|
The projects
included in the Cancer Research Survey are
classified in many different ways. Classification
has many merits as it improves the ease with which
data can be grouped into meaningful categories, and
it reduces the need for ad hoc, text-based searches
to locate relevant information. For this survey,
classification is based on summary descriptions
of the research provided by the funding organizations. It is not based on detailed
project information, and does not involve
consultations with investigators.
Within the CCRA
Cancer Research Survey, a number of classification
tools are used:
The CSO is a
classification system specific to cancer research,
which is organized around seven broad areas of
scientific interest (see
Common Scientific Outline).
It has become an international standard, being the
principal classification framework used by the
International Cancer Research (ICR) Partners. Use
of this classification system facilitates comparison
with other funding organizations within the
partnership.
Each project within
the Cancer Research Survey is assigned CSO codes (a
maximum of four) which represent the primary focus
of the research as articulated in the project
abstract or summary. In addition, where more than
one CSO code is assigned, the project budget is
distributed equally among the codes.
CSO codes are
assigned by two different coders who work
independently. Projects where there are
non-reconcilable disagreements between coders are
referred to the Expert Advisory Working Group, whose
members will make a final decision about coding.
(For more information, see
Expert Advisory Working Group.)
For the purposes of
the CCRA Cancer Research Survey, the cancer site(s)
relevant to each project are coded using the
International Statistical Classification of Diseases
and Related Health Problems, 10th Revision, Version
for 2007 (ICD-10). The ICD is the international
standard diagnostic classification used for general
epidemiological and health management purposes. The
ICD codes permit a great deal of specificity, and
can be easily mapped to other disease-based
classification systems.
The number of ICD
codes that can be assigned to a project has no upper
limit. For projects with more than one site code,
the project budget is allocated accordingly to each
code so that it sums up to 100% of the total.
This classification
focuses on how the research was conducted. It was
developed specifically for the Cancer Research
Survey, and consists of 5 broad categories
(epidemiological, basic/fundamental,
clinical/developmental, evaluative/applied,
methodological), and 24 subcategories. A maximum of
two codes is provided per project, with the aim
being to classify the major research method(s) used
in the project.
Research projects
with broad ranging relevance to many disease
conditions and not exclusively focused on cancer are
important to include within the CCRA Cancer Research
Survey as many contribute significantly to our
understanding of the causes, prevention and
treatment of cancer.
For this reason,
each project in the Cancer Research Survey is
classified as having either primary or secondary
cancer relevance. The former includes all projects
that were funded by an agency with a cancer mandate
(e.g. National Cancer Institute of Canada) as well
as projects supported by general research funders
(e.g. Canadian Institute for Health Information)
where the project description clearly identifies
cancer as the primary research focus. Projects that
are cancer-related but that do not have the study of
cancer as a primary focus are designated as having
secondary cancer relevance, and the budgets for
these projects are weighted at less than 100%.
|
