The FAIR database: facilitating access to public health research literature

Objectives In public health, access to research literature is critical to informing decision-making and to identify knowledge gaps. However, identifying relevant research is not a straightforward task since public health interventions are often complex, can have positive and negative impacts on health inequalities and are applied in diverse and rapidly evolving settings. We developed a "living" database of public health research literature to facilitate access to this information using Natural Language Processing tools.Materials and Methods Classifiers were identified to identify the study design (eg, cohort study or clinical trial) and relationship to factors that may be relevant to inequalities using the PROGRESS-Plus classification scheme. Training data were obtained from existing MEDLINE labels and from a set of systematic reviews in which studies were annotated with PROGRESS-Plus categories.Results Evaluation of the classifiers showed that the study type classifier achieved average precision and recall of 0.803 and 0.930, respectively. The PROGRESS-Plus classification proved more challenging with average precision and recall of 0.608 and 0.534. The FAIR database uses information provided by these classifiers to facilitate access to inequality-related public health literature.Discussion Previous work on automation of evidence synthesis has focused on clinical areas rather than public health, despite the need being arguably greater.Conclusion The development of the FAIR database demonstrates that it is possible to create a publicly accessible and regularly updated database of public health research literature focused on inequalities. The database is freely available from https://eppi.ioe.ac.uk/eppi-vis/Fair.NETSCC ID number NIHR133603. When people are making decisions about which services to commission, or which treatments to use, it is important to consider the research evidence. Even well-intentioned interventions can increase inequalities in health between different population groups, so understanding potential impacts here is critical given the need to reduce inequalities. Unfortunately, it is very difficult to find research about inequalities, owing to the way that research is published and stored in large databases (often containing millions of articles).To address the above problem, we built the FAIR database that locates public health research and identifies those papers that are about inequalities. It then uses machine learning to catalog the papers and makes them easy for decision-makers to find. This research paper reports the work involved in developing the database, and the research we undertook to evaluate how accurate the machine learning components are. We found that the machine learning was quite good at identifying the type of research reported but was much more variable when identifying specific types of inequalities. This was due to a lack of data for training the machine learning. We would expect better performance if more training data were available.