Social-emotional Under 4's Screening & Intervention S.U.S.I.

The purpose of the feasibility study is to evaluate the impact of specific mental health interventions for the children, parents and carers in three high risk groups of children under the age of 4 years.

Recruitment This research will take place in the Carelink, CAMHS Looked after Children's team.

A brief introduction to the research will be sent with relevant appointment information to the parents in all groups stating that they may be approached by a researcher at a particular time (e.g at a scheduled Contact, a CP conference, appointment with their keyworker). The letter will say that the researcher will be making contact to discuss details of the research at an agreed and convenient date for the parents. It will make clear that participation is voluntary. We have made the letter as brief and clear as possible because parents may have difficulty processing information like this at a stressful time.

If the birth parents agree to meet with a CAMHS research practitioner, it will give them an opportunity to ask any questions and to discuss the study in detail to facilitate informed consent. Also at the meeting we will set up a time for the first screening visit to take place once parent have confirmed their participation.

We will inform all keyworkers when a family gives informed consent to participate in the study.

For all the groups, screening will take place over two home visits (unless parents or carers particularly wish to meet in an alternative location) that will last approximately one hour each. The screening visits will take place in the child's current home setting.

Procedure:

In Group 1 birth parents will be invited to complete the Ages and Stages Questionnaire: Social Emotional (ASQ:SE) to gain their view on their child's needs.

The home visit screening will include the ASQ-SE interview with the child's foster or kinship carer (Group 1) or parent (Groups 2 and 3) followed by the Keys to Interactive Parenting Scale (KIPS) free play observation of the child and carer/parent. The KIPS observation will be videoed, if we gain specific consent to do so, or will be coded from the live observation if the parent or carer does not wish to be videoed. We will also administer the Mind-mindedness measure (MM) and a language scale. The DASS will also be completed with the birth parents in Groups 2 and 3.

Additional information may be gained regarding the child's health and development from the Child Health service, gathered in either the child statutory initial health assessment (IHA) for Group 1, or if a child has been referred for a health or developmental check for Groups 2 and 3.

Interventions:

The detailed information from the screening process will be used to inform specific strengths and needs in relation to both the individual child and the parents or caregivers which will then guide the intervention plan to be agreed with the parent or caregiver according to the cohort (Groups 1-3). The intervention will primarily aim to promote healthy child development, responsive parenting behaviours and positive cycles of carer-child relations.

Six month review screenings will take place with all families; the children and their carers who join the study and have completed the screening in the first 9 month period will have two review screenings and those joining after this time will have one review. The review screenings will follow the same structure as the initial screening and all the measures will be completed in order that outcomes for the child and child-carer relationship can be obtained.

In addition, information on the child's use of health and social care services will be collected from families, using the CA-SUS, at the 6 and 12-month follow-up points, to allow assessment of the cost of the screening and intervention and exploration of factors that influence the economic burden.

Evaluation:

We hypothesize that screening may pick up undetected or misunderstood social-emotional, developmental and attachment difficulties, and in a small number of cases specific mental health disorders in the children. We also expect to gain a greater understanding of the areas of difficulty in key aspects of the child's relationship with primary caregivers. By intervening early to specifically address the issues identified in the screening we hope the intervention will impact positively on both the quality of the carer-child relationship as well as the child's social-emotional development. In addition, we anticipate that the close interagency collaboration developed in the study will positively impact on the care planning and decision-making around the child taking place in a timely fashion.

Data already collected from the parent or the child will be retained in identifiable form and used in the research. This is important as it will give the research team further information about this high risk group and their vulnerable parents. This will be made clear in the consent procedure.

Study participants will be assigned an alphanumerical study identification number at the time of providing consent. The ID number will be used on study electronic databases, questionnaires and report forms. The study identification number will be recorded on the consent form (which will have the participant's full name). Participant contact details will be needed for the duration of the study and will be kept together with the paper study records in locked cabinets of locked offices of a secure NHS building.

Copies of written reports, correspondence and process recordings from the screening and intervention will be stored on the password protected 'u' drive of Trust computers.

A Research Consultant and Health Economics Consultant will coordinate the analysis of the data by research staff employed in the study. All information will be stored on a secure data base.

Differences on level of uptake of the screening, the proportion of families and carers who engage in treatments offered and the prevalence and types of difficulties in all three groups will be examined using Chi Square tests of association. Comparison of scores on the KIPS, ASQ:SE and Mind mindedness measures will be made between high risk groups before and after the intervention using t-tests and both univariate and multivariate analyses of variance. Comparisons of scores will be made between type of intervention offered using t-tests. Other outcomes, such as a comparison of the views of contact workers before and after the new system is implemented and decision timescales before and during the project will be made using the Chisquare tests of association and t-tests.

Qualitative feedback from birth parents, foster carers, kinship carers and professionals on the usefulness of the screening process and interventions will be analysed using information collected on feedback questionnaires designed for the study.

The economic evaluation will involve: 1) Calculation of the cost of all interventions provided as part of the proposed work; 2) Description of and calculation of the total cost of all other health and social care services used; and 3) Exploration of characteristics of the participants and their family that influence the economic burden.

Data on the use of all hospital and community health and social care services will be collected using the Child and Adolescent Service Use Schedule (CASUS), administered in interview with parents/carers at the 6 and 12month followups.

Information on the use of the intervention in each of the three groups will be collected directly from therapist records. The interventions under evaluation will be costed using a standard microcosting approach (Drummond et al, 2005) on the basis of the salary of the therapist plus appropriate overhead expenses (administrative, managerial and capital). Calculation of indirect time, including preparation and supervision, will be based on information provided by the therapists on the ratio of direct face to face contact to other activities Nationally applicable unit costs will be applied to all other services used (Curtis, 2012).

Service use and cost for each of the three populations will be reported descriptively (mean, standard deviation, median and range) to explore the cost burden of each. Prior to statistical analysis we will generate hypotheses about factors likely to influence service use and cost drawing on evidence from a review of the literature and expert opinion from the research team. Only a limited number of hypothesis driven predictors will be selected from the available baseline characteristics of study participants in order to avoid finding associations by chance. In line with previous research (Byford et al, 2001; Barrett et al, 2012), univariate associations between the specified factors and total health and social care costs will be examined using linear regression. Multiple regression will then be used to reduce the variable set to those factors independently associated with cost. This will initially include those variables found to have significant univariate associations and discarding those that cease to be important. Each variable that does not have a univariate association with costs will be added, one at a time, and retained if it adds significantly to the model or otherwise discarded. This will ensure that none of the excluded factors would add significantly to the model. We will use a significance level of p≤0.05 (5%). Standard ordinary least squares regression will be used on untransformed costs, despite the skewed distribution of the cost data, because this enables inferences to be made about the arithmetic mean (Barber \& Thompson, 1998). The results of the model will be checked through comparison with a nonparametric bootstrap regression, to assess the robustness of the confidence intervals and p values to non-normality of the cost distribution, and with a generalised linear model, with a non-normal gamma distribution assumed for costs (Blough et al, 1999).