Access to Allogeneic Transplant in Canada:

A Canadian Blood and Marrow Transplant Group/Canadian National Transplant Research Program Study

Introduction: Allogeneic hematopoietic stem cell transplant (aHSCT) is a potentially curative treatment for patients with blood cancers and disorders of blood/immune system, but due to the complexity of the procedure, is only offered in a limited number of medical centres. Many barriers might exist that prevent patients from receiving an aHSCT, including physical and social geographic barriers. We sought to understand how access patterns to aHSCT varied across Canada, a country with a government-funded universal health care system.

Methods: The Canadian Institute for Health Information (CIHI) Discharge Abstract Database (DAD) is a national record of all hospital admissions in all Canadian provinces other than Quebec. We identified all Canadians under the age of 65 admitted to hospital in provinces other than Quebec with a diagnosis of acute myeloid leukemia (AML) between 2004 and 2015, using the CIHI DAD and ICD-9 diagnosis codes for AML. We determined which of these patients subsequently were admitted to hospital for an aHSCT, using ICD-9 procedure and diagnosis codes for AML. Residence at the time of admission with AML was identified using their forward sorting area (FSA) component of the postal code. Socio-demographic attributes of the FSAs were identified using data from the 2006 Canadian Census, including whether the FSA was rural or urban, the proportion of the population that was a visible minority, aboriginal, or low income after tax . Logistic regression was used to investigate potential associations between these factors and the odds of receiving a transplant. Two sensitivity analyses were conducted (age less than 18 at time of diagnosis, and age between 18 and 65 at the time of diagnosis).

Results: 6119 non-Quebec Canadians were admitted to hospital with a diagnosis of AML between 2004 and 2015. Of these, 1745 (28.5%) received aHSCT. Several variables were significantly associated with receiving a transplant in univariable analyses (Table 1; time period, province of residence, gender, age, and proportion of low income families), but after accounting for other variables in the model, only time period, province of residence, gender, and age remained significantly associated. In the pediatric subgroup, similar results were seen, except province of residence and gender were not significant. The results of the adult sensitivity analysis were identical to the main cohort, with province of residence, time period, gender, and age predictive of receiving a aHSCT.

Discussion: In contrast to previous studies done in the United States with similar methodology, in non-Quebec Canada, low income level was not associated with inferior access to aHSCT. This might suggest that Canada's universal health care insurance program is protective against socioeconomic barriers. In addition, in contrast to previous studies, rural location was not associated with the odds of receiving a transplant, and reassuringly, the proportion identifying as aboriginal was not significantly associated with the odds of receiving a transplant.

Dramatic differences were seen in aHSCT rates by province, with residents of Alberta (third largest province in this cohort) diagnosed with AML more than twice as likely to receive an aHSCT compared to residents of Ontario, the most populous province in Canada (OR 0.45, p < 0.01). The reasons for this are unclear, but likely include practice patterns at the leukemia/transplant sites, center resources, and provincial health care budgets. Notably, the province with the highest per-capita GDP (Alberta) had the highest proportion of individuals receiving a aHSCT. Thus, is it possible that in Canada, regional wealth is more predictive of access to health care than individual wealth. The reasons for these dramatic regional differences in access to transplant in Canada should be studied further.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution