It’s estimated that approximately 2 in 5 Canadians will develop cancer during their lifetime, and that 1 in 4 will die from the disease1. Cancer affects or touches almost everyone in this country in some way, and a lot of research has gone into preventing and treating the disease. The overall 5-year survival rate for all cancers in Canada was 63% for 2006-20081. It’s well known that cancers that are caught early have a higher chance of successful treatment and survival. One of the ways that we can diagnose these early-stage cancers is through screening.
Cancer screening is performed on apparently healthy individuals to find early, asymptomatic disease. It’s employed on a population level – for example, it’s recommended that women aged 50-74 in Ontario receive breast cancer screening in the form of a mammogram every 2 years2. An important factor to note is that we’ll never be able to detect 100% of cases 100% of the time. Our ability to detect disease is limited by technology.
Assessing screening tests
A screening test can be evaluated based on its ability to correctly separate healthy and diseased individuals. Sensitivity measures the proportion of truly diseased individuals that the test identifies (i.e. the true positive rate), while specificity measures the proportion of truly non-diseased individuals the test identifies (i.e. the true negative rate). In practice, a new screening test is compared to a ‘gold standard’ that is assumed to represent the truth.
Test to be evaluated Gold standard test (‘truth’)
Table 1: a 2×2 table to evaluate a screening test. Sensitivity= a/a+c; specificity=d/d+b; PPV= a/a+b; NPV= d/d+c
We can also look at positive predictive value, which is the probability that a person with a positive test truly has the disease. Negative predictive value is the probability that a person with a negative test result truly does not have the disease.
Other considerations around screening are ethical in nature; for example how invasive is the procedure? What is the impact of false positive test results (worry, stress, additional tests)? What is the likely uptake (how many will people actually get screened)? How many people need to be screened to save 1 life? Do we have the resources (financial etc.) needed to implement the screening program on a population level? Sidenote: it’s unethical to implement a screening program if there is no effective treatment for the disease being screened for, or if early detection provides no tangible benefit.3
Example: colon cancer
Let’s walk through the ‘who,’ ‘how,’ and ‘why’ of cancer screening, using colon cancer as an example. The Canadian Task Force on Preventive Medicine recently updated their guidelines on screening for colon cancer, so this is a good time to take a look at what goes into these recommendations.
For colon cancer screening, colonoscopy is the gold standard test. Its sensitivity has been estimated to be about 94%4. Sigmoidoscopy, a test that looks at only the sigmoid/lower half of the colon, is another screening test that can be used to detect colon cancer. The sensitivity of this test varies based on whether it’s performed alone or in conjunction with other tests. For a single fecal occult blood test, sensitivity is 30%, but when the test is repeated three times, sensitivity increases to 92%5.
Although colonoscopy has very high sensitivity and specificity, it’s a very invasive screening test requiring sedation, and serious complications occur in about 25 per 10,000 procedures5. It’s also an expensive screening test, costing an estimated $320 in Ontario. Applied to the population level, this quickly adds up6. However, gastroenterologists currently prefer colonoscopy to other options because it’s also therapeutic – precancerous polyps can be removed during a colonoscopy.
The Canadian Task Force on Preventive Health Care doesn’t recommend colonoscopy to be used as a screening test on individuals aged 50-74 who aren’t at high risk for colon cancer (i.e. no family history or symptoms such as a change in bowel habits or abdominal pain)7. This is based on weak evidence stating that the benefits of the procedure outweigh its risks. This doesn’t mean that colonoscopy isn’t the best screening test, it just means that we currently don’t have the evidence to support its use on a population-level. Instead, the Task Force recommends screening adults aged 50-74 with a fecal occult blood test every 2 years, or a sigmoidoscopy every 10 years7.
In today’s environment of budget cutbacks, jurisdictions have to balance costs and benefits when considering whether or not to implement a cancer-screening program. It’s one thing to detect cancer early; it’s another to be able to effectively treat all the cases you find.
- Canadian Cancer Society. (2016). Cancer statistics at a glance. Retrieved from http://www.cancer.ca/en/cancer-information/cancer-101/cancer-statistics-at-a-glance/?region=on
- Ontario Ministry of Health and Long-Term Care. (2016). Cancer screening program. Retrieved from http://www.health.gov.on.ca /en/public/programs/cancer/screening/guidelines.aspx
- Webb, P., & Bain, C. (2011). Essential epidemiology: An introduction for students and health professionals. Glasgow: Cambridge University Press.
- Whitlock EP, Lin J, Liles E, et al. (2008)/ Screening for Colorectal Cancer: An Updated Systematic Review. Rockville (MD): Agency for Healthcare Research and Quality (US);Evidence Syntheses, No. 65.1.) Available from: http://www.ncbi.nlm.nih.gov/books/ NBK35179/
- US Preventive Services Task Force. (2008). Final recommendation statement: colorectal cancer screening; Retrieved from http://www.uspreventiveservicestaskforce.org/ Page/Document/RecommendationStatementFinal/colorectal-cancer-screening
- Ontario Ministry of Health and Long Term Care. (2016). Colon cancer check. Retrieved from http://www.health.gov.on.ca/ en/pro/programs/coloncancercheck/colonoscopy_faq.aspx
- Canadian Task Force on Preventive Care. (2016). Colorectal cancer. Retrieved from http://canadiantaskforce.ca/ ctfphcguidelines/2015-colorectal-cancer/
Latest posts by Nicole Haywood (see all)
- Applied epidemiology: Investigation of outbreaks of foodborne illness - 15 / 12 / 2018
- Personal Health Technology: Potential, perils, and privacy - 22 / 03 / 2018
- Mortality prediction with a single question - 11 / 12 / 2017
- Choice in childbirth: VBAC - 08 / 08 / 2016
- Ins and outs of cancer screening - 10 / 03 / 2016