Abstract Details

Title: Gender differences in work disability duration across three workers' compensation systems in Canada
Authors: Robert Macpherson; Mieke Koehoorn; William Quirke; Jonathan Fan; Benjamin Amick; Cameron Mustard; Sheila Hogg-Johnson; Allen Kraut; Christopher McLeod
Session: Poster Session - Colony Ballroom, 16 Oct 2016, 5:30 PM - 7:00 PM

Background & Objectives

Previous research has suggested that women experience longer work disability duration than men after adjusting for confounding factors. This study investigated whether differences in disability duration by gender were consistent across provinces in Canada.

Methods

Comparable cohorts of injured workers in British Columbia (BC), Manitoba (MB) and Ontario (ON) were created using individual-level claims data for injuries occurring between 2007 and 2011. Disability duration was measured using cumulative wage-loss benefit days during the first year (260 work days) following work-related injury. Cox proportional hazard models were used to estimate the interaction effects of gender and jurisdiction on the cessation of wage-loss benefits, while adjusting for age, occupation, injury type (back strain, non-back strain, fracture, concussion, and other), and injury year.  A model was constructed for all injuries and stratified models for strain and non-strain injuries.

Results

The number of claims for men and women in each jurisdiction was 168,130 men and 87,114 women in BC, 46,659 men and 22,770 women in MB, and 177,376 men and 114,884 women in ON. Using BC men as the reference category, the adjusted model for all injuries showed that BC women were less likely to exit wage-loss benefits within one year (Hazard Ratio [HR]: 0.87, 95% Confidence Interval [CI]: 0.86-0.88). This pattern remained the same across all the models stratified by injury type. However, both men and women in ON were more likely to exit wage-loss benefits compared to BC men (HR: 1.13, 95% CI: 1.12-1.14 and HR: 1.16, 95% CI: 1.15-1.18, respectively).  The greatest gender differences by jurisdiction were found in Manitoba with men more likely to exit wage-loss benefits (HR: 1.19, 95% CI: 1.18-1.20) compared to women (HR: 0.96, 95% CI: 0.95-0.98). This effect was also consistent across the models stratified by injury type.

Conclusions

Persistent differences in disability duration by gender are not fully explained by the socio-demographic or socio-economic variables available in our data. Building on recent research, there is also evidence of effect modification of gender by jurisdiction. This finding highlights the importance of considering gender as both an individual and contextual characteristic for disability duration.