Transitions from child and adolescent to adult mental health services for young people involved with social care services in Wales: A national population-based retrospective cohort study using linked administrative data

Transitions from child and adolescent to adult mental health services for young people involved with social care services in Wales: A national population-based retrospective cohort study using linked administrative data | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Transitions from child and adolescent to adult mental health services for young people involved with social care services in Wales: A national population-based retrospective cohort study using linked administrative data Louisa M. Roberts, Sophie Wood, David Wilkins, Katherine H. Shelton This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9105839/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Background Transitions from Child and Adolescent Mental Health Services (CAMHS) to adult services (AMHS) are often complex and negative experiences for young people and those supporting them. Young people involved with social services are at greater risk of mental ill-health but little is known about their mental health service transitions. Methods This retrospective population-based cohort study used Secure Anonymised Information Linkage (SAIL) data to examine whether young people in Wales with CAMHS contact transitioned to AMHS. Social care status was the primary exposure. Gender, ethnicity, deprivation, health board, and mental health related diagnosis codes in health records were included as covariates. Stratified logistic regression compared odds of CAMHS with odds of transition, and estimated differences in transition by social services involvement. Results Results showed young people receiving care and support (YPRCS) and young people looked after (YPLA) had higher prevalence of diagnosis codes in health records than those with no social services involvement. YPRCS and YPLA were more than twice and nearly four times as likely to access CAMHS (OR = 2.03, 95% CI [1.86, 2.21], p < .001; OR = 3.74, 95% CI [3.07, 4.56], p < .001). However, likelihood of transition was much smaller for YPRCS (OR = 1.18, 95% CI [1.03, 1.35], p = 0.020) and not statistically significantly different for YPLA (OR = 1.27, 95% CI [0.96, 1.98], p = 0.095) when compared with their peers with no social services involvement. Recorded diagnosis codes were more strongly associated with CAMHS access than transition to AMHS. Associations between diagnosis codes and CAMHS were weaker among those involved with social services. Neurodevelopmental condition diagnosis codes influenced transition differently according to social services involvement. Conclusions Despite higher levels of diagnosis codes in health records and greater CAMHS access, young people involved with social services were only marginally or no more likely to transition to AMHS. This suggests potential discontinuity in care. The findings highlight the importance of considering how social services involvement interacts with diagnostic profiles in shaping transition outcomes. Further research is needed to explore mechanisms underlying these differences and to inform targeted support strategies. Transition mental health CAMHS AMHS young people social services involvement administrative data Figures Figure 1 Background The transition from childhood to adulthood is characterised by instability and increased support needs (Arnett, 2000; Meleis et al., 2010) particularly for young people ‘aging out’ of child-focussed mental health services (Singh et al., 2010) and social care services (Häggman-Laitila et al., 2018). Furthermore, transitions from Child and Adolescent Mental Health Services (CAMHS) to Adult Mental Health Services (AMHS) are often experienced negatively by young people and those supporting them (Dunn, 2017) leading to an increased risk of disengagement from support (Meleis et al., 2010). Difficult transitions are exacerbated by poor planning and can negatively impact individuals’ mental health and wellbeing (Appleton et al., 2020; Dunn, 2017; Singh and Tuomainen, 2015). The experience of changing ethos, culture and practice from person-centred holistic child-focussed approaches to more medicalised and individualised adult mental health services with higher thresholds of need can be especially hard for young people (McLaren et al., 2013). Whilst navigating multiple, complex, inter-related transitions young people can ‘fall through the net’ when their support needs are greatest (Islam et al., 2016). Mental ill-health in childhood is a predictor of mental ill-health in adulthood (Johnson et al., 2018; Kessler et al., 2007; McKenna et al., 2021) and can negatively influence other outcomes including health (Prince et al., 2007), education (Dalsgaard et al., 2020) and employment (Copeland et al., 2015; Hale et al., 2015). In Wales, the prevalence of treated mental health conditions for young people aged 4-to-24 years increased by approximately 50% between 2007-2014 (Pitchforth et al., 2019) with sustained high demand and increased presentations of anxiety and depression amongst adolescents (Welsh Government, 2025b). Mental ill-health places significant burdens on individuals, families, workplaces, societies and economies (Doran and Kinchin, 2017) making mental health a prominent feature of health and social policy across nations. Young people involved with social services face an increased likelihood of mental ill-health (Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003). In Wales, where the state places a child in foster care, a residential setting or with relatives under a legal order to ensure their safety and wellbeing, a child is said to be ‘looked after’. A broader category of ‘children receiving care and support’ refers to all children receiving support from a local authority under a statutory care and support plan which can be due to issues such as disability or family difficulties. For young people looked after or receiving care and support, at age 18 their social care provision will change either to adult social services, leaving care support, or independent living. The Mental Health and Wellbeing Strategy (Welsh Government, 2025a) cites the relationship between social care and mental health services as critical and commits to assessing the mental health support available to transition age youth with complex needs. There remains little research about the mental health service transitions of young people who are also transitioning from and between social care services (Butterworth et al., 2017; Jones, 2012; Tarren-Sweeney, 2013; Fledderjohann et al., 2021). This research aims to address this gap. Sociodemographic factors are known to influence mental health and access to support. Rates of probable mental disorder were found to be twice as high for young women than young men in England (Trethewey et al., 2023) and racially minoritised people in the UK face poorer mental health outcomes as well as complex barriers to accessing support (Alam et al., 2024). There are well documented associations between poverty and mental ill-health (Public Health England, 2019; Ridley et al., 2020; Zaneva et al., 2022). In Wales, more than twice as many people over 16 experienced mental health problems in the fifth most deprived areas than in the fifth least deprived areas (Public Health Wales, 2020). Intergenerational impacts mean adolescents exposed to persistent poor parental mental health and poverty are at increased risk of poor mental health themselves (Leijdesdorff et al., 2017; Reupert et al., 2022). Furthermore, the relationship between deprivation and social care need is complex and contested, with debate over whether higher levels of social services involvement in deprived areas reflect greater underlying need, structural inequalities, or differential system responses (Bywaters et al., 2015). There is also limited research as to how social services involvement may interact to influence mental health service use. In Wales, health and social care policy are devolved to the Welsh Government with health care delivered through seven health boards. Differences in localised service delivery and urban/rural geographies could be of importance regarding young peoples’ transitions from CAMHS to AMHS. Clinical severity indicators, including a diagnosis of a severe mental illness, ongoing medication, and previous hospitalisation, have all been linked with greater likelihood of transition from CAMHS to AMHS. However, evidence also suggests that older adolescents and those with either complex or lower-level needs may be at heightened risk of referral not resulting in service access (Crenna-Jennings and Hutchinson, 2020) indicating that clinical need alone does not guarantee successful transition. This study examined whether access to outpatient mental health services among young people in Wales at transition age (16–20 years) differs according to their involvement with social services. Linked Welsh health and social care administrative datasets were used to construct a population-based cohort of young people born April 1993 to March 2001 and registered with a Welsh General Practice (GP) between ages 16 and 20. We hypothesised that young people who were concurrently involved with social serviecs would have different patterns of mental health service access compared with their peers not involved with social services. Ethics This research was approved by the Information Governance Review Panel at the Secure Anonymised Data Linkage (SAIL) Databank SAIL Databank which reviews proposals as ethical and in the public interest. The routinely collected administrative data used were fully anonymised and accessed within the SAIL Trusted Research Environment. No identifiable human participants were involved and therefore informed consent was not required. Methods This was a retrospective national population-based cohort study. Table 1 shows the anonymised datasets accessed through SAIL (Jones et al., 2019; Lyons et al., 2009). More detail about these datasets is available on the Health Data Research Innovation Gateway (2023). Table 1 : Datasets accessed through the SAIL Databank The study sample of n=223,458 was derived from the Welsh Demographic Service Dataset (WDSD) and contained individuals born April 1993 to March 2001 who had lived within six of the seven health boards in Wales from age 16-to-20 (see Figure 1). Outcome variables The primary outcome variable was transition from CAMHS to AMHS. Using the consultant specialism code in the Outpatient Dataset Wales (OPDW), the transition variable reflected whether a young person who had an appointment made with CAMHS 16-plus also had an appointment made with AMHS before age 21. Welsh Government recommendations highlight transition planning should be in place by age 16 (Welsh Government, 2022) and a cut-off of age 21 included three years of possible transition. For context, a variable for whether an individual had a CAMHS appointment 16-plus, or not, was also investigated. Both variables were based on appointments made, whether seen or missed, to better reflect individual need over service access. Primary exposure The primary exposure was social services involvement 16-plus. This includes whether a young person was receiving care and support 16-plus (YPRCS) or was looked after 16-plus (YPLA) and was derived from the three social care census returns (Table 1). For clarity, young people identified in either the Children in Need Wales (CINW) or the Children Receiving Care and Support (CRCS) returns at age 16-plus are collectively referred to as the YPRCS 16-plus cohort. Those additionally recorded in the YPLA cohort were excluded from this cohort to ensure mutually exclusive exposure groups. Covariates Covariates included gender, ethnicity, area-level deprivation, health board of residence, and mental health related diagnosis codes in health records. The variable name gender is used in the WDSD (and is therefore used throughout this study). This variable reflects sex as recorded in GP records that can be different from sex at birth. Only male or female gender are recorded and therefore non-binary and self-identified genders were not available. Ethnicity was derived from the Education Wales (EDUW) dataset and recoded to a six-level categorisation of Asian, black, mixed, other, white and unknown. Ethnic identities in the EDUW dataset reflect the ethnic group with which the pupil identifies (Welsh Government, 2024). Because there were very small numbers of individuals from ethnic groups other than White in Wales, a binary variable was created for some analyses, categorising individuals as either “White” or “Asian, Black, Mixed, Other or Unknown”. This is for statistical purposes and recognises that in Wales these ethnicities are minoritized but cannot reflect the diverse health and social care experiences of ethnic minority young people in Wales. The deprivation variable was derived from the Welsh Index of Multiple Deprivation (WIMD) 2019 and reflected the lowest quintile of relative deprivation a young person had lived in age 16-to-17. The WIMD is based on eight domains of area-level