HomeJuly 2015The limitations of regulation

The limitations of regulation

The goal of medicine is to alleviate suffering and enhance longevity. A brutal paradox arises when medicine does the opposite. Although there is limited evidence on the scale of harm inflicted by medical error, available data suggest this is a vast problem, write Drs. Padhraig Ryan and David Vaughan.

Drs. Padhraig Ryan and David Vaughan
Drs. Padhraig Ryan and David Vaughan

In the United States, the number of deaths due to medical errors in hospitals may be in the region of 90,000 per year. As a frame of reference, imagine a Boeing 747 crashing and killing everyone on board, each day of the year. Efforts are underway to quantify the scale of harm in Irish hospitals.

There has been systematic effort to enhance clinical quality in Ireland, most notably in the HSE Clinical Care programmes. Further advancement of the quality agenda should draw on international experience, to avoid repeating the mistakes of others and to learn from best practice. This article explores lessons from divergent approaches to quality improvement in Canada and the United States.

Efforts are underway to quantify the scale of harm in Irish hospitals.

Regulatory intervention

One approach to care redesign relies on regulatory intervention. The Canadian province of Ontario regulated to enforce the adoption of a promising innovation, the surgical safety checklist, in 2010. When applied properly in other settings, this safety checklist appears to reduce mortality by 50% and the rate of major adverse events by 20%. Given these significant benefits, shouldn’t government regulate to enforce adoption?

The experience of Ontario indicates that more sophisticated tactics are needed. At the three month post-intervention point there was no significant change in surgical mortality or complications. It is possible that the concept of a surgical safety checklist is redundant in the Ontario context, due to a high pre-intervention standard of quality. But the checklist appears effective in a number of other high-income countries, and there remains scope for quality improvement in Canada, therefore this seems unlikely.

Two other factors may explain the disappointing results. First, the checklist may require further modification for the Canadian system. Clinicians frequently modify healthcare innovations to meet local needs, and this may enhance clinicians’ sense of ownership of an innovation. But in Ontario, around 90% of hospitals reported using an unmodified version of the checklist, suggesting this did not occur.

Evidence suggests that punitive interventions such as regulation are not an optimal mechanism to instil behaviour change.

The second possibility is that clinicians may not have implemented the checklist reliably. Indeed, the lack of checklist modification may reflect limited commitment from clinicians. Evidence suggests that punitive interventions such as regulation are not an optimal mechanism to instil behaviour change. Effective programmes, by contrast, have adopted more sophisticated interventions.

Towards improvement

The predominant reason for quality defects is that the design and management of health systems are not aligned with optimal care. For the most part, clinicians are highly trained and strive to provide excellent care, but the organisational context can obstruct these efforts.

Simplistic promotion of a tool such as a checklist fails to recognise the complexity of successful improvement strategies. In Michigan, United States, clinicians significantly reduced the rate of central line infections when a checklist formed part of a more complex sociological intervention. The intervention nurtured commitment to shared goals, teamwork, and interdisciplinary communication.

Based on the success of Michigan and other programmes, we can identify key elements that should underpin improvement strategies. The HSE’s Clinical Care programmes offer good examples of some of these elements, but there is scope for improvement in others.

  1. Collect data to measure improvement from baseline, and to compare performance against other clinical teams.
  2. Visualise data for easy comprehension (e.g. a performance dashboard).
  3. Use narrative, as well as quantitative data, to convey the need for change. The input of Sorrel King, whose 18 month old daughter Josie died following a series of avoidable medical errors, is perceived as a key factor in Michigan’s success.
  4. Governance can enhance credibility and encourage clinician engagement.
  5. Tailor innovations for local contexts. For example, team members in a small hospital with one operating room may not need to introduce themselves before each surgery. Furthermore, tailoring enables clinicians to own and believe in the intervention.
  6. Minimise the burden of implementation and monitoring (for example, use electronic medical record data to monitor performance rather than bespoke data collection).
  7. Share practical lessons and motivation for improvement across clinical teams and organisations.
Conclusion

Complex problems in health care are unlikely to be solved by simple interventions, and it is important to temper such expectations. Ineffective change strategies raise and dash public and political expectations, waste resources, and lead to clinician scepticism.

Ireland possesses a clinically sophisticated workforce, and the HSE Clinical Care programmes leverage this in pursuit of improved services. Future interventions should build on this promising start by learning carefully from the evidence base. By examining international evidence on the impact of multifaceted change initiatives, we can improve Irish health care for the benefit of clinicians and patients.

Dr. Padhraig Ryan, Health Research Board Interdisciplinary Capacity Enhancement Fellowship Programme, Trinity College Dublin

Dr. David Vaughan, Executive Director of Quality and Safety, Hamad Medical Corporation, Qatar

References: on request

Thanks to Dr. James O’Mahony for helpful comments.