Research Focus: Design, Development, Implementation and Evaluation of Infrastructural Support for Team Science

My research is focused on the coordination and support of collaborative biomedical research, especially on the design, development, implementation and evaluation of infrastructural support for team science.

Some of my specific questions include:

  • How can we improve the coordination and support of multi-site biomedical research such that we offload the burden of collaboration from scientists and allow them to spend more time on their science?
  • How do successful Coordinating Centers enhance collaboration and can we apply their knowledge in other realms?
  • What elements of "Human Infrastructure" are essential to collaborative biomedical research and what types of training do participants need?
  • How do multi-site studies share and harmonize heterogeneous Small Data datasets to do pooled statistical analyses?

As a Fellow at the NCI, I am developing a research program to answer some of these questions. While a PhD student, I led two main research projects around these topics:

The Role of Coordinating Centers in Collaborative Cancer-Epidemiology Studies (R03CA150036)

PIs: Betsy Rolland (lead), Charlotte P. Lee

A Coordinating Center (CC) is a central body tasked with coordination and operations management of a multi-site research project. Its main purpose is to facilitate collaboration among the diverse sites with the aim of completing all research objectives and producing impactful science. NIH spends millions of dollars each year supporting CCs, yet little research has been done about how to run a successful CC.

Using qualitative research methods, including field observations and interviews, we investigated the work practices of several CCs at the Fred Hutchinson Cancer Research Center (FHCRC). FHCRC houses some of the most successful and longest-running CCs in the country, collectively representing decades of experience coordinating and facilitating collaborative research. Our aim was to capture as much of this knowledge as possible and package it in a way that it can be easily applied by other consortia.

The specific aims of this NCI-funded project were to:

  1. Identify and describe models of Coordinating Centers (CCs) and corresponding models of collaborative cancer-epidemiology research projects
  2. Develop metrics that CCs can use to predict and evaluate their performance
  3. Build a downloadable toolkit to help CCs ramp up their operations efficiently and effectively
  4. Produce a modular course to develop and train CC staff

Dissertation: A subset of this research became my dissertation, which focused on answering the question, How do Coordinating Centers facilitate collaboration in multi-site studies? For my thesis, I focused on two CCs at FHCRC, both led by the same PIs.At the time of my research, one of these CCs had begun its third 5-year grant term, while the second had just entered its second year. By focusing on these two very different collaborative projects being run essentially by the same personnel, I was able to mitigate (to the greatest degree possible) the effect of personality and experience on faciliating collaboration. This allowed me to highlight the actual work practices in which the staff and PIs engage.

My committee members were:

  • Charlotte P. Lee, PhD, Human Centered Design & Engineering, University of Washington (advisor and chair)
  • John D. Potter, MD, PhD, Epidemiology, University of Washington, Fred Hutchinson Cancer Research Center
  • Matthew J. Bietz, PhD, Informatics, University of California-Irvine
  • Mark Zachry, PhD, Human Centered Design & Engineering, University of Washington
  • Wanda Pratt, PhD, Information School, University of Washington (Graduate School Representative).

Reuse of Data by Cancer-Epidemiology Post-Doctoral Researchers

CCs are often tasked with managing the collection and harmonization of disparate datasets from participating sites. To do so, they often must interpret the meaning of variables in a given dataset, yet we know very little about how such data are interpreted by researchers who were not involved in the original data collection. This study, a subset of our research on CCs, set out to answer the question, How do cancer epidemiology postdoctoral researchers determine how to use a variable from an existing dataset appropriately for their own analyses? Developing an understanding of these work practices allows us to begin to think about ways to provide better support, both social and technological, for data sharing and reuse.

Post-docs in our study were conducting new analyses of existing datasets under the guidance of mentors who had a direct connection to the dataset being used, either as the PI of the original study or as an experienced user of the dataset. Because post-docs are just one step removed from the original study staff, post-docs' experience working to understand an unfamiliar dataset gives us insight into the work practices involved in developing such an understanding. Studying post-docs, who work closely with a mentor, also allows us to bypass the issues of trust and reliability so frequently mentioned as a stumbling block to data sharing. The results of this study have been accepted to the 2013 Computer Supported Cooperative Work (CSCW) conference, and are available in the ACM Digital Library.

Non-Traditional Roles for Information Professionals in Biomedical Research

PIs: Betsy Rolland and Emily Glenn

I am also interested in how librarians and other information professionals can be engaged in the process of supporting collaborative research. Along with my colleague Emily Glenn of Seattle BioMed, I received the 2008 Special Libraries Association Research Grant for our proposal "Experimenting Outside the Information Center: Non-Traditional Roles for Information Professionals in Biomedical Research." Our research investigated the ways in which librarians are using their traditional library-based skills in non-traditional ways in biomedical research.

 

   
    brolland *at* uw.edu