Blog

Discussing Work Contribution: Part 2

Blog - Discussing Work Contribution: Part 2

Dr Mary Wyatt | Published: June 30, 2015

Today I want to talk about Occupational Epidemiology.

That might include studying an epidemic, such as the spread of the AIDS virus. It might involve studying the occurrence of asthma, which groups of people get asthma, and what factors might increase the risk of asthma in the various demographics.

Occupational epidemiology studies what happens to the population of people who are working. For example, do brick layers get more shoulder problems than bank managers? Which chemicals tend to induce asthma? Which body movements increase the likelihood someone will experience neck or back problems?

Occupational epidemiology is often a challenging field and the resulting information is often poorly used.

Research information can be hard to access, hard to read, and it is hard to bring together all of the disparate research study types and results available in the research literature.  

To access the full text of studies, one either needs to pay a significant amount of money or have access to a university database. As an adjunct lecturer at Monash University I am fortunate to have the access, however the vast majority of doctors do not.

Abstracts or summaries of research are available online through the publicly available Pubmed database, however you cannot rely on the abstract alone.

Anyone using the available studies needs to look at the breadth of research, as there can sometimes be conflicting results depending on how studies are conducted.  If one study says work is contributing to an ailment and eight say it is not, you need to review all of the studies to get a handle on the topic. It also needs to be understood that studies are done in different ways, with different levels of reliability.

The most reliable study type is where people are followed forward over time. This is known as a prospective study. An example of a prospective study might be when participants are asked each year if they have had any soreness in their back during that year. An even better quality study asks people every week if they have experienced soreness in their back. This can easily be achieved through new technologies such as SMS tracking, where people receive a text and respond with one word. The study provides a solid contemporaneous record. 

Cross-sectional studies look at people at a certain point in time. They might be used to match symptoms with risk factors. 10,000 people might be asked on one particular day in 2011, “Have you ever experienced any back problems over the course of your life?”  

The limitation of this type of study is what we call recall bias.  Many factors will influence whether a person recalls a past event. For example, somebody who works as a labourer is more likely to recall a back problem from ten years ago, as it is more likely he was not able to do his job for a certain period due to the issue.  On the other hand, a payroll officer with the same problem may well have been able to continue with her normal job so is less likely to recall that episode.

This is the reason cross-sectional studies are a poor second when it comes to occupational epidemiology. Despite their drawbacks however, they are easier to conduct and the studies are much less expensive.  As a result, there are many more cross-sectional studies than there are good quality prospective studies which follow people forward in time.

Over the past ten years there have been an increasing number of prospective studies, enhancing our knowledge in this area.  

We are only now getting to the point where we have sufficient information to understand work contribution in certain limited areas. We need many more good quality studies to have a solid basis from which to make inferences on a broader range of areas.

Some doctors rely on their own experience to make inferences about work contribution to health conditions, but this is no better than relying on our own experience for assessing the benefit of the treatments we recommend.  

For example, if a surgeon is caring and engaging, patients are more likely to say they have done well after an operation, as they want to do the ‘right thing’ by the surgeon.  

We also, like most people, have a tendency to hear what we want to hear. This is why we use evidence based medicine to assess the effectiveness of various treatments. Proper studies remove the bias.  

It is not appropriate to rely on our own experience when it comes to assessing work contribution. We need to temper research evidence with experience, but it is insufficient to rely on it in assessing work contribution. Different doctors have hugely different beliefs regarding this area, and this is generally because the beliefs are not based on research evidence.

Changing this will require significant interest from policy makers. It is a vexed area, and perhaps at a policy level it’s a bit of a hot potato. As a result it is too often put in on the back burner.  

If we are going to bring credibility and integrity to compensation schemes, this is an important area which needs to be tackled.

It might take five or ten years to get this to a level where it is dealt with reasonably most of the time, but I believe that it is worth the investment.