Paper: Something might be missing from OHS audits: findings from a content validity analysis of five audit instruments

Author(s) and Affiliation(s):
Lynda Robson, Institute for Work & Health
Sara Macdonald, Institute for Work & Health
Dwayne Van Eerd, Institute for Work & Health, University of Waterloo
Garry Gray, Harvard University
Philip Bigelow, University of Waterloo
Download Presentation PDF:
Day/Time: Saturday at 14:00
Room: St. Patrick Room, 3rd Floor
Objectives:

The specific objective of the project was to examine the content validity of OHS management audit methods. This was part of a broader objective to contribute to the gap in the research literature regarding the measurement properties of OHS management audit instruments.

Methods:

Evaluating content validity requires a definitional standard against which concepts or items can be compared. The standard for this study was the OHS management standard developed by the Canadian Standards Association, CSA Z1000. There are five major elements in this standard (Commitment, Leadership, and Participation; Planning; Implementation; Evaluation and Corrective Action; Management Review and Continual Improvement) and these were subdivided into 163 content units for the purpose of analysis. Using the documentation for an audit instrument, two independent raters determined, for each unit, whether the instrument, when used in the field, assesses the concept described by the unit (fully, partially, or not at all). Results for the elements of CSA Z1000 were derived by aggregating the results for the constituent content units. This analysis was carried out for five of the more comprehensive OHS management audit instruments used by public and not-for-profit OHS organizations when serving workplaces in Ontario.

Results:

A relatively high proportion of CSA Z1000’s content (74%) was partially or fully represented on average in the audit methods. However, six management elements were found to be incompletely represented in three or more of the methods: general [OHSMS] (i.e. integration with other management systems), objectives and targets, documentation, internal audits, management review input, and management output. The most extreme example is the internal audits element whose content was completely missing for three of the audit methods.

Conclusions:

Some OHS management audit instruments in current use are incomplete relative to a recent OHS management standard. It may be that some instruments warrant revision in order to better reflect current expert consensus.