This post is also available in: हिन्दी (Hindi)
A lady in her 60s recently consulted me. She was extremely concerned because she was facing a dilemma and she did not know what to do. Basically she had been to see her GP and because her cholesterol was found to be a little on the higher side, she was being strongly advised by her GP to take a Statin. My patient did not want to take Statins. Her argument was that she had never had a problem with her heart, did not have a family history of heart disease and therefore did not want to take medications that could cause side effects. Her GP told her that by not taking a Statin, she was being irresponsible because her risk of a heart attack or a stroke was going to be much higher on account of her having a higher cholesterol and the statin would reduce this risk. My patient was left very troubled. She wanted to do the right thing but her GP did not really have the time or patience to discuss her concerns with her. As far as he was concerned, it was either his way or the highway. So she came to me and said ‘Dr Gupta, could you please help me make sense of this. What should I do? Am I being irresponsible and should I go on Statins?’
Today, I am going to tell you what I told her. This blog is entitled ‘The Great Medication Lottery’.
All health conditions have to either impact your quality of life adversely or predispose you to some kind of future risk I.e impact your length of life adversely – otherwise they aren’t medical conditions at all. High Cholesterol is not really a condition unless it impacts on the patient’s quality of life or length of life.
Therefore all treatments that are offered have to either improve quality of life or reduce future risk and improve prognosis.
One important difference between quality of life and quantity of life is that your quality of life can be measured and only measured by your yardstick. No-one else can measure your quality of life like you can. Quantity of life on the other hand can’t be measured by you – only other people can measure your quantity of life because to measure something, it has to come to an end.
Deciding on taking medications that improve quality of life is easy because you can measure your quality of life and therefore if you take a medication that is given to you to improve your quality of life and it does then you know its worth taking. If it doesn’t, then there is absolutely no need to take it.
Deciding on medications that are supposed to improve your length of life is far more difficult because you will never know whether the medication has made you live longer. You could take it all your life and may think it has but you will never know for sure because you don’t know for sure as to what would have happened if you did not take the medication.
So how do you make the decision to take something for the rest of your life when you will never know if it will help or not?
The only way to gauge if it is worth taking is if it has helped other people like you and more importantly you want to know if it has helped a population of patients just like you. This is where data from research studies can help. What you want to know is if there have been any studies which have studied a large enough population (the larger the better) of patients just like you for a long enough duration (the longer the better) and compared the effect of taking the medications with a similar population who were given placebo and you want to know how many patients in the population taking the medications outlive all of the population not taking the medications by the end of the study period and based on that number you can decide whether it is worth you taking the medications or not.
When we look at studies that test the impact of a medication on longevity, we will never see that every single person taking the medications will outlive every single person not taking the medications. What we will always see is that bad things happen in both groups and the chances of bad things happening in both groups will increase with time – it’s just that more bad things happen in the group that are taking the less effective medication. So the more effective the medication, the larger the number of people standing at the end of the study compared to the group that has the less effective medication or placebo. So by calculating the difference in the number of people alive and well in both groups, we can work out how effective that medication is. This is called the number to treat – NNT.
Let me give you an example:
We have a study which studies the benefit of Medication A (which is supposed to prevent heart attack-related deaths) compared to placebo. We take 100 people and give them medication A for 5 years We take 100 almost identical people and give them a placebo for 5 years. At the end of 5 years, we find that 10 people in the group taking medications have had a heart attack related death. On the contrary, 20 people in the group taking the placebo have had a heart attack related death. We can therefore conclude that 10 people would have had a heart attack related death regardless of whether they took the medications or not and 80 people would have lived regardless of the medications but it does appear that the medication did make a difference to 10 people who may have otherwise died if it were not for the medication. So in essence we would have to treat 100 people for 5 years in order to save 10 lives which means that we would have to treat 10 people for 5 years to potentially save one heart attack hence the number needed to treat is 10 (over 5 years of treatment). The lower the NNT the more effective the medication.
In the same way, medications may also have harmful side-effects and therefore it is possible using the same method to calculate a number needed to harm. The lower the NNH, the more harmful the medication.
As every medication that is supposed to prolong life or reduce future risk should have been studied in a quantitative study, you would be forgiven for thinking that there should be a number needed to treat and a number needed to harm easily available for us to make a judgment on whether we feel that the medication is worth taking or not. Unfortunately these numbers are incredibly difficult to find. The reason for this is that often the numbers needed to treat tend to be underwhelming and to try and push their products, pharmaceutical companies rely on relative percentages to make their product appealing.
