the conventional method found no difference.

Over the past 10+ years, I have worked on a variety of projects in HIV prevention and treatment. I am currently the Associate Director for the Biostatistics Core in the University of Rochester's Center for AIDS Research.

References:

  • Sharma A, Johnson BA, Sullivan PS. (2015) Evaluating interventions to promote routine preventive screenings: a comparison of analytical outcomes. Contemporary Clinical Trials. 41,152-159.
  • Wall K, Kilembe W, Vwalika B, Htee Khu N, Brill I, Chomba E, Johnson BA, Haddad L, Tichacek A, Allen S. (2015) Hormonal contraception does not increase women's HIV acquisition risk in Zambian discordant couples, 1994-2012. Contraception. 91, 480-487.
  • Colasanti J, McDaniel D, Johnson B, Del Rio C, Sunpath H, and Marconi VC. (2015) Novel predictors of poor retention following a down-referral from a hospital-based ART program in South Africa. AIDS Research and Human Retroviruses. (In Press)
  • Berg C, Nehl E, Wang X, Ding Y, He N, Johnson BA, Wong FY. (2014) A comparison of HIV-positive and negative MSM smokers in Chengdu, China: Quit attempts and healthcare provider intervention on smoking. AIDS Care. 26(9), 1201-1207
  • Hare AQ, Ordonez CE, Johnson BA, Del Rio C, Kearns RA, Wu B, Hampton J, Wu P, Sunpath H, Marconi VC. (2014) Gender-specific risk factors for virologic failure in KwaZulu-Natal: Automobile ownership and financial insecurity. AIDS and Behavior. 2014 Jul 19 [Epub ahead of print].
  • Carnathan D, Wetzel KS, Yu J, Lee ST, Johnson BA, Paiardini M, Yan J, Morrow M, Sardesai NY, Weiner DB, Ertl HCJ, Silvestri G. (2014) Activated CD4+CCR5+ T-cells in the rectum predict increased SIV acquisition and higher viremia in SIVGag/Tat vaccinated macaques. Proceedings of the National Academy of Sciences. 112(2), 518-523.
  • Wu P, Johnson BA, Nachega J, Wu B, Ordonez C, Hare A, Kearns R, Murphy R, Sunpath H, Marconi V. (2013) The combination of pill count and self-reported adherence is a strong predictor of first-line ART failure for adults in South Africa. Current HIV Research. 12(5), 366-375.
  • Marconi VC, Johnson BA, Wu B, Quinn-Hare A, Ordonez C, Kuritzkes DR. (2013) Individual-level Early Warning Indicators for first-line virologic failure independent of adherence measures in a South African urban clinic. AIDS Patient Care and STDs. 27, 657-668.
  • Khosropour CM, Johnson BA, Ricca AV, Sullivan PS. (2013) A randomized trial of text messaging to enhance retention in an internet-based cohort study of men who have sex with men (MSM). J Med Internet Res .25(8):e194. doi:10.2196/jmlr.2756.
  • Johnson BA, Ribaudo H, Gulick RM, Eron JJ. (2013) Modeling clinical endpoints as a function of time of switch to second-line ART with incomplete data on switching times. Biometrics. 69, 732-740.
  • Li L, Eron J, Ribaudo H, Gulick RM, Johnson BA. (2012) Evaluating the effect of early versus late ARV regimen change after failure on the initial regimen: results from the AIDS Clinical Trials Group Study A5095. Journal of the American Statistical Association. 107, 542-554.
  • Sunpath H, Wu B, Hampton J, Johnson BA, Moosa Y, Ordonez C, Kuritzkes DR, Marconi VC. (2012) High rate of K65R for ART naive patients with subtype C HIV infection failing a TDF-containing first-line regimen in South Africa. AIDS. 26, 1679-84.
  • Garber D, O'Mara L, Gangadhara S, McQuoid M, Zhang X, Zheng R, Gill K, Verma M, Yu T, Johnson BA, Li B, Derdeyn C, Ibegbu C, Altman J, Hunter E, Feinberg M. (2012) Deletion of specific immune modulatory genes from modified vaccine ankara-based HIV vaccines engenders immunogenicity in Rhesus macaques. Journal of Virology. 86(23):12605-12615.

Environmental Health

I was introduced to statistical problems in environmental and occupational health while I was a postdoctoral fellow at UNC-CH. I worked with Professors Larry Kupper and Stephen M. Rappaport. Dr. Rappaport is an expert in exposure biology and now a professor at UC-Berkeley. The three of us thought about statistical methods for nonlinear models with potentially subject-specific parameters, but only few repeated measurements. The basic idea is to estimate the biomarker response curve as a function of occupational exposure. In contrast to classic pharmacokinetic data (in Davidian and Giltinan, 1995, for example), we do not see multiple outcomes per subject over time, dose, or exposure. Rather, we get to observe outcome measurements for a single exposure dose; hence, the subject-specific rate parameters are no longer estimable.

