Validation of QPrediction scores in QResearch and CPRD (QResearch ref SPCR 93)
What is the aim of the study and why is it important?
The QResearch database and the Clinical Practice Research Datalink (CPRD) databases are two large general practice databases which are used for research. The CPRD was set up in 1988 and is of similar nature and size to QResearch although it is derived from practices using a different clinical computer system and covers a different population.
A series of risk prediction models have been developed using the QResearch database to predict a wide range of important clinical outcomes for patients. The algorithms were originally developed using a 2/3rd sample of practices contributing to the QResearch database and validated on the remaining third. None have been validated on the CPRD.
The objectives of this study are to compare the QResearch and CPRD databases for the following a range of measures over time:
Demographic characteristics: age/sex/ethnicity/deprivation
Recording rates of data for clinical values such as smoking status, body mass index, blood pressure and laboratory data
Incidence, prevalence, mortality and survival of a range of chronic diseases
To examine the linked ONS deaths, HES and Cancer Registry data on both databases and evaluate its usefulness e.g. what % of total cases of a serious outcome (such as heart disease, cancer, renal failure, thrombosis, diabetes) can be identified using the GP record alone and what % using only the ONS linked death record, HES data and the Cancer Registry data
To validate the performance of a series of risk prediction scores which have been developed on the QResearch database
Professor Julia Hippisley-Cox
Location of research
Date on which research approved
Project reference ID
Generic ethics approval reference
Are all data accessed are in anonymised form?
Brief summary of the dataset to be released (including any sensitive data)
Risk factors and outcomes for multiple clinical conditions on GP, HES, cancer registry and mortality data
National School for Primary Care Research
Professor Julia Hippisley-Cox, Professor Carol Coupland, Dr Peter Brindle
Department of Health Disclaimer
The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the NIHR School for Primary Care Research, NIHR, NHS or the Department of Health.
The performance of seven QPrediction risk scores in an independent external sample of patients from general practice: a validation study
Authors: Hippisley-Cox J, Coupland CA, Brindle P
Ref: BMJ Open 2014;4(8):e005809