September 1, 2017
Learn how microsimulation modelling can project outcomes as a result of changes to colorectal cancer screening programs
What you need to know
Data modelling tools can help determine the most cost-effective approaches to screening. This article explores different thresholds for what is considered a “positive” test for colorectal cancer screening using microsimulation modelling.
What is this article about?
Cancer screening is the use of medical tests or procedures to detect cancer in patients before they have symptoms of the disease. Studies show that colorectal cancer (CRC) screening reduces both the number of people diagnosed with CRC and the number of people who die from CRC.
The most commonly used CRC screening test used for average-risk individuals in Canada is the fecal immunochemical test (FIT). The FIT measures the concentration of blood in stools, which can be an early sign of cancer. If the concentration of blood measured in a FIT exceeds a specified value, then the patient is recommended to undergo further investigation, usually by colonoscopy. In Canada, the value which is considered “positive” ranges from 50, 75, 100, and 175 nanograms per millilitre (ng/mL). However, no studies have investigated the effect of these different thresholds on Canadian patients’ long-term health outcomes and health system costs.
Microsimulation modelling is used by researchers to predict a medical intervention’s or policy change’s impact on health outcomes, such as rates of diagnosis, treatments and death. Microsimulation can also predict an intervention’s financial costs or savings.
OncoSim is an online tool that uses microsimulation modelling to assess the impact of cancer interventions on the Canadian population or specific subpopulations. The tool has several modules that simulate the effects of interventions on different types of cancers, including colorectal cancer.
In this study, the researchers used OncoSim to predict the costs and effectiveness of CRC screening at different FIT thresholds. Effectiveness was measured by combining 1) quality-adjusted life-years (QALYs) – a health outcome measure which considers a person’s quantity and quality of life, with 2) estimated costs to calculate cost-effectiveness in Canadian dollars.
What was done?
The researchers did a literature review to determine how accurate FIT was at correctly detecting whether a person had cancer or did not have cancer at different positivity thresholds.
The team used the CRC module in OncoSim to test threshold values of 50, 75, 100, 125, 150, 175, 200 and 225 ng/mL.
What were the key findings?
Compared to no screening at all, OncoSim projected that the effectiveness of FIT screening improved as threshold levels declined. In all scenarios, lower thresholds led to increased QALYs and reduced CRC rates.
Lowering the FIT threshold from 100 ng/mL to 50 ng/mL increased the demand for colonoscopy by 20-77%. Increasing the FIT threshold from 100 ng/mL to 225 ng/mL decreased the demand for colonoscopy by 13-40%.
OncoSim projected that regardless of the FIT threshold level, CRC screening and treatment would reduce the population’s CRC health-care costs compared to no screening at all. This finding indicates that CRC screening at any threshold is more cost-effective than no screening.
While lower CRC screening thresholds are more effective, it also increases the demand for colonoscopy resources because more people will require follow-up and ongoing surveillance. This can negatively impact the quality and availability of colonoscopy services within a region. As colonoscopies are resource-intensive and invasive procedures, determining the most appropriate FIT positivity threshold must be carefully considered.
Why is this important?
This research shows that OncoSim is an important tool that can be leveraged by decision makers to understand how different FIT thresholds for CRC screening affect a population’s future health and quality of life, as well as associated resource implications.