Over the years, I have collaborated on a number of other projects with different colleagues from multiple institutions.

References:

  • Kim Y-M, Zhou Y, Gao Y, Fu JS, Johnson BA, Huang C, Liu Y. (2014) Spatially-resolved estimation of ozone-related mortality in the United States under two representative concentration pathways (RCPs) and their uncertainty. Climatic Change. 128, 71-84.
  • Wu J, Zhou Y, Gao Y, Fu JS, Johnson BA, Huang C, Kim Y-M, Liu Y. (2013) Estimation and uncertainty analysis of impacts of future heat waves on mortality in the Eastern United States. Environmental Health Perspectives. DOI:10.1289/ehp.1306670.
  • Raysoni AU, Stock TH, Sarnat JA, Montoya-Sosa T, Sarnat SE, Holguin F, Greenwald R, Johnson B, Li W-W. (2013) Characterization of traffic-related air pollutant metrics at four schools in El Paso, Texas, USA: Implications for exposure assessment and siting schools in urban areas. Atmospheric Environment. 80, 140-151.
  • Rappaport SM, Johnson BA, Bois FY, Kupper LL, Kim S, Thomas R. (2013) Ignoring and adding errors do not improve the science. Carcinogenesis. 34, 1689-1691.
  • Greenwald R, Johnson BA, Hoskins A, Dworski R. (2013) Exhaled breath condensate formate after inhaled allergen provocation in atopic asthmatics in vivo. J Asthma. 50, 619-622.
  • Rappaport SM, Kim S, Thomas R, Johnson BA, Bois F, and Kupper LL. (2012) Low-dose metabolism of benzene in humans: science and obfuscation. Carcinogenesis. 34, 2-9.
  • Zora J, Sarnat SE, Raysoni AU, Greenwald R, Johnson BA, Stock TH, Holguin F, Li WW, Sarnat JA. (2012) Associations between urban air pollution and pediatric asthma control in El Paso, Texas. Science of the Total Environment. 448, 56-65.
  • Greenwald R, Sarnat JA, Li WW, Raysoni AU, Johnson BA, Sarnat SE, Stock TH, Holguin F, Montoya-Sosa T. (2012) Associations between source-indicative pollution metrics and increases in pulmonary inflammation and reduced lung function in a panel of asthmatic children. Air Quality, Atmosphere, and Health. DOI 10.1007/s11869-012-0186-3.
  • Sarnat S, Raysoni AU, Li W, Holguin F, Johnson BA, Flores-Luevano S, Garcia J, and Sarnat JA. (2012) Air pollution and acute respiratory response in a panel of asthmatic children along the U.S.-Mexico border. Environmental Health Perspectives. (In press)
  • Taylor DJ, Kupper LL, Johnson BA, Kim S, Rappaport SM. (2008) Parametric methods for evaluating nonlinear exposure-biomarker relationships when the predictor and the response variables are measured with error. Journal of Agricultural, Biological, and Environmental Statistics. 3, 367-387.
  • Johnson BA and Rappaport SM. (2007) On modeling metabolism-based biomarkers of exposure: a comparative analysis of nonlinear models with few repeated measurements. Statistics in Medicine, 26, 1901-1919.
  • Kim S, Lan Q, Waidyanatha S, Chanock S, Johnson BA, Vermeulen R, Smith MT, Zhang L, Li G, Shen M, Yin S, Rothman N, Rappaport SM. (2007) Genetic polymorphisms and benzene metabolism in humans exposed to a wide range of air concentrations. Pharmacogenetics and Genomics. 17, 789-801.
  • Kim S, Vermeulen R, Waidyanatha S, Johnson BA, Lan Q, Rothman N, Smith MT, Zhang L, Li G, Shen M, Yin S, Rappaport SM (2007) Modeling human metabolism of benzene following occupational and environmental exposures. Cancer Epidemiology Biomarkers and Prevention. 15, 2246-2252.
  • Kim S, Vermeulen R, Waidyantha S, Johnson BA, Lan Q, Rothman N, Smith MT, Zhang L, Li G, Shen M, Yin S, Rappaport SM. (2006) Using urinary biomarkers to elucidate dose-related patterns of human benzene metabolism. Carcinogensis. 27, 772-781.
  • Johnson BA, Kupper LL, Taylor DJ, Rappaport SM. (2005) Modeling exposure-biomarker relationships: applications of linear and nonlinear toxicokinetics. Journal of Agricultural, Biological, and Environmental Statistics. 10, 440-459.
  • Pleil JD, Vette AF, Johnson BA, Rappaport SM. (2004) Air levels of carcinogenic polycyclic aromatic hydrocarbons after the World Trade Center disaster. Proceedings of the National Academy of Sciences. 101, 11685-11688.