Spatial variation in deceased organ donation registration in Ontario, Canada

Abstract

Introduction The demand for organs for transplantation exceeds supply in almost every country in the world, including Canada. As only a small fraction of deaths are eligible for organ donation due to type of death or the presence of a disease or health condition, there is a need to increase a number of people who may eventually become deceased donors. One of the most promising mechanisms to achieve this goal is to increase organ donor registration rates. In order to improve the registration rate, more information is needed on the nature of regional variation in donor registration rates across Ontario. The objective of this study is to assess the magnitude of regional variation in rates of deceased organ donation registration across Ontario. Given that the rates of organ donation registration may be correlated in different areas, using spatial analysis allows to determine which areas in Ontario have higher (i.e., hot spots) or lower rates (i.e., cold spots) of organ donation registry. Methods Donor registration information and geographic information for each year from 2006 to 2013 was obtained through two administrative databases from the Institute for Clinical Evaluative Sciences: the Organ Donor Registry (ODR) and the Ontario Registered Persons Database (RPDB). Forward sortation areas (FSA), geographic units based on the first three characters of the postal code, were used to identify geographic units. Organ donor registration rates were computed for each FSA. In total, there were 519 FSAs included. To examine the overall spatial clustering of organ donor registration rates, the global Moran's I statistic was computed to measure spatial autocorrelation. The local indicator of spatial autocorrelation (LISA), was computed to assess the specific contribution of each FSA within Ontario. Results Rates of organ donation registry varied across FSAs from 6.27% to 56.59%, with the overall average rate of 27.94%. Assessing for global spatial autocorrelation using Moran's I indicated significantly positive autocorrelation among the population (I = 0.811; SD = 27.786; P-value ≤ 0.001). LISA maps were plotted after significance testing and only clustering of FSAs with high rates of donor registration (i.e., positive spatial autocorrelation) were found. Conclusions This study illustrates the marked variability of donor registration in Ontario at the FSA-level. Additionally, certain areas were identified to be ‘hot spots’ for organ donor registration, particularly in northern Ontario. Further research to investigate factors related to why individuals choose or do not choose to register in certain areas of Ontario would provide more context to the current findings.