deprivation, and attributes relative deprivation to 1896 Lower Super Output Areas (LSOAs) in Wales (Welsh Government, 2014). Due to frequencies under 10 for some analyses a binary variable comparing the 60% least deprived with the 40% most deprived areas was used. The health board area of residence variable was derived from the WDSD and whilst health board area was retained in models as a control, results are not reported due to frequencies under 10. One health board did not return outpatient mental health appointment data for the period of the study and young people who had lived in that health board age 16-to-20 were removed from the sample (see Figure 1). Diagnosis code variables were derived from the three health datasets (Table 1) which contain ‘Read’ codes and ICD-10 codes of the primary diagnosis for health events (prescriptions, consultations or inpatient episodes). Validated lists from the Adolescent Mental Health Data Platform (2026) were used to search health events age 11-to-17 for diagnosis codes of severe mental illness, anxiety, depression, eating disorders, alcohol use, drug use, self-harm, and neurodevelopmental conditions. The range of diagnosis codes investigated reflected data availability and diagnosis codes relevant to CAMHS and transition. To prevent disclosure of frequencies under 10, in some analyses diagnosis codes were grouped as: 1) severe mental illness / anxiety / depression / eating disorder, 2) alcohol / drug use, 3) self-harm and 4) neurodevelopmental conditions. Data access and analysis Datasets were accessed within the secure SAIL platform and SQL in Eclipse IDE (Version 2024-12) and StataSE (Version 17) were used for analysis. Descriptive statistics, Chi-square tests and logistic regression models were used. Separate logistic regression models were used for the primary outcome of transition and the contextual outcome of CAMHS 16-plus, and odds ratios were used as the measures of association. Both models were stratified by the primary exposure of social care status age 16-plus. Diagnostic tests did not indicate specification errors or significant collinearity of reported predictor variables. To formally test differences in odds ratios between the primary outcome of transition and the contextual outcome of CAMHS 16-plus, pooled logistic regression models were fitted using a stacked dataset including both outcomes to establish relative odds ratios (RORs) of the interaction terms between outcome types. Findings Study sample characteristics by cohort Of the 223,458 young people in the study sample, 5,933 (2.7%) had received care and support 16-plus and 852 (0.4%) had been looked after 16-plus. Study sample characteristics by cohort are shown in Table 2. There were slightly more males in the sample than females. Higher percentages of YPRCS (39.2%) and YPLA (50.1%) had lived in areas within the quintile of greatest deprivation compared with young people with no social services involvement (SSI) (22.9%). There was also variation in the cohort proportions by different health board areas. YPRCS and YPLA were over-represented across all diagnosis codes investigated compared with young people with no SSI. For example, 8.3% of YPRCS and 17.0% of YPLA had a recorded self-harm code compared with only 2.1% of those with no SSI. In addition, higher proportions of YPRCS (18.0%) and YPLA (24.2%) had two or more of the eight diagnosis codes present than those with no SSI (8.9%). Table 2 : Sociodemographic characteristics and diagnosis code presence for the study sample, by social services involvement *frequencies < 10 supressed CAMHS and transition sub-sample characteristics by cohort For 92.9% of CAMHS appointments in the study dataset the patient was under 18, and for 93.5% of AMHS appointments the patient was 18 or over reflecting the expected transition age of approximately 18 years. Characteristics of the sub-samples who had CAMHS 16-plus and who transitioned are shown in Table 3 and Table 4 respectively. Higher percentages of YPRCS (19.1%, n=1,136) and YPLA (29.2%, n=249) had CAMHS 16-plus than young people with no SSI (4.3%, n=9,389). The differences were smaller for transition where 35.7% (n=406) of YPRCS who had CAMHS 16-plus and 35.3% (n=88) of YPLA who had CAMHS 16-plus had transitioned compared to 28.2% (n=2,645) of young people with no SSI. Chi-square tests confirmed these differences were statistically significant (p < 0.05). Females were overrepresented among those with a CAMHS appointment 16-plus, although gender differences in transition were minimal, meaning more males than females transitioned. Small but significant differences were observed in CAMHS 16-plus appointments by ethnicity, with minimal ethnic differences in transition. Across both the CAMHS 16‑plus and transition sub‑samples, young people who had lived in the most deprived areas were over‑represented, with the pattern most pronounced among YPRCS and YPLA. This was especially evident in transition, where 45.6% of YPRCS and 56.8% of YPLA had lived in the most deprived quintile compared with 33.2% of those with no SSI. Rates of CAMHS appointments 16-plus and transition also varied by health board. Chi-square tests indicated that all sociodemographic differences described were statistically significant. Comparing between sub‑samples, neurodevelopmental condition diagnosis codes were present at noticeably higher percentages in the transition sub-sample than in the CAMHS sub-sample for all cohorts. Between cohorts, YPRCS 16-plus and YPLA 16-plus consistently showed lower percentages of those with severe mental illness, anxiety, depression and eating disorder diagnosis codes, and higher percentages of those with alcohol or drug use, self‑harm and neurodevelopmental diagnosis codes when compared with young people with no SSI. Chi‑square tests indicated that these differences were statistically significant in both sub‑samples, except for anxiety, eating disorder and alcohol use codes in the transition sub‑sample Table 3 : Sociodemographic characteristics and diagnosis code presence for young people with a CAMHS appointment age 16+, by social services involvement *frequencies < 10 supressed Table 4 : Sociodemographic characteristics and diagnosis code presence for young people who transitioned from CAMHS to AMHS before age 21, by social services involvement *frequencies < 10 supressed Association between social services involvement and mental health service use Logistic regression models adjusted for gender, ethnicity, deprivation, health board and diagnosis codes in health records showed that the odds of a CAMHS appointment 