The impact of work injury and permanent impairment on the probability of poverty

Abstract

Objectives: We investigate the prevalence of poverty across different workers compensation programs using large representative samples of workers’ compensation claimants who have sustained a permanent impairment from a work injury. The programs, which have existed in the provinces of Ontario and British Columbia, Canada over the last 25 years, are the Permanent Disability (PD) program, the Future Economic Loss (FEL) program, the Loss of Earnings (LOE) program, and the Bifurcated Benefits (BB) program. The nature of benefit determination and the return to work supports provided by the four programs are very different. The focus of the study is on evidence of programmatic impact on the probability of poverty in the nine years post injury.

Methods: The study included claimants sampled from each of the four programs who sustained a permanent impairment from a work injury. Claimants were identified in a Revenue Canada tax database know as the Longitudinal Administrative Databank (LAD), which is a longitudinal 20% simple random sample of all Canadian tax filers. Each claimant was matched with similar uninjured controls that were also in the LAD, based on sex, age, labour-market earnings amounts and trajectories in the four years prior to injury, family income, marital status, number of children, and a propensity score. Descriptive analysis was undertaken to compare near poverty, poverty and deep poverty levels of claimants relative to their match controls using data on family and individual earnings over a ten-year period post injury. Statistical modeling was used to determine the probability of poverty and near poverty for claimants versus controls. A key issue of interest was to determine was whether the probability of poverty differed between programs.

Results: Based on after-tax adjusted family income, the level of poverty was quite low, less than 2% in every program over a ten-year period. The level of poverty was also lower for claimants than their matched controls, but only nominally so. The BB program had the lowest proportion of poverty followed by the PD program, the FEL program and then the LOE program. In the statistical modelling analysis male claimants did not have a higher probability of poverty compared to controls, though female claimants did. Both male and female claimants had a higher probability of near poverty.

Conclusions: Poverty levels are very low for workers’ compensation claimants who sustain permanent impairments from a work injury across different programs and time periods in Ontario and British Columbia. Overall the Bifurcated Benefits program from British Columbia had the lowest proportion of claimants in poverty in absolute terms and relative to non-injured workers. Increased levels of poverty due to work injury and permanent impairment are particularly a concern for female claimants, though both female and male claimants have a higher chance of near poverty compared to non-injured workers.