My trust is thinking about entirely re-labelling all our medical devices with bar codes.
External trusts we do work for also express a preference for bar codes on equipment on account of the belief that
(a) they are useful.
(b) they are idiot proof.
Given the basic rule of databases, "garbage in garbage out", bar codes seem perilous to me because once you put a dud number onto an item I think it would be relatively difficult to spot the error if the bar code is large and the human readable bit is tiny. The tendency would be to just assume it's ok & keep on merrily scanning rubbish into the system.
Thinking about this though, I can't see why nowdays we aren't instead using text scanner pens to read in the ordinary text on the current generation of labels
I will be looking into this possibility over the next month but wondered if any out there had either tried using this route to automating the data input.
Our current alphanumeric labels were put on over a 6 month period a few years back. 3,000 were applied. About one in 20 labels were made up wrongly and spotted before application.
About 25 of the 3,000 actually applied were subsequently found to contain errors.
I doubt those 25 errors would have been spotted with bar coding.
Marc
