Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

Our prognostic modelling team performs external validation of existing predictive models and builds new models with clinical data

Prognosis is the forecast or estimate of the risk of something happening in the future. In medicine, we are interested in predicting a person's future health state using their current characteristics, such as their height, weight, history, blood markers, or X-ray results. We may be interested in whether they will develop a particular disease, whether they will have recovered from a current disease, or whether their pain levels will have changed, for example.

Prognostic models are mathematical models that relate a person's characteristics now to the risk of a particular future outcome. Prognostic models can take into account one or many current characteristics (multivariable). These models are useful tools for doctors, as they provide objective estimates of the probability that something will happen, to use alongside other clinical information.

There are three stages to prognosis research:

1. Development

The ideal model uses multiple pieces of information from the patient to give accurate estimates of risk, but is also easily used by clinicians during their day-to-day work. These two criteria are difficult to balance.

We are involved in the development of new prognostic models. Sometimes we use existing published datasets of patient characteristics and disease outcomes, and sometimes we are involved with a study to collect specific information for building a model. We are currently developing prognostic models in the following therapeutic areas:

  • Diabetes and metabolic research
  • Gastrointestinal research
  • Maternal, foetal, and newborn health
  • Orthopaedics and musculoskeletal research
  • Perioperative care

2. Validation

Building and publishing a prognostic model is just the first step. The model has to be tested to make sure that it can accurately predict outcomes using data that it has never seen before. Depending on where the data came from that the model was built with, the model may not be generalisable to all people.

External validation is the process of checking a published model using different data. We internally validate all of the models that we develop. We are also involved in external validation studies in the following therapeutic areas:

  • Excretory system research
  • Orthopaedics and musculoskeletal research

3. Impact assessment

A validated model can be used by clinicians. We need to follow up with clinicians to check whether using the model improves their diagnosis and patient outcomes, or whether the model is unhelpful.

Improving prognostic modelling

We also conduct applied statistics research to improve the methods used in prognostic modelling.