16-plus for YPRCS were approximately two times higher (OR = 2.03, 95% CI [1.86, 2.21], p < .001), and for YPLA were nearly four times higher (OR = 3.74, 95% CI [3.07, 4.56], p < .001) compared with young people with no SSI. However, the odds of transition for YPRCS were only modestly higher (18%) (OR = 1.18, 95% CI [1.03, 1.35], p = 0.020) and YPLA did not have significantly higher odds of transition compared to those with no SSI (OR = 1.27, 95% CI [0.96, 1.98], p = 0.095). Stratified logistic regression models were used with both sub-samples to enable between-cohort comparison and the results are presented in Table 5 and Table 6 showing odds ratios (OR), P-values and 95% confidence intervals (CI). Table 5 : Odds ratios (OR) and 95% confidence intervals (CI) for having a CAMHS appointment 16+ by social services involvement * = p <0.05 ** = p <0.01 *** = p <0.001 Table 6 : Odds ratios (OR) and 95% confidence intervals (CI) for transitioning from CAMHS to AMHS by social services involvement *p<0.05 **p<0.01 ***p<0.001 Sociodemographic covariate effects within cohorts Table 5 shows that gender was only associated with CAMHS 16-plus and transition for young people with no SSI, increasing odds of CAMHS for females by 7% and reducing the odds of transition for females by 15% (Table 6). Having lived in the 40% most deprived areas in Wales increased the odds of CAMHS 16-plus for young people with no SSI but was not associated with CAMHS 16-plus for YPRCS and YPLA. Deprivation was also not associated with transition for any cohort. Although descriptive statistics suggested young people who had CAMHS 16-plus and transitioned were disproportionately from the most deprived areas in Wales, logistic regression indicated only a modest deprivation effect on CAMHS 16-plus and only for those with no SSI (OR = 1.21, 95% CI [1.15, 1.27], p < .001). Diagnosis code effects within cohorts Diagnosis codes present in health records were the strongest predictors of CAMHS 16-plus but stratified models indicated that diagnosis codes increased the odds of CAMHS 16-plus most strongly for young people with no SSI. Differences between cohorts were minimal for alcohol or drug use diagnosis codes. Considering the ORs for transition, diagnosis codes had substantially weaker associations than for CAMHS 16-plus across all cohorts, with the exception of alcohol or drug use diagnosis codes for YPLA. For example, among young people with no SSI, a neurodevelopmental condition diagnosis code increased the odds of CAMHS 16-plus 49-fold but only fourfold for transition. For YPLA, a self-harm code increased CAMHS 16-plus odds by over sixfold but was not significantly associated with transition. The only diagnosis code category with a relatively stronger association with transition than CAMHS 16-plus was alcohol or drug use for YPLA (OR = 2.24, 95% CI [1.01, 5.00], p = 0.048). Formal comparison of CAMHS vs transition effects To test whether the effect of social services involvement differed between CAMHS 16-plus and transition outcomes relative odds ratios (RORs) were calculated. Two pooled logistic regression models were fitted using a stacked dataset including both outcomes. Interaction terms between outcome type and SSI cohort were significant for YPRCS (χ² = 36.30, p < 0.001; coefficient = −0.63, 95% CI: −0.83 to −0.42) and YPLA (χ² = 31.14, p < 0.001; coefficient = −1.21, 95% CI: −1.63 to −0.78). Relative odds ratios (RORs) confirmed these differences showing that the odds of transition were 47% lower than the odds of CAMHS 16-plus for YPRCS (ROR = 0.53, 95% CI: 0.44–0.66) and 70% lower for YPLA (ROR = 0.30, 95% CI: 0.20–0.46). To examine whether the effect of diagnosis code presence differed by social services involvement further interaction tests were carried out. These showed no significant differences in transition odds for those with severe mental illness, anxiety, depression, eating disorders, substance use, or self‑harm diagnosis codes. In contrast, neurodevelopmental conditions diagnoses showed a significant interaction (χ 2 = 13.84, p = 0.001). The ratio of odds ratios (ROR) for neurodevelopmental conditions was 0.55 (95% CI: 0.40-0.77) for YPRCS indicating approximately 45% lower odds of transition compared to those with no SSI. The ROR for neurodevelopmental conditions did not differ significantly between YPLA and those with no SSI (ROR = 0.59, 95% CI: 0.29-1.21). Overall, diagnosis codes strongly predicted CAMHS 16-plus but had weaker and less consistent associations with transition, except for neurodevelopmental condition diagnosis codes among YPRCS. Sub analysis of sample with no diagnosis codes present Given the known increased mental health needs of YPRCS and YPLA (Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003), it could be hypothesised that these cohorts would be more likely to transition to AMHS even when diagnosis codes were not present. To examine this, the logistic regression models were repeated on a sub-sample of young people with none of the investigated diagnosis codes recorded in health records age 11-to-17. Whilst the odds of CAMHS 16-plus were six times higher for YPRCS (OR = 5.66, 95% CI [4.87, 6.58], p < .001) and 11 times higher for YPLA (OR = 11.12, 95% CI [8.05, 15.36], p < .001), the odds of transition were not significantly different for YPRCS (OR = 1.37, 95% CI [0.97, 1.94], p = 0.074) or YPLA (OR = 1.33, 95% CI [0.65, 2.73], p = 0.441) compared with young people with no SSI. This indicates that YPRCS and YPLA were not more likely to transition when diagnosis codes were not present. Discussion The transition from Child and Adolescent Mental Health Services (CAMHS) to Adult Mental Health Services (AMHS) is widely recognized as a high-risk period for service disengagement. The findings from this study show that this risk may be especially great for young people involved with social services, raising pressing questions about equity and continuity of care. The results indicated that, compared to young people with no social services involvement 16-plus (SSI), higher proportions of young people receiving care and support (YPRCS) and looked after (YPLA) 16-plus in Wales had mental health related diagnosis codes recorded in their health records age 11-to-17. In addition, YPRCS and YPLA 16-plus had higher prevalence of multiple (two or more) diagnosis codes investigated present in their health records. This finding aligns with longstanding evidence of the increased mental health needs and complex support requirements of young people involved with social services, (Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003). It