Niko Kaciroti, PhD

• PhD, Biostatistics, University of Michigan, 2002
• MSc, Biostatistics, University of Michigan, 1994
• BSc, Applied Mathematics, University of Tirana, 1991

My research interest is twofold, the application of statistical methods in health related research and the development of statistical methodologies related to it. More specifically, I am interested in using Bayesian modeling techniques for analyzing longitudinal multilevel data from randomized clinical trials with dropouts. I have developed and implemented Bayesian models for sensitivity analysis in randomized trials where the dropout process is potentially informative. Such models apply to a range of outcome types including normal, binary, ordinal, Poisson, and time-to-event data. I am also interested in using Bayesian methods to model nonlinear and dynamic models in a multilevel setting. Bayesian approach has the flexibility and the computational power to address complex and dynamic models and at the same time it can incorporate prior knowledge into the data when making inferences. Some examples include using person-centered or variable-centered approaches, mixture-modeling and cross-lagged analysis to better understand complex relationships and how they evolve over time. This work is applied to pediatric research related to obesity, media use and their impact in developmental outcomes on children, as well as to health outcomes for general population including infection diseases and cardiovascular disease.

Radesky JS, Kaciroti NA, Weeks HM, Schaller A, Miller AL: Longitudinal associations between use of mobile devices for calming, emotional reactivity, and executive functioning in young children.  JAMA pediatrics 177(1): 62-70, 2023.

Shafie Khorassani F, Taylor JMG, Kaciroti NA, MR Elliott MR. Incorporating Covariates into Measures of Surrogate Paradox Risk. Stats 6 (1): 322-344, 2023.

Kaciroti NA: Letter to the Editor in response to ‘Z-Score Burden Metric: A Method for Assessing Burden of Injury and Disease’. American Journal of Preventive Medicine 64 (2): 301, 2022. 

Kaciroti NA, Little RJA: Bayesian sensitivity analyses for longitudinal data with dropouts that are potentially missing not at random: A high dimensional pattern-mixture model. Stat Med 40(21): 4609- 4628, 2021. 

Kaciroti NA, Lumeng C, Parekh V, Boulton ML: A Bayesian Mixture Model for Predicting the COVID-19 Pandemic in the United States. Am J Trop Med Hyg 104(4), 1484-1492, 2021. 

Brook RD, Kaciroti NA, Jamerson T, Jamerson KA: Cardiovascular Benefits of Combination Angiotensin- Converting Enzyme Inhibition Plus Calcium Channel Blockade in Black Hypertensive Patients. Hypertension 78(4): 1150-1152, 2021. 

Kaciroti N, DosSantos MF, Moura B, Bellile EL, Nascimento TD, Maslowski E, Danciu TE, Donnell A, DaSilva AF: Sensory-Discriminative Three-Dimensional Body Pain Mobile App Measures Versus Traditional Pain Measurement With a Visual Analog Scale: Validation Study. JMIR Mhealth Uhealth 8(8): e17754, 2020.

Shah PE, Weeks HM, Richards B, Kaciroti NA. Early childhood curiosity and kindergarten reading and math academic achievement. Pediatric research 84 (3), 380-386, 2018.

Mosli RH, Kaciroti NA, Corwyn RF, Bradley RH, Lumeng JC: Effect of Sibling Birth on BMI Trajectory in the First 6 Years of Life. Pediatrics 137(4): e20152456, 2016. 

Lumeng J, Kaciroti N, Sturza J, Krusky A, Miller A, Peterson K, Lipton R, Reischl T. Changes in body mass index associated with head start participation. Pediatrics 135(2): 449-456, 2015

Kaciroti NA, Raghunathan T: Bayesian sensitivity analysis of incomplete data: bridging pattern-mixture and selection models. Stat Med 33(27): 4841-57, 2014. 

Kaciroti NA, Raghunathan TE, Taylor JM, Julius S: A Bayesian model for time-to-event data with informative censoring. Biostatistics 13(2): 341-54, 2012. 

Kaciroti NA, Schork MA, Raghunathan T, Julius S: A Bayesian sensitivity model for intention-to-treat analysis on binary outcomes with dropouts. Stat Med 28(4): 572-85, 2009. 

Kaciroti NA, Raghunathan TE, Schork MA, Clark NM: A Bayesian model for longitudinal count data with non-ignorable dropout. J R Stat Soc Ser C Appl Stat 57(5): 521-534, 2008. 

Julius S, Nesbitt SD, Egan BM, Weber MA, Michelson EL, Kaciroti NA, Black HR, Grimm RH, Messerli FH, Oparil S, Schork MA, Trial of Preventing Hypertension (TROPHY) Study In: Feasibility of treating prehypertension with an angiotensin-receptor blocker., N Engl J Med. 354(16): 1685-1697, 2006. 

Kaciroti NA, Raghunathan T, Schork MA, Clark N, Gong M: A Bayesian approach for clustered longitudinal ordinal outcome with nonignorable missing data: Evaluation of an asthma education program. Journal of the American Statistical Association 101: 435-446, 2006.

https://scholar.google.com/citations?user=ITyIYD0AAAAJ&hl=en&oi=ao

Email: nicola@umich.edu 
Office: 734-936-9714
Fax: 734-936-9288

Address:
300 N. Ingalls Bldg, 10th Floor
CHGD, #1027NW
Ann Arbor, Michigan 48109