Let me explain. If I treat 10 people with a drug and 8 are alive at the end of 1 year but in the placebo group only 6 are alive then we could say that the number of deaths has been reduced by 50% by the medication although we have only saved 2 people. Here the number needed to treat is low i.e 5 (which suggests that that is an effective medication) .Now if we treated 1000 people with a drug for 1 year and 998 people are alive at the end of the 1 year and only 996 people are alive in the placebo group then the pharmaceutical company can still claim that number of deaths has been reduced by 50% although this time you have had to treat 1000 patients to save 2 lives but the NNT is much higher at 500 (this suggests that the medication is not particularly effective at all). So when you get prescribed a medication, and you even ‘dare’ ask the doctor, what the benefits are – at most you will be given a relative percentage reduction such as it will reduce your risk of a heart attack by 30% which sounds impressive but is actually very far from the correct picture because it doesn’t tell you how many people you had to treat to get that benefit. The problem with treating lots of people is that they are having to take the medication unnecessarily and are subject to possible harm from the medication and have an exceptionally low chance of benefiting from the medication.
We can use a lottery analogy here. If someone comes to sell you a lottery which offers a million pounds as a prize, there are 2 important questions that you should always ask. How much is the ticket (ie how much discomfort do I have to go through to get a chance at winning the prize but more importantly how many tickets are going to be sold. If there are only 10 tickets being sold, then I would be tempted to buy the ticket, even if it is a lot more expensive. On the other hand, if a million tickets are going to be sold then there is little merit in me investing, especially if the expense of the ticket is going to make me uncomfortable. Unfortunately, the pharmaceutical companies response to the question, how many people bought the ticket is -don’t worry about that – the more important point is that if you buy two tickets you will double your chances of winning. Yes and no. If 10 people are buying the ticket then buying 2 tickets does meaningfully increase my chance of winning the lottery but if a million tickets are being sold then buying two tickets is not really going to increase my chances of winning.
All doctors who prescribe medications that are supposed to improve longevity should have the NNT and NNH for the medications that they are prescribing at their fingertips. Unfortunately hardly anyone does – and I can say this with confidence because even I don’t and the real reason is that these numbers tend to be hidden away from the public eye and relative risk reduction percentages tend to be promoted everywhere.
This just goes to highlight the absurdity of modern day medical practice which is doctors prescribe medications to patients to promote longevity and neither the doctor nor the patient even knows what the perceived benefit is likely to be and therefore you just end up taking medications without any good reason. Using the lottery analogy, the lottery seller can’t tell you the cost of the ticket, or what your chances of winning are (based on how many tickets are sold) for a prize which you wont know anything about even if you did win!
Next time you are with your doctor and he recommends a statin, ask him or her what the NNT and NNH are and I can guarantee they won’t know! You will get an answer like it will lower your cholesterol and high cholesterol is bad for us. The reality is that you are not taking a cholesterol lowering agent to lower the cholesterol – you are taking it because you don’t want to have a heart attack or a stroke and therefore just being told that it will lower your cholesterol and this is good for us is not an adequate explanation – you want to know exactly how many people like you would need to take a statin to save one person from having a heart attack or stroke ie the NNT for a heart attack or a stroke and you will be unpleasantly surprised that your doctor will not know this value.
Because the NNTs are so difficult to find, I use a website called theNNT.com – on this website a bunch of statisticians and researchers try and work out the NNT for commonly used interventions and I would encourage everyone who is contemplating a potentially life-prolonging medication to check it out.
Here are some examples of NNTs (taken from the website)
- Aspirin for 1 year to prevent first heart attack or stroke causing death (no benefit)
- Aspirin to prevent a first non-fatal heart attack – 2000
- Aspirin to prevent a first non-fatal stroke – 3000
- Aspirin causing bleeding- 3333 (NNH)
- Statins for 5 years to prevent a first heart attack causing death – no benefit
- Statins for 5 years prevent a first non-fatal heart attack 104
- Statins for 5 years to prevent a first non fatal stroke 154
- Statins for 5 years to cause muscle pains/muscle damage – 10
- A Mediterranean diet for preventing death -30
- A Mediterranean diet for causing harm – 0
So as you can see, the NNTs for medications are in general underwhelming. .
My patient seemed a lot more comfortable after our discussion. She finally decided that she was going to decline the statin and adopt a Mediterranean diet.
Here is a link to the video
My aim in this video is not to say that medications are not beneficial – my aim is to empower you to ask questions and be as well informed as possible so that you can make the decisions that are right for you
Please do not make any changes to your medication regime without consulting your own doctor first.
I hope you found this useful. I would love to hear your thoughts.
Keywords: Big pharma; NNT; NNH; Overmedication; Statins; Yorkcardiology; Dr Sanjay Gupta
This post is also available in: हिन्दी (Hindi)