Project Description

The Applications of Infectious Disease Surveillance Methodologies to Humanities Data

Nearly all data contain some degree of uncertainty. These uncertainties manifest themselves through a number of different mechanisms such as an imprecise or biased measurement or an incomplete observation, as two examples. My work focuses on the development, testing, and distribution of novel statistical methodologies to improve inference from observed health data, in the presence of such uncertainties. However, recognizing how these uncertainties affect one’s data, and how to correct for these uncertainties can be a difficult problem.

Currently, my work falls into two broad categories, (1) The spatial analysis of data with uncertainty regarding either the location of exposure to a causative agent, or the effects of imperfect diagnostic tools on our ability to accurately characterize the distribution of disease within a community; and (2) Estimating the effect of worker mobility on the ability to effectively design worksite-level health interventions within industries dominated by a temporary or mobile workforce.

While these may seem like two very distinct problems, consistent themes surrounding uncertain exposures and subject mobility present themselves across both settings. My current research investigates methods that allow us to address many of these common sources of data uncertainty. Specifically, I employ flexible weighting schemes to quantify, adjust for, and visually express uncertainties arising from (1) individuals changing residences with frequencies shorter than the incubation period of the disease (2) multiple locations of potential exposure to a disease causing agent (3) imprecise knowledge around the outcome or exposure status corresponding to a large proportion of individuals (4) individuals in worksite health interventions changing jobs with frequencies shorter than the length of the designed interventions.

Viral Culture Lecture



 Justin Manjourides is an Assistant Professor in the Department of Health Sciences and is Co-Director of the Biostatistics Service Center at Northeastern University. Dr. Manjourides received his PhD in Biostatistics at Harvard University and completed post-doctoral training at the Harvard School of Public Health. His current research interests involve developing new statistical methodologies to better analyze spatial and temporal health data in the presence of missing or misspecified information, with specific applications to disease surveillance. He is currently a Co-Investigator on several NIH and CDC funded grants involving research ranging from the estimating the health effects of environmental exposures, risk mapping of drug-resistance tuberculosis, and occupational health interventions for construction workers.



Cohen T, Manjourides J, and Hedt-Gauthier B. Linking surveillance with action against drug-resistant tuberculosis. American Journal of Respiratory and Critical Care Medicine. 186(5):399-401. 2012. doi:10.1164/rccm.201203-0394PP.

Manjourides, Lin, Shin, Jeffery, Contreras, Santa Cruz, Jave, Yagui, Asencios, Pagano, and Cohen. Identifying multidrug resistant tuberculosis transmission hotspots using routinely collected data. Tuberculosis, 92(3):273-279, 2012.

Manjourides J and Pagano M. Improving chronic disease surveillance by incorporating residential history. Statistics in Medicine. 2011; 30: 2222-2233.

Ozonoff A, Jeffery C, Manjourides J, White LF, and Pagano M. Effect of spatial resolution on cluster detection: a simulation study, International Journal of Health Geographics 2007 6:52.