Indicators of acute deterioration in adult patients nursed in acute wards: a factorial survey

Abstract

The primary objective of the study was to determine which professional, situational and patient characteristics predict nurses' judgements of patient acuity and likelihood of referral for further review. A secondary aim was to test the feasibility of the factorial survey method in an acute area. There is increasing recognition that indicators of deterioration in acutely unwell adults are being missed and referrals delayed. The reasons for this are unclear and require exploration. Assessing nurses' clinical decision-making or judgements in a 'real-world' situation is problematic. The study used a factorial survey design where participants completed randomly generated paper-based vignettes on one occasion. The dependent variables were assessment of patient acuity and likelihood of referral. Independent variables consisted of a number of patient characteristics, i.e. heart rate, blood pressure, nurse characteristics, i.e. clinical experience, and situational characteristics i.e. staffing. Participants were registered nurses working in acute areas excluding intensive care and theatre. Ninety-nine participants responded resulting in 1940 completed vignettes. An early warning score was the single most significant predictor of referral behaviour accounting for 9.6% of the variance. When this was not included in the vignette, nurses used physiological characteristics e.g. respiratory rate, urine output, neurological status. These explained 12% of the variance in the model predicting assessment of patient acuity and 9.4% or the variance predicting likelihood of referral. When given a series of vignettes, nurses appear to use appropriate physiological parameters to make decisions about patient acuity and need for referral. Our results support the use of early warning scoring systems. Education and professional development should focus more on developing and maximising clinical experience and expertise rather than knowledge acquisition alone. A factorial survey method is feasible to explore decision-making in this area. This study has several implications for practice. The emergence of an early warning scoring system as a significant individual predictor supports the use of such systems. However, the small amount of explained variance suggests that there are other influences on nurses' assessment of patient acuity and referral decisions that were not measured by the factorial survey approach. Educational provision might focus not just on knowledge acquisition but include educational delivery methods that incorporate or mimic real-ward settings.