Tackling bias when handling a sick member of your staff

" I run a small business and one of my employees was recently diagnosed with a terminal illness. I have since become emotional whenever I am dealing with this employee and I don't know how to deal with the risk of bias. Help.'

***

You have, in one short question introduced a number of concepts in both science and the use of language, and as you will see shortly, your question is neither easy nor as it seems on the surface, without danger of introducing much scientific jargon. We will avoid the latter to the extent possible.

You have asked us to consider the matter of ‘Risk bias’. At the simple linguistic level, (and I suspect this to be your question), what you intend to avoid, is the chances that in your relationship with this and other members of staff, you might act in favour or against this member of staff.

This might happen if, for example, you denied him time off to go to see his doctors or by making working conditions so difficult that he is unable to discharge his duties. In the alternative scenario, you might treat him with so much compassion that you end up antagonising other members of staff, who might feel upset because all the work is left to them, as “this sick person is paid to do nothing”.

This type of being biased happens to human beings in daily life and is best understood as the ‘risk of being biased’.

To the scientist, ‘risk bias’ is a whole new and different ball game that is to be found in medical research and involves at its most basic level a discussion on the quality of the research work being reported on. So, same words, different concept. Now for another situation.

As we come closer to the general elections in Kenya, many scientific studies will be done to “prove” which candidate is doing better than the other. In the process of interpreting the results, we already know in advance the response of those in contention.

Those in the lead in the particular poll, (and their supporters) will agree with the findings (whatever the biases) and those who are behind will oppose the results despite the veracity of the study design. This type of situation is not uniquely Kenyan and is found in many countries in times of political contest.

In science, we cannot allow this type of problem to entrench itself and this is where your question raises, without intending to, the question of bias as a scientific concept.

If for example, you want to study the effectiveness of a particular drug in the treatment of malaria, you might find your results being challenged if you do not distinguish between the groups that you consider between those who live in tea-growing areas and those who live near the lake. Similarly, you may want to consider the children and or pregnant mothers and or the elderly.

Your study must be without any bias.

In this understanding of the word bias, many different types of bias are included, for example, selection bias, information bias and confounding bias. Complicated? So the point is made.

To make this point differently, let us look at two commonly used words that have both lay and technical usage. The first is the word depression. You might for example have told us that upon learning of the illness of the member of staff, your reaction was that of depression.

This could mean on the one hand that you became sad knowing that he was the sole breadwinner in the larger family or indeed because he was one of the star performers in your company.

You might also have stated that after learning of the illness of that member of staff, you were taken to a psychiatrist who made a diagnosis of the clinical entity called depression and that you are now on medication for the depression. One word used correctly in different contexts.

The word anxiety has very similar usage. Your anxiety about the possibility of losing an important cog in the wheel of production in your company might have been your reaction as opposed to the anxiety that you might catch the life-threatening condition from your employee. Both are states of anxiety, one lay usage, the other technical and medical.

As you can now see, your simple question has taken us to language, politics and science in a number of easy steps. This, I hope has illustrated the fact that yours is a deeper question than seems on the surface, but the fact remains that because you have taken time to think of the possibility that you might be biased one way or the other, you are on the right path to finding the correct way out of your conundrum.