We have several interests:
- disease mapping and clusters
- time-space analysis
- ecologic bias
- the use of combinations of individual- and group-level data (multi-level studies)
- the use of Geographic Information Systems (GIS) in exposure assessment
Disease Mapping and Clusters
Disease registry data are often mapped by town or county of diagnosis and contain limited data on possible confounders. These maps often possess poor spatial resolution, the potential for spatial confounding, and the inability to consider latency. Population-based case-control studies can, on the other hand, provide detailed information on residential history and covariates. We are developing and applying methods for mapping case-control and cohort data while adjusting for risk factors and latency. We map point patterns using generalized additive models (GAMs), a statistical framework which allows us to analyze binary outcome data, adjust for covariates, smooth on space (with optimal degree of smoothing), and perform hypothesis tests. For example, the map on the left above shows breast cancer risk at the time of diagnosis; the map on the right shows maps residential location twenty years prior to diagnosis. We are also working on time-space clustering.
Individual-level studies collect information on exposure, outcome and covariates for each individual; purely ecologic studies collect group-level (aggregate) information for these variables. Ecologic bias can occur when aggregate data are used to make inferences about individuals. We are interested in comparing the direction and magnitude of ecologic bias compared with biases occuring on the individual-level. Such information is useful in designing ecologic studies, doing sensitivity analyses of ecologic studies, and understanding what happens when a group-level variable is used in an otherwise individual-level study (e.g., as a proxy for exposure on the individual-level). For more information, look here.
Many diseases are associated with an individual's socioeconomic status (SES), also known as socioeconomic position (SEP). Community level SES is often associated with disease risk as well. However, despite our knowledge of these separate associations, most previous research has not examined individual and community SES simultaneously. As a result, it is unclear if the greater disease incidence in certain communities is related to the SES of the individuals who live there (composition) or because some aspect of living in a the community confers a greater risk of disease, regardless of their SES (context). A recent paper investigated this question for breast cancer on Cape Cod.
News & Awards (For details, look here)
- September 2011: A series of four papers on the power and type I error rates of our mapping approach (using gams) have now been published--see papers below with first author Bliss/Young.
- 28 August 2010: We ran a half day workshop on Disease Mapping using Generalized Additive Models at ISES/ISEE 2010.
- 19 July 2010: Two new papers examine the statistical properties of using GAMs for mapping of disease. GAMs compare favorably with the widely used spatial scan statistic SaTScan in many cases.
- 26 March 2010: Graduate student Robin Young defends her dissertation "Properties of Hypothesis Tests Using Generalized Additive Models with Smoothers of Geographic Location in Spatial Statistics"
- 25 March 2010: New paper on the spatial distribution of rheumatoid arthrititis in the US.
- 13 October 2008: Our recent time-space analysis of breast cancer was covered in the Cape Cod Times.
- 22 September 2008: Our multilevel analysis of breast cancer was featured in a National Institute of Environmental Health Sciences (NIEHS) press release on spatial epidemiology.
- 13 August 2008: New publication on time-space clustering of breast cancer on Cape Cod, including a cool animation, Vieira et al 2008.
- 25 April 2008: New publication on multilevel modeling of breast cancer and socioeconomic status on Cape Cod, Webster et al 2008.
- Workshop on ecologic inference: 28-30 November 2007, DIMACS, Rutgers University
The workshop is part of the DIMACS Special Focus on Computational and Mathematical Epidemiology. The focus will be on study designs combining individual and group level data. More information on the workshop
- 5 July 2007: New publication. Webster TF. Bias magnification in ecologic studies: a methodological investigation. Environmental Health 2007; 6:17. The full text is freely available here.
- Hoffman K, Aschengrau A, Webster TF, Bartell SM, Vieira VM. Associations between residence at birth and mental health disorders: a spatial analysis of retrospective cohort data. BMC Public Health 2015; 15:688 [accepted 1 July 2015] doi:10.1186/s12889-015-2011-z
- Bliss RL, Weinberg J, Vieira VM, Webster TF. Adjusted significance cutoffs for hypothesis tests applied with generalized additive models with bivariate smoothers. Spatial and Spatio-temporal Epidemiology [accepted]
- Gallagher LG, Vieira VM, Ozonoff DM, Webster TF and Aschengrau A. Risk of breast cancer following exposure to tetrachloroethylene-contaminated drinking water in Cape Cod, Massachusetts: reanalysis of a case-control study using a modified exposure assessment. Environmental Health 2011, 10:47 [21 May 2011]. doi:10.1186/1476-069X-10-47.
- Bliss RY, Weinberg J, Vieira VM, Ozonoff A, Webster TF. Power of Permutation Tests Using Generalized Additive Models with Bivariate Smoothers. J Biomet Biostat 2010; 1:104. [Online 11 November 2010]. The full text is freely available doi:10.4172/2155-6180.1000104.
