Embed & repeat
Make it business as usual. Watch your lagging indicators between surveys, review what the interventions changed, embed psychosocial risk management into how the organisation already plans and reviews its work - then re-run the full survey in 1-2 years to confirm the shift.
Psychosocial risk management is not a one-off. Work, people, and hazards change. A repeating cycle of measurement, action, and review is both a regulatory expectation and the only approach that produces sustained improvement.
ISO 45003:2021 - what it is and why it matters
ISO 45003:2021 is the first international standard for psychological health and safety at work, supplementing ISO 45001:2018 (the broader OHS management standard) with practical guidance for managing psychosocial risks.
The standard isn't prescriptive about specific actions, but it's clear on process: identify hazards, assess risk, implement controls, monitor effectiveness, continuously improve. It treats participation (workers involved, not just surveyed) and leadership commitment as non-negotiable preconditions.
ISO 45003 and Australian regulation
ISO 45003 isn't a legal requirement in Australia, but its framework aligns closely with WHS obligations across all jurisdictions. Demonstrating alignment is strong evidence of a systematic approach - relevant to regulatory defence and due diligence.
The ISO 45003 management cycle
Understand the work environment, consult workers, identify hazards using multiple sources (surveys, incident data, consultation, observation). Prioritise by likelihood and potential severity.
Apply the hierarchy of controls, starting with elimination and substitution. Document actions, assign ownership, communicate the plan.
Review whether controls are working. Combine quantitative measures (repeat surveys, claims, turnover, absenteeism) with qualitative methods (check-ins, informal consultation). Compare against baseline.
Assess what worked and what didn't - be honest with yourself and with workers when interventions fell short. Reflect on what got in the way and keep trying. Revise ineffective controls, identify new hazards that emerged, and feed findings into the next planning phase.
What to measure
No single metric tells the full story. Combine leading indicators (conditions predicting future risk) with lagging indicators (outcomes that have already occurred).
Leading indicators
- ·Hazard and near-miss reports
- ·Repeat survey scores (Prior Work constructs)
- ·Pulse survey results
- ·Manager support skill assessments
- ·Action plan completion rate
- ·Worker consultation frequency
Lagging indicators
- ·Workers' compensation claims (psych)
- ·Sick leave rates and trends
- ·Voluntary turnover rate
- ·Complaints and grievances lodged
- ·Presenteeism proxy measures
- ·Regulatory notifications or investigations
Recommended review cycle
Action plan status check. Any new hazards or incidents? Any changes to work that create new risk?
Review lagging indicators (claims, turnover, absenteeism). Brief leadership update. Adjust action plan as needed.
Run a Pulse survey - track the priority factors from your model without repeating the full survey. Use results to check whether actions are producing measurable shifts, and to update the model.
Formal risk review. Compare Pulse findings against the action plan. Assess whether controls reduced hazard levels. Document and report to senior leadership.
Full survey re-run. Compare to baseline. Rebuild the model if conditions have changed substantially. Update the risk register and report to Board.
Closing the loop with workers
The same "communicate back" discipline from Action Planning applies at every review milestone, not just once: a short, honest update on what changed, what didn't, and why.
Returning in two years
Aim for a full re-survey every one to two years, depending on how stable your environment is. Major changes (restructures, new leadership, significant growth or contraction) are triggers for an earlier review, not reasons to wait.
Between full cycles, the Pulse tool provides a lighter-weight check - a short, model-guided survey that tracks only the priority factors from your baseline, run with a random sample around the 6-12 month mark. Pulse results compare against baseline and update model parameters with what's been observed since.
Intermediate check: Prior Work Pulse
At 6-12 months, run a targeted follow-up with a random sample. The tool picks the factors worth tracking from your model, accepts partial responses, and compares against baseline. Pulse data can also update model parameters - giving the next cycle a revised picture of the workforce.
See step 7: Pulse survey.
Starting your next full cycle
Upload the new CSV to Data Analytics and rebuild the model. The model may shift between cycles - expected as the environment evolves. Ideally, hazard responses move from H/M towards L, with outcomes following - the clearest signal that interventions are working at population level.