is therefore unsurprising that in this study YPRCS and YPLA 16-plus in Wales were over-represented in CAMHS and had significantly higher odds of having a CAMHS appointment 16-plus after adjusting for sociodemographic factors and diagnosis codes present in health records. However, among young people who had a CAMHS appointment 16-plus, those involved with social services 16-plus were far less represented in transition to AMHS. Transitions from childhood to adulthood, from living in care to leaving care, and from CAMHS to AMHS are well-established points of instability that may heighten the risk of worsening mental health as well as disengagement from services (Dunn, 2017; Singh et al., 2010). At a time when their support needs may be greatest, concurrently being in receipt of local authority care and support had only a modest effect of increasing the likelihood of transitioning to AMHS, and being looked after had no significant effect when compared with those not involved with social services. Overall, only 29% of young people who had a CAMHS appointment 16-plus also had an AMHS appointment by age 21 years. These results suggest that the well-documented reduction in mental health support at transition age may be most pronounced for young people concurrently involved with social services. However, given that social services involvement encompasses a diverse range of circumstances and levels of need, the factors influencing transition are unlikely to be uniform and future research should seek to identify the specific mechanisms operating within this broad category. Descriptive statistics suggested that higher proportions of males and young people who had lived in the most deprived areas in Wales had transitioned to AMHS. These kinds of proportional representations may reflect what practitioners experience as they support young people at transition age, and thus feed into narratives around which young people do or do not transition. However, stratified logistic regressions showed that gender and deprivation did not independently influence the likelihood of transition when adjusting for social care status, ethnicity, health board area of residence and diagnosis codes in health records. This demonstrates the importance of moving beyond descriptive patterns to understand the drivers of transition outcomes Previous research has shown that having mental health related diagnoses, prescription medications, or periods of inpatient care are all strongly associated with mental health service usage (Appleton et al., 2023; Singh et al., 2010). In this study, diagnosis codes in health records captured these clinical needs. Relative odds ratios showed that diagnosis codes in health records had a much weaker effect on the likelihood of transition than on the likelihood of CAMHS 16-plus across all cohorts. Diagnosis codes did not have statistically significantly different effects on the odds of transition between cohorts, except for neurodevelopmental condition diagnosis codes which had a weaker association with the likelihood of transition for YPRCS than for young people with no SSI. Given the rising prevalence of ADHD and ASD diagnoses among young people in the UK (Department of Health and Social Care, 2025; McKechnie et al., 2023) it is concerning to see that whilst these diagnoses increase the likelihood of transition to adult mental health services nearly fourfold for young people not concurrently involved with social services, the same types of diagnosis codes only increase the likelihood of transition for young people concurrently in receipt of care and support from their local authority less than twofold. To plan effective support for young people transitioning from CAMHS, further research is needed to explore how different diagnoses interact with social services involvement to shape continuity of mental health care. Strengths and limitations To our knowledge this is the first study to explore how the effects of sociodemographic factors and recorded diagnosis codes on mental health service usage may differ for young people who are concurrently involved with social care services as they transition to adulthood. Using population level administrative data enables more comprehensive mapping of young people’s mental health service access nationally than using localised service or survey data. However, administrative data are subject to restrictions and biases, including how information is captured, how variables are conceptualised, and errors in recording. Large sample numbers add strength to statistical testing and generalisability, but in this study the sample is substantially reduced by mental health usage sub-samples, social services involvement cohorts, and low numbers of some predictor variables such as minority ethnic categories, health board areas of residence, and specific diagnosis codes. Adapting existing data to research questions, however, can be considered a more ethical approach compared to using resources to gather or re-gather study specific data. This is particularly important when considering the data of vulnerable populations and those who may be over-researched. The WDSD uses a gender code reflecting sex as GP-registered, which can be different than sex at birth, but does not reflect non-binary or subjective gender identities. It was not possible to maintain six categories of ethnic identities for all the analyses, and even where six categories were used, these cannot reflect heterogeneity within each group. The read code lists used for types of diagnosis have been validated through a rigorous process combining expert clinical input with systematic review of coding standards. Relevant diagnostic, medication, and treatment codes identified are cross-checked against established classification systems and tested for accuracy using real-world data to ensure validity and completeness. Clinical diagnoses provide more complete case ascertainment than surveys or cohort studies, which can be susceptible to selection bias, low recruitment and high attrition. However, routinely collected data commonly underestimate prevalence as not all individuals in need will access services, and some conditions might not be recognised or recorded. Conversely, including GP records and using a simple binary variable for the presence or absence of diagnosis codes means that individuals presenting once in relation to a particular condition will be identified with presence of that diagnosis code, even with no subsequent contacts or diagnoses. Thus, whilst in many instances diagnosis code records will reflect a formal diagnosis, in some