Return to work: Clinical judgements and evidence-based decisions

It’s not often that we find an article that draws on clinical knowledge rather than directly from experimental findings, but when we do, it can add something really helpful as in this article by Heidi Muenchberger, Elizabeth Kendall, Peter Grimbeek and Travis Gee.

Now I’m definitely a proponent of evidence-based management – but in very complex situations such as the return to work setting, the insights gained from our clinical experience can be very helpful.

What these researchers did was review the literature on factors known to be associated with returning to work for people experiencing musculoskeletal injury. From an initial group of over 1000 articles, only 55 met their inclusion criteria, which is surprising in some ways, and could suggest that perhaps they were rather too restrictive in their choices. Despite this, 38 unique predictors were identified, which were expanded by a group of researchers to a total of 85. Some of the predictors that were expanded included: Dependents, Occupation, Employment status, Nature of injury etc.

The authors of this report commented that, as in many meta-analytic studies, many of the terms used were rather loosely defined and could cover a multitude of factors. For example, ‘rehabilitation intervention’ ranged from medical treatment, physical rehabilitation, educational rehabilitation, multidisciplinary rehabilitation, vocational rehabilitation… ahh do you get the picture?!

A separate group of rehabilitation practitioners were then brought in to participate in the clinical examination of the relevant factors. From a large number of individuals, only 12 participants completed the study, and sadly there were no occupational therapists (in New Zealand, occupational therapists and physiotherapists appear roughly in equal numbers providing vocational rehabilitation).

This groups’ job was to rate each of the 85 predictors using three scales according to importance, modifiability and categorisation. Importance and modifiability were rated using a three-point scale, while they were asked to classify each predictor according to a seven-point nominal scale, with ‘each value representing one of seven categories of the typology proposed by Krause and colleagues’
1 = Individual Worker Factors,
2 = Injury Factors,
3 = Medical and Vocational Rehabilitation Factors,
4 = Job Factors,
5 = Organisational Factors,
6 = Insurer Factors
7 = System Factors
The most useful aspect of this study probably also took the most time: ‘practitioners were asked to provide a written rationale for each of their 85 ratings of importance. Specifically, participants were asked to describe why each predictor was important for rehabilitation and how it influenced return-to-work. This qualitative data was analysed separately to the survey results.’

And the findings?
Well, I was interested in just how much agreement there was between practitioners – and it came out as ‘fair’ (Kendall tau = 0.27). Curiously, the authors suggest that this is ‘acceptable’, but in my mind it’s probably a little low…Despite this, the results of internal reliability testing at the individual predictor level showed that practitioners were in agreement for importance of predictor. Only four predictors were considered the most important to rehabilitation and return-to-work by almost all practitioners (over 90%). These were:

  • timeliness of rehabilitation,
  • clear return-to-work goals,
  • communication between GPs and injured workers and
  • rehabilitation in the workplace

‘Predictors that were considered least important to rehabilitation included gender, cultural background and whether the person was a member of the union. Interestingly, some of these predictors (e.g., gender, cultural background) are among those represented most frequently in the return-to-work literature.’

What can we make of this discrepancy? Well with the small reference group, not a lot really. What would be interesting would be to see what other groups such as employers or people undergoing rehabilitation would say!

Two predictors, namely ergonomic strategies within the workplace and timeliness of rehabilitation,
were agreed by most (at least 80%) practitioners as having the greatest possibility for modification.
I’m not sure that the first, ergonomic strategies, is particularly helpful especially for the long term management of a musculoskeletal problem, however, it’s nice to know that practitioners acknowledge that they’re relatively easy to modify – perhaps that ease of modification (and the visibility of modifications) is why they get done, while other aspects such as timeliness and communication do not.

Judgements of clinical utility were made using ratings of importance and modifiability (is that even a word?!). Nine factors were identified as having sufficient agreement:

  • Rehabilitation in the workplace
  • GP and injured worker communication
  • Clear return-to-work goals
  • Timeliness of rehabilitation
  • Proactive response by employer
  • Workplace accommodations
  • Elimination of risk factor from workplace
  • Modified work
  • Intensity of rehabilitation

Cultural factors were commented upon by the research group: ‘‘injured workers from
non-English speaking backgrounds often have cultural pressures on them not to complain or to keep
working even if injured’’, and, ‘‘some cultures have a particularly strong work ethic and other cultures (on the whole) cope less well with the change an injury brings, which deeply affects their pride and their ability to be the breadwinner’’.

To accommodate this, the authors state that ‘culture was more likely to be inherently considered in the design of intervention rather than as a single predictor of outcome.’

I’m not entirely sure how this was determined from this piece of research – and indeed, I’m not sure whether it is well considered. Perhaps this is yet another area for research from the patient’s perspective.

I’m going to continue commenting on this study tomorrow – the qualitative comments provide some interesting findings, and deserve more space than I can give today!
So, come on back tomorrow for more!!

Muenchberger, H., Kendall, E., Grimbeek, P., Gee, T. (2007). Clinical Utility of Predictors of Return-to-work Outcome Following Work-related Musculoskeletal Injury. Journal of Occupational Rehabilitation DOI: 10.1007/s10926-007-9113-0

Krause N, Frank JW, Dasinger LK, Sullivan TJ, Sinclair
SJ. Determinants of duration of disability and returnto-
work after work-related injury and illness: Challenges
for future research. Am J Ind Med 2001;40(4):464–84.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.