- Young RL, Weinberg J, Vieira V, Ozonoff A, Webster TF. A power comparison of generalized additive models and the spatial scan statistic in a case-control setting. International Journal of Health Geographics 2010, 9:37. [Online 19 July 2010]. The full text is freely available doi:10.1186/1476-072X-9-37
- Young RL, Weinberg J, Vieira VM, Aschengrau A, Webster TF. A multilevel non-hierarchical study of birth weight and socioeconomic status. International Journal of Health Geographics 2010, 9:36 [Online 9 July 2010]. The full text is freely available doi:10.1186/1476-072X-9-36.
- Young RL, Weinberg J, Ozonoff A, Vieira V, Webster TF. Generalized Additive Models and Inflated Type I Error Rates of Smoother Significance Tests. Computational Statistics & Data Analysis 2011; 55:366-374. doi:10.1016/j.csda.2010.05.004.
- Vieira VM, Hart JE, Webster TF, Weinberg J, Puett R, Laden F, Costenbader KH, Karlson EW. Association between Residences in U.S. Northern Latitudes and Rheumatoid Arthritis: A Spatial Analysis of the Nurses’ Health Study. Environ Health Perspect. 2010; 957–961 [Online 25 March 2010]. The full text is freely available doi:10.1289/ehp.0901861.
- Gallagher LG, Webster TF, Aschengrau A, Vieira VM. Using Residential History and Groundwater Modeling to Examine Drinking Water Exposure and Breast Cancer. Environ Health Perspect. 2010; 118(6):118:749–755 [Online 17 February 2010]. The full text is freely available doi:10.1289/ehp.0901547.
- Hoffman K, Webster TF, Weinberg JM, Aschengrau A, Janulewicz PA, White RF, Vieira VM. Spatial analysis of learning and developmental disorders in upper Cape Cod, Massachusetts using generalized additive models. International Journal of Health Geographics 2010, 9:7. The full text is freely available doi:10.1186/1476-072X-9-7.
- Vieira VM, Webster TF, Weinberg J, Aschengrau A. Spatial analysis of bladder, kidney, and pancreatic cancer on upper Cape Cod: An application of generalized additive models to case-control data. Environmental Health; 2009, 8:3. [Online 10 February 2009]. The full text is freely available doi: 10.1186/1476-069X-8-3
- Vieira VM, Webster TF, Weinberg JM, Aschengrau A. Spatial-temporal analysis of breast cancer on Upper Cape Cod, Massachusetts. International Journal of Health Geographics 2008, 7:46. The full text is freely available here.
- Webster TF, Hoffman K, Weinberg J, Vieira V, Aschengrau A. Community and Individual-Level Socioeconomic Status and Breast Cancer Risk: Multi-level Modeling on Cape Cod, MA. Environ Health Perspect 2008; 116(8):1125-1129. doi:10.1289/ehp.10818. [Online 25 April 2008]. The full text is freely available here.
- Webster TF. Bias magnification in ecologic studies: a methodological investigation. Environmental Health 2007; 6:17 (5 July 2007). The full text is freely available here.
- Webster T, Vieira V; Weinberg J; Aschengrau A. Method for mapping population-based case-control studies using Generalized Additive Models. International Journal of Health Geographics 2006, 5:26 (9 June 2006).The full text is freely available here.
- Ozonoff A, Webster T, Vieira V, Weinberg J, Ozonoff D, Aschengrau A. Cluster detection methods applied to the Upper Cape Cod cancer data. Environmental Health: A Global Access Science Source 2005, 4:19 (15 September 2005). The full text is freely available here
- Vieira V, Webster T, Weinberg J, Aschengrau A, Ozonoff D. Spatial analysis of lung, colorectal, and breast cancer on Cape Cod: An application of generalized additive models to case-control data. Environmental Health: A Global Access Science Source 2005, 4:11 (14 June 2005). The full text is freely available here.
- Webster, T. Commentary: Does the spectre of ecologic bias haunt epidemiology? International Journal of Epidemiology 2002; 31:161-162. The full text is freely available here.
How do we make these maps?
We subscribe to the philosophy of the BU Superfund Research Program (BUSRP) to which we are also connected, i.e., making results and products freely available where possible through open-access publications and open source software available under general public license. We are currently creating maps using R; get our R package MapGAM. Our code is freely available, as is some synthetic data for trying it out: here. See also our BUSRP research translation core.
View a movie of breast cancer time-space analysis
The movie shows the risk of breast cancer diagnosis 1983-1993 based on residential history on upper Cape Cod discussed in our publication, Vieira et al 2008. The movie is a Windows Media file (1 mb). Click to view
Where are we?
For more information:
email: Dr. Tom Webster, Dept. Environmental Health