What level of medical device failure is acceptable? Brad Abbey argues that the industry needs to arm itself against the threat of litigation – not with lawyers, but with the right kind of evidence.
I was somewhat taken aback by a letter in the British Medical Journal last December, where under the unlikely-sounding title ‘FDA is gold standard of review’, Mark B. Leahy, president and CEO of the US Medical Device Manufacturers Association, while singing the praises of the industry, said that a recent study of FDA-approved medical devices from the past 5 years showed that fewer than 1% had been recalled.
Most recalls, he said, were due to manufacturing and design problems in a post-marketing setting. This was in response to an article that had been highly critical of the safety surveillance of medical devices in the US and the low hurdles that have to be jumped to get a device approved (seemingly true on both sides of the Atlantic).
I am a generalist in the healthcare industry, mainly involved with medicines, but to say I was surprised by that figure is an understatement. I realise that the number of car recalls may be higher – but with a surgically implanted device, the owner cannot check it in at the service centre and pick it up at the end of the day.
I know that MHRA gives daily warnings about medical devices, from wheelchairs to drug-eluting stents; but given that the level of adverse event risk that is acceptable to the public for a medicine is somewhere between 1 in 10,000 and 1 in 100,000, the 1% risk seems difficult to accept.
A lead article in the May 2011 BMJ, by Peter Wilmshurst of STARFlex fame, opened with the comment that the regulation of medical devices is (in his opinion) unsatisfactory, unscientific and in need of a major overhaul. Pretty damning stuff.
Duty of care
The registration of medicines requires data on the safety, efficacy and quality of products, and the numbers of patients needed to demonstrate an acceptable risk/benefit profile can be dauntingly high. The same level of scrutiny does not happen in Europe for medical devices, where a single approval can trigger cross-community acceptance.
With the increasing complexity of devices and the high levels of patient expectation, it is hardly surprising that when seemingly good devices go wrong the patients want compensation – and, where there is a suitable arena, for punishment to be meted out.
In the US, where many complex medical devices are developed and initially marketed, the ‘learned intermediary’ doctrine has been used by healthcare product manufacturers in recent times to protect themselves in the event of something going wrong. This doctrine, used in the US legal system, states that the manufacturer of a product has fulfilled their duty of care when they provide all the necessary information to a ‘learned intermediary’, who then interacts with the consumer.
This doctrine has been used primarily by pharmaceutical and medical device manufacturers in defence against tort suits. In a majority of American states, the courts have accepted this as a liability shield for pharmaceutical companies.
However, drug and medical device manufacturers sustained an unexpected blow in August 2008 in Rimbert v Eli Lilly and Company: in a federal court decision, for the first time, there was a rejection of the learned intermediary doctrine in its entirety. The decision rejected the notion that the manufacturers of drugs and medical devices do not have to make the patient fully aware of the risks associated with them and that this can be delegated to the prescriber.
The idea underlying the ‘learned intermediary’ doctrine is that the prescriber, who has expert knowledge and skill, should make the decision about risk. But changes in the consumer environment whereby prescription products can be advertised directly to potential patients have rendered this justification obsolete, and so it was predictable that for medical devices – some of them traditionally never coming ‘into the hands’ of the patient – the risk scenario would be influenced by the lesser amount of risk/benefit information needed before approval for marketing. While the doctrine has not been used in Europe, the risk information relating to devices is lagging behind that for medicines.
In order to be vigilant about the risks of medical devices, companies will be best served by surveillance systems that monitor the risk/benefit profile of products from the moment they are first evaluated (even if that takes place in animal models). This is not always easy.
A letter in the BMJ (in the same issue as Leahy’s letter) from a Welsh group of doctors highlights the problems of post-marketing surveillance for medical instruments, and in particular the use of single-use devices for tonsillectomy from 2001 in the wake of the variant Creuzfeldt-Jakob disease that followed the ‘mad cow’ scare of the 1990s.
Widespread adverse events were associated with these non-reusable instruments despite their CE marking, and they were deemed not fit for purpose. The case for reform of medical device regulation therefore seems a given.
