Volume 199 - Issue 8
Patient safety in hospitals — can we measure it?
Author: Neville Board
Med J Aust 2013; 199 (8): 521-522. || doi: 10.5694/mja13.10626
Published online: 21 October 2013
Published online: 21 October 2013
Collection of data on adverse events is challenging, and part of the challenge is ensuring that the data are put to good use.
Patient safety measurement tools all have limitations, so monitoring a range of data sources is recommended
Patient safety has been described as “the reduction of risk of unnecessary harm associated with healthcare to an acceptable minimum”.1 It is the hardest dimension of quality to measure. Three landmark reports — from the Harvard Medical Practice Study,2 the US Institute of Medicine3 and the Quality in Australian Health Care Study4 — quantified hospital adverse event rates derived from different methods. Safety researchers remain concerned that health care may not be getting safer5 and that robust measurement of safety in health care is still some way off.6
Understanding the quality of health care in hospitals means knowing how appropriate, effective and safe that care is. Clinicians, hospital boards and chief executive officers want to know complication rates and profiles in their hospitals, but few managers can put their hands on their hearts and say they have systems in place for routine and accurate measurement of patient safety. The difficulties lie in the capture, classification and integration of such information.
In this issue of the Journal, Trentino and colleagues contribute new evidence to the field.7 In a study of five Western Australian hospitals in 2010 and 2011, they report a correlation between complications experienced by hospital inpatients (detected through administrative datasets) and length of stay in hospital. By extrapolation, decreased incidence of complications could result in shortened lengths of stay, and therefore improve access to hospital for elective patients and those admitted through emergency departments.
No clinician will be surprised by the correlation. However, some will worry on several counts: different patient populations are not equally vulnerable to complications, the measurements rely on documentation of adverse events in clinical notes, and coding resources vary between hospitals.
In correlating incidence of complications with length of stay, the authors sought to quantify patient safety. However, hospital adverse event rates are generally characterised by low numbers of many different types of harm, in categories that few would collapse into a single, countable pile.
What are the relative values of falls in older patients, obstetric trauma and central line infections? What is the meaning of a low rate of pressure injuries in a hospital where few patients are diabetic, old and bedridden? We need to remember that measurement of safety is not amenable to the epidemiological model:
The majority of events leading to both serious and minor outcomes occur too infrequently to be characterised or tracked at a hospital level and require large scale collections of events.8
The lesson I take from the study by Trentino et al is not to take useful, but inaccurate, counting systems to create more punitive reporting. Rather, the study reaffirms the need to ensure sustainable systems of safety measurement are in place to help clinicians and managers understand the breadth and types of harm in an institution, drive improvement and prioritise safety programs.
Hospitals have invested heavily in incident reporting systems since the landmark safety studies.2-4 Incident reporting systems are workflow systems that support logging, grading and reviewing of adverse events data and responding to the findings. They work best when adverse events are documented objectively by clinicians, and when responses are developed thoughtfully and communicated widely in a generative hospital culture.
Incident reporting systems, along with administrative datasets (as analysed by Trentino et al7) and chart reviews, provide viable starting points for hospital safety initiatives. However, all have limitations as measurement tools. Adverse event rates generated from administrative data and voluntary reporting systems underenumerate events; chart reviews show greater sensitivity, but are costly and limited to detecting documented adverse events.9
So how can we monitor and prevent complications? One suggestion is that hospitals adopt a balanced measurement framework, in which multiple data sources (including patient experience and staff attitude surveys) would be used to drive learning and action for safety improvement.10
Recently, the UK Health Foundation published a multi-faceted approach to patient safety measurement.11 The report reminds us that there are no simple answers to patient safety measurement, and cites a collective mindfulness process, termed “commitment to resilience”, in which a generative organisation monitors and analyses adverse events to strengthen resilience to harm and harm-inducing practices. The process involves rapidly communicating the results of reviews and using simulation to seek out potential harm.
Commitment to resilience is manifested by:
Detection and containment of errors at early stages through anticipation but intelligent monitoring, improvisation and recovery. Responding to unexpected disturbances to gain rapid real-time learning and feedback on the effectiveness of their responses.Rapid and accurate communication processes along with experience in varied operational scenarios.Training through simulated scenarios that allow people to practice recovery, improvisation and response to variation.12
Australian hospitals are accredited against National Safety and Quality Health Service Standards.13 The Standards specify that health services have “Systems ... in place to analyse and report on incidents” and provide “Feedback on the analysis of reported incidents ... to the workforce”. Administrative datasets, infection surveillance systems, patient experience surveys, and reported incidents, complaints and claims can all inform hospitals of where risks can become harm. Hospital boards and clinicians are encouraged to regularly monitor a range of signals of adverse events and near misses to review hospital practice and improve patient safety. The report–review–feedback loop is at the core of an active safety culture.
Competing interests
I work full-time at the Australian Commission on Safety and Quality in Health Care. This article reflects my analysis, but has been approved for publication in the Medical Journal of Australia by the Commission Executive.
References
- Runciman W, Hibbert P, Thomson R, et al. Towards an international classification for patient safety: key concepts and terms. Int J Qual Health Care 2009; 21: 18-26. 0_i1139910
- Brennan TA, Leape LL, Laird NM, et al. Incidence of adverse events and negligence in hospitalized patients — results of the Harvard Medical Practice Study I. N Engl J Med 1991; 324: 370-376. 0_i1139912
- Kohn LT, Corrigan JM, Donaldson MS. To err is human: building a safer health system. Washington, DC: National Academy Press, 2000. 0_i1139914
- Wilson RM, Runciman WB, Gibberd RW, et al. The Quality in Australian Health Care Study. Med J Aust 1995; 163: 458-471. 0_i1139916
- Shojania KG, Thomas EJ. Trends in adverse events over time: why are we not improving? BMJ Qual Saf 2013; 22: 273-277. 0_i1139918
- Pham JC, Frick KD, Pronovost PJ. Why don’t we know whether care is safe? Am J Med Qual 2013; Mar 24 [Epub ahead of print]. 0_i1139920
- Trentino KM, Swain SG, Burrows SA, et al. Measuring the incidence of hospital-acquired complications and their effect on length of stay using CHADx. Med J Aust 2013; 199: 543-547. 0_i1139922
- Runciman WB, Edmonds MJ, Pradhan M. Setting priorities for patient safety. Qual Saf Health Care 2002; 11: 224-229. 0_i1139924
- Classen DC, Resar R, Griffin F, et al. “Global trigger tool” shows that adverse events in hospitals may be ten times greater than previously measured. Health Aff (Millwood) 2011; 30: 581-589. 0_i1139926
- Wakefield JG, Jorm CM. Patient safety — a balanced measurement framework. Aust Health Rev 2009; 33: 382-389. 0_i1139929
- Vincent C, Burnett S, Carthey J. The measurement and monitoring of safety. London: Health Foundation, 2013. 0_i1139931
- Weick KE, Sutcliff KM. Managing the unexpected: assuring high performance in an age of complexity. San Francisco: Jossey-Bass, 2001. 0_i1139933
- Australian Commission on Safety and Quality in Health Care. National Safety and Quality Health Service Standards. Sydney: ACSQHC, 2011. 0_i1139935
Provenance: Commissioned; externally peer reviewed.