Directorates
·	Director General's Office
·	Data Development
·	Strategic Evaluations
·	Partnership and Continuous Evaluations
Publications
·	Evaluation Reports
·	Evaluation Briefs
·	Labour Market Development Agreement Series
·	Lessons Learned Series
·	Evaluation Tool Kit Series
·	Evaluation Forum Newsletter
·	From Theory To Practice

5.1 Limits of Interpretation

A number of issues limit the interpretation of results for this study. This section briefly describes these issues and how they limit the interpretation of results.

The data used for comparison with the performance of funded departments provides information on only one year post program (1999). These data do not include detailed information on the related costs of accidents and fatalities. Thus no comparative analysis of the costs of work related accidents and fatalities can be performed at this time. This limits the interpretation of results because costs would provide a more direct measure of the success of funded departments in meeting the objectives of the program. In particular, the program's impact on case and claims management approaches in funded departments cannot be assessed by analysing the number of accidents per FTE. The analysis of the number of accidents per FTE may provide an indication that departments have implemented programs and policies that have resulted in fewer accidents and illnesses and thus lower workers' compensation costs.

According to departmental representatives from funded departments interviewed early in this study, the Canada Labour Code (2000) has had a larger impact on their department's approach to occupational health and safety than the Workers' Compensation Cost Recovery Program. Based on this, it was decided that the comparison group used for this study must also be subject to the Canada Labour Code. In addition, data on the number of Full-time Equivalents was also required for this group. As a result of this, it was not possible to find a "perfect" comparison group in the sense that employees in the comparison group were employed under the Canada Labour Code and their employers were not subject to this cost recovery. All federal jurisdiction employers (i.e. governed by the Canada Labour Code) for which employment data are publicly available are included in the cost recovery program. Thus the comparison group chosen for this study are also included in the Workers' Compensation Cost Recovery Program although they were not funded. We note that with departments now on Cost Recovery, all Federal Jurisdiction employers are on some form of cost recovery for workers' compensation costs.

Interviews with departmental representatives, results of the survey of funded and unfunded departments and program data appear to be providing conflicting results. Results conflict because the information comes from very different sources. Interviews and the survey results provide data/information on the opinions or impressions of individuals whereas program data provides an actual measure of the number of accidents, costs, and average days lost per accident. This makes it difficult to draw strong conclusions at this time. Ideally, a multiple lines of evidence approach will serve to reinforce conclusions.

Data on work-related accidents and illnesses is subject to variability. Accidents are by their nature random and so a large number of data points are required in order to identify clear patterns. When more data points will be available, a more thorough analysis will be possible. The analysis work in this evaluation gives only an indication of the current situation.

Representatives from funded departments were asked to estimate the impact of cost recovery on the time spent on case and claims management. These estimates must be interpreted with caution because in general, there are no records kept of the amount of time spent on a particular case or activity. For many departments the individual responsible for case or claims management currently was not the same person responsible prior to April 1, 1998. Further, nine departments indicated that case and claims management is decentralized to the regional offices. These nine departments were Agriculture and Agri-Food, Defence, Fisheries and Oceans, the National Research Council, RCMP, Transportation, and Veteran's Affairs. Other departments may also have decentralized case and claims management but interviewees did not explicitly cite this during interviews.

As a result of these limitations, strong conclusions cannot be drawn at this time. The Worker's Compensation Cost Recovery Program has been in place for three years (although this evaluation was able to assess only the first two years of the program). The program's objectives require changes in the approach to work-related accidents and illness that are expected to evolve over the long term. The experience of Crown Corporations such as Canada Post has shown that it may take as long as 7 to 10 years in order for consistent improvement in case and claims management to appear and for workers' compensation costs to decrease.

5.2 Development and Implementation of Occupational Health and Safety (OHS) Programs and Policies

Based on views expressed by departmental representatives during interviews and the responses to the Survey of Funded and Unfunded Departments, there are indications that funded departments have increased their efforts at returns to work (case management) and claims management. Nevertheless, there is no clear indication that these changes were the direct result of the Workers' Compensation Cost Recovery Program.

