Despite the significant presence of health research in Canada, the benefits of health research and its ability to impact patient outcomes is largely debated. While some assert that health research has little impact on patient outcomes, others indicate that health research can positively impact patient outcomes through a variety of different mechanisms. To date, little empirical work has been conducted on this subject. As such, the purpose of the present study is to gain a more comprehensive understanding of the nature of the relationship between health research and patient outcomes, and the possible mechanisms linking research and outcomes in the Canadian context. We will address this question in the form of a case study focusing on stroke research, as investments and advances in stroke research have increased in Canada in recent history.
There are a number of policy issues associated with the field of stroke in Canada. Stroke is one of the leading causes of death, adult disability and hospitalization in Canada. Current estimates suggest that costs of stroke to the Canadian economy surpass $3.6 billion annually in physician services, hospital costs, lost wages and decreased productivity. Particularly concerning is the higher prevalence of stroke among the population above the age of 65, as the size of this age group is expected to increase in the next twenty years as a result of the aging population. This suggests that the aging population could result in a greater impact of stroke on Canadians in the coming years. Stroke patient outcomes also vary largely between Canadian provinces, creating the need to understand the factors that contribute to these differences.
Given that stroke research activity varies between provinces, it is possible that stroke research could be one of these factors contributing to differences in patient outcomes. As such, our analysis will aim to explore the relationship between stroke research and stroke patient outcomes to deduce whether or not stroke research can impact outcomes. To guide the analysis, our study first compiled a brief literature review of studies that have investigated the relationship between health research and patient outcomes to date. The literature review indicated the necessity in using proxy measures for patient outcomes and research activity, given the inherent difficulties in measuring these variables directly. Additionally, most studies found that research activity had a positive impact on patient outcomes, and that a greater extent of research activity correlated with a greater improvement in patient outcomes. These results created the hypotheses for our analysis.
The next part of our study involved the collection of stroke research funding and 30-day stroke in-hospital mortality rate data as proxy measures for research activity and patient outcomes. This data was collected for all Canadian provinces from the fiscal year of 2000-01 to 2014-15.
Stroke research funding data were obtained from online databases from three federal funding agencies: the Canadian Institutes of Health Research, the Heart and Stroke Foundation and Natural Sciences and Engineering Research Council of Canada. A total of 1,455 projects were funded in Canada over the time period of the study, distributed among seven provinces. Mortality rate data were collected from the Canadian Institute for Health Information online indicator library.
We then graphically represented compiled research funding and mortality rate data as trends over time for each Canadian province. Research funding was shown to have an overall increase over the study period, whereas mortality rates were shown to have an overall decline. The extent to which these changes occurred varied between provinces, creating difficulties in elucidating a specific relationship between research funding and mortality rates. Nonetheless, we proposed several mechanisms that may explain how research funding impacts mortality in this particular context, including the potential lack of an impact, the influence of the CSN as a research network, and attraction of specialized personnel that enhance stroke care.
To add clarity to the specific nature of this relationship, we completed two regression analyses for provinces with at least one funded project per fiscal year for the duration of the study period. These provinces were Alberta, British Columbia, Ontario and Saskatchewan. We first completed a linear regression analysis for each province. Alberta and Saskatchewan exhibited a weak and non-significant positive association between research funding and mortality rate, whereas Ontario and British Columbia exhibited a strong and significant positive association. A fixedeffects linear regression including these four provinces was then completed, which allowed us to quantify the relationship between the two variables while accounting for inherent differences between provinces. This regression indicated a moderate and significant positive association between research funding and mortality rate for all four provinces. These analyses allowed us to conclude that to some extent, stroke research activity positively impacts stroke patient outcomes in Canada.
Our results are an important first step in understanding the benefit to society in investing in health research. These types of analyses are important in guiding policy decisions regarding allocation of funding to research programs, as they can allow policy makers to strategically allocate research dollars towards fields of research that have a proven benefit to society. Our results also indicate research networks as a potential important tool in improving patient outcomes, creating a role for government in forming and facilitating these networks in various areas of health research. Moving forward, it is clear that the role of research should not be overlooked or undervalued in Canada anytime soon, and that additional work in this area is key to guiding policy and improving health of Canadians.