Data + statistics
The survey redesign gained employers and lost the question about harm
The National Return to Work Survey was rebuilt for its 2025 round, and the change did two things at once. It added employers for the first time, which produced the single most striking number in the strategy’s mid-term review. It also removed the question that asked injured workers whether returning to work had hindered their recovery. When that question was last asked, in 2021, 22 per cent said it had. One headline indicator was discontinued outright and four more now carry comparability warnings. This is the review that decides how the second half of a ten-year strategy is run.
What the redesign added, and it is not small
Start with the case for it, because it is a good one. Safe Work Australia describes a survey
program reviewed in 2022, with resulting enhancements implemented in 2025
, and says the
new elements include streamlined online delivery and, for the first time, an employer
cohort. Historically, only injured workers had been covered in the survey.
That addition is why the Year 6 scorecard contains the finding we reported on 30 August 2026: employers say they supported the injured worker 94.1 per cent of the time and workers say 49.8 per cent. Nobody could have published that gap before 2025, because only one side had ever been asked. It is the most valuable thing in the report and the redesign is the reason it exists.
What it cost
The same rebuild removed or degraded six measures, and the scorecard is candid about each one in its own tables and footnotes.
| Measure | What happened | Last figures |
|---|---|---|
| Work Role Functioning (indicator d) | Discontinued. Reported as N/A for Year 5 and Year 6 | 55.3% baseline, 52.9% at Year 3 |
| Whether returning to work hindered recovery (supplementary) | Question removed from the survey in the redesign | 2021: 22% hindered, 36% helped, 42% no effect |
| Employer contact (indicator l) | In 2025 asked only of workers who had returned to work at some point | 61.9% baseline to 68.8% at Year 6 |
| Perceived fairness (indicator g) | Differences in the questions making up the aggregate | 74.6% baseline to 60.1% at Year 6 |
| Return to work self-efficacy (indicator o) | Differences in the questions making up the aggregate | 74.9% baseline to 63.1% at Year 6 |
| Workplace accommodations (indicator n) | Wording change the report says may explain a large result shift | Different duties 53.1% baseline to 29.8% at Year 6 |
Compiled by us from the Year 6 scorecard’s national performance measures table, its explanatory notes and its supplementary indicators table. The grouping is ours; every entry and figure is the report’s.
The four comparability cautions are Safe Work Australia’s own. Its note says changes to
the asterisked indicators should be interpreted with a higher degree of caution due to
changes in the NRTWS questionnaire between 2021 and 2025
, and it names them: indicator l was
only asked of workers who had returned to work at some point
, indicators g and o
have some differences in the individual questions that contribute to the aggregate
measure
, and for indicator n A small change in wording of the survey question informing
indicator 3n may have contributed to the large change in results in 2025 when compared with
previous surveys.
The one that is different in kind
Five of those six are measurement problems, and the report handles them properly by flagging them rather than presenting a clean series. The sixth is not a measurement problem. It is a question that no longer gets asked.
The supplementary indicator existed to test something the strategy asserts about itself. Its
stated relevance is that it Indicates whether injured workers are staying in ‘good’
work
, and that good work should not hinder a person’s recovery
. In the 2021
survey, 22 per cent of injured workers said returning to work had hindered their recovery,
36 per cent said it had helped, and 42 per cent said it had no effect. The 2018 results were
close: 21 per cent hindered, 34 per cent helped. The scorecard then records the outcome in one
line: Data from the 2025 is not available for this question, as it was removed from the
survey during the redesign.
So for two consecutive rounds, about one in five injured workers said the thing the strategy exists to encourage had made their recovery worse. There is now no way to know whether that number moved.
What we are not saying
We are not saying the questions were dropped because the answers were unwelcome. The report gives no reason for any of the removals, and we have not found one anywhere else, so any account of motive would be invention. Survey redesigns drop questions for length, for cost, for mode changes and because a program was told to add an employer cohort without doubling its respondent burden, and every one of those is an ordinary reason.
What we are saying is narrower and does not depend on motive. The mid-term review of the National Return to Work Strategy 2020-2030 reports that worker experiences have generally deteriorated. At that same review, the measure asking whether the return-to-work process itself harms people no longer exists, and four of the measures that do exist come with a warning against comparing them to their own baselines. A strategy with four years left to run has less instrumentation than it had at the halfway mark, and the piece it lost is the one aimed squarely at harm.
That is a fixable thing rather than a scandal. The question was asked in 2018 and 2021 and could be asked again in the next round.
How we did this
Everything here comes from the NRTW Strategy Year 6 Key Achievements and Scorecard report, published by Safe Work Australia on 28 April 2026, read in full. The six affected measures are drawn from three separate parts of it: the national performance measures table, the explanatory notes carrying the asterisk cautions, and the supplementary indicators table. Bringing them into one list is our editorial work; every entry, figure and quotation is the report’s own.
We describe indicator d as discontinued because the table records
N/A (discontinued)
against it for both Year 5 and Year 6. We have not gone behind the
scorecard to the survey instruments themselves, so we cannot say what a removed question was
replaced with, if anything, and we have not implied that it was replaced with nothing.
The section headed What we are not saying is there because the obvious reading of these facts is an accusation, and the facts do not support one. We have stated the limit of what the evidence carries rather than leaving a reader to assume the rest.
Sources
- Safe Work Australia, NRTW Strategy Year 6 Key Achievements and Scorecard report (PDF, 30 pages), published 28 April 2026, read in full: the redesign highlight and its description of the 2022 review and 2025 enhancements, the employer cohort, the national performance measures table including indicator d, the explanatory notes naming indicators g, l, n and o, and the supplementary indicator on whether returning to work hindered recovery together with its 2018 and 2021 figures and its removal.
- Safe Work Australia, NRTW Strategy Year 6 Key Achievements and Scorecard report (publication page): the publication date of 28 April 2026.
Spotted an error? Request a correction and we will check it against the source and log the outcome here.