Behavioural interventions in the treatment of drooling for children with cerebral palsy

Abstract

Anterior drooling is common in children with cerebral palsy (CP) and other neurodisabilities (Reid et al., 2010), and has a significant impact on children’s physical and psychosocial health (Chang et al., 2012). Causative factors for drooling in children with CP include oromotor dysfunction (Hockstein et al., 2004) and impairment of swallowing (Erasmus et al., 2009; Senner et al., 2004; Tahmassebi & Curzon, 2003). Behavioural intervention is a conservative non-invasive treatment that is frequently recommended before invasive approaches and routinely practiced by allied health professionals (Meningaud et al., 2006; Van der Burg et al., 2007a; Walshe et al., 2012). Previous experimental research has shown behavioural intervention to have positive effects on drooling in children with CP, but studies are few and the collective evidence is weak (Van der Burg et al., 2007b; Walshe et al., 2012). Given the emphasis on using interventions that are feasible and supported by high quality research evidence (Dollaghan, 2004; Hoffman, 2010; Sackett, 1997); further research on the effectiveness of behavioural intervention for drooling is needed to optimise the outcomes of children with neurodisability, including CP. The purpose of this thesis was to contribute to the evidence about the treatment of drooling in CP through the conduct of three studies. The first study was a registered systematic review, undertaken to investigate and synthesise the effects of behavioural interventions on Body Functions & Structures (BFS) level and other outcomes for children with CP and neurodisability. A detailed search in eight databases sought studies that (1) included participants 0 -18 years with neurodisability and drooling; (2) provided behavioural interventions targeting drooling or a drooling-related behaviour; and (3) used experimental designs. Two xix reviewers extracted data from full-text papers independently. Results were tabulated for comparison. Eight studies were found that described seven behavioural interventions, including reinforcement and prompting, that mostly targeted BFS level outcomes, but the risk of bias was high in all studies. In the context of low-level research evidence to support the use of behavioural interventions in practice, a cross-sectional online survey exploring Australian speechlanguage pathologists’ practice decisions was conducted. Descriptive data were analysed using non-parametric statistics. Informal assessments were preferred over the available validated measures. Several behavioural intervention methods were selected as treatments of choice, mainly due to client-related factors. Australian speech-language pathologists perceived drooling in children with CP as a complex practice area, in which there were several reported barriers to evidence-based practice. The third study used a concurrent multiple baseline single case experimental design to explore the therapeutic effects of a novel intensive behavioural intervention, the Lee Silverman Voice Treatment LOUD (LSVT LOUD®), on drooling severity, swallowing and speech with five participants with CP who drooled (n=5). This study was designed and implemented using guidelines from the Risk of Bias in n-of 1 trials (RoBiNT) Scale (Tate, 2015). Strategies to optimise internal validity included: randomisation of phase onset, using an independent data collector and blinding of all the outcome assessors involved. A systematic visual analysis was supplemented with statistical analysis, where appropriate, to analyse the data. The results were mixed across participants, with each of the participants showing improvement on at least one outcome. Immediately following treatment, small but positive effects on drooling severity occurred for three participants (P1, P3, P4) with P4 having a statistically xx significant reduction in drooling. There were positive effects on swallowing and feeding competency in three participants (P1, P2, P4) including a large effect for P1. Medium effects on sentence intelligibility (P2 and P3) and to a lesser degree, conversational intelligibility, were found. For P1, P2, P4, and P5, small negative and mixed effects were also found across outcomes. The topic of anterior drooling is substantially under-researched in children with CP. There are significant gaps in research evidence on behavioural intervention effectiveness but also in the integration of evidence-based assessments into routine clinical care. Evidence-based practice in the field could be advanced by research to explore the lived experience of individuals with CP who drool; further examine the covert non-visible causal mechanisms of drooling in children with CP; and by undertaking further experimental behavioural intervention-based research on treatment effectiveness, informed by future perspectives.