Medication safety and the real world of general practice

Drug development is a pretty costly process – roughly $1 billion over a ten year period for a single new chemical entity. This process covers everything from the initial drug discovery phase, pre-clinical in-vitro and animal studies, through to clinical trials. But then, drugs can be costly in another important way – safety.

Clinical trials are designed to establish both the clinical efficacy of a drug, and its safety. Many of the modern procedures regulating drug development were a response to the thalidomide tragedy of the 1960s, where thousands of infants were born with severe malformations, most notably of the limbs. There are typically four phases of clinical research in humans. Phase I involves testing the drug in a relatively small number of healthy volunteers to establish safe doses, as well as various aspects of how the body metabolises the drug (pharmacokinetics) and how the drug affects human physiology (pharmacodynamics). These trials are often the first time drugs are used in humans; many will remember the disastrous case of the testing of drug TGN1412 by Parexel at London’s Northwick Park hospital in 2006, although fortunately such events are vanishingly rare. Phase II trials follow, evaluating the therapeutic dose of the drug in a larger number of patients. This is turn leads to Phase III, which generally establishes whether a drug has therapeutic benefit. Such studies, in perhaps 1000 patients, are essential for the drug to gain approval from the regulatory authorities. Subsequent post-marketing studies – Phase IV – may then examine the drug in different patient populations or for new clinical indications, often in much larger populations and over longer time-periods.

All these studies contribute valuable safety data. Problems identified in the earlier stages will prevent a drug reaching market, and adverse effects detected during post-marketing surveillance may lead to withdrawal. Indeed, a number of high profile drugs have had approvals revoked due to concerns about safety once the drug has found its way into routine practice, such as the anti-inflammatory rofecoxib and the diabetes drug rosiglitazone.

So clinical trials are clearly a fundamental part of drug development. They also need to achieve a difficult balance, establishing safety whilst avoiding excessively protracted evaluations which delay the population gaining from important therapeutic advances. Obviously some compromises may therefore have to be made. Don’t get me wrong – we’d all like to know that a drug has been tested successfully in 10,000 patients and long-term safety established over 20-odd years, but I don’t believe that is a realistic expectation. Let’s get real.

Ah – but clinical triallists do not live in the real world. They live in a world of rigorously conducted randomised controlled trials, with strict inclusion and exclusion criteria, carefully defined outcomes and meticulous monitoring procedures. Standard operating procedures and protocols are the norm. The watchful eye of the data monitoring committee and regulatory authority are ever-looming in the shadows. Now let’s contrast this with a typical example from clinical practice. It’s the end of a busy three-hour surgery, and Dr Merlin’s final patient Mrs Mim walks through the door of the consulting room. Mrs Mim is an 80-year-old lady who keeps fit and active. She’s got hypertension and a couple of other medical conditions which don’t limit her functionally, although she enjoys complaining to Dr Merlin about how she rattles after her usual morning handful of pills. This time round she has a new problem – malignalitaloptereosis (look it up if you must). Fortunately, GSK have come to the rescue with a new antimicrobial, Disnefloxamicin (don’t bother looking that one up). All doctors know that Disnefloxamicin is a brilliant drug – it tastes of strawberries and the free pens that the drug reps hand out have four different colours. And it kills all known germs. Everyone knows this. But few people know that it was never tested in people aged over 75 years, or that people with high blood pressure were excluded from the original trial, or that there is little experience of its use in people on the drugs that make Mrs Mim rattle in the morning. Yet there are no alarm bells ringing on Dr Merlin’s computer as he prints out the prescription, and Mrs Mim goes off happy in the knowledge that she’ll be better in time for her Zumba class at the end of the week.

Of course, this is probably all pretty reasonable stuff. After all, we can’t test drugs in everyone, particularly given the increasingly multimorbid nature of the general practice patient population and the fiendishly complex nature of many prescribing regimens. So the results of trials are necessarily generalised to the broader population. And on the off-chance that something does go wrong with Mrs Mim, Dr Merlin can always fill in a Yellow Card and post it to the MHRA, or report an incident to the National Patient Safety Agency.

But here lies a problem. We know that this often doesn’t happen. Less than 2% of reports on prescribing incidents to the NPSA are from primary care. And Yellow Cards are also under-used. Sure – if Mrs Mim grows a tail, turns bright blue, or has a cardiac arrest, then this is likely to spark concerns. But if her blood pressure goes up significantly or her renal function worsens suddenly, there’s a high likelihood this may not even be noticed. Even if it is, a transient deterioration may be dismissed as clinically unimportant.

So post-marketing surveillance is crucial, and primary care has a vital role to play in this regard. How can general practice help with drug safety research? I would propose a number of key areas:

  • Firstly, generate new sources of safety information. Yellow Card data can be mined for subtle drug safety signals. Work also needs to be done to understand why medication problems are under-reported in primary care, and how we can develop a better culture of safety.
  • Secondly, practices should be more closely involved with clinical trials. Pragmatic, primary care based trials are more likely to generalise to the populations GPs treat. Efforts are also being made to integrate recruitment processes into routine clinical care, and randomisation at the practice as opposed to patient level potentially simplifies the process of trial participation.
  • Thirdly, we should encourage the sharing of routine data. This is something advocated by the recent Caldicott 2 review of information governance. If there are concerns about the legitimacy of data sharing, explicit consent should be sought from all patients as a matter of course, and legislation must support this. Large primary care datasets already exist, such as CPRD, and practices should be encouraged to contribute to them; they provide excellent opportunities for post-marketing surveillance. Linkage of data to other sources should also happen, such as hospital records and registries.
  • Fourthly, there needs to be qualitative evaluations conducted examining prescribing behaviours in more detail. We need to understand the educational needs of GPs in relation to safe use of medicines – the RCGP curriculum has a section “Patient Safety and Quality of Care”, but more emphasis needs to be placed on prescribing. We also need a better understanding of why certain drugs are used in potentially harmful ways, and how we can successfully intervene to reduce harm.

There are undoubtedly other areas. What is perhaps paramount is that we better acknowledge the importance of research in general practice. It is an accepted norm in secondary care, but less so in the GP surgery. Clinicians have a duty of improving the health of the population as a whole, which includes patient safety, and research is key to achieving this. “We’re too busy” is a common, and entirely understandable, excuse, and so efforts need to be made to facilitate research, perhaps by reducing or reprioritising other commitments, or by automating research processes. Only by doing these things can new pharmaceutical developments be made more relevant – and more safe – for patients in the real world.

This entry was posted in Blog and tagged , , . Group: . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • The Cambridge Centre for Health Services Research (CCHSR) is a thriving collaboration between the University of Cambridge and RAND Europe. We aim to inform health policy and practice by conducting research and evaluation studies of organisation and delivery of healthcare, including safety, effectiveness, efficiency and patient experience.