May 8, 2011
Medicines are sometimes withdrawn from the market or heavily restricted in their use, when the risks are judged to outweigh the potential benefits. Is this just a case of bad luck or an unavoidable risk when it happens, or is it something more?
According to a fascinating article in the American Journal of Public Health (Am J Public Health. 2011;101: 399–404. doi:10.2105/AJPH.2010.199844) there may be a more systematic reason. The authors argue that there is an inverse benefit law in operation – that the overall benefits of a medicine are inversely proportional to the extent to which it is marketed.
The key to their argument lies in the number and condition of people treated with the medicine. Good clinical trials may indicate (correctly) that a medicine is beneficial for the patients in the trials but when the medicine reaches the market, does the population treated reflect the same condition and degree of severity as the trial patients? If a very low proportion of the population are as “sick” as the trial patients, the market for the medicine will be small – unless the numbers to be treated can be increased. The authors cite six ways in which this can be achieved through marketing.
Small changes in the indications for which a medicine is prescribed may greatly expand the population to be treated – and therefore the profits of the pharmaceutical company. The numbers of people to be medicated for high blood sugar, for example, can be greatly increased by lowering the threshold for treatment. (To be slightly technical, the article shows a normal distribution curve where the threshold to treat is towards the extreme right hand end of the curve. If the threshold can be “nudged” to the left towards the centre of the curve, the population to be treated, and the market for the medicine, can be increased dramatically.)
The use of surrogates instead of hard outcomes can drive the same process. High blood sugars, high cholesterol and high blood pressure are not diseases; they are surrogates for a risk of certain bad outcomes. Too much focus on surrogates can give the illusion of greater efficacy and help to drive down the thresholds for treatment.
The safety of new medicines can be over-estimated easily; the risks of medicines on the market for some time may be relatively well known and may operate to make prescribers very cautious in their use – the newer medicines can easily seem to be safer by comparison, particularly if certain aspects of safety are stressed in the marketing.
The efficacy of new medicines can also easily be over-estimated. The authors take the example of a new medicine that is a good second line treatment for a small minority of patients who react badly to an existing medicine. In these circumstances, marketing can operate so as to move the new medicine towards first line use – because the prescriber may not wish to take the risk of a known if infrequent adverse reaction to the old medicine.
The fifth mechanism cited in the article is selling sickness or creating new diseases. The authors cite “pre-diabetes” or “pre-hypertension” as examples, and I think there are many others.
The sixth mechanism for increasing the numbers treated with a particular medicine is off-label use – prescribing the medicine for conditions for which it has not been approved. There are a myriad of ways in which marketing can encourage off-label use without actually breaking the law against promoting such use. (I will do a separate post on this sometime.)
Reducing thresholds for treatment, and the other mechanisms mentioned above, will usually alter the benefit risk balance, adversely. The number needed to treat will increase, the number benefiting may not, or at any rate not to the same extent.
The arguments in the article apply most perhaps to medicines for conditions that vary in their severity, such as blood sugar rates, rather than to conditions (many infections?) that simply exist or do not exist in a given patient.
This is my summary – the actual article is far more rigorous (and qualified) in its argument and should be read by anyone who can get their hands on it. It should be read especially by policy makers, including those working at EU level on “Information to Patients”. It deals with marketing (to doctors in this particular case) and displays a subtle understanding of marketing that is far removed from the simplistic arguments as to whether marketing does or does not “make” doctors prescribe things they shouldn’t.
Now that DG Sanco has responsibility for EU policy on medicines, the argument about information to patients must be seen primarily in the context of health and health policy and with a clear understanding of the very wide scope of marketing, its effects and how it actually works in practice. ENDAuthor : Jim Murray