A risk assessment model integrates information from many sources including biological, clinical, genetic, epidemiological and behavior studies to predict your disease risk. Such model can answer questions like what is the probability that you will get lung cancer if you continue smoking. The risk assessment models are widely used as tools to identify individuals with high risk, hence, they have significant public health implications.
Risk assessment has gained widespread attention from the oncology community in recent years. Although treatment of cancer has greatly improved due to advances in pharmacology and biotechnology, the survival rate of advanced cancer patients is still low. For instance, only 10-15% of patients with advanced lung cancer survive 5 year following diagnosis. Thus, it is important to identify individuals with high risk to apply preventive activities. Alternatively, these individuals can be screened for cancer if they have cancers, the cancers will be detected at early, curable stages rather than at later, terminal stages.
According to Johnson et al (2007), more than 3000 studies related to assessing risk cancer were published between 2003 and 2007. In the last decade, the number of risk models of cancer has increased rapidly, covering 15 common cancers: breast cancer, colorectal cancer, lung cancer, prostate cancer, ovarian cancer, testicular cancer, skin cancer, and gastric cancer. So far there has been no systematic effort to unify all these models within a consistent framework.
Many models published in literature only focuses on one aspect of the natural history of cancer, either about the incidence, mortality or recurrence of cancer. Risk is often reported in terms of absolute risk (i.e. the probability of a certain event over a specified period) or relative risk (i.e. the ratio between the absolute risks of the exposed and unexposed groups). An example of absolute risk is the lift-time risk of developing colorectal cancer is 5.5%. An example of relative risk is the relative risk of having colorectal adenomas in obese individuals (BMI 35 or greater) in comparison to individual with normal weight (BMI = 20-25) is 1.5.
The main issue of communicating risk to public is that half of Americans do not have sufficient basic mathematical skills to make use of the numbers given to them. According to Yamaghishi (1997), many Americans think 1286 out of 10,000 is greater than 24.4 out of 100. People also have a difficult time to interpret what a low probability, like 6% lifetime risk, mean to them, as in the case of colorectal cancer. Should they worry about 6% or should they do something to reduce their risk to 3%? Clearly, risk communication is a challenging problem that public health educators are currently facing.
Johnson and Molenski, Risk assessment model to predict cancers, Current Oncology Reports, 2007