Approved Research
Detecting high-risk smokers in Primary Care Electronic Health Records
Approved Research ID: 74471
Approval date: February 24th 2022
Lay summary
The aims of the project are to:
- Evaluate the quality and validity of EHRs in Scotland to identify and characterise patients with lung cancer.
- Develop a risk model using information from EHRs.
With the main research question:
What factors/data features, contained in electronic health records (EHRs), are associated with and produce estimates of risk for lung cancer in individuals that smoke?
Rationale and impact
Although lung cancer screening has been recommended to aid in the early detection of lung cancer, it is yet to be implemented. To enable implementation, further research exploring methods to identify eligible screening participants and optimum risk thresholds for inclusion would be required. As this project will examine whether Electronic Health Record (EHR) data can be used for the identification of smokers at high risk of developing lung cancer, it will address the need for further research.
EHRs can provide researchers with data on metrics and features that may be challenging to obtain using other methods. There is also compelling evidence that EHRs can provide researchers with information on smoking behaviour in individuals. The accuracy and validity of this data is yet to be identified, as this will vary depending on the reference used and the accuracy with which data is extracted. Regardless, use of EHRs in epidemiological research has the potential to reduce late lung cancer diagnosis and explore data features such as symptoms and pack years that would have otherwise been difficult to accurately obtain. As such, this project will aim to expand knowledge around lung cancer risk research by seeking to identify smokers in EHRs and subsequently utilising this information in modelling risk of incidence. If EHR information can be used to accurately predict risk of lung cancer, a risk prediction score can be developed which GPs can feasibly use to aid them in their identification of at risk patients.
Duration
The project will last 3 years.