According to WorkCover’s latest public relations handouts to medical practitioners, “work generally is shown to be good for health and wellbeing”, and “Research shows that early return to work is important to recovery”.
Let’s just pass over the amusing irony inherent in the claim that the very activity being performed by a patient at the time of injury, i.e. “work”, was in fact beneficial for that patient. Instead, let’s look at the evidence that WorkCover assures GP’s “overwhelmingly” supports these surprising statements. I’ll deal with the early return to work claim first since it is the most ludicrous.
Early Return to Work
Here’s the claim from WorkCover WA’s GP Support page http://gpsupport.workcover.wa.gov.au/healthbenefitsofwork.html :
Health benefits of work
Good work is beneficial to health and therapeutic to recovery. Compelling evidence shows that supporting injured workers to return to suitable work early improves the likelihood of optimal recovery.
The converse has also been found – where patients remain off work long-term, this can have significant side effects, including poorer physical health, mental health and psychological wellbeing.
When workers remain away from work, there is increased likelihood of not returning to work at all. If a worker is absent from work for:
20 days the chance of ever getting back to work is 70%
45 days the chance of ever getting back to work is 50%
70 days the chance of ever getting back to work is 35%
Therefore, assisting injured workers to return to suitable work earlier will likely increase their chances of optimal recovery, promoting physical and mental health and wellbeing.
The source for quoted statistics is a 2002 paper by Johnson & Fry “Factors affecting return to work after injury: a study for the Victorian WorkCover Authority” https://minerva-access.unimelb.edu.au/handle/11343/33726.
So, the claim being made is that time off work causes time off work, because while they are away from work, workers are denied the “health benefits of work”. Denial of these benefits prejudices recovery, so workers take longer to get better. That this is so, is indicated by the table which shows a negative association between time off work and the chance of returning to work: greater absence – lower probability of return to work.
However, the claim is fallacious as (for a start) “correlation or association is not causation”. There are numerous examples available to illustrate this fallacy. An amusing one is as follows:
A correlation is when two variables vary together, whereas. For instance, ice cream sales may increase in the summer and decrease in the winter. The same may be true for drowning accidents. Does this mean we can draw the conclusion that drowning accidents causing ice cream sales? Does this mean that people have become so selfish and morally vile that they prefer to buy ice cream and watching people drown than trying to save them!? Fortunately, not really. Just because two variables vary together does not mean that one caused the other. It might be that the other caused the first, that they both cause each other or that a third factor causes both. In the case of ice cream sales and drowning accidents, a third factor that probably explain the correlation is season. In the summer, more people eat ice cream and go bathing, but fewer to these things in the winter. http://debunkingdenialism.com/2011/11/17/the-top-five-most-annoying-statistical-fallacies/
Increasing absence from work may be associated with a lower probability of returning to work, but this does not show a causative relationship. In this case the “third factor” that explains the correlation is most likely a worker’s of severity of injury. That is, the length of absence from work is explained by the severity of a workers injury. Pretty obvious really.
Another example of the fallacy could cobbled from be the data presented at Figure 5.2 of Johnson & Fry, where an association is shown between hospitalisation and length of absence from work. The table shows a “base” worker could expect to face a probability of 50% of remaining on benefits after about 50 days following injury. However, if a worker undergoes hospitalization, the time taken for the probability of returning to work to fall to 50% is about 100 days, twice as long. Could we therefore conclude that hospitalization “causes” injured workers to remain off work twice as long? Perhaps it’s not hospitalization itself, but malign exposure to medical staff in hospitals that causes work absence! Perhaps a public education campaign is needed to alert GPs to this compelling evidence of the dangers to workers posed by hospitals.
I should say that my comments are in no way critical of Professors Johnson or Fry, or of their paper. They certainly do not claim that their paper provided compelling evidence of the health benefits of work. In fact, they outline, quite properly, in their concluding remarks, the limitations of their study and the need for further investigations.