- HIT has been slow to embrace the concepts and practices of safety critical computing. There doesn't need to be FDA level of scrutiny, but there has to be recognition that HIT is analogous in some ways to a new drug or device. I would add that the current approach to certifying the fitness for use of HIT in the NHS appears to be User Acceptance Testing, which is wholly inadequate. The current consultation document Information Revolution says nothing about safety critical computing. The final version could of course, but it would require a more centrally directive approach than seems to be envisaged at the moment, a consequence of NPfITs failure.
- The logic that since 'to err is human', computerisation will reduce error is too simplistic. Errors are generally produced by the interactions between multiple systems/people. So it's wrong to assume that computerising manual processes necessarily reduces error. Computerisation may introduce error.
- Measurements of usage are not in themselves an adequate measure of the success of an implementation. I think a lot of people would settle for usage as a measure of success. The point made here is that in a complex system any single measure doesn't tell the whole story.
- The messy desk fallacy - it is tempting to think of clinical processes as somewhat chaotic and therefore ripe for rationalisation. Some truth in this of course, but only some - healthcare is complex and non-linear. I would add that some of key techniques of computerisation, such as Business Process Modelling, look pretty inadequate in relation to complex systems.
- HIT is too focused on delivering benefits of an administrative nature, and not enough on delivering direct benefits to clinicians and patients. With a few exceptions, this appears to be true. Think for example about the huge efforts that go into standardisation and terminologies, which are largely irrelevant to clinicians and patients, compared to functionality and user experience. The basis for this error is the belief that a central aim of healthcare is the efficient production of good records. That it is not was demonstrated by Garfinkel in the 1950s. But major areas of HIT activity, for example HL7, appear to have overlooked this.
- The field of dreams fallacy - if we provide new IT systems they will attract users
- 'Sit-stay'. The fallacy that computers are in a smarter than humans. A very deep seated error, almost one of the founding stones of HIT - medicine is a cognitive/information processing profession (make a diagnosis then apply the right treatment), so lets see if computers can do better. The NHS PRODIGY programme was an example of this erroneous way of thinking.
- HIT is often designed as if a single clinician were dealing with a case, or at most a team of clinicians sharing a common perspective on a case. Again, this may be too simplistic.
- "Paper forms are not simple data repositories that, once computerized, could be eliminated. Rather such ‘scraps’ of paper are sophisticated cognitive artifacts that support memory, forecasting and planning, communication, coordination, and education"
- "Teams of well-intentioned clinicians and software engineers may believe that understanding of clinical processes coupled with clever programming can solve the challenges facing healthcare. But such teams typically will not have the requisite breadth and depth of theories, tools, and ideas to develop robust and usable systems."
Individually and collectively, these criticisms of the current approach to developing HIT have some force. The NPfIT can be seen as suffering from most of them. But there is nothing to suggest that the incoming government's market led strategy has recognised these issues yet.
Related Post:
Widget by [ Iptek-4u ]
