The range of values within which, 95% of the time, the true value would
fall. The wider the range, the fewer the number of cancer cases and the more
the numbers fluctuate. For example, if the 95% confidence interval is 1.0-15.0,
the uncertainty is larger than if the confidence interval is 1.5-1.7.
A procedure where weighted averages of age-specific
rates are used to modify rates to a standard
population in order to minimize the effects of differences in the age
composition of given populations (such as provinces or census divisions)
when comparing rates for these populations. Since cancer is more common in
older age groups, a population that is older will have a higher crude incidence
rate. The purpose of this rate is to compare groups of people from different
backgrounds and age structures, for example when comparing breast cancer
between countries, a world population is used, so that the difference in
incidence rates is not due to one country having older citizens. The age-standardized
rates for both sexes combined also adjust for possible differences in the
gender distribution.
The ratio between the number of deaths and the number of new cases of
a particular cancer. This provides a crude measure of potential survival.
If the case fatality ratio is 50%, then one would expect half of those diagnosed
with that disease to eventually pass away.
A general term applying to counties, regional districts, regional municipalities,
etc. In Newfoundland, Manitoba, Saskatchewan, and Alberta, the term describes
geographical areas that have been created by Statistics Canada in co-operation
with the provinces.
A ratio of the age-standardized incidence
rate for a disease in a specific area compared with the incidence rate for
all of Canada. Those areas with a CIF less than one have an incidence rate
that is less than the Canadian average. If the CIF is above one, then the
area has a higher rate of disease than the rest of Canada.
A ratio of the age-standardized mortality rate
for a disease in a specific area compared with the mortality rate for all
of Canada. Those areas with a CMF less than one have a mortality rate that
is less than the Canadian average. If the CMF is above one, then the area
has a higher rate of disease than the rest of Canada.
Confounding bias is caused by the presence of an extraneous factor associated with both an exposure under study and a disease outcome. A commonly used method to adjust for a potential confounder is stratification in which the comparison between exposure and disease is done at specific levels of the potential confounder. When a study mentions that they controlled for a factor, they have tried to remove the effect of that variable.
While confounding is a bias which an investigator wishes to eliminate, effect modification refers to a difference in the magnitude of an effect measure across levels of another variable. An example is throat cancer (disease) and high alcohol use (risk factor). An effect modifier is smoking, since the relative risk for alcohol has been reported to be greater at higher levels of smoking. More accurately this should be called risk-ratio modification since effect modification also depends on the risk scale for the outcome, for example risk ratio or risk difference.
The number of new cases or deaths due to a disease over the total population
that could be affected, without considering age or other factors. It is usually
expressed as a rate per 100,000 persons per year.
A geographic term for significantly inhabited regions. Populated areas
are shaded in their appropriate colour providing they have a minimum population
density of about 0.4 person per square kilometres (approximately 1 person
per square mile).
See Variation. When looking at cancer
incidence by Census Division, one will
notice that some rates in areas are higher than in others. The inherent difficulty
is in finding out if there is a reason for the higher or lower rates other
than chance. Provincial registration practices can also play a role.
The number of people who leave a hospital either through a completed procedure,
discharge or death. It is often used to examine the trends in morbidity from
a disease. In this context, it does not include any out-patient procedures.
International Classification of Diseases, 9^{th} edition. It is no longer the standard method of coding of disease. It was used for coding cancer incidence to 2000 and for coding mortality to 1999.
International Classification of Diseases for Oncology, 3^{rd} edition. It is the current standard method of coding cancer incidence, and is used on this website.
The conclusions that one is able to draw from the data. Sometimes the
numbers do not tell the whole story. Please see the section on How
to use Canadian Cancer Surveillance On-Line. For instance it may seem
that one area on a map has a particularly high cancer rate, when in fact
it could be just be chance that the cases occurred that year. By checking
the rates before and after, one would notice that the rates are more like
the average over time. The statistical significance of the rate should also
be considered.
A misleading factor that may lead one to think that screening causes increased
survival when it does not. If an average person only survives 5 years with
a particular cancer, but if diagnosed 5 years earlier, they may show 10 years
survival when in fact they are not surviving longer with the cancer, they
are just aware of it longer.
A mathematical scale used for examining the rate of change. The units
are based on the power of ten, i.e. 0.1, 1, 10, 100, 1,000, etc. It is useful
when the rates under study vary considerably.
The proportion of a group affected over a period of time such as a year.
It is usually expressed as new cases (or deaths, separations, etc.) per 100,000
people per year.
The chance that a person may eventually develop a particular disease.
When comparing two groups this can be expressed as a relative risk of disease
or the odds of a disease.
A risk factor is a factor associated with an increase in the chances of
getting a disease; it may be a cause or simply a risk marker. Factors associated
with decreased risk are known as protective.
The testing of an apparently healthy group of people to separate those
who are likely to have a disease from those who probably don't; e.g. with
a Pap smear or mammogram. Screening must be followed up with more complicated
diagnostic testing to confirm that the disease is truly present.
A population distribution that is used to create rates that have the same
age structure, so that different rates can be properly compared. See age-standardized
rates.
A method that tests whether the result given is so rare that it is unlikely
to be due to chance alone. Examples include a p-value (for probability) or
a T-test. The most common cut-off is 5%, that is if this result would occur
by chance only one in twenty times, it would be considered to be significant.
Cancer surveillance includes the collection of data, and the review, analysis
and dissemination of findings on incidence (new cases), prevalence, morbidity,
survival and mortality. Surveillance also serves to collect information on
the knowledge, attitudes and behaviours of the public with respect to practices
that prevent cancer, facilitate screening, extend survival and improve quality
of life.
A mathematical test that will help determine how likely an increase or
decrease in incidence or mortality rates is greater than expected due to
random chance.
For given areas, how different the rates are from each other, or from
the national value. Whenever you examine a large group of numbers, such as
cancer rates across Canada, there will always be some variability in the
numbers due to chance, some will be higher and some will be lower.
Please use your browser's [BACK] button to come back to the previous page