cases they do not. Diagnoses may therefore be both over-represented and under-represented when using diagnosis codes as a proxy. Finally, there are no validated measures of the clinical severity in these data. The study dataset sub-sample contains a wide range of mental health service access experiences from young people with a single CAMHS appointment that was never attended, through to those with frequent and ongoing contact with CAMHS and AMHS across the five-year period. Further research is needed to examine different patterns or typologies of mental health service access and how these relate to the predictor and outcome variables of interest. Conclusion This study of Welsh administrative health data concurs with previous research showing that young people involved with social services have an increased risk of mental ill-health and complex mental health needs. Whilst these groups are over-represented in CAMHS 16-plus, the same is not true of their transition to AMHS. Across all cohorts, most young people accessing CAMHS 16-plus did not transition to AMHS, highlighting a significant gap in continuity of care. The ‘cliff edge’ of formal mental health support at transition age in Wales is therefore of considerable concern, particularly given the rising levels of mental ill-health among children, and the financial pressures facing public services. Furthermore, the finding that young people concurrently involved with social services have distinct patterns of mental health service use at transition suggests that they may experience the greatest reduction in support at this critical point. This study also indicates that established predictors of mental health transitions such as diagnoses, medication and inpatient care, may have different impacts on transition for young people concurrently involved with social services. Greater attention from mental health services to the social care contexts of young people is essential to ensure their needs are identified, continuity of care is maintained, and they are supported to flourish through their transition to adulthood and beyond. Abbreviations AMHS Adult mental health services CAMHS Child and adolescent mental health services SSI Social services involvement YPRCS Young people receiving care and support YPLA Young people looked after Declarations Ethics approval and consent to participate: This research was approved by the Information Governance Review Panel (IGRP) at the Secure Anonymised Data Linkage (SAIL) Databank which reviews proposals as ethical and in the public interest. Approval was granted under SAIL project 1418. Once IGRP approval is in place further institutional ethical approval is not required. The routinely collected administrative data used were fully anonymised and accessed within the SAIL Trusted Research Environment. No identifiable human participants were involved and therefore informed consent was not required. All procedures were conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication: N/A Availability of data and materials: The data that support the findings of this study are available from the Secure Anonymised Data LinkageDatabank. The data are accessible via a two-stage application process consisting of scoping and governance review to assess user and project approvals Apply to work with the data - SAIL Databank. Competing interests: The authors declare that they have no competing interests Funding: This PhD study was funded by the Economic and Social Research Council and supported by the CASCADE Partnership which receives funding from Health and Care Research Wales. Author’s contributions LR conceptualised the study, developed the methods, conducted the data analyses and wrote the manuscript. SW, DW and KS supported the conceptualisation of the study, the development of methods, the data analysis and read and approved the final manuscript. References ALAM, S., O’HALLORAN, S. & FOWKE, A. 2024. What are the barriers to mental health support for racially-minoritised people within the UK? A systematic review and thematic synthesis. The Cognitive Behaviour Therapist, 17. APPLETON, R., CANAWAY, A., TUOMAINEN, H., DIELEMAN, G. C., GERRITSEN, S., OVERBEEK, M., MARAS, A., VAN BODEGOM, L., FRANIĆ, T., DE GIROLAMO, G., MADAN, J., MCNICHOLAS, F., PURPER-OUAKIL, D., SCHULZE, U., TREMMERY, S. & SINGH, S. 2023. Predictors of transitioning to adult mental health services and associated costs: a cross-country comparison. BMJ Mental Health, 26 , article e300814. APPLETON, R., ELAHI, F., TUOMAINEN, H., CANAWAY, A. & SINGH, S. P. 2020. "I'm just a long history of people rejecting referrals" experiences of young people who fell through the gap between child and adult mental health services. European child & adolescent psychiatry, 3 , 401–413. ARNETT, J. J. 2000. Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55 , 469. BRONSARD, G., AUQUIER, P. & BOYER, L. 2016. Links between early child maltreatment, mental disorders, and cortisol secretion anomalies. Journal of Physiology-Paris, 110 , 448–452. BUTTERWORTH, S., SINGH, S. P., BIRCHWOOD, M., ISLAM, Z., MUNRO, E. R., VOSTANIS, P., PAUL, M., KHAN, A. & SIMKISS, D. 2017. Child and Adolescent Mental Health | ACAMH Journal | Wiley Online Library. Child and Adolescent Mental Health, 22. BYWATERS, P., BRADY, G., SPARKS, T., BOS, E., BUNTING, L., DANIEL, B., FEATHERSTONE, B., MORRIS, K. & SCOURFIELD, J. 2015. Exploring inequities in child welfare and child protection services: Explaining the ‘inverse intervention law’. Children and Youth Services Review, 57 , 98–105. COPELAND, W. E., WOLKE, D., SHANAHAN, L. & COSTELLO, E. J. 2015. Adult Functional Outcomes of Common Childhood Psychiatric Problems. JAMA Psychiatry, 72. CRENNA-JENNINGS, W. & HUTCHINSON, J. 2020. Access to child and adolescent mental health services in 2019. Education Policy Institute. DALSGAARD, S., MCGRATH, J., ØSTERGAARD, S. D., WRAY, N. R., PEDERSEN, C. B., MORTENSEN, P. B. & PETERSEN, L. 2020. Association of Mental Disorder With Subsequent Educational Achievement. JAMA Psychiatry, 77. DEPARTMENT OF HEALTH AND SOCIAL CARE 2025. Independent review into mental health conditions, ADHD and autism: terms of reference. Available at https://www.gov.uk/government/publications/independent-review-into-mental-health-conditions-adhd-and-autism-terms-of-reference. Accessed [9 Mar 2026]. DORAN, C. M. & KINCHIN, I. 2017. A review of the economic impact of mental illness. Australian Health Review, 43 , 43–48. DUBOIS-COMTOIS, K., BUSSIÈRES, E.-L., CYR, C., ST-ONGE, J., BAUDRY, C., MILOT, T. & LABBÉ, A.