National Institute for Health Research (NIHR) Cardiovascular Health Informatics Collaborative

The NHIC Cardiovascular Project in the United Kingdom was established to enable the sharing and repurposing of routinely captured clinical data for re-use in research. Data sharing for this study has been enabled by establishing a data sharing agreement between each of the collaborating hospitals. The data sharing agreement allows data to be shared between National Health Service (NHS) hospitals in accordance with an anonymisation and de-identification profile approved by each information governance department, ensuring that patient identities are protected. Ethics approval for each dataset has been obtained which detailed the further de-identification steps to ensure that research datasets are fully anonymised.

The cardiovascular theme clinical leads developed a standard target data model specification to capture the longitudinal record for patients who presented with a suspected acute coronary syndrome, characterised by the request of a troponin test (NHIC-Troponin Study). The data model was patient centric allowing for minor discrepancies between dataset areas such as missing discharges or test results returning outside of episodes of care. The model included 156 data points, grouped into demographics, cardiovascular risk factors, emergency department attendance and inpatient episodes, biochemistry, revascularisation and mortality. All patients were followed up on the NHS Spine Application, Summary Care Record until death or censoring on 1st April 2017.

This dataset comprised all patients who had a troponin measured at each of the five major academic centres between 2010 (2008 for University College Hospital) and 2017. The investigators identified a total of 257948 patient records who underwent troponin testing during the study period.

A Committee of Experts validates the protocol methodology and supervises the data management. A Steering Committee oversees study proposals including hypotheses, study design and statistical analyses including planned outcome measures.