Owen Yang
Clinicians or medical scientists should be encouraged to explore psychosocial health of patients in every opportunity, but most of us do not have the training to explore them efficiently and appropriately. Some of us are trained to explore the psychosocial health of our patients, but exploring this in our research is not something that has regularly come to our mind.
Psychosocial health does not need to cause a disease
Because of lack of investment, there is limited evidence that psychosocial health causes diseases through biological pathways. Therefore, I do not want to waste time debating on this point.
Even if there is no biological pathway, our psychosocial condition could affect our lifestyle and our ability to access to health care. Those who struggle psychosocially may not have the space and time to engage a healthy lifestyle, and may present disease at a late stage. They might not be followed up in the same way, and therefore could cause detection bias. Some medications may have interaction with other medications, and therefore the management of other diseases could be affected. It is also important to demonstrate our study is representative, and so can be applied to patients at different stage of life.
In brief, we should not be surprised that the effect of our treatment could depend on their psychosocial health. This would mean some effect modification (i.e. interaction) by psychosocial wellbeing. This is quite similar to how we treat socioeconomic status. Although we do not necessarily hypothesise a biological pathway between socioeconomic status and a disease, we do hypothesise that the impact of socioeconomic status is broad, and its potential bias on our study should not be dismissed. It would be the same for psychosocial wellbeing in the 21st century.
One-liner is generally unacceptable
There are more than one way to collect the information of psychosocial wellbeing. Most people would like to explain this and go all academic and enthusiastic. As much as I want to, I feel it is more useful here to just give some over-simplified advice.
If there is no specific hypothesis, but we feel generous about our project, which we should do, we should first make sure we prioritise collecting information about smoking (current, past, never), alcohol consumption (times a week or amount a week), and body mass index (height and weight), and at least one measure of socioeconomic status (income or education) before we even think about psychosocial wellbeing.
After this, I would say find a most generic wellbeing index of 4-5 items (5 is best) that covers general wellbeing in different words. Examples would be WHO-5 or PSW-4. Believe me, although you might not feel they are as sophisticated as your other aspects of shiny study such as MRI or flow cytometry, a lot of time incorporating this in your project could branch out as a standalone paper.
Without giving an academic speech or debate, please just take my word: do not use a one-liner to collect psychosocial wellbeing. Do try to have a least 4-5 items, and try to take them as a whole whatever is being asked.
A few more words on the interpretation of these wellbeing measure
The medical research society is heterogenous. A few of us are so scientifically trained and have this belief of objectivism. Some people with this view would feel that asking subjective feeling and quantify them is not science. This is a big topic and a debate here is futile. What I would draw your attention is that we could not pretend something does not exist just because we cannot measure it perfectly.
We should also not pretend we have measured something just because it looks scientific. Think about body mass index, which is very easy to measure in a way that is so ‘scientifically’ that we forgot the fact that most of the time body mass index is just a number that represents a complex biological process that we absolutely have not measured. We should not pretend we have measured obesity just because we have a reproducible number from height and weight.
It is also important not to take the words of the wellbeing items literally. There are a few doctors insist on patients to be a logic master, and felt patients should not say they are happy and worried at the same time. The psychosocial scores is trying to take into account people’s different personal experience on life and on language use. We do not say the individual is happy by directly asking them whether they are happy. We try to use a range of information to judge whether the individual is ‘happy’ as to what happy means to that individual and to the society. Just like we do not ask patients whether they are addicted to smoke, but ask them in a number of different ways that the society may define as a problematic smoker.
But again this is itself a big topic.