-P. 2021. Are children and adolescents in foster care at greater risk of mental health problems than their counterparts? A meta-analysis. Children and Youth Services Review, 127. DUNN, V. 2017. Young people, mental health practitioners and researchers co produce a Transition Preparation Programme to improve outcomes and experience for young people leaving Child and Adolescent Mental Health Services (CAMHS. BMC Health Services Research, 17 , 12. ENGLER, A. D., SARPONG, K. O., VAN HORNE, B. S., GREELEY, C. S. & KEEFE, R. J. 2022. A systematic review of mental health disorders of children in foster care. Trauma, Violence, & Abuse, 23 , 255–264. FLEDDERJOHANN, J., ERLAM, J., KNOWLES, B. & BROADHURST, K. 2021. Mental health and care needs of British children and young people aged 6–17. Children and Youth Services Review, 126 , 106033. FORD, T., VOSTANIS, P., MELTZER, H. & GOODMAN, R. 2018. Psychiatric disorder among British children looked after by local authorities: Comparison with children living in private households. British Journal of Psychiatry, 190 , 319–325. HÄGGMAN-LAITILA, A., SALOKEKKILÄ, P. & KARKI, S. 2018. Transition to adult life of young people leaving foster care: A qualitative systematic review. Children and Youth Services Review, 95. HALE, D. R., BEVILACQUA, L. & VINER, R. M. 2015. Adolescent health and adult education and employment: a systematic review. Pediatrics, 136 , 128–140. ISLAM, Z., FORD, T., KRAMER, T., PAUL, M., HARLEY, K., WEAVER, T., MCLAREN, S. & SINGH, S. P. 2016. Mind how you cross the gap! Outcomes for young people who failed to make the transition from child to adult services: The TRACK study. BJPsych Bulletin, 40 , 142–148. JOHNSON, D., DUPUIS, G., PICHE, J., CLAYBORNE, Z. & COLMAN, I. 2018. Adult mental health outcomes of adolescent depression: a systematic review. Depression and Anxiety, 35 , 700–716. JONES, K. H., FORD, D. V., THOMPSON, S. & LYONS, R. A. 2019. A profile of the SAIL databank on the UK secure research platform. International journal of population data science, 4. JONES, R. 2012. The effectiveness of interventions aimed at improving access to health and mental health services for looked after children and young people: A systematic review. Families, Relationships and Societies, 1 , 71–85. KESSLER, R. C., BERGLUND, P., DEMLER, O., JIN, R., MERIKANGAS, K. R. & WALTERS, E. E. 2007. Age of onset of mental disorders: a review of recent literature. Current Opinion in Psychiatry, 20 , 359–364. LEIJDESDORFF, S., VAN DOESUM, K. T. M., POPMA, A., KLAASSEN, R. & VAN AMELSVOORT, T. 2017. The impact of parental mental illness on children and adolescents: a systematic review. European Child & Adolescent Psychiatry, 26 , 869–879. LYONS, R. A., JONES, K. H., JOHN, G., BROOKS, C. J., VERPLANCKE, J. P., FORD, D. V., BROWN, G. & LEAKE, K. 2009. The SAIL databank: linking multiple health and social care datasets. BMC Medical Informatics and Decision Making, 9. MCKECHNIE, D., O’NIONS, E., DUNSMUIR, S. & PETERSEN, I. 2023. Attention-deficit hyperactivity disorder diagnoses and prescriptions in UK primary care, 2000–2018: population-based cohort study. BJPsych Open, 9(4). MCKENNA, S., DONNELLY, M., ONYEKA, I. N., O’REILLY, D. & MAGUIRE, A. 2021. Experience of child welfare services and long-term adult mental health outcomes: a scoping review. Social psychiatry and psychiatric epidemiology, 56 , 1115–1145. MCLAREN, S., BELLING, R., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., HOVISH, K., ISLAM, Z., WHITE, S., SINGH, S. P., MCLAREN, S., BELLING, R., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., HOVISH, K., ISLAM, Z., WHITE, S. & SINGH, S. P. 2013. ‘Talking a different language’: an exploration of the influence of organizational cultures and working practices on transition from child to adult mental health services. BMC Health Services Research 2013 13:1, 13. MELEIS, A. I., SAWYER, L. M., IM, E., MESSIAS, D. K. H. & SCHUMACHER, K. 2010. Transition theory. Transitions theory: middle-range and situation specific theories in nursing research and practice. New York: Springer Publishing Company , 52–83. MELTZER, H., GATWARD, R., CORBIN, T., GOODMAN, R. & FORD, T. 2003. The mental health of young people looked after by local authorities in England , London: The Stationery Office. PITCHFORTH, J., FAHY, K., FORD, T., WOLPERT, M., VINER, R. M. & HARGREAVES, D. S. 2019. Mental health and well-being trends among children and young people in the UK, 1995–2014: analysis of repeated cross-sectional national health surveys. Psychological medicine, 49 , 1275–1285. PRINCE, M., PATEL, V., SAXENA, S., MAJ, M., MASELKO, J., PHILLIPS, M. R. & RAHMAN, A. 2007. No health without mental health. The Lancet, 370. PUBLIC HEALTH ENGLAND 2019. Health Profile for England 2019. Available at https://www.gov.uk/government/publications/health-profile-for-england-2019. Accessed [9 Mar 2026]. In: ENGLAND., P. H. (ed.). London. PUBLIC HEALTH WALES 2020. Mental Wellbeing in Wales 2020. Available at: https://phw.nhs.wales/services-and-teams/observatory/data-and-analysis/mental-wellbeing-in-wales-2020/. [Accessed 9 Mar 2026]. In: WALES, P. H. (ed.). Cardiff. REUPERT, A., MAYBERY, D., NICHOLSON, J., GOPFERT, M. & SEEMAN, M. V. 2022. Children of parents with a mental illness: a systematic review of interventions. International Journal of Mental Health Nursing, 31 , 6–25. RIDLEY, M., RAO, G., SCHILBACH, F. & PATEL, V. 2020. Poverty, depression, and anxiety: causal evidence and mechanisms. Science, 370. SINGH, S. P., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., MCLAREN, S., HOVISH, K., ISLAM, Z., BELLING, R. & WHITE, S. 2010. Process, outcome and experience of transition from child to adult mental healthcare: Multiperspective study. The British Journal of Psychiatry, 197 , 305–312. SINGH, S. P. & TUOMAINEN, H. 2015. Transition from child to adult mental health services: Needs, barriers, experiences and new models of care. World Psychiatry, 14 , 358–361. TARREN-SWEENEY, M. A. V., A. EDS., 2013. Mental health services for vulnerable children and young people: Supporting children who are, or have been, in foster care , Routledge. TRETHEWEY, S. P., MATHEWS, F., RUSSELL, A. & NEWLOVE-DELGADO, T. 2023. Mental health of children and young people aged 5-16 in England: socio-demographic and clinical characteristics associated with support and service contact. European Psychiatry, 66 , S582–S582. WELSH GOVERNMENT. 