Index of suspicion of cancer using prescriptions
Research type
Research Study
Full title
Computational inference from population datasets to improve the early diagnosis of cancer
IRAS ID
246385
Contact name
J Rashbass
Contact email
Sponsor organisation
Public Health England
Duration of Study in the UK
2 years, 0 months, 1 days
Research summary
Cancer survival in the UK trails behind that in other developed healthcare economies. Many studies have suggested that one of the most important contributing factors is that patients in England are detected later, by which time the cancer is at a more advanced and less curable stage. This is particularly true for those cancers with the worst survival – pancreatic, stomach, ovarian and brain cancer - as patients with those patients often present with non-specific.
Recently, Public Health England and the NHS Business Services Authority have created a novel public health data source comprising anonymous, record-level information on nearly all the primary care prescription data in England; for patients with cancer this information is linked to their cancer registration record. This dataset has created a completely new resource ripe for population-based research, of which the proposed research is an example.
Our Pioneer Award is a proposal to develop computational methods to identify patterns in the medication and interactions with secondary care (including imaging) prior to the diagnosis of cancer. This should allow us to create an “index of suspicion” that an individual patient is at increased likelihood of presenting with a cancer. The index could then be used to provide an alert to help the GP recognise patients with cancer at an earlier stage, particularly in those cancers with the worst survival.
The results of this work, once in the peer-reviewed literature, can be rapidly transferred to front-line patient care. For example, our findings may help to enrich clinical decision support algorithms already in use in English primary care, supporting decisions for either referral for investigation for suspected cancer, or active monitoring of patients at risk levels not mandating immediate specialist investigations/referral.
REC name
London - Surrey Research Ethics Committee
REC reference
18/LO/1078
Date of REC Opinion
11 Jun 2018
REC opinion
Favourable Opinion