Thursday, October 11, 2007

Hospital Computing – Why is it so Hard?

The following article was developed for Pulse + IT. The new print edition will be out in a week or so.

-----

All over the country it seems we have Hospital IT projects that are struggling to actually get started (e.g. WA, SA and to some extent QLD), are running very much behind the initially planned time lines (e.g. NSW and VIC), or are failing to satisfy their users (almost everywhere).

The first thing to be said about this situation is that we are not alone.

In 1998, Scott Silverstein M.D. launched a Web site devoted to shining light on healthcare IT failures. Hospital leaders, IT vendors and the media have swept the topic under the rug, he says. “IT failure is a serious problem, but people are reluctant to study it,” says Silverstein, the director of the Philadelphia-based Institute for Healthcare Informatics at Drexel University College of Information Science and Technology. “We like to talk about success, not failure.”

His site can be reviewed at the following URL:

http://www.ischool.drexel.edu/faculty/ssilverstein/failurecases/?loc=home

It now covers over thirty project which would appear, by all accounts, to have come seriously adrift!

Why is it that these projects seem to go badly so often – especially in the public sector.

I would suggest there are a range of reasons.

First , many believe a key point is that managerial and organisational instability is a major cause of failure. I agree this is really important and, indeed, when one reflects on the Public Health Sector it is really a relative rarity to have an Area Health Service CEO or CIO serve out their full five year contract. This flux is due, in part at least, to a combination of Government and Ministerial changes, changing policy priorities, some being perhaps promoted beyond their capabilities and the unexpected events that precipitate management change. Conducting any significant project in the absence of continuing stable senior management support is a recipe of disaster.

Second, especially in the public sector, there is often a disconnect between the managerial responsibility placed on a project manager and the freedom to act they are accorded. At times this leads to the “wrong” staff being retained in roles for which they are no longer suited, to the detriment of the project as a whole. The disconnect (and budget inflexibility) also often leads to difficulty in attracting and retaining suitably skilled staff as well as excessive delay in staff acquisition. The other problem that is almost universally encountered in Hospital projects in my experience is the “drip feed” of funds and the difficulties in getting suppliers paid. More than once I have seen competent project managers just resign in disgust when they realise they have neither the spending authority, money or the staff to deliver the project they are required to make happen.

Third, because executive health-care management are often uncomfortable regarding many aspects of Health IT, frequently associated with a fairly limited understanding of what is required, at an executive level, for project success, the quality of project sponsorship and support is less than is needed. Senior executives, like everyone else, prefer to stay within their “comfort zone” and, if the Health IT project is not within that zone, real difficulties are almost inevitable. The project manager has a difficult responsibility to carry the project sponsor along on the journey, and to make it clear what they must do for the project to be a success on their watch!.

Fourth, clinicians inevitably see a new system as a very low priority in their “caring for their patients” activities. This will lead to all sorts of difficulties with change management, training and effective use of a new system, unless both executive management are fully committed and real “clinician” evangelists and enthusiasts are recruited to work with their peers.

Fifth, involvement of all relevant categories of clinicians in the selection and later configuration of systems is crucial. The clinicians really have to be confident the system will work for them and be convinced of its value and utility or the project will be at extreme risk before it even starts.

Sixth, there is a real tendency to underestimate the complexity of and the effort required to implement say a new laboratory or patient management system – to say nothing of clinician facing systems such as Computerised Physician Order Entry or Computerised Nursing Documentation which involve virtually all key staff changing the way they work. Careful planning and an really adequate emphasis on education and change management are vital as is developing real clinician ownership of the project.

Seventh is it clear that all organisations need to develop organisational competence and teamwork with Health IT. I think the best way to do this is to choose one or two easily “doable” projects and get them done on time and within budget. Only once this capability is proven should an organisation try the larger and more complex implementations. Success, as they say, builds on success.

Eighth it is clear that when implementing systems in hospitals size really does matter. It is a relatively straightforward process to put basic systems in a 100 bed regional hospital in 3-6 months with very little difficulty. The 1000 bed tertiary teaching referral hospital is a horse of a totally different colour. The budget is likely to be in the millions, the complexity of what is needed much higher and the work practices more entrenched. All this means both risk and duration are much higher. Additionally these organisations cannot be fed a ‘one size fits all’ solution. The systems that are deployed must not only be flexible but be flexibly implemented in consultation with ALL involved.

Last it is vital to work hard to develop an open and frank relationship between the system vendor and the organisation which is implementing the new system. No contract will prevent a disaster but work on ensuring a constructive, frank and balanced relationship will make a huge difference.

