Spatial Uncertainty Data, Modeling, and Communication (R21) | Apply Online

Frequently Asked Questions (FAQs)

What is the Spatial Uncertainty Data, Modeling, and Communication (R21) funding opportunity (PA 15-009)?

PA 15-009 is a National Institutes of Health (NIH) Funding Opportunity Announcement (FOA) using the R21 research grant mechanism. It supports new, exploratory research focused on spatial uncertainty in public health and environmental health research, including how spatial uncertainty arises in real datasets, how to model it directly in analysis methods, and how to visualize and communicate its effects on results and decisions.

What does the FOA mean by "spatial uncertainty" in public health research?

In this FOA, spatial uncertainty refers to ways that location-based information can be inaccurate, imprecise, unstable over time, or otherwise unreliable in ways that affect analysis. Examples include incomplete or incorrectly geocoded addresses, exposure surfaces modeled at mismatched resolutions, shifting administrative boundaries, or health events assigned to a residence even when relevant exposure occurred elsewhere (such as a workplace, school, or commuting route).

What is the main purpose or theme of this opportunity?

The central theme is that "place matters" in public health, but recorded "place" in datasets is often messy. The FOA encourages projects that treat this messiness as a first-class scientific problem by identifying sources of spatial uncertainty, building that uncertainty into statistical and computational methods, and creating better ways to communicate how spatial uncertainty affects conclusions and real-world actions.

What types of work does the FOA specifically aim to support?

The FOA focuses on innovative work that: (1) identifies where spatial uncertainty comes from in real public health data, (2) incorporates uncertainty directly into statistical and computational methods rather than ignoring it, and (3) develops visualization and communication approaches that make the presence and consequences of spatial uncertainty understandable to technical and non-technical audiences.

Why is spatial uncertainty considered a major issue for public health research?

Spatial uncertainty can change scientific conclusions and downstream decisions. If locations, boundaries, or exposure estimates are uncertain, then analyses such as disease mapping, cluster detection, exposure assessment, cancer surveillance, and neighborhood effects studies may produce different results depending on how uncertainty is handled. The FOA is intended to push the field beyond simply noting uncertainty in limitations and toward methods that account for it throughout the analysis pipeline.

What are examples of real-world data problems this FOA is concerned with?

Examples highlighted include incomplete or incorrect geocoding of addresses, modeled exposure surfaces that do not match how people move and live, changes in administrative boundaries over time, and misalignment between where a health event is recorded (often a residence) and where relevant exposure actually occurred (workplace, school, commuting routes).

Does the FOA support projects that catalog and quantify sources of spatial error?

Yes. The FOA encourages studies that systematically catalog sources of spatial error or instability, quantify their magnitude, and determine how they propagate through common public health analyses.

Which public health analyses are mentioned as being affected by spatial uncertainty?

The FOA lists several examples where spatial uncertainty can propagate through analysis, including disease mapping, cluster detection, environmental exposure assessment, cancer surveillance, and studies of neighborhood effects and disparities.

What is the modeling emphasis in this FOA?

The modeling emphasis is on methods that incorporate spatial uncertainty explicitly. Rather than assuming a single "true" location or perfectly measured geographic covariates, projects are expected to explore statistical approaches that reflect uncertainty in inputs and evaluate how conclusions change when location is uncertain.

What kinds of statistical or computational approaches are mentioned as relevant?

The FOA mentions measurement-error frameworks, hierarchical or Bayesian approaches that treat location as a latent variable, simulation-based sensitivity analyses, and other techniques that adjust inference when geocodes, boundaries, or exposure estimates are imperfect.

What does it mean to "treat location as a latent variable" in this context?

As described in the FOA, this refers to approaches (often hierarchical or Bayesian) that do not assume observed locations are exact. Instead, the "true" location is treated as unobserved (latent) and uncertainty in location is represented directly in the model, allowing inference to reflect that uncertainty.

Is the FOA encouraging applicants to move beyond acknowledging uncertainty in the limitations section?

