| Title: | Validation and comparative analysis of novel prediction models focussed on modifiable lifestyle factors for the risks of common, preventable diseases and all-cause mortality: a cohort study |
| Lead Institution: | Huma Therapeutics Ltd |
| Principal investigator: | Mr Sam Nikbakhtian |
Application 55668
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
This project will use the UK BioBank to measure how effectively a new general health score can predict the risk and incidence of a basket of common preventable diseases such as Type II Diabetes, Heart Disease and Stroke. The score is made up of metrics that are easily measured on a smartphone (user input, or sensor-based), known to be strongly linked to well-established and changeable risk factors, which are combined through a weighted algorithm to give a single health score. Health scores which work in this way are often referred to as 'risk models' or 'risk scores' and are an established way of adapting existing scientific evidence for use in public health or clinical settings. To date, risk scores have been used to predict a large number of diseases, most notably cancer, diabetes and depression. These population-level predictive indices are often created by reviewing past research and are validated (shown to be accurate) through large statistical studies on biomedical databases such as the UK BioBank. To do this we plan to calculate a general health score for a selected number of BioBank participants using information from the existing dataset on a fixed date. This will be called 'baseline'. Any gaps in the available data will be filled using statistical methods to ensure the information needed for the algorithm to produce a score is complete for all participants. We will compare the baseline health scores of participants who have preventable diseases such as Type II Diabetes, Heart Disease and Stroke with that of those participants who do not. This should indicate whether the health score is able to predict preventable disease at baseline. The second phase of the research will investigate whether or not people develop diseases over longer periods of time is estimated reliably by the app-based health score. The general health score, if it is shown to be accurate, could be very useful for doctors, nurses and clinical researchers, as it could improve and inform the advice and treatment people receive, as well as health policy decisions. If delivered as part of a preventative health campaign or digital tool, a new health score could be used to educate the general public on risk of different preventable diseases, drive engagement in health, and ultimately act as a platform for behaviour change reducing preventable disease incidence.2 Publications
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 12662 | Accelerometer-derived sleep onset timing and cardiovascular disease incidence: a UK Biobank cohort study | Shahram Nikbakhtian (+6) | 2021 | European Heart Journal - Digital Health |
| 7439 | Development of an accessible 10-year Digital CArdioVAscular (DiCAVA) risk assessment: a UK Biobank study | Nikola Dolezalova (+6) | 2021 | European Heart Journal - Digital Health |
Enabling scientific discoveries that improve human health