For almost a decade researchers have investigated the relationship between the quality and/or quantity of nursing care and adverse patient outcomes (Aiken, Clarke, Sloane, Sochalski, & Silber, 2002; Kane, Shamliyan, Mueller, Duval, & Wilt, 2007; Lankshear, Sheldon, & Maynard, 2005; Needleman, Buerhaus, Mattke, Stewart, & Zelevinsky, 2002). Interest in quantifying this relationship emerged in response to a trend of reducing numbers of Registered Nurses per patient in an effort to curb rising health care costs (Aiken, Clarke, & Sloane, 2000; McCloskey & Diers, 2005; Shindul- Rothschild, Berry, & Long-Middleton, 1996). Some of the challenges in this field of research are yet to be fully resolved, for example, defining and accurately measuring nursing care (Reinier et al., 2005) and deciding upon suitable and measurable indicators of quality nursing care (Doran et al., 2006; Jennings, Staggers, & Brosch, 1999; Spetz, Donaldson, Aydin, & Brown, 2008). In 2002, a team of researchers constructed fourteen clinical indicators (listed in Table 1) that could be both conceptually linked to nursing care and feasibly measured using routinely collected hospital discharge information from administrative data sets (Needleman et al., 2002).
Following on from the work of Aiken, Smith, and Lake (1994) and Needleman, Buerhaus, Mattke, Stewart, and Zelevinsky (2001) there has been growth in the literature in the field. A systematic review of studies relating to hospital nurse staffing and patient outcomes was commissioned by the Agency for Healthcare Research and Quality, published in 2007 (Kane et al., 2007). Of the 94 eligible studies included in the metaanalysis, the pooled results showed that: every additional registered nurse full-time equivalent (FTE) per patient day was associated with a 'relative risk reduction in hospital-related mortality by 9% in intensive care units and 16% in surgical patients' (Kane et al., 2007, p. 2); the death rate decreased 1.98% for each additional nurse hour per patient day (95% CI 0.96-3.0%; Kane et al., 2007); and there was a significant negative correlation between the percentage of nurses with Bachelor of Nursing Science degrees and the incidence of death related to health care (r = -0.46, p = 0.02; Kane et al., 2007). The systematic review concluded that increased nurse staffing in hospitals is associated with better patient outcomes, but acknowledged that the relationship does not necessarily imply causation (Kane et al., 2007).
Despite the concept of nursing sensitive outcomes (NSO) being embraced internationally (International Council of Nurses, 2007) and a growing research focus in Australia there is limited evidence to support the conceptual validity of Needleman's 14 NSOs for practising nurses in Australia (Duffield et al., 2007; Twigg, Duffield, Bremner, Rapley, & Finn, 2011). There is a lack of research elucidating practising nurse's perceptions of how NSO have been defined and how NSO may be affected by measures of nursing care. Moreover we were interested in whether characteristics of the nurse, such as age, education level or job description, influence a registered nurse's perception of the impact of nursing care on NSOs. Aiken et al. (2002), argue that nurses are at the front line of patient care delivery and as such are well placed to influence patient outcomes, but is this potentially compromised if nurses do not recognise the effect of their care on patient outcomes? In addition, nurses at the 'front line' of patient care may have novel ideas about additional indicators that might reflect the quality of nursing care delivered.
The aim of our study was to explore registered nurses' perceptions of the impact of nursing care on 13 of the 14 clinical outcomes defined by Needleman et al. (2001). We targeted Registered Nurses attending a nursing practice conference where the keynote speaker was an international researcher into nurse sensitive outcomes. …