If all this is taken into account – and experienced project managers are engaged and then supported a good outcome is more than possible. Other things that can increase the likelihood of success are:

1. Making sure a strong educational project that gets to all the hospital’s staff is conducted early to explain to everyone involved what is needed from them and how their lives will be easier once implementation is complete.

2. Preparedness on the part of both executive and technical management to seriously address issues raised by grass roots staff and to ensure there are real working processes to gather honest feedback before problems fester and then spin out of control.

3. Making sure that consultation is more than token. There is a tendency for project managers to exhibit a rigidity regarding goals and processes, that often means some involved get the feeling that their being consulted is little more than an unnecessary formality.

4. Being prepared, from time to time, to offer small incentives to reward success, and to acknowledge that change is never easy with some fun and interesting occasions, awards etc.

5. Working to identify the inevitable ‘organisational opinion leaders’ that exist in all large organisations outside the formal hierarchy and work very hard to have these people on side and supportive.

If you ignore any of these points you do so at your peril!

David.

ps. Please visit the Pulse + IT website – richest Health IT Content in OZ…well maybe other than my blog !

D.

6 comments:

Dr Ian Colclough said...

I agree that a great deal can be learned from a thoughtful review of Scott Silverstein’s site. There is much to be said in favour of routinely adopting the ‘review and learn’ cycle which underpins the philosophy of ‘constant improvement’. Although it doesn't work for the ostrich.

I would add two more reasons to your list:
1. Corporate memory is often lacking in many organisations where there is a high staff turnover.

2. A deep and perceptive understanding of the multiplicity of cultures prevailing throughout the health system and how to deal with them is an essential prerequisite to implementing the highly complex systems deployed in health.

Oliver Frank said...

David More said: "All over the country it seems we have Hospital IT projects that are struggling to actually get started (e.g. WA, SA and to some extent QLD)"

The South Australian CareConnect project (formerly Oacis) is not struggling to get started. It is well-established, is being used increasingly and seems to be one of the few large-scale implementations that is actually succeeding and whose users are asking for its functions to be expanded.

From a GP view, the main visible output is semi-automatically generated discharge summaries which are available by email as encrypted PDFs, and soon are to be delivered via State Health's pathology result delivery system (IMVS). We are pressuring IMVS to deliver via Argus and I think that we will get there eventually.

I believe that the CareConnect implementation leads the nation in terms of its range of functions and its ability to share information between the State's public hospitals.

Anonymous said...

Health is quite a hot topic this election. Last night the ABC’s ‘Difference of Opinion” focused on Australia’s health system. Michael Wooldridge made some confessions about his time as Health Minister! That’s two confessions in 24 hours I‘ve heard - must be an election in the wind.

The entire panel seemed pretty much in agreement that the focus should be first and foremost on the Primary Care sector, and on progressively rejigging the service delivery model as the starting point in order to ‘fix’ the system. That is not to say that hospitals did not receive plenty of comment. They did, for they too were deemed in need of attention, but not at the expense of the Primary Care sector where it was unanimously agreed far more can be achieved by focusing on what is needed to help keep people out of hospital.

This all leads to how we can better use information technology to make improvements in health service delivery. Kevin Rudd, John Howard, Peter Costello, and Tony Abbott should study today’s blog as should Premiers and Health Ministers everywhere and also all senior and middle level bureaucrats involved in making decisions around the use of IT in health.

Anonymous said...

The bit I found most telling was Dr Wooldridge's comments about the bureaucracy. While the software suppliers bear the brunt of the bureaucracy's ineptitude they aren't without blamel as all too often they overpromise on what they can safely deliver.

surturz said...

The reason is that Health IT is a developing industry, yet the state governments favour centralised procurement. Centralised procurement is fantastic for buying established products such as pencils, milk, computer hardware, etc where there is strong competition and the products are nearly identical.

For developing products like Health Software, it is better to have decentralised procurement, so that different approaches can be tried and compared. The systems can start tailored to individual departments in individual hospitals, and then borrow features from each other as they grow.

Decentralised procurement seems inefficient because there is a lot of duplication and "reinventing the wheel", but it is better than the government picking winners and finding themselves locked in to a single behemoth supplier.

Dr Ian Colclough said...

And also Surturz .... we should not overlook the fact that decentralised procurement not only stimulates much needed market competition, but also helps contribute to ensuring Australia retains a viable health IT community.

It is to be hoped the proposal that every hospital have its own Board fails to get up, although, concentrating all control centrally is also folly. Area or Regional Health Service Boards or Health Networks, with a full complement of hospitals, Primary Care, community health and other services, each under the direction of a Governing Board and with Community representation, is the probably the more practical structure under which to deliver fully integrated health services.

Such structures also provide some freedom of choice in the selection of IT, thus bringing a balanced amount of competition to the equation. Government has an important role to play in ensuring that different approaches can be tried and compared