1. QCancer www.qcancer.org Professor Julia Hippisley-Cox
Clinical practice
Research questions
Research
Innovate
Implement
Professor of Clinical Epidemiology & General Practice
Professorial Fellow St Annes, Oxford
GP Woodstock
Director QResearch (non-profit collaboration with EMIS Health)
Director ClinRisk Ltd (software company)

2. GP Data QResearch database founded in 2003
Collaboration with EMIS Health
One of the largest databases of linked electronic health records internationally
Anonymised data from > 1500 practices
Individual data > 30 million patients
Historical Data > 25 years
World beating resource for innovative medical research
Many tools integrated into EMIS Web
GP Data
ONS mortality
Hospital data
Other data sources
Cancer Registry
Generating new knowledge to improve patient care

3. Generating new knowledge to improve patient care
Clinical practice
Research questions
Undertake Research
Innovate
Implement
Generating new knowledge to improve patient care

4. Risk Prediction Tools
Developed, validated and implemented suite of novel risk prediction tools
Prognostic models - Risk of future disease
Diagnostic models - Risk of current disease
1st time national GP data used to develop risk tools
Enables informed consent - improved risk communication
Enables risk stratification – targeting resources to highest risk & those most likely to benefit from interventions
Screening & prevention
Early diagnosis
Independent, external and international validations undertaken
Recommended in NICE guidelines & implemented into NHS computer systems
Generating new knowledge to improve patient care

5. QTools-absolute risk of major outcomes
QCancer - 3 aspect
risk of current cancer
10 year risk of cancer
prognosis for patients with cancer
QRISK3
QStroke
QDiabetes
QFrailty & QMortality
QAdmissions
QBleed
QKidney
Generating new knowledge to improve patient care

6. Early diagnosis of cancer: The problem
300,000 new cancers each year
UK has relatively poor track record when compared with other European countries
Partly due to late diagnosis with estimated 7,500+ lives lost annually
Later diagnosis due to mixture of
late presentation by patient (lack awareness)
Late recognition by GP
Delays in secondary care
Earlier diagnosis cancer improves chances of survival

7. NICE guidance on suspected cancer (2015)
Symptom based approach
Lower threshold for intervention to 3% PPV of individual cancers
More investigation and referral
Didn’t make explicit reference to use of risk assessment tools
Used research underpinning QCancer to develop guidance
Qcancer provides practical way to identify patients at risk to improve early diagnosis
QCancer now in EMIS Web

8. Why symptoms based approach?
Many patients present with symptoms
GPs need to decide which patients to investigate and refer
Decision support tool must mirror setting where decisions made
Symptoms based approach needed (rather than cancer based)
Must account for multiple symptoms
Must have face clinical validity eg adjust for age, sex, smoking, Family history
Need to be able to update to meet changing requirements, populations, recorded data

9. QCancer– what it needs to do
Accurately predict individual level of risk of multiple cancers for based on multiple risk factors and multiple symptoms
Discriminate between patients with and without cancer
Help guide decision on who to investigate or refer and degree of urgency.
Educational tool for sharing information with patient. Sometimes will be reassurance.
QCancer combined model published BJGP January 2013
Cancer in women
Cancer in men

10. Methods – development algorithm
Representative cohort from QResearch 2.5 million men & women aged years
cancer outcome - all new diagnoses on GP record or linked record in 2 years
Identify key risk factors
Identify key symptoms
Established methods to develop risk prediction algorithm
Measure of absolute risk of any cancer as well as by cancer type

11. QCancer predicts global cancer risk & risk12 types cancers
Lung
Pancreas
Renal tract
Ovary
These accounts for 85% cancers
Plan to extend to rarer cancers when sufficient data
Colorectal
Gastro
Testis
Cervix
Breast
Prostate
Blood
Uterus

12. Key Risk Factors in QCancer
Demographics
life style
Co-morbidities
Age
Sex
Smoking status
Alcohol use
Deprivation score
Family history of cancer
COPD
Endometrial hyperplasia/polyp
Chronic pancreatitis
Type 2 diabetes
Anaemia (HB < 11g/DL)
Venous thromboembolism

13. Key symptoms in model general Bleeding Loss of appetite
Unintentional weight loss
Indigestion +/- heart burn
Dysphagia
Abdominal pain or swelling
Constipation
Night sweats
Cough
Neck lump
Haemoptysis
Haematemesis
Haematuria
Rectal bleeding
Unexplained bruising
Men
women
Vaginal Bleeding
Breast lump
Breast pain
Nipple changes
Genito-Urinary symptoms (men)
Retention
Nocturia
Frequency
Impotence
Testicular lump

14. Don’t forget QCancer works out cancer risk
Validation of QCancer
Essential to demonstrate the tools work and identify right people in an efficient manner
Tested performance
separate sample of QResearch practices
external dataset (Vision practices)
Good at identifying those who do and don’t have cancer
Good at estimating level of risk
Real world testing with Macmillan and CRUK to evaluate its use in > 250 practices across the UK.
Currently used in the NHS England ACE program
Don’t forget QCancer works out cancer risk
NOT diagnosis

15. Using QCancer in practice
Standalone web calculator
Integrated into clinical system
Within consultation: GP with patients with symptoms
Within consultation QOF style alert
Run in batch mode to risk stratify populations

16. QCancer Protocol Alert
Alert on all sites now (version 5.5)
Deployed in INACTIVE form
Practices need to activate it:
Alert will appear if codes added during consultation on fly e.g. if add haematuria, it will pop up
Click on score will take you to template

17. QCancer symptom checker template
Template to capture relevant symptoms & risk factors

18. QCancer site specific scores

19. QCancer follow up template

20. QCancer Batch Add
Similar to QRISK2 which is in 95% of GP practices– automatic calculation of risk for all patients in practice based on existing data.
Safety netting - Identify patients with symptoms/adverse risk profile without follow up/diagnosis
Enables systematic recall or further investigation
Systematic approach - prioritise by level of risk.
score

21. Thank you for listening & any questions