Using Routine Clinical Datasets to Develop Risk Algorithms in Oncology
Research type
Research Study
Full title
The use of routine clinical datasets to develop decision support rules and risk algorithms in cancer patients on treatment
IRAS ID
121988
Contact name
Geoff Hall
Contact email
Sponsor organisation
Leeds Teaching Hospital NHS Trust
Research summary
The treatments used to treat cancer, including both chemotherapy and radiotherapy, have significant toxicity. The side-effect with the greatest potential to harm the patient is the effect on the function of the bone marrow and resulting reduction in the numbers of circulating blood cells. A reduction in the white cells, particularly the neutrophils which fight bacterial infections, places the patient at significant risk of life-threatening infection. The identification of this reduction in neutrophils (neutropenia) with a simple blood test is required in all patients prior to the next cycle of treatment and in all patients who develop signs and symptoms of infection.
This toxicity may lead to treatment delay, episodes of febrile neutropenia requiring hospitalisation, neutropenic sepsis and ultimately death. Significant neutropenia also often leads to dose reductions with subsequent cycles, which may adversely affect the efficacy of cancer therapy. However, not all patients will develop these complications and there is currently no method to accurately predict which patient is going to experience severe, potentially life-threatening, neutropenia until they become unwell.
The research proposed will use data from the oncology patient electronic patient record (EPR) from Leeds Teaching Hospitals Trust (LTHT) to provide detailed information on patients receiving cancer treatment. This will allow the development of mechanisms to (i) predict how a particular patients’ white blood cell count will change between treatments, (ii) to predict adverse events in patients and (iii) to develop algorithms for incorporation into Clinical Decision Support software. By using predictive modelling and baseline risk assessment, the profile of persons at the highest risk of complications could be defined and management altered to reduce rates of complication including treatment delay, dose reduction and neutropenic complications.
REC name
North of Scotland Research Ethics Committee 2
REC reference
13/NS/0128
Date of REC Opinion
17 Sep 2013
REC opinion
Favourable Opinion