March 9th, 2013

Data Sources for Young Investigators: Blogging from ACC.13

Several Cardiology Fellows who are attending ACC.13 in San Francisco this week are blogging for CardioExchange. The Fellows include Tariq AhmadMegan CoylewrightJeremiah DeptaKumar Dharmarajan Payal Kohli, and  Sandeep Mangalmurti. View the previous post here and the next one here.  

I just attended a great session entitled “Getting Access to Data: Where Can Young Investigators Get Started in Clinical Research.” The session was co-moderated by Tracy Wang from Duke and Robert Yeh from MGH. The panel included Fred Masoudi from UC Denver, Lesley Curtis from Duke, and Gregg Fonarow from UCLA.

Here are some important take home points:

  1. Any trainee can submit a research proposal to the National Cardiovascular Data Registry (NCDR). If the proposal is approved, the trainee may be able to receive full analytic support directly from the NCDR. No experience in outcomes research is needed to submit a proposal, but keep in mind that the selection process is merit-based, so proposals will need to be of high quality.
  2. Access to Medicare fee-for-service claims data entails significantly more challenges. Depending on the patient cohort, costs can range from approximately $3,000 to over $40,000 per year of data. Data use agreements are relatively narrow, and new proposals that leverage already purchased data will require a re-use fee. Data files may require significant processing following their receipt. To conduct research with Medicare claims data, young investigators will almost always need to collaborate with more senior investigators with access to the data and expertise in its analysis.
  3. The Healthcare Cost and Utilization Project (HCUP) data set from the AHRQ can be a great data source to understand trends over time in hospital-based care for the diversity of patients within 44 states, not just Medicare fee-for-service beneficiaries. The data set is relatively inexpensive and easy to use for young investigators still learning how to program in SAS, STATA, and other languages.
  4. The Get with the Guidelines (GWTG) registries are rich data sources similar to the NCDR databases. As with NCDR, there is a formal mechanism for proposal submission and review. Data resides at an analytic center, and proposals can receive biostatistical support. GWTG also has a young investigator seed grant program to help trainees fund study analysis and travel to conferences to present their results.
  5. Older prospective clinical trials and longitudinal patient cohorts can be great sources of data depending on the clinical question. In some cases, this data may be publicly available and/or linked to longitudinal claims data. Examples include data from the DIG study in heart failure and the Cardiovascular Health Study.
  6. Don’t be discouraged if your institution does not have strong mentorship in outcomes research. The panelists presenting today have all helped young investigators at other institutions work through the processes of study design, data analysis, paper writing, paper submission, and publication. Don’t hesitate to reach out to senior outcomes researchers at other institutions who have expertise in the database and study domain you intend to study.

For more of our ACC.13 coverage of late-breaking clinical trials, interviews with the authors of the most important research, and blogs from our fellows on the most interesting presentations at the meeting, check out our Coverage Headquarters.

2 Responses to “Data Sources for Young Investigators: Blogging from ACC.13”

  1. Saurav Chatterjee, MD says:

    Excellent post-also to note, NHANES has all of its data publicly available.

  2. Kumar, This collaborative culture, exemplified by your detailed post, defines the outcomes community…thank you fro sharing!