The aim of the study was to evaluate the feasibility of setting up a UK multicentre registry study to document CMR use in patients who activate the PPCI pathway. There were three elements to the study, listed below with the associated objectives.
- A. To determine whether or not it is feasible to set up a CMR registry in this patient population by linking sources of routinely collected data.
- Objective A1: to implement consent and establish patient consent rate.
- Objective A2: to provide evidence of whether or not data linkage and extraction can be carried out across PPCI centres.
- Objective A3: to determine whether the registry can be compiled from hospital episode data [HES and Patient Episode Database Wales (PEDW)] rather than multiple HISs.
- B. To estimate the proportion of the target population who have CMR at two centres with dedicated CMR facilities and to describe resource use and associated costs of having CMR after PPCI.
- Objective B1: to estimate the proportion of the target population who get a CMR scan following PPCI pathway activation.
- Objective B2: to develop simple cost-effectiveness models in relevant subgroups of patients who activate the PPCI pathway in order to identify key drivers of cost-effectiveness for CMR imaging in patients who activate the PPCI pathway.
- C. To identify an outcome measure representing a definitive change in clinical management, conditional on having undergone CMR, that would be credible to cardiologists and other stakeholders as an interim measure of the value added by doing CMR.
- Objective C1: to define, using formal consensus methods, a treatment/process outcome that will constitute a definitive change in clinical management arising from having undergone CMR.
- Objective C2: to identify patient subgroups in whom CMR use is indicated using formal consensus methods and illustrative data from the registry.
- Objective C3: to determine whether or not hospital episode data adequately capture the main changes in clinical management that are identified using formal consensus.
- Objective C4: to pilot the implementation of a primary outcome that is based on the changes in management, for which consensus was achieved using hospital episode data.
Conditional on the success of achieving objectives A, B and C, there was a final objective, D, for which we envisaged that we would define the research objectives and calculate the sample size needed for a cohort study to achieve the aim of evaluating the clinical effectiveness and cost-effectiveness of CMR after PPCI pathway activation, with respect to the outcome identified by objective C.
Changes to objectives during the study
The objectives described above comprise all objectives that were studied rather than the objectives at the outset. The following objectives were added or modified during the course of the study.
With respect to objective A3, we faced several challenges when identifying the eligible population and implementing patient consent in the PPCI centres. These are described in detail in Chapter 3. We therefore wanted to determine whether or not we could identify ACS patients who activate the PPCI pathway in hospital episode data in order to determine whether or not a future registry in this patient population could be set up solely from hospital episode data (Chapter 4, Identifying the eligible study cohort and cardiovascular magnetic resonance exposure in hospital episode data).
With respect to objective B2, we originally planned to identify key cost drivers empirically, by describing resource use and associated costs of having CMR after PPCI pathway activation from HES and PEDW data for participants during follow-up of the cohort. We experienced considerable delays in obtaining HES data, which prevented us from doing this. Therefore, the economists used an alternative approach, seeking secondary sources of data as inputs into economic decision models to identify the key drivers of cost-effectiveness, that is the model parameters that most influence estimates.
With respect to objective C4, we had initially intended to collect follow-up data from each participating hospital to identify the changes in clinical management that could result from CMR (defined through the formal consensus). However, we realised early on that this would not be possible for a proportion of our cohort because many patients whom we recruited had their treatment at the PPCI centres but were quickly repatriated to their local hospital. We therefore decided that following the cohort through hospital episode data would provide more complete data because follow-up data would be available for all patients regardless of which hospital (within England or Wales) they attended after their index admission/procedure.
With respect to objective C3, some of the cardiologists in the research team were sceptical about our ability to pick up potential changes in clinical management resulting from CMR using HES and PEDW data. They highlighted that changes in clinical management may be too subtle to be captured in a data set that records only information about inpatient, outpatient and accident and emergency (A&E) attendances. In contrast, electronic data sources in many hospitals capture a range of additional information, for example prescribing data and patient-level nursing observation data. We therefore designed a questionnaire requesting specific information from the referring cardiologists about how CMR changed management in all patients in our study who were referred for a CMR. We determined whether or not changes in management described in the questionnaire were adequately reflected in hospital episode data [see Chapter 6, Implementation of a primary outcome based on the changes in management for which consensus was achieved (objectives C3 and C4)].
Major delays with the study
The major delay to the study was receiving the HES/ONS linked data set from NHS Digital. Below is a timeline of our data request and the processing of our application:
- 24 June 2013 – application was submitted to NHS Digital. We were told that our application would be processed closer to the time at which we required the data (September 2014), because our request was straightforward and the data set would not take long to assemble.
- March 2014 – NHS Digital commissioned an independent review into data release and stopped processing all applications.
- March 2015 – NHS Digital began processing our application.
- August 2016 – linked HES/ONS data received.
This process added a 23-month delay to the study.