Hip or lame?
In the meantime it seems that the visible portion of the iceberg of device regulation-related problems is giving rise to a stream of litigation that could possibly become a tide. Recent Medtech Business news reports have followed the fate of orthopaedic company DePuy and its ASR hip replacement.
Hip replacement is one of the clinical successes of the marriage between orthopaedic surgeons and the medical device industry, and it was estimated (before this year’s NHS rationing) that about 70,000 patients were undergoing total hip replacement each year in the UK.
I remember metal-on-metal hip replacements from the 1970s (I have one in a drawer at home that came to me as a result of its breaking), and they became popular again in the 1990s. However, the most recent generation have not fared so well, with higher than expected rates of failure and concerns about excessive levels of metal ions (cobalt and chromium) in the blood of patients.
According to 2010 data from the National Joint Registry of England and Wales, the DePuy ASR Hip Resurfacing System has a revision rate of 12% at 5 years after surgery and the DePuy ASR XL Acetabular System has a revision rate of 13%.
That means that during the first 5 years after a hip replacement with the DePuy ASR hip, at least 1 in 8 patients will experience hip failure requiring painful and expensive revision surgery. With more than 90,000 DePuy ASR hips in patients worldwide, over 11,000 people could require additional surgery due to the defective design of this implant and DePuy’s failure to remove it from the market earlier.
One of the questions that remains unanswered was whether there were potential conflicts of interest between the supplier and the healthcare professionals who developed and were involved in promoting these devices. The key issue in litigants’ minds is that the device did not perform as well as such a device might be expected to, and it seems that the device’s registration in the US was obtained without clinical trials to prove its long-term safety and efficacy. In a litigious society such as the US, where someone must pay for any mistake, the supplier appears to have suffered with the rolling of heads and the decision to remove the offending brand from the market.
Don’t get the idea that this case is a one-off: the recent history of medical devices suggests that arrivals on the market may sometimes be premature, as real risks may not have become apparent. Whether this is related to inappropriate endorsements from the medical profession is difficult to judge, but there are known examples of high-level payments to medical inventors who ‘sell’ their developments to industry and subsequently endorse them.
On the other hand, everyone is aware of what happened to Peter Wilmshurst when he took the opposite stance against a device manufacturer: there was a serious attempt to punish his critical views (which seemed to be well founded) and personally break him through the English court system.
Evidence is strength
Litigation against medical device companies is nothing new. However, in an age when people with problems can readily find lawyers willing to take on their problems, and some lawyers (particularly in the US) go looking for people who did not even know they had problems, access to litigation seems to be easier – and it is oiled by the possibility of compensation (which may be deserved when devices turn out to be inadequate or unsafe).
A Google search for the term ‘medical device litigation’ returned 640,000 hits; most of the leading ones were to do with lawyers offering their services in the pursuit of such litigation, or training sessions for lawyers who want to become involved in such cases, or training for companies who want to avoid them. I don’t believe a wake-up call about the risks of being sued is necessary, but what is well worth thinking about is the possible root causes of the current danger, which can ruin a company that believed it had a good product.
The message I am offering is consistent. The products of the healthcare industry must be subject to close and continuous scrutiny for their risk/benefit profile, and this should be done prior to marketing and continue in a structured manner post-marketing. NICE advisory policy on the best devices to use is still in its early stages.
There seems to be a raft of opinion supporting the idea that the regulation of medical devices (in Europe, and probably also in the US) needs to be overhauled to eliminate the placing of devices on the market with inadequate safety and efficacy monitoring.
Rather than finding ways of avoiding expenditure during a product’s development and launch by minimising the collection of such data, companies need to embrace the need for resilient data sets and continual risk/benefit signal monitoring. The competent authorities will wake up to this need, and those with effective systems in place will withstand the culture change best.
Brad Abbey is an industry observer, or the pen-name of an industry observer. The views expressed in this article are those of Brad Abbey, and do not necessarily reflect the views of Medtech Business.