Yes. The intent is to move beyond simply acknowledging uncertainty and instead incorporate spatial uncertainty into estimation, hypothesis testing, prediction, and risk mapping so the analysis results reflect the uncertainty in geographic inputs.

How important are communication and visualization in this funding opportunity?

They are described as equally important. The FOA supports development of tools and visual methods that communicate not only mapped outcomes (such as disease rates) but also uncertainty around locations, boundaries, and exposure estimates, and the downstream consequences for hotspots, associations, and intervention priorities.

What kinds of visualization or communication products does the FOA mention?

The FOA mentions interactive or decision-support visualizations, new map symbology, uncertainty-aware dashboards, and other approaches that make uncertainty understandable without creating misleading overconfidence.

What is the NIH R21 mechanism and why is it used here?

The R21 mechanism is generally intended for early-stage, high-risk/high-reward, exploratory projects. In this FOA, it is used to encourage new methodological directions and proof-of-concept tools related to spatial uncertainty.

What is the award ceiling for this FOA?

The award ceiling listed is $275,000.

Is cost sharing or a matching requirement required?

No. The FOA states there is no cost sharing or matching requirement.

What is the funding activity category for this opportunity?

The funding activity category is listed under Education, Environment, and Health.

Which CFDA numbers are associated with this FOA?

The FOA lists CFDA numbers 93.113 (Environmental Health), 93.393 (Cancer Cause and Prevention Research), and 93.399 (Cancer Control).

Who is the sponsoring agency?

The sponsoring agency is the National Institutes of Health (NIH).

When was the announcement originally posted, and what are the closing and archive dates?

The announcement was originally posted on October 10, 2014. The closing date is listed as January 7, 2018. The opportunity was archived on February 7, 2018.

Where can the full announcement be found?

The FOA is hosted in the NIH Grants Guide at: http://grants.nih.gov/grants/guide/pa-files/PA-15-009.html

Who is eligible to apply?

Eligibility is broad and includes many types of U.S. organizations such as public and private institutions of higher education, nonprofits (including 501(c)(3) and non-501(c)(3)), small businesses, for-profit organizations (other than small businesses are explicitly listed as eligible), state and local governments (state, county, city/township), special district governments, public housing authorities/Indian housing authorities, independent school districts, and Native American tribal governments and organizations.

Are specific institution types and community-based entities explicitly included?

Yes. The FOA explicitly includes Historically Black Colleges and Universities (HBCUs), Hispanic-serving Institutions, Tribally Controlled Colleges and Universities (TCCUs), Alaska Native and Native Hawaiian Serving Institutions, Asian American Native American Pacific Islander Serving Institutions (AANAPISIs), and faith-based or community-based organizations. Eligible federal agencies are also included.

Are non-U.S. (foreign) organizations eligible to apply?

Yes. The FOA explicitly states that foreign organizations and foreign institutions may apply, non-U.S. components of U.S. organizations may apply, and foreign components are allowed as defined in NIH policy.

Does the FOA explain why eligibility is so broad?

Yes. The description notes that spatial data issues and uncertainty-aware methods are globally relevant and that innovation can come from diverse institutional settings, which aligns with the wide eligibility language.

What kinds of dataset issues does the FOA want applicants to treat as "first-class scientific problems"?

The FOA points to issues such as geocoding errors, incomplete addresses, exposure misalignment with daily movement, changing boundaries, and residence-based assignment of health events when exposure may occur elsewhere. Applicants are encouraged to treat these as central research problems rather than peripheral nuisances.

What kinds of outcomes or impacts are implied by the FOA's goals?

The FOA aims to produce methods and tools that help researchers and decision-makers understand when, where, and how spatial uncertainty changes scientific conclusions and real-world actions, including policy decisions, resource allocation, and risk communication.

Who should be contacted for access or technical issues with the electronic announcement?

For access or technical issues with the electronic announcement, the contact provided is the NIH Office of Extramural Research (OER) webmaster at FBOWebmaster@OD.NIH.GOV.