Owen Yang
I think I might have been generalising a little too much. When I say the UK, it is probably more so in England than in other UK countries (i.e. Scotland, Wales, and Northern Ireland).
The hidden quality assumption on the healthcare data for research investigation
For a long time, I have felt and heard that people in the UK have this myth, thinking the evidence generated from routine healthcare data in the UK, Sweden, and Denmark is qualitatively better than those generated from US and Taiwan. In all of these countries, there are some public or private mechanisms to track an individual’s medical records overtime to some extent. This allows some feasibility to set up a central office that keeps all these data updated, and therefore can be used for nationwide or large scale research at a relatively low cost. Other key countries include Canada, Australia, and perhaps Korea, but I am not sure what their attitudes are and what others’ attitudes are towards them.
The dismissal of so-called ‘claim data’
To give an example, let us just say there is a study to investigate whether prescribing benzodiazepine may cause dementia. One can try to get the prescribing and diagnosis data from routinely collected healthcare records in the UK, US, Taiwan, or many other countries that this may be available. Doing as many countries as possible is great, because it provides a good diversity so that we have better insights.
What people in the UK may feel special, cutely, is that we like to use an umbrella term ‘claim data’ as opposed to ‘data not for claiming purposes’. The UK runs a health system (the NHS) that is free at the point of care (which is not entirely true but it is generally believed so). Because the recorded data are not used to claim insurance money, it is commonly thought to be less biased than the data, say in the US or in Taiwan, where the data are recorded to claim money. This is true in a way that, in Taiwan at least, some doctors may provide ‘free’ services or perks that they do not claim back from the national budget. Although these services would not be reimbursed by the health system, doing them help to keep patients happy and loyal. There are other capitalist and non-capitalist reasons why an item may not be recorded.
Every record is somewhat about money
What is not correct here is that all recorded data, UK or otherwise, are used for assessment, and the assessment is almost always about money. In the UK, hospitals are assessed by analysing outcomes based on patient demographics. There are henceforth capitalistic mechanisms in place to reward hospitals that ‘perform’ well, and to punish hospitals that perform badly. I call it capitalistic because non-capitalistic policies may consider doing the opposite, giving more money to support worse-performing hospitals. GP units (general practice, i.e. the primary care doctors) are assessed by their performance based on routinely recorded data, and so GPs who ‘play the games well’ should know which ‘quality indicator’ they should be focussed on this year in order to achieve a best overall assessment, and a good assessment is linked to the larger pay. This is not evil. This is just one common way how health system works. A health system does not run on love. It runs on money and human resource. With a larger pay, a GP surgery can hire more people and provide a better care.
It is really not evil. Just like a student, in order to do well, would try to make a strategy on which subject they do well or not do well, and how to maximise their total score.
Are doctors more noble in the UK?
Here is where the trick is. In the UK (or western countries) someone could argue ‘although the records are used indirectly for incentive purposes, GPs or doctors have to follow the national or professional guidelines, and cannot deviate from these guidelines. The doctor’s salaries are not (are they not?) linked to the incentives, and the budget is not completely determined by these incentives. In other words, doctors in the UK are more ‘noble’ and are not affected by these quality indicators.
Well, this argument cannot be true. If doctor’s behaviour is not affected by selection of quality indicators, then there is no point to use quality indicators to drive national policies and determine budget. Importantly, there is nothing infamous to try to use the effective resource to achieve care quality. I am not convinced.
And when you make these argument, do you think doctors in other countries not prescribe according to clinical guidelines? There is something seriously going on in the NHS as a monopoly. It narrows our mind a little bit. But this is the topic for another day.
Fundamentally, it is really not appropriate to think in any reason doctors or patients in the UK (or in the Western countries) has a higher moral standard. Sometimes I think this is what drives the underlying argument, and whoever implies this in a peer-review process should be called out and justify themselves.
If you are a peer reviewer for an international journal, is it appropriate to accept a paper from the UK, but not accept the same paper from Taiwan because it is ‘not generalisable’? What is in the back of your mind when you say this? Who do you want to generalise to?
