Joel Ackerman outlines the embedded chip problem in the health care field. His conclusion: you must check every system with a chip.
This applies to all other industries. Problem: it's impossible to do. There are too many of these systems and too many chips.
* * * * * * * *
The majority of computing processors currently in use do not reside within what we would normally consider to be computers, i.e., mainframe computers, mini-computers, and personal computers. Embedded processors are found everywhere within health care organizations. Some of the more obvious examples include elevators, security systems, phone systems, parking systems, heating/ventilation/air conditioning systems, fax machines, copiers, and printers. Less obvious examples include voice mail systems and backup power systems for buildings, mainframe, LANs, and PCs. Many of these systems are expected to fail if uncorrected. The problem is that we donít know which ones, or how and when they will fail. Will they just stop working, or will they give wild and unpredictable results?
Now consider the processors embedded in medical devices. Think about all the places that chips are used in a hospital, laboratory, or clinic: infusion pumps, lab equipment, MRI, CT Scan, dialysis, chemotherapy, intensive care, and on and on. How many medical devices do a date, age, or timing calculation? Probably more than you know. How many of these are going to fail or give erroneous results? No one is sure. But think of the potential impact on quality of care. Now think about the legal and social liabilities, liabilities for which you might be held personally accountable.
I recently had the opportunity to meet with a group of health insurance CEOís from around the world to discuss health care information technology issues. After discussing the year 2000 issues, one of the CEOís said, "Anything with a chip is suspect and needs to be checked out, right?" Bingo!