Paper: Evaluation of the Ontario High Risk Firm Initiative

Author(s) and Affiliation(s):
Sheilah Hogg-Johnson, Institute for Work & Health
Lynda Robson, Institute for Work & Health
Roman Dolinschi, Institute for Work & Health
Donald Cole, University of Toronto
Emile Tompa, Institute for Work & Health
Peter Smith, Institute for Work & Health
Philip Bigelow, University of Waterloo
Dwayne van Eerd, Institute for Work & Health
Cameron Mustard, Institute for Work & Health
Day/Time: Friday at 13:30
Room: Ballroom, 2nd Floor

The Ontario Ministry of Labour (MoL) ran the High Risk Firm Initiative (HRFI) from 2004 to 2008 – a targeted inspection and consultation program aimed at reducing work-related injuries. Targeting was based on recent work injury statistics. In 2006, two health & safety associations randomized their lists of firms into three study arms. Our objectives were to evaluate the 2006 HRFI program year in the manufacturing and service sectors by examining trends in work injury statistics.


The study population includes all firms in manufacturing or service sectors registered with the Ontario Workplace Safety and Insurance Board (WSIB) and in business in 2005. The study sample consists of those firms ranked by the HRFI algorithm to be in the worst performing 2% (high risk firms) and the next worst performing 8% (last-chance firms) for HRFI program year 2006. The last-chance firms were randomized to one of the three study groups (HSA consultation, priority inspection, referent). Information was extracted from the Ontario WSIB on characteristics of the firms (size, age, region, sector, closed for business) and their claims experience (counts of claims and disability days) from 2002 to 2008 inclusive. Trends by study arm and year were examined graphically. Negative binomial generalized estimating equation models were used to model claim counts and disability day counts by study arm and year, while controlling for firm attributes.


There were 6814 firms in the study sample: (3616 manufacturing and 3198 service). Among the randomized study arms within each sector, firms were very similar with respect to age, size, region and prior counts of claims and disability days, but the 2% high risk firms showed differences in these attributes. When trends were examined from 2002 to 2008, there were significant differences in claims and disability days counts over time, but these trends were similar across the three randomized study arms. The 2% high risk firms showed higher claims and disability rates than last-chance firms prior to intervention, but very similar rates after intervention.


The targeted consultation and reduced inspection schedule had little impact on the immediate post-intervention years relative to the referent group of firms. It is more difficult to draw conclusions about intensive inspection given the lack of comparability of this group of firms with the last-chance firms. Changes in the high risk firms could be due to the intensive inspection or due to other sources of variation over time, such as regression to the mean.