Clinical and cost effectiveness of a parent mediated intervention to reduce challenging behaviour in pre-schoolers with moderate to severe learning disability: a randomised controlled trial

DESIGN: Two arm multicentre pragmatic single blind randomised controlled trial of level 4 Stepping Stones Triple P (SSTP) and TAU versus TAU alone. SETTING: NHS, social care and third sector services in England TARGET POPULATION: Children aged 30 to 59 months with moderate to severe intellectual disability (ID) and challenging behaviour. INCLUSION CRITERIA: Parents aged 18 years and over; child aged 30-59 months with moderate to severe ID (ABAS general adaptive composite 20-69); parent/service concerns about challenging behaviour over 6 but no less than 2 months; informed consent by parent EXCLUSION CRITERIA: Child has mild, profound or no LD; parent/carer has insufficient English language to complete study questionnaires; another sibling is taking part in the study. HEALTH TECHNOLOGY BEING ASSESSED: Weekly 6 x 2.5 hours group sessions followed by 3 x half hour individual contacts of level 4 SSTP delivered by trained practitioners in addition to TAU. TREATMENT AS USUAL: Any type of care currently available to families and their children with challenging behaviour. The control group will also receive a list of local and national resources and the Contact a Family guide to challenging behaviour. MEASUREMENT OF COST & OUTCOMES: The primary outcome is parent reported reduction in challenging behaviour at 12 months post randomisation using the parent completed preschool Child Behaviour Checklist (CBCL). Secondary outcomes measured at baseline, 4 and 12 months are: Direct parent-child observations (Revised Family Observation Schedule, FOS-RIII); caregiver reported child behaviour (Child Behaviour Checklist Caregiver-Teacher Report Forms; C-TRF); parent psychiatric morbidity (General Health Questionnaire, GHQ-12); parenting activities (Questionnaire on Resources and Stress, QRS-F); frequency of behaviour severity during care-giving tasks (Caregiving Problem Checklist-Difficult Child Behaviour); satisfaction and efficacy as parent (Parenting Sense of Competence Scale, PSOC), health related quality of life (Pediatric Quality of Life, PedsQL), health and social care service use (study specific Child and Adolescent Service Use Schedule, CA-SUS); Parent health related quality of life (EQ-5D). Other: child level of disability (Mullen Scales of Early Development, if not previously assessed); parent intervention acceptability (Client Satisfaction Questionnaire). Sociodemographic and clinical information will be captured by the Case Report Form. The measures are appropriate to this type of study and are meaningful to parents and clinicians. SAMPLE SIZE, RECRUITMENT and INTERNAL PILOT A sample of 258 children is required to detect a low to moderate (standardised) effect size of 0.40 for the primary outcome CBCL at 12 months, at 5% significance level with 90% power; this is equivalent to detecting a clinically important difference of 8 points, (standard deviation=20). We assume a correlation of 0.5 between baseline and follow-up measurements, 10% attrition and clustering in the SSTP arm only (ICC=0.05 and group size=7). The sample size is based on 3:2 randomisation which allows for clustering and is relatively easy to implement. A 7 month internal pilot will be carried out prior to embarking on the full trial in order to test the study processes, acceptability of the intervention and flow of participants into the study.