What is QAdmissions®?
QAdmissions® is a clinical risk prediction tool which calculates an individual's risk of emergency admission to hospital over the next one or two years. The algorithm can be used to identify patients at high risk of an emergency admission so that they can be assessed in more detail to help reduce their risk.
Why has QAdmissions® been developed?
Emergency hospital admissions cost approximately £11,000,000,000 a year in England; and they are also potentially distressing to individuals. In 2013, NHS England announced a new Enhanced Service Specification to reward GP practices for the identification and case management of patients identified as seriously ill or at risk of an emergency admission.
As part of this, GPs need to undertake regular risk profiling and risk stratification of their patients. One problem which has beset all existing risk algorithms for emergency admissions is the practical difficulty in implementing them in primary care since they have not been designed to run off routinely collected data already in GP computer systems or validated in that setting.
We therefore decided to develop and validate a new risk prediction algorithm to predict the absolute risk of emergency admissions to hospital which could run off data that:
- can be automatically populated using data solely from GP computer systems and so provide an expedient practical alternative where primary care data are not routinely linked to secondary care data and
- incorporate clinically relevant variables which the health professional can then follow up.
Who can have a QAdmissions® score calculated?
QAdmissions® calculate a risk of emergency admission for any person aged 18-100 years.
How can I work out the risk of an emergency admission to hospital?
The QAdmissions® web calculator qadmissions.org can be used to work out your risk of an emergency admission to hospital. Software is also available to build QAdmissions® into existing clinical systems.
If you do not know some of the information needed for the calculator (like body mass index) then it will substitute population average values for someone of the same age, sex and ethnicity.
What data are needed to calculate a QAdmissions® risk?
The QAdmissions® calculation is based on variables stored in the patient's electronic health record including age, sex, and ethnicity, as well as clinically relevant variables such as diagnoses, medication, prior admissions and laboratory measurements.
How can a GP identify patients at high risk of an emergency admission?
QAdmissions® risk can be calculated using data already recorded in GP computer systems. This is a unique feature of QAdmissions® as it means it can be integrated into the GP's computer system and used without the need for disclosing identifiable data outside the practice setting.
QAdmissions® can also be calculated using GP data linked to hospital data where this is available, (e.g. a CCG's data warehouse), although this would require a legal basis for use of identifiable data. This would also be possible to implement with pseudonymised data.
How can QAdmissions® be used in a clinical setting?
QAdmissions® works in three modes:
- It can be used for an individual patient in the surgery setting to calculate their admission risk. This could be done either via qadmissions.org or an equivalent tool integrated into the GP computer system.
- It can be applied to the entire practice population in 'batch mode' to generate an admission risk for each patient so that patients at highest risk can be identified and prioritized for interventions to reduce their risk. This is best when QAdmissions® is integrated into the heart of the GP computer system, or within the data warehouse.
- Some of the underlying parameters of the algorithm are ones that the patient, with the right help and support, can do something about. The algorithm can therefore be used to target appropriate interventions across a population, with the intention of reducing emergency hospital admission within a high risk patient group.
Can I have some examples of interventions?
QAdmissions® includes the following modifiable factors:
- Smoking status - smokers have higher risks than ex-smokers or non-smokers
- Alcohol intake - heavy drinkers have higher risks
- Anaemia
- Raised platelets
- Abnormal liver function tests
- Obesity
For example, a patient with anaemia, raised platelets or abnormal liver function tests, has a higher risk than a patient without these features. The risk is likely to be related to the underlying cause for the abnormal blood test results and require further investigation or re-checking.
Some types of medicines are associated with a higher risk of emergency admission and this is why medication has been included in the QAdmissions® algorithm. The QAdmissions® tool can identify a patient with high risk who may be suitable for a medication review, e.g. to check whether the dose or type of medication needs changing, or, indeed, whether additional measures might help.
QAdmissions® will also identify patients with multiple co-morbidities who may be helped by a community care package or virtual ward.
Has QAdmissions® been scientifically reviewed and published?
Yes. The research underpinning QAdmissions® has been scientifically reviewed and published in BMJ Open.
How has QAdmissions® been validated?
QAdmissions® has been tested on two separate populations of primary care patients and shown to be good at identifying high risk patients. This validation was done using a section of the QResearch database not used to develop the model and also using the Clinical Research Practice Data Link (which uses data from a different GP system supplier to QResearch, so it is doubly independent). The results will be published in BMJ Open.
How often will QAdmissions® be updated?
QAdmissions® will updated annually to account for evolving requirements, improvements in data quality and changes in the population to which is it applied.
Ten key advantages of QAdmissions®
- Practical: QAdmissions® is the only score which can run entirely on GP data.
- IG safe: QAdmissions® does not require identifiable data to be extracted from primary care thereby avoiding unnecessary breaches in patient confidentiality.
- Clinically relevant: QAdmissions® has good clinical validity including factors known to increase admission risk and those which can be investigated or changed.
- Scientifically robust: The research has been undertaken by leading experts using established scientific methods used for closely related risk prediction tools such as QRISK®2, QDiabetes® and QFracture®. QAdmissions® has been scientifically reviewed and published in a peer reviewed journal.
- Validated: The algorithm has been validated on a separate set of practices and patients and shown to be reliable and accurate.
- Transparent: QAdmissions® is transparent both for the factors used to calculate the risk and how the calculation itself works. The heart of the algorithm is published as open source software.
- Suitable for use across England: QAdmissions® is up-to-date and modelled on a population of primary care patients across the country. The weightings for each Strategic Health Authority and deprivation help tailor the score for local use.
- Purpose designed: QAdmissions® has been designed by GPs for GPs and can be integrated into their computer system.
- Updatable: QAdmissions® will be regularly updated to take account of changes to the population, data or policy.
- Available: QAdmissions® can be implemented by any GP system supplier, CCG or CSU by licencing our QAdmissions® Software Development Kit.
What thresholds are appropriate to use with QAdmissions®?
See QAdmission-Thresholds-1.1.pdf.
Updates
The 2015 update is described here: QAdmissions-2015-Annual-Update-Information.pdf
The 2016 update is an update to the Townsend deprivation table.