How well are outcomes within Core Outcome Sets (COS)

How well are outcomes within Core Outcome Sets (COS) relevant to child health captured within Electronic Health Records during routine clinical care at Great Ormond Street Hospital?

Background:
Core outcome sets (COS) represent the minimum set of patient health outcomes that should be measured and reported for a specific condition (1). They are developed to improve the potential for clinical effectiveness research, including the ability for results from clinical trials to be combined where appropriate. Whilst driven by an aim to improve the evidence base from clinical trials, they are also an opportunity to benchmark routine clinical care. There is increasing interest in the potential for COS to be collected in routine care via electronic health records (EHR) (2). COS may also include patient-reported outcomes. By integrating data from electronic health records at GOSH (using its EPIC software system) with any relevant electronically captured PROs, a dashboard for COS could be developed to summarise patient progress at an individual level and in comparison to the wider patient cohort with the same age and clinical diagnosis at GOSH. This is likely to be highly attractive for clinical care teams to benchmark care, but the data pipeline developed for this process would support clinical trials in routine care, e.g. using a Trials within Cohorts design (3) or platform trial.

Aims/Objectives:
Development of a pipeline to summarise clinical and patient-reported data in core outcome sets (COS) collected during routine care. The student will then utilise this to a) highlight areas where COS could be better-captured within usual clinical care at GOSH, and b) investigate its utility for quality improvement and in supporting future clinical trials embedded within routine care.

Methods:
1. Explore ‘what is needed’ and ‘what is available’ from routine clinical care data in relation to evidence based COS, using 1-2 conditions as example (e.g. asthma, cystic fibrosis, obstructive sleep apnoea). Literature review of trial endpoints and frequency of outcome reporting in comparison to granular patient-level EHR data. In absence of relevant COS, develop concept of ‘minimum dataset’ that could be captured in routine care, in line with COMET guidance (1). Explore feasibility of this from perspective of clinical care teams – what is done already, what else would be required, is it only new ways of processing/summarising data that is already captured.
2. Develop pipeline to populate both a clinical dashboard and a research database with outcomes for clinical effectiveness studies including prospective trials. Investigate factors such as: case definition, outcome/variable definition, granularity, frequency of data capture in routine care (e.g. spectrum and relationship with age, disease ‘severity’). Ensure all code relevant to above satisfies interoperability standards in order that this could be deployed beyond GOSH/systems using EPIC.
3. Evaluate using existing EPIC data within 1-2 conditions, and validate with period of prospective data collection.
4. Compare results for exemplar condition(s) with linked data from NHS Digital, in terms of areas of overlap/potential benefits of the wider COS approach within routine care.

References:
1. COMET Initiative. Core outcome measures in effectiveness trials. http://www.comet-initiative.org/
2. Dodd S et al. Core outcome sets through the healthcare ecosystem: the case of type 2 diabetes mellitus. Trials 2020; 21:570, https://doi.org/10.1186/s13063-020-04403-1
3. Relton C et al. Rethinking pragmatic randomised controlled trials: introducing the “cohort multiple randomised controlled trial” design. BMJ 2010;340:c1066, doi:10.1136/bmj.c1066