Medical device companies must balance a drive for innovation and new product features with an equally intense focus on thorough verification and validation. Hunting a haystack of operating conditions for an obscure failure mode needle may not be sexy, but it’s as critical to the success of a venture as innovative product concepts and technology.
In April 2013, Abbott announced a recall of their Freestyle InsuLinx Blood Glucose Meter. The recalled device reports incorrect (and much lower) values when measuring very high glucose levels of more than 1024 mg/dL. Glucose levels that high are rare, but dangerous.
Eight days later the Lifescan OneTouch Verio IQ glucose meter was recalled for slightly different behavior but the same fundamental failure mode – turning off at 1024 mg/dL.
Some of the most professional companies in the medical device industry were affected. Fortunately, nobody seems to have died as a direct result. The question is what can we learn from the event?
It appears that inadequate attention to verification and risk analysis was paid in these cases. I believe a team of engineers that embodies innovation, thorough execution and testing, and a mature quality management system are all critical for medical device success.
Let’s examine the recalls above. Both meters failing in similar ways at the same time may indicate that one device was designed to be similar to the other or they used the same OEM module.
The number 1024 is exactly 10 bits of data. A 10 bit analog to digital converter may have been used without controlling for out-of-range conditions. Or the rollover of an integer value may be related to being very economical with memory usage and allocating no more room than thought necessary.
From a risk analysis standpoint it appears this failure mode was not considered or tested in their design control. Here are a few places to capture this problem before it gets into shipping product:
- While creating the specifications, it’s possible the glucose range was not anticipated properly.
- Identifying and managing boundary conditions of critical components is always a good element to include during risk analysis. Per ISO 14971 and mandated by IEC60601 edition 3, the possibility that an incorrect reading could take place should have been identified and specifically mitigated.
- Verification testing the whole range of blood glucose levels should have identified this error. Was a corner cut during testing? Often verification work seems expensive, happens late and incurs a lot of pressure to compromise.
- During validation, the failure could have been caught. However, such a very high level of blood glucose is rare, so it may also have been missed.
When two highly respected companies like Abbott and Johnson & Johnson get caught up in recalls, we are reminded that thorough verification and validation are fundamental to successful medical device development.