Recording Disabled Children Within Scotland's Child Protection System: A Test Collection

Abstract

Disabled children have been found to be at three to four times greater risk of harm and abuse than their non-disabled peers (Jones et al. 2012; Sullivan and Knutson 2000). Applying this to the 9% of Scotland's children who are reported as disabled (Scotland's Census 2022), disability among Scotland's children in need of care and protection would be expected to be in the range of 27%–36%. This increased level of risk is not, however, reflected in Scotland's children's social work statistics. Just 5% of children on the child protection register and 10% of ‘looked after’ children were recorded as having a disability (Scottish Government 2025).

While it is possible that the harm and abuse of disabled children are being missed by services, the more likely scenario is that disabled children at risk of or experiencing harm and abuse are being identified but are not consistently being recorded as disabled in children's social work management information systems. Disability is a contested term and consequently a complex area of measurement (Franklin et al. 2020). There are many reasons for an under-recording of disability, including differing definitions of disability, variable practitioner skill and confidence in identifying and reporting disability (particularly among younger children) and waiting lists for specialist health and disability assessments (Cappa et al. 2015; Loeb et al. 2018; McTier 2024). However, without improvements to the collection of statistics, the number and needs of disabled children will only be partially visible. In response, this article reports on a data improvement project that has developed and tested a revised set of disability indicators in one part of Scotland's child protection system, specifically joint investigative interviews using the Scottish Child Interview Model.

A joint investigative interview is a forensic interview, conducted jointly by a specially trained police officer and social worker, with a child who may have been a victim or witness of a crime or who may be at risk of significant harm. The interview forms part of a multi-agency child protection investigation.

The Scottish Child Interview Model, which began delivery in 2020, is a new approach to joint investigative interviewing which is trauma-informed, focused on the child's needs, and seeks to achieve best evidence through improved planning and interviewing techniques. As part of the approach, interviewers develop a ‘Plan for the Child's Needs’ that draws on information about each child's strengths, resources, previous experience of trauma, relationships, disability, health and wellbeing to help ensure the interview is tailored to each child.

Development and implementation support for the Scottish Child Interview Model has been led by the National Joint Investigative Interviewing Team, which is a collaborative partnership between Social Work Scotland, Police Scotland and Convention of Scottish Local Authorities (COSLA).

The data improvement project stemmed from a desire among partners to improve the recording of disability within Scotland's joint investigative interviews. Led by the Centre for Excellence for Children’s Care and Protection (CELCIS) at the University of Strathclyde, the project brought together CELCIS, the National Joint Investigative Interviewing Team and social work and police managers and practitioners from seven local multi-agency partnerships to develop and test an enhanced collection. The seven local partnerships were Ayrshire, Dumfries and Galloway, Glasgow, Highland, Lanarkshire, North Strathclyde and Western Isles, with these collectively spanning 13 of Scotland's 32 local authority areas. Two other national partners—the Scottish Government and the Health and Social Care Alliance Scotland (of third sector organisations)—agreed to support the project in a critical friend capacity.

Partners were very alert to the ethical dimensions of the project, but ethical approval was not sought. The test collection was agreed to be consistent with local authorities' and Police Scotland's duty to collect disability data as a protected characteristic under the UK Equality Act 2010. Furthermore, only aggregate, anonymised data were shared for analysis in keeping with General Data Protection Regulations and the Data Protection Act 1998, meaning no child could be identified from the data shared.

The test collection data for August 2024 to January 2025 found that 175 children (36.8%) of the 475 children subject to a Joint Investigative Interview were recorded as disabled in line with the Equality Act 2010 definition. Across the seven partnerships, the percentage ranged from 25.0% to 50.9% of children recorded as disabled (M = 38.2%; SD = 9.18).

The 36.8% prevalence was a significant increase on partnerships' baseline data. Previously recording disability using the options of ‘yes’, ‘no’ and ‘not assessed or known’ without a supporting definition of disability, 6.6% of 710 children subject to a Joint Investigative Interview in 2023–2024 were recorded as disabled. Partners felt the 2023–2024 data to be an under-recording of disability, brought about by disability not being defined and practitioner reticence to record without a medical diagnosis.

By condition, and noting that multiple conditions could be recorded for each child, Table 1 shows that 21.3% of the 475 children subject to a Joint Investigative Interview were recorded as having a social interaction condition (e.g., related to autism or ADHD). Mental health (15.8% of children) was the second most recorded condition, followed by learning difficulty (9.5% of children) and speech, language and communication (5.4% of children). The importance of being able to record without a medical diagnosis is evident with ‘present but without medical diagnosis’ featuring for at least half of the children recorded as having these four conditions.

The qualitative feedback from the local partnerships endorsed the test indicators and the statistical data generated. The interviewers found the indicator definitions to be clear and felt able to record children's different conditions using the indicator options available. Interviewers also felt their practice had improved because they were having to consider each condition, meaning they were asking for further information from families and practitioners to better understand each child's needs. These insights were then used when planning and conducting the interviews to best support each child to share their experiences.

The data must be interpreted with the understanding that the recording of disability is a sensitive and complex area of practice, dependent on practitioner skill and the information available to them at that time. Furthermore, recording was based on each child's presenting conditions and behaviours, noting that these can change over time and could be brought about by the child's experience of trauma rather than a disability. The data should therefore be interpreted with both a developmental and trauma-informed lens, and work with health practitioners is planned to further refine the indicators and definitions.

The project focused on children subject to a joint investigative interview in seven local partnership areas. The findings may not therefore reflect disability prevalence among children involved in other child protection stages and processes or in other parts of Scotland. The design of the project also meant that only aggregate, anonymised data were shared, meaning further analysis by child characteristics (e.g., age, gender and whether interviewed as a victim or witness) could not be undertaken.

The finding that 36.8% of children subject to a joint investigative interview had a disability is in line with the range that can be derived from Jones et al. (2012) and Sullivan and Knutson (2000). It is also consistent with Govenden et al. (2024) findings from an audit of child protection medical examinations across three Scottish health board areas, where 32% of the children examined were recorded as having a disability. The proportion is significantly higher than the 5%–10% prevalence reported within Scotland's Children's Social Work Statistics (Scottish Government 2025), noting the collection uses the same ‘disability: yes, no and not assessed or known’ question as the Joint Investigative Interview partnership 2023–2024 baseline data.

The data generated, along with practitioners' positive experience of using the indicators, open up the prospect of their use within other statistical collections. While exciting, it is important to acknowledge that joint investigative interviews are a highly conducive ‘part of the system’ for using these indicators because of the national and local support infrastructure around the social work and police interviewers. There is also assurance in the quality of the disability data recorded because interviewers engage with the family and other practitioners to plan for interviews and then directly interact with each child during interview. The implication is that disability data are best recorded where practitioners have the time, space and skills to truly understand the child.

This article reports on the development and testing of indicators that can support joint investigative interviewers record disability among the children they interview. The project found a higher prevalence of disability than existing statistical collections in Scotland. The data therefore increase the visibility of disabled children within Scotland's child protection services and point towards the importance of specialist disability training and support for practitioners.

This study was funded through the University of Strathclyde’s Economic and Social Research Council Impact Acceleration Account.

Ethical approval for the research project was not sought. The test collection was agreed by partners (University of Strathclyde, COSLA, Social Work Scotland and Police Scotland) to be a continuation of local partnerships’ existing collection of data about the children interviewed. Furthermore, only aggregate, anonymised data were shared for analysis, meaning no child could be identified from the data shared.

The authors declare no conflicts of interest.

The data that support the findings of this study are available from the corresponding author upon reasonable request.