In Canada, injuries are the greatest single contributor to potential years of life lost before age 65 and remain a major health problem in the country. As the largest province by population, Ontario represents a significant proportion of the national injury burden, with thousands of lives lost, and hundreds of thousands of injuries sustained annually. To increase public awareness about the threat of injury, the Injury Prevention Research Office of St. Michael's Hospital in Toronto, Ontario, collaborated with GeoConnections to develop a web-based injury atlas for the province.

St. Michael's worked with The University of Toronto, Ryerson University, and a group of end users throughout the province, including health care providers, public health agencies, non-governmental organizations, and provincial ministries. The datasets used include hospitalization rates for injury, census data from Statistics Canada, and map data layers from Natural Resources Canada. Public health units can use the atlas to determine areas and populations at risk, enabling them to provide appropriate injury prevention programs and make informed decisions about policies and programming.

Some of the datasets used in the online atlas are served with Web Map Services (WMSs) coming from a number of sources. For example, a public road map layer from the CGDI  (WMS) and injury rates from a protected, high-security server at the Centre for the Study of Commercial Activity at Ryerson University (WMS) are overlaid with household income data from a protected Census map server at the Cartography Office, University of Toronto. Users can view the distributions of injuries by type (e.g., falls, motor vehicle, assault, self-inflicted) and by age ranges, according to census subdivisions or local health integration network (LHIN) areas. The application provides the ability to display census or injury data on either of two side-by-side maps. A top-level selection tool was incorporated based on end-users consultations during the development of the online atlas, and provides a useful means for displaying more data (such as emergency room visits) in future versions of the web-atlas. 

The Online Injury Atlas for Ontario has made significant contributions to the Canadian Geospatial Data Infrastructure by combining a number of existing WMSs with health, injury, and demographic data published as new WMSs for this project. As expected, injury data (and health data in general) cannot easily be published and integrated in the CGDI due to their confidentiality. In order to protect the confidential datasets, access to the entire web-atlas is limited to authorized users with a user name/password log-in process. The user name and password were frequently altered to maximize data privacy and security. The online atlas demonstrates the combination of public and protected datasets within one web mapping application, and the need to provide levels of access within spatial data infrastructure that includes password-protected sections. Thus, while this project can only contribute new services to the CGDI if the data providers agree to share their data, the atlas demonstrates the use of certain public WMS in combination with protected WMS to support decision-making in public health.

How Framework Data was Accessed and Used

Several layers of geographic data were needed to meet the functional requirements of the online atlas, including framework data from the Atlas of Canada and other base map layers. The layers being used and their sources are identified in the table below.

A number of challenges were encountered and addressed during the development of the online atlas, including:

• Reaching agreement with the owner of the injury data took longer than anticipated, and this limitation needs to be taken into consideration when developing timelines for projects involving data privacy concerns.
• The open source web mapping software used for the online atlas, CartoWeb, proved successful in achieving the goals of the project, but the resources required for customization were significantly higher than originally planned and there were many gaps in the documentation provided.
• It was more difficult than expected to find useful general base data layers that would serve up efficiently and could be relied upon. For example, one of the planned WMS services was not a dependable data source.  It may be necessary to test services for reliability in particular applications (a guide for service testing can be found here).