Search inside of Supercourse and lectures in HTML and PPT format
Cancer Trends over Time
The best measure for time trends is incidence (number of new cases per 100,000 population per year), but incidence data are available directly only in settings where there is proper population-based cancer registration or cancer surveys have been done. Since this is relatively recent it means that cancer mortality data, which have been available in many countries for most of the 20th century, are used to study geographic and temporal patterns of the disease. Cancer deaths reflect both the incidence of the disease as well as the effects of treatment. Mortality drops relative to incidence as the effectiveness of treatment improves. Not all patients diagnosed with cancer die; some begin to survive their disease. Where treatment is poor deaths can reflect incidence quite well.
Fatality Ratio. A comprehensive follow-up of registered cases in order to compute survival is not possible for many registries. But a fatality ratio can be used as an indication of survival instead. It is the ration of new cases to reported deaths from the same diagnosis occurring within a specified period. It is also called "deaths in period" or the "mortality/incidence ratio". It looks at the fraction (or percent) of new cases that die of the disease over the given time period (usually a year). Incidence cases and mortality do not refer to identical cases, just to identical diagnoses, so the ratio is only an indirect description of the survival experience. But 5-year relative survival rates (as % surviving) are directly proportional to the fatality ratio as deaths/cases (as a fraction from 0.0 to 1.0) for 24 major cancer sites within the same registry. Thus fatality ratio is a reasonably close reflection of 5-year survival (fatality ratio of 0.5 corresponds with 5-year survival of 50%; 1.0 with 0%; 0.25 with 25%; 0.75 with 75%; 0.0 with 100%, etc.) within the same registry. It is not clear, however, that one can use fatality ratios to compare survivals between different registries.
This is the case for both lung and stomach cancers. Lung cancer has risen significantly throughout the 20th century by more than 10-fold in North America. In men it is peaking and may be starting to decline in younger men. But in women the rise in smoking in younger women continues to drive the rise in lung cancer that is now surpassing breast cancer as the commonest cancer in women. In contrast stomach cancer has declined in incidence by about 50% from 1950 through 1970 in many countries.
Thus temporal trends in cancer tend to show epidemics that rise to a peak and then recede over very long time periods of decades. The long time periods hide the epidemic nature of the disease.
The cancer burden in Canada is expected to increase by 50% through the first 2 decades into the new millenium, in large part reflecting the aging of the "baby boomers" who make up a large proportion of its population.
Projections of incidence, prevalence and mortality are important for planning cancer control interventions.
to return to slide 26