IT Benefits in Healthcare Performance and Safety

Introduction

The earliest work on accurately monitoring the success and failure of healthcare systems is attributed to Florence Nightingale and Lord Moynighan in the UK and Ernest Codman in the US. The pioneering of the professional role of nurses by Florence Nightingale in the 19^th Century is well known, but her extension of healthcare analysis into the audit of hospitals is less well publicised. She worked with the British Home Secretary Lord Moynighan in producing these early audits, which, in the absence of computing power, were time consuming and extraordinary for their time (Chambler & Emery, 1997; Nuttall, 1983). Later, in the early 20^th Century, a New England physician, Ernest Avery Codman, attempted a similar analysis by defining the “outputs” of a hospital and attempting to relate these to different practices. Again all of Codman’s analyses were undertaken without the assistance of computers and represent a large number of calculations that can currently be undertaken by modern computers in fractions of a second. However his work was uniformly unpopular with his colleagues and he was ostracised by the local medical community (Codman, 1914; Kaska & Weinstein, 1998). The importance of these early pioneers of audit and performance monitoring was that they assumed that individual and system performance could be measured and then optimized (Spiegelhalter, 1999). The tacit assumption that ensured their lasting unpopularity with the medical profession was that healthcare performance was not already perfect; or that system or individual error was occurring in the hospitals and health systems that they examined. Largely due to the work of these pioneers and the application of sophisticated computing programs and devices we now know much more about the frequency and cost of these adverse events in healthcare.

Errors in medicine are expensive, not only in terms of human life and the suffering they impose on patients, relatives and carers but also the financial burden that they impose on healthcare systems in the developed world (Kohn, Corrigan, & Donaldson, 1999; Lesar, Lomaestro, & Pohl, 1997; Michel, Quenon, de Sarasqueta, & Scemama, 2004; Vincent, Neale, & Woloshynowych, 2001; Wilson et al., 1995; Woods, 2000). Information about these episodes in complex healthcare delivery has only become widely available through the use of large computing and IT systems designed to collect clinical information. The most recent estimates indicate that this cost was $17-29 billion, in the US alone, in 2000 and may have contributed to between 48,000-98,000 deaths in the same period (Fenn, 2002; Kohn et al., 1999). Despite this human and financial cost, rates of adverse events do not appear to be falling, certainly in Australia and the US there is little evidence that adverse event rates are decreasing (Ehsani, Jackson, & Duckett, 2006; Wilson & Van Der Weyden, 2005).

Studies of adverse events have defined generic failings that have contributed to these identified errors (Thomas et al., 2000; Thomas et al., 2000; Wilson, Harrison, Gibberd, & Hamilton, 1999). Incident reporting in clinical medicine improves patient safety by reducing future errors. This occurs as incident reports lead to the identification of underlying generic and systemic errors and measures are put in place to prevent their recurrence (Barach & Small, 2000; Bolsin, Faunce, & Colson, 2005; Kraman & Hamm, 1999; Wolff, 1996; Wolff, Bourke, Campbell, & Leembruggen, 2001).