Layout 1


ISDS Annual Conference Proceedings 2012. This is an Open Access article distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and
reproduction in any medium, provided the original work is properly cited.

ISDS 2012 Conference Abstracts

Using Surveillance Data to Identify Risk Factors for
Severe H1N1 in First Nations
Kathryn Morrison*1, Yanyu Xiao2, Seyed Moghadas2 and David Buckeridge1

1Epidemiology & Biostatistics, McGill University, Montreal, QC, Canada; 2York University, Toronto, ON, Canada

Objective
We sought to measure from surveillance data the effect of prox-

imity to an urban centre (rurality) and other risk factors, (e.g., age,
residency on a FN reservation, and pandemic wave) on hospitaliza-
tion and intensive care unit admission for severe influenza.

Introduction
Research has shown that Canadian First Nation (FN) populations

were disproportionately affected by the 2009 H1N1 influenza pan-
demic. However, the mechanisms for the disproportionate outcomes
are not well understood. Possibilities such as healthcare access, in-
frastructure and housing issues, and pre-existing comorbidities have
been suggested. We estimated the odds of hospitalization and inten-
sive care unit admission for cases of H1N1 influenza among FN liv-
ing in Manitoba, Canada, to determine the effect of location of
residency and other factors on disease outcomes during the 2009
H1N1 pandemic.

Methods
We obtained surveillance data on laboratory confirmed cases of

pandemic H1N1 influenza from the province of Manitoba. These data
described demographic characteristics, residence location, and dates
of hospital and ICU admission. We measured the rurality of each case
using a pre-exiting scale (Rambeau & Todd, 2000). We tabulated the
number of hospitalizations (and ICU admissions) stratified first by
reservation residency and second by rurality and calculated unad-
justed odds ratios. We then used logistic regression to calculate the
odds of hospitalization given infection (and the odds of ICU admis-
sion given hospitalization), adjusting for age, reservation residency,
rurality, and pandemic wave. We also investigated the effect of ru-
rality and reserve residency on time to hospitalization from infection.

Results
FN individuals diagnosed with influenza and living on-reserve

were more likely to be hospitalized than those living off-reserve, even
after controlling for the effects of rurality (OR: 2.16, 95% CI: 1.15,
4.05) . FN living in rural areas were hospitalized more frequently and
experienced longer delays between infection and hospitalization than
FN residing in more urban areas. Rurality and reserve residency had
less effect on ICU admissions once an individual was hospitalized.

Conclusions
While it is established that FN individuals had disproportionately

high rates of severe outcomes from H1N1, the causal mechanisms at
work are not well understood. Reasonable possibilities include bar-
riers to healthcare access, lack of proper housing and infrastructure,
and pre-existing comorbidities. This research using surveillance data
suggests that geographic location has an effect on healthcare access,
including both on vs. off reserve residency as well as rurality.

Keywords
Influenza; First Nations; Severe outcomes

References

Rambeau, Sheila, & Kathleen Todd. January 2000. Census Metropolitan
Area and Census Agglomeration Influenced Zone (MIZ) with Cen-
sus Data. Ottawa: Statistics Canada, Geography Division.

*Kathryn Morrison
E-mail: kt.morrison@mail.mcgill.ca

Online Journal of Public Health Informatics * ISSN 1947-2579 * http://ojphi.org * 5(1):e42, 2013