JCPSLP Vol 20 No 2 July 2018

Entrepreneurship in speech-language pathology

Clinical judgement just as reliable as an explicit prioritisation tool A comparison of three prioritisation approaches for inpatient speech pathology Jo Brady and Katherine Harding

Speech-language pathologists (SLPs) often need to prioritise services when demand outstrips supply. This study compared three approaches to prioritisation. SLPs (n = 14) were randomised into 3 groups. Each group prioritised the same hypothetical caseload using either: (a) the original prioritisation tool emphasising diagnosis; (b) a redesigned prioritisation tool emphasising general patient factors and interventions provided; or (c) clinical judgement alone. Interrater agreement and differences in prioritisation patterns were compared. Agreement was fair to moderate among SLPs using the original tool (K = 0.39), and higher in the redesigned tool (K = 0.5) and clinical judgement groups (K = 0.48). There were no significant differences in prioritisation of specific patient groups. Findings do not support the assumption that prioritisation tools increase consistency of decision-making. The implicit clinical reasoning of clinicians with greater than 3 years’ experience may be an equally reliable method for allocating resources. S peech-language pathologists (SLPs) and health managers frequently make decisions about the allocation of limited resources. It is not always possible to provide every patient with optimum care at the ideal time, and SLPs are not alone in facing the challenge of balancing demand against available supply (Kreindler, 2010). Such decisions about the prioritisation of services may be implicit based on prior experience and clinical judgement, or explicit using prioritisation tools based on defined categories and protocols (Rice, 1998). It has been suggested that use of explicit prioritisation tools can reduce in-hospital delay times (Mosley, Morphet, Innes, & Braitberg, 2013), support efficient and effective services with limited resources (Gauthier, Straathof, & Wright, 2006), and increase fairness in the distribution of resources by increasing transparency of prioritisation criteria (MacCormick, Collecutt, & Parry, 2003). Prioritisation tools can also be used to help to translate organisational priorities for clinical service provision to frontline clinicians (Harding & Taylor, 2013).

Despite wide discussion in the allied health literature about the use of prioritisation systems there has been relatively little investigation into their effectiveness (Brown & Pirotta, 2011; Harding, Taylor, Leggat, & Shaw-Stuart, 2009; Porter & Jamieson, 2013). Studies that have been conducted suggest that there are a range of potential problems associated with their use, including a lack of reliability; limited evidence of effectiveness (Brown & Pirotta, 2011; Harding & Taylor, 2013); and the potential in some settings for low priority patients to never be seen (Raymond, Demers, & Feldman, 2016). Low levels of reliability are particularly problematic as reliability is an important prerequisite to establishing validity (Streiner & Norman, 2003). A SLP department providing acute and subacute services in a large metropolitan public health service in Melbourne, Australia, was experiencing many of these challenges with prioritising patients. The tool used for prioritising SLP services emphasised diagnosis as the basis for priority decisions, prioritising patients with dysphagia over other groups due to risks associated with aspiration pneumonia and malnutrition (Lorinczi, Denheyer, Pickard, Lee, & Mager, 2012). This approach, however, failed to recognise emerging evidence of adverse health outcomes associated with communication and cognitive deficits (Bartlett, Blais, Tamblyn, Clermont, & MacGibbon, 2008). The tool also had limitations in its application to all settings across the care continuum and its validity was untested. As a consequence, resources were invested in redesigning the prioritisation tool to improve consistency of priority decisions in both acute and subacute inpatient settings, with a more equitable approach to different diagnoses. The primary aim of this study was to compare the interrater reliability of priority decisions made by SLPs using the original tool and the redesigned tool (explicit methods), and also to compare the interreliability of both methods with a control group using clinical judgement alone (an implicit method). The secondary aim was to determine whether the removal of automatic priority for dysphagia in the redesigned tool would lead to a change in practice when prioritising these patients. Method Participants SLPs (n = 14) employed in either the acute or subacute services with greater than 3 years of clinical experience working with adult inpatients consented to participate in the

KEYWORDS ACUTE AND SUBACUTE INPATIENT

CLINICAL DEMAND PRIORITISATION TOOL SPEECH AND LANGUAGE SERVICES TRIAGE TOOL THIS ARTICLE HAS BEEN PEER- REVIEWED

Jo Brady (top) and Katherine Harding

57

JCPSLP Volume 20, Number 2 2018

www.speechpathologyaustralia.org.au