[Previous]
[Table of Contents] [Next]
Data Limitations
The data presented in this report must be interpreted with
caution for the following reasons:
- The data represent cases of newly diagnosed infections in
individuals for whom serum specimens and corresponding
epidemiologic information are provided to Health Canada from
provincial partners participating in the CHSDRSP. The data are
based on convenience sampling and therefore do not include all
newly diagnosed cases in a given population for any given year.
While we do not anticipate any biases introduced as a result of the
convenience sampling, we need to bear in mind that the data are not
representative of all newly diagnosed cases in the
population.
- The data presented are only from those individuals who are
infected with HIV and who then seek testing. They do not represent
individuals who do not know their HIV status or choose not to seek
testing. Furthermore, they do not represent individuals from whom
there is insufficient sample for strain and drug resistance
genotyping.
- Since some mutations conferring resistance may not be stable
over time, the ability to detect resistance is highly dependent on
the time since drug pressure was withdrawn. In the case of
treatment naVve individuals with newly diagnosed infection, time of
drug withdrawal is presumably reflected by time since infection. It
is therefore possible that mutations associated with drug
resistance are no longer detectable in the older, prevalent
infections.
- Wherever possible and before submission of sample and
epidemiologic data, provinces participating in the CHSDRSP review
and assess the presence of duplicate positive test reports in order
that the data accurately reflect the number of new individuals who
test positive for HIV infection. However, because of the nature of
HIV reporting in certain jurisdictions (e.g. non-nominal
reporting), the removal of all duplicates may not have been
possible.
- The data do not include information from two provinces that
bear the burden of HIV infections – Ontario and Quebec. Work
is already under way on mechanisms to include data from these
provinces, and it is anticipated that one or both of them will be
included in the next report on strain and primary drug resistance
surveillance in Canada.
- Since this report deals solely with primary drug resistance,
analysis was conducted on the laboratory specimens collected from
treatment naïve individuals at the time of initial testing for
HIV. However, treatment history cannot always be verified. For
example, at least 5% of laboratory specimens from British Columbia
are likely to have been collected from individuals who have had
treatment.
- The report presents major mutations associated with drug
resistance and not minor mutations, which, in combination, may be
associated with drug resistance. The prevalence of primary drug
resistance is therefore likely an underestimate of the true
prevalence.
- Subtype and drug resistance analyses are conducted on 1053 base
pairs in the pol gene and reflect what is observed within
this small region of the viral genome.
- At the time of writing, all data had been received
retrospectively, so the report actually describes what has happened
historically. While these data are still useful in informing
program planning and policy formulation, their utility could be
enhanced with “real-time” data.
- Missing or unknown epidemiologic data remain problematic,
particularly with respect to information on previous HIV testing,
date of first positive HIV test, ethnicity, risk behaviour, CD4 and
viral load at diagnosis, and previous antiretroviral treatment.
This may have resulted in misclassification (in the variables
presented in this report) or inability to assess the association of
a variable of interest (e.g. viral load at diagnosis).
- Wherever possible, epidemiologic data submitted through this
enhanced HIV surveillance initiative have been linked to routinely
submitted data on HIV infections. However, because of the nature of
reporting in certain jurisdictions, this has not always been
possible. Therefore, there may be some discrepancies between the
epidemiologic data reported in the routine HIV surveillance report
when compared with the data in this report.
[Previous]
[Table of Contents] [Next]