In Archives of physical medicine and rehabilitation ; h5-index 61.0
OBJECTIVE : To summarize the progress towards the National Institutes of Health (NIH) Research Plan on Rehabilitation goals and the methods by which tracking occurred.
DESIGN : Each grant award was manually coded by NIH staff for research plan goals, type of science categories (e.g., basic, applied, infrastructure, etc.), and if applicable, training, and then validated by NIH Institute/Center (IC) experts. Data for years 2015 through 2017 were used to develop a coding algorithm to automatically code grants in 2018 for validation by NIH IC experts. Additional data for all years (2015-2018) were also analyzed to track changes and progress.
SETTING : The research utilized administrative data from NIH Reporter and internal NIH databases.
PARTICIPANTS : (or Animals, Specimens, Cadavers): The data sample included research grants and programs funded from fiscal years 2015 through 2018. The year 2015 was considered a baseline year as the research plan was published in 2016.
INTERVENTIONS : Not applicable.
MAIN OUTCOME MEASURE(S) : The primary outcome measures were substantial growth in NIH funding and numbers of awards for rehabilitation research, across most research plan goals and types of science, as well as validation of an automatic algorithm for coding grants.
RESULTS : Number of grants, funding dollars, funding mechanisms, patent data, scientific influence and translational science, research plan goals, and type of science categories were tracked across years (2015 - 2018). Algorithm validation is presented for 2018 data.
CONCLUSION : (s): NIH advanced the goals stated in the Research Plan on Rehabilitation, but gap areas remain. Though funding in this portfolio is growing, continued focus and participation by the field is needed to advance rehabilitation science.
Jackson Jennifer N, Cernich Alison N
algorithm, machine learning, national institutes of health, rehabilitation research