Clinical Informatics

Information System Failures and the NHS


  1. Introduction and Context of Information Systems Failure at large and in the NHS

  2. The Definition of Information System Failure

  3. The causes of Information System Failure

    1. Pre-implementation
    2. During Implementation
    3. Post-implementation

  4. Managing Individual and Organisational Change

  5. Conclusions

  6. References

The causes of Information System Failure

Multiple causes have been cited in the literature. They can be viewed as coming from the aspects of Project Management, Engineering Principles, Organisational Change and Change Management.

Specific issues that have seemed to be recurring themes are accountability, professionalism, formal training, research, good communications, leadership, risk management and ownership.

These factors may have a prominent role before, during or after the implementation of the Information Systems and will be discussed in turn.

The Challenges in Managing Change in the NHS

In addition, one must always remember that the NHS is a unique organisational entity. It is massive in terms of size, centrally funded by public funds and yet run locally to meet local needs and priorities.

It is multi-tiered and consists of many different groups of professional and non-professional workers. Despite this plurality, all come together and only exist because of a common goal.

That common goal is also unique. It is not profit or fame but equal distribution of common wealth in the pursuit of the best patient care possible with the available resources. As such, the work is highly unpredictable and improvisation the norm rather than the exception.

The different professional groups are also unique, in the sense that they are fully accountable to their peers and are fully regulated by their respective governing professional bodies like the General Medical Council or the Nursing Council.

There is high autonomy among these groups, whose only focus is effective and efficient patient care.

Last but not least, the organisation itself is a unique entity. It employs the second largest public workforce in the world after the Chinese army, but was created based on the idea of equitable distribution of wealth after the sufferings experienced by all classes of the society during and immediately after the Second World War.

These factors are in stark contrast with the hierarchical top down world that exists as the norm in business companies, where their sole purpose is to make profit and their creation in an environment dictated by the concept of survival of the fittest (i.e. "free market").

The political structure of the NHS is also very unique as the clinical members are held accountable in their clinical areas by their peers, and not by a central point of authority. Yet, their actions are guided in totality by one principle:

Hence, the explanation for why some managers and politicians may view the NHS as "unmanageable", "messy", "unorganised"; has "rebellious and recalcitrant clinical staff" and is "inefficient" or "profit losing" and "economically non-viable" when viewed from the traditional business or political model point of view.

Essentially, the NHS staff, in the original model as proposed by Aneurin Bevan, has only one duty:


Pre-implementation factors

Lack of research, risk management and long-term commitment:

Information Systems implementation are getting more complex and immense in terms of their costs and required functions. It is no longer sufficient to assume that one can competently manage the project without prior research, formal training or the necessary expertise.

This would apply equally to the management, the Information Technology staff and the leadership as well.

Project risks, including budget over runs and delays, must be fully researched and equally distributed and managed prior to the initiation of the project.

To facilitate the above, the management and their political masters must be prepared for a more realistic time-frame for the delivery of the Information Systems and willing to be part of long-term commitment to the project.

The "acceptability" of failures:

Despite the huge amounts of money involved, there is no obvious traceable accountability in the project management process in the past.

Even if the project fails, the management teams and vendors seem to only "melt away" in the background and re-group again for the next project.

The questions of "who", "how" and "why" are seldom if ever asked, let alone answered.

For example, one company have won 60% of the government contracts in the UK despite being repeatedly involved in high profile failures. The only "loser" in these failures appears to be the funding party (i.e. the public tax-payer).

It is therefore staggering to read in the British Computing Society and The Royal Academy of Engineers report that they do not recommend regulation of software engineers and of the Information Technology industry despite acknowledging that many whom are involved in "high consequence" systems are not professionally accredited, nor need to be, in order to practise.

Neither are they required to be formally trained in view of the high demand for their services at present (9). Failures seem to be "acceptable" or even "expected", in spite of the huge financial and opportunity costs borne by each and every failed project.

Lack of user buy-in and ownership:

In addition to making a case for change, users must also be convinced that the new system would lead to better work practices and better patient care in the unique environment of the NHS.

Clinicians are on the whole very pragmatic people, and the above factors would be much more valued than the inherent "convenience" of using an Information Technology system.

Too often, assumptions are made on behalf of the end users without consulting them. To quote Ash et al, "we often blame the users for not embracing new systems; yet a system may embody perspectives that may not be meaningful to or appropriate for their intended users" (10).

An example in the real world would be the implementation of Computerised Physician Order Entry Systems.

Although these systems benefit the management by providing them with a rich source of coded data and enable the close scrutiny of physician prescribing patterns, the extra work required with its unintended frustrations from systems slowdowns and crashes, does not add any extra benefit to the physicianÂ’s practise.

The much publicise reduction of adverse drugs reactions does not hold much water in reality on close study as there already exist ward pharmacists, pharmacy-based pharmacists and discharge nurses, who will check each and every prescription along the patient care pathway in the current system.

Although this may seem like an inefficient process at first glance, it has an inherent flexibility in prescribing for special circumstances and has the advantage of multiple layers of inspection by clinical staff involved in the direct care of the specific patient (The "Swiss-cheese" safety model). The staff in this chain of care are able to say:

"We know who this patient is"

Something that not many CPOE systems can claim to do.

To-date, there are no head-to-head randomised comparative trials in this area so the comparative benefits are only presumed, not tested.

However, in a poorly staffed ward or hospital, it would not be inconceivable that any additional support from the Computerised Physician Order Entry System may reduce errors.

Whether this is cost-effective in the context of good overall clinical care is unclear, as "good clinical care" is more qualitative than quantitative and thus, difficult to define and study.

Work satisfaction derived from human interactions must also be taken into account in any analysis of "good clinical care", as this would be necessary for sustained high performance among clinical staff.

Inappropriate use of technology:

Technology is a tool and nothing more. Information Technology cannot exist and solve problems on its own, nor can it increase productivity or profit without the right work processes already in place and with adequately trained staff whom are ready and able to use it appropriately.

Worst still, if a tool is "hijacked" and manipulated to achieve a secondary goal, it would be very likely then that it would fail to achieve its original aims.

The "Choose and Book" initiative is a good example. The original project was designed to facilitate patient referral and booking into local outpatients clinics by General Practitioners.

However, once the pilots started showing promise, the focus was changed to meet the political agenda of "Patient's Choice" and "Payment by Result".

Most NHS patient surveyed have not preferred choice over a good local service, where they can be visited by their friends and family, and General Practitioners do not see additional benefit in patient care nor the logic in referring patients to Consultants whom they have not established a professional relationship with.

There is also the additional time-consuming work to explain the benefits of each option over the others, when one may not have all the necessary information to do so in the first place, especially when the hospital is hundreds of miles away or is a private entity.


This website is certified by Health On the Net Foundation. Click to verify. This site complies with the HONcode standard for trustworthy health information:
verify here.


arrow left arrow right

Page Updated: 17 May, 2017

About Us | Site Map | Privacy Policy | Contact Us | Disclaimer | Acknowledgement | Top of Page |