It’s often rather strange coming back from holiday, and this week was doubly so as I left the snowy mountains and arrived later that morning to British summer time! What amazing weather here! The BBQ has been out on 2 consecutive days, and if I were on my deathbed and allowed one last meal, it would be a BBQ. The whole family is intact without a skiing injury between us, and we were so lucky with the weather. It really was a fabulous holiday, and I’m super grateful for it.
Personalised medicine is a term we hear increasingly frequently these days. There are a number of interpretations of this, but in my mind, it speaks to the fact that we humans are all rather different, and as such, what works for one will not work for others. This is patently obvious from medical trials. If a particular drug, operation or other intervention worked in 100% of individuals, it would be a miracle. This is never the case. Medical trials often quote a ‘number needed to treat’ ie, how many patients need to be treated a certain way to observe the desired outcome. In Cardiology, we are really lucky to have a great many drugs, operations, and other interventions that are backed up by amazing bodies of evidence. I was at medical school when ‘evidence based medicine’ really started becoming established (ie using robust randomised controlled trials to base decisions on rather than expert opinion) and in fact, was one of the reasons I picked Cardiology as my specialty. However even in the most successful Cardiology trials, the number needed to treat will still be in double figures.
I’m going to slightly digress here, but a notable exception is exercise and fitness. It beats drugs hands down.
Part of the difficulty of delivering personalised medicine is of course, time. You really need to know a patient personally to give yourself the best chance of doing it well. Maybe in time we will each have our own health AI that can do this for us, but whilst I think there are many areas of a doctor’s job that will be made easier using AI, I believe speaking to patients and building a rapport is not one of them. Would I want to see an AI doctor online? No. However, I would like for example to use an AI to help make diagnoses of conditions like rare eponymous syndromes though.
For the most part, I expect people would consider personalised medicine to be about advising on specific aspects of a patient’s treatment that are personal to them, and tweaking according to a patient’s wishes. I expect people would also expect this to be about promoting healthy behaviour.
I consulted with 2 patients in recent weeks who met the first criteria, but probably not the second! Both were in their 80s, and as you would expect had a number of heart problems. We were towards the end of the consultation, when they both asked about alcohol. Both liked a drink, but had been told by other doctors that they must not drink. I feel uncomfortable with this for a few reasons.
Firstly is not wishing to be a hypocrite. I do believe that there are considerable health benefits to not drinking alcohol (not least its impact on sleep), but whilst I don’t drink much these days, I can’t imagine a weekend where my wife and I don’t enjoy a bottle of good red wine (unless I’m on call!), and I had a glass most nights whilst skiing.
Secondly, patients will often do what they want anyway (and doctors who think otherwise are kidding themselves!) so far better to work with patients and try and meet them half-way, or compromise, even if a behaviour is not ideal.
Thirdly, personalised medicine for a man in his 80s with a chronic heart condition looks very different to a patient for example of my age. The look of delight on these chaps faces when I said I would be fine with them having a drink or two was priceless. Yes, it’s not what a textbook would say, but oh so important for their quality of life, and isn’t that what it’s all about?