Despite the early signs of success, two departmental representatives have not been successful at implementing new policies and programs because of a lack of commitment on the part of their superiors. This may be an indication that not enough is being done to educate senior managers on the importance of occupational health and safety. Recent changes to the Canada Labour Code (2000) with regard to due diligence may encourage some change on the part of senior managers.

Program data indicate that the number of days lost per new claim decreased by 11 percent in 1999-2000. This may be the result of more proactive management of cases and claims on the part of funded departments. For example, an active return to work policy will have a direct impact on the number of days work lost for each claim. The decrease in the number of days lost per time loss claim may be early indication of success of the return to work and claims management programs implemented by funded departments since April 1998. More data points will be required in order to identify this as a trend.

Despite indications of success of funded departments at decreasing the number of days lost per claim, there are a small number of departments that have had a consistently higher number of days lost per claim since 1995-96. It has been beyond the scope of this study to examine the reasons for this. Some possible explanations suggested by interviewees include:

• Job class of employees in the department. Some job categories may be subject to specific injuries that are more severe and/or require more time for recovery and rehabilitation;
• Labour relations within departments. During interviews some departments expressed frustration at the adversarial relationships that have developed between management and unions over the years limiting the implementation of proactive return to work policies and programs;
• Approach of individual departments and/or managers to case management. For example, one manager interviewed expressed the view that employees are entitled to 130 days of Injury on Duty Leave (IODL) and it is not up to them to deny their employees this right;
• Department infrastructure. A small number of department representatives interviewed expressed the view that the quality and age of buildings contributes to work-related injuries and illnesses.

In the case of one of these departments, Canada Customs and Revenue, the number of accidents per 1,000 FTEs has consistently been lower than the average for the rest of the public sector. However, the average days lost per claim has been consistently higher than the average number of days lost per claim for all funded departments. This is an indication that the department has been successful at preventing accidents from happening but less successful at getting injured or sick employees back to work.

5.3 Adequacy of the Allocations

The analysis of allocations versus actual workers' compensation costs in funded departments indicate that allocations do not keep pace with actual costs. HRDC Program staff and representatives from funded departments agree that this is largely due to costs associated with carry over claims. According to program objectives, funded departments are expected to control their workers' compensation costs by implementing return to work, claims management and other work-related health and safety programs. As departments become more proactive at decreasing the number of carry over cases, it is expected that their workers' compensation costs will decrease.

Cost and allocation data and opinions of departmental staff indicate that a small number of departments should not have been funded. Funding to departments was based on cost data for only two years, 1995 and 1996. In the case of two departments, (Clerk of the Senate, and Solicitor General) the average costs for these years were unusually high due to a particular claim in each of these departments. Similarly, funding to other departments may have been more, or less, than the amount they would receive had allocations been based on an average across more years. As more data points become available, it is expected that allocations will be more closely aligned with actual costs of new claims.

There is some early indication of success of funded departments. The cost of new claims decreased by 2.8 percent in the first year of the cost recovery program, indicating that some departments may have been successful at decreasing the average cost per new claim.

During the first year of the program, 7 departments spent more than their allocation on workers' compensation. In the second and third year of the program, 18 departments spent more than their allocation. The overall ratio of allocation to costs in the second and third year of the program was 0.5 and 0.4 respectively. This may be an indication that funded departments are successfully decreasing their actual workers' compensation costs. Additional post-program data points will be required in order to identify this as a trend.

5.4 Impact on Accident Rates and Costs

Program data do not indicate any measurable change in the number of disabling and fatal accidents per 1,000 FTEs as a result of the Workers' Compensation Cost Recovery Program. This result is not robust because only a single year of post program data were available, making it impossible to discern any trends.

However, interviews with departmental representatives indicate that there is more awareness in funded departments of the dollar costs of work-related injuries and illnesses. Prior to the Cost Recovery Program, there was little incentive for managers to keep track of workers' compensation costs paid to workers in their departments. Cost recovery has encouraged managers to keep track of these costs and, as a result, they have become aware of the financial implications of work-related accidents and illnesses. This provides some indication that the program has met the objective of increasing awareness on the part of senior and front-line managers of the consequences (in terms of cost) of work-related accidents and injuries.