Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Developing statistical approaches to analyse and report reinterventions in children who have heart surgery for national benchmarking and quality improvement.

   GOSH BRC Applied Child Health Informatics Theme (Non-Clinical) PhD Studentships

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr D Ridout, Prof Katherine Brown  No more applications being accepted  Funded PhD Project (UK Students Only)

About the Project

Congenital heart disease (CHD): CHD affects approximately 5600 live-born children annually in England and Wales and is the commonest cause of infant death due to congenital anomalies in the United Kingdom (UK). Annually, 7000 paediatric cardiac procedures are undertaken in the UK, 58-60% of them in children under a year old. In very complex CHD, with only one functional ventricle, children need serial surgeries to achieve stable palliation consisting of at least two in infancy. For many CHD types with two functional ventricles, one-stage repair in infancy is standard practice and more than one operation may be required for smaller or more complex patients since they need a staging surgery; or as children grow, they may need reintervention to accommodate their increasing size.

The national congenital heart diseases audit (NCHDA): The research analysis will focus on the NCHDA data with linkage to other important dataset providing long term survival status and hospital care episodes. NCHDA data is of high quality, as evidenced by annual independent validation of children’s heart surgeries at the 11 specialistic centres in the UK. CHD and paediatric cardiac operations are very diverse and complex and are described in NCHDA using a special coding scheme called the International Pediatric Congenital Cardiac Code (IPCCC). The specific primary CHD diagnoses and paediatric cardiac operations are described in terms of their complexity or risk of death based on code combinations, using previously developed and tested methodology from our research group.

Paediatric cardiac surgery outcomes: In children undergoing paediatric cardiac surgery the early post-operative mortality rate is currently <2%, and this means that to understand disease and treatment impacts better, with a view to further care improvements, we need to look beyond this early post-operative time window. In our recent CHAMPION project (to which this proposal is aligned), key datasets were linked for ~ 60,000 children with CHD and longer-term survival was ascertained. For the most complex CHD, hypoplastic left heart syndrome survival is 54% at 10-years of age and for a less complex condition, tetralogy of Fallot survival is 95% at 10-years of age. We found that the range of procedures that children undergo varies. For hypoplastic left heart syndrome children have a median (IQR) of 3 (2,4) procedures by the age of 5-years and 44% have at least one ‘off pathway re intervention’. With tetralogy of Fallot, children have a median (IQR) of 1 (1,2) procedure by the age of 5-years, and 20% have at least one re intervention. As we detailed in the PPI section, patients and users have identified the occurrence of reinterventions as a key metric for people living with CHD. Reinterventions lead to more time in hospital and important health burdens for patients. Quality improvement initiatives directed at reducing rates and impacts of reinterventions require that these are measured and recorded in a coherent and robust way. There are important methodological and statistical challenges to meet before this can be undertaken.


The aims are to describe the burden of cardiac reinterventions for the national cohort of children with complex CHD;to explore applicable statistical approaches for future analyses of cardiac reinterventions as competing events in national audit and benchmarking, and to assess the impact of cardiac reinterventions early in life on subsequent survival.


1. To identify cardiac unplanned reinterventions meaning off pathway surgeries and interventional catheterisations undertaken over and above the interventional treatment pathway amongst children with complex sentinel CHDs   

2. To explore a range of statistical techniques applicable to the analysis of unplanned reintervention, accounting for competing events, for children surviving with complex sentinel CHDs

3. To identify risk factors for cardiac reinterventions undertaken over and above the established treatment pathway for children with complex sentinel CHDs

4. To explore links between early cardiac re interventions undertaken over and above the established treatment pathway in complex sentinel CHDs and the outcome of long-term survival


Study design:

Retrospective cohort study based on linked electronic health care record data for ~ 60,000 children and adolescents with congenital heart disease.

a) National Congenital Heart Diseases Audit, (NCHDA)

c) Hospital Episode Statistics (HES),

d) Deaths from Office of National Statistics (ONS).

Inclusion criteria: The CHD types were selected (by clinicians and patients) as part of the CHAMPION project, as major conditions that affect young children, necessitate intervention at a very early age to survive, and commonly require more than one procedure: single ventricle - hypoplastic left heart syndrome, functionally univentricular heart conditions (double inlet left ventricle and tricuspid atresia), single and two ventricle - pulmonary atresia all types, two ventricle defects - transposition of the great arteries, tetralogy of Fallot, atrioventricular septal defect, congenital aortic stenosis, significant ventricular septal defect, and coarctation of the aorta.

Study period: We will include children who were born and were recorded as having any surgical procedure for one of the selected CHD types between 2000 and 2022, with complete follow up data available to 2023.

Exposures and comorbidities: gender, ethnicity, deprivation, region, age, preterm birth, non-cardiac underlying health conditions as previously defined.

Study outcomes: for objectives 1-3) the outcome is cardiac reintervention and for objective 4) the outcome is survival.

Statistical methods

Objectives 2 and 3 – Reinterventions will be explored using both time to event analysis and competing risks analysis (competing events include death, heart transplant etc.). It is likely that conditional models will be explored, based on whether patients have achieved or completed different stages of their planned clinical pathway and the use of multi-phase models can be investigated. Models will be adjusted for known, pre-specified covariates and time varying covariates will be considered as appropriate.

Objective 4 - the use of novel causal mediation methods will be explored, to understand the relationship between case complexity, reintervention as a mediator and the outcome of long term survival.


Year one – descriptive analysis objective 1)

Year two – develop time to event and competing risk analysis objectives 2) and 3)

Year three – exploring links between reinterventions and outcome objective 4)

Mathematics (25) Medicine (26)
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

 About the Project