2014. Welsh Index of Multiple Deprivation 2011 to 2019 [Online]. Available: https://www.gov.wales/welsh-index-multiple-deprivation-2011-2019 [Accessed]. WELSH GOVERNMENT 2022. Transition and handover from children’s to adult health services: Guidance. Available at: https://www.gov.wales/transition-and-handover-childrens-adult-health-services/ [Accessed 9 Mar 2026]. Cardiff: Welsh Government. WELSH GOVERNMENT 2024. Pupil Level Annual School Census (PLASC) Census day: 20 January 2026. Available at: https://www.gov.wales/sites/default/files/publications/2026-01/technical-completion-notes-pupil-annual-school-census-jan-2026.pdf. [Accessed 9 Mar 2026]. WELSH GOVERNMENT. 2025a. Mental health and wellbeing strategy 2025 to 2035 | GOV.WALES [Online]. Available: https://www.gov.wales/mental-health-and-wellbeing-strategy-2025-2035 [Accessed]. WELSH GOVERNMENT. 2025b. Mental health statistics: interactive dashboard [Online]. Available: https://www.gov.wales/mental-health-statistics-interactive-dashboard [Accessed 09 Mar 2026]. ZANEVA, M., GUZMAN‑HOLST, C., R., A. & BOWES, L. 2022. The impact of monetary poverty alleviation programs on children’s and adolescents’ mental health: a systematic review and meta‑analysis. Journal of Adolescent Health, 71 , 147–156. Tables Tables 1 to 6 are available in the supplementary files section Additional Declarations No competing interests reported. Supplementary Files Tables.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 24 Apr, 2026 Reviews received at journal 24 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 15 Apr, 2026 Reviewers invited by journal 15 Apr, 2026 Editor assigned by journal 13 Apr, 2026 Editor invited by journal 24 Mar, 2026 Submission checks completed at journal 20 Mar, 2026 First submitted to journal 20 Mar, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9105839","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":626042468,"identity":"f5c8aa7e-b029-4c34-a942-b3d39a357e83","order_by":0,"name":"Louisa M. Roberts","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAUlEQVRIie3RP2rDMBiH4Z8wyIV+B7CJqa8QE3AoDuQqAkOz+ADtYlQCyWJfJ1MHGYGnNHPWEsjUIVmKQzvUTtfK7thB76BB8PDpD2Cz/cvougahu6xOgEPuUl13xv1EgKKiTr2OUC3+SLDP4puWYJBM12V1eGpyYjL7wOUlCYiDnRroiYkE210abYUmF68bVhwX1BLHL6BjE/G8LPalUMSeyw1IaZqHEiNAz3rI9CJFTtB0ZF8t6aZ8DpCYSeEQauK39EN4N8V8MNqlvnzQ7SPzyShQ3V3Y6r4YL4zX99yyOstZfheGh7fzu0rmxB29bx6TSJrMLzHZ+5E2m81mG+4bPQFMnB5OX+sAAAAASUVORK5CYII=","orcid":"","institution":"Cardiff University","correspondingAuthor":true,"prefix":"","firstName":"Louisa","middleName":"M.","lastName":"Roberts","suffix":""},{"id":626042469,"identity":"cda5735b-b59f-4fe1-bb3e-dc99b57727aa","order_by":1,"name":"Sophie Wood","email":"","orcid":"","institution":"University of Alberta","correspondingAuthor":false,"prefix":"","firstName":"Sophie","middleName":"","lastName":"Wood","suffix":""},{"id":626042470,"identity":"770e32dc-e87f-4f6a-875b-b19143a6b768","order_by":2,"name":"David Wilkins","email":"","orcid":"","institution":"Cardiff University","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Wilkins","suffix":""},{"id":626042473,"identity":"fe71eb60-e548-4ee0-8740-0734e9f6029b","order_by":3,"name":"Katherine H. Shelton","email":"","orcid":"","institution":"Cardiff University","correspondingAuthor":false,"prefix":"","firstName":"Katherine","middleName":"H.","lastName":"Shelton","suffix":""}],"badges":[],"createdAt":"2026-03-12 14:23:07","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9105839/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9105839/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107515259,"identity":"556faf1a-4d1b-485d-86f9-ebc2c2b57caf","added_by":"auto","created_at":"2026-04-22 08:28:13","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":136622,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlow chart showing sample derivation\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9105839/v1/2c4083b2a6614f15b9674c8a.png"},{"id":107705427,"identity":"7efa015d-f780-45e7-87db-aa0516cee431","added_by":"auto","created_at":"2026-04-24 09:12:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":378508,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9105839/v1/b5b71714-579d-4d7c-9018-622902ca564b.pdf"},{"id":107515008,"identity":"93cb414e-b0e4-4b8c-b9fa-a7d5c262f916","added_by":"auto","created_at":"2026-04-22 08:27:57","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":36514,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-9105839/v1/27f333ed939a9b5b35f6dd4d.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Transitions from child and adolescent to adult mental health services for young people involved with social care services in Wales: A national population-based retrospective cohort study using linked administrative data","fulltext":[{"header":"Background","content":"\u003cp\u003eThe transition from childhood to adulthood is characterised by instability and increased support needs (Arnett, 2000; Meleis et al., 2010) particularly for young people \u0026lsquo;aging out\u0026rsquo; of child-focussed mental health services (Singh et al., 2010) and social care services (H\u0026auml;ggman-Laitila et al., 2018). Furthermore, transitions from Child and Adolescent Mental Health Services (CAMHS) to Adult Mental Health Services (AMHS) are often experienced negatively by young people and those supporting them (Dunn, 2017) leading to an increased risk of disengagement from support (Meleis et al., 2010). Difficult transitions are exacerbated by poor planning and can negatively impact individuals\u0026rsquo; mental health and wellbeing\u0026nbsp;(Appleton et al., 2020; Dunn, 2017; Singh and Tuomainen, 2015). The experience of changing ethos, culture and practice from person-centred holistic child-focussed approaches to more medicalised and individualised adult mental health services with higher thresholds of need can be especially hard for young people (McLaren et al., 2013). Whilst navigating multiple, complex, inter-related transitions young people can \u0026lsquo;fall through the net\u0026rsquo; when their support needs are greatest\u0026nbsp;(Islam et al., 2016).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMental ill-health in childhood is a predictor of mental ill-health in adulthood (Johnson et al., 2018; Kessler et al., 2007; McKenna et al., 2021) and can negatively influence other outcomes including health (Prince et al., 2007), education (Dalsgaard et al., 2020)\u0026nbsp;and employment\u0026nbsp;(Copeland et al., 2015; Hale et al., 2015). In Wales, the prevalence of treated mental health conditions for young people aged 4-to-24 years increased by approximately 50% between 2007-2014 (Pitchforth et al., 2019) with sustained high demand and increased presentations of anxiety and depression amongst adolescents (Welsh Government, 2025b). Mental ill-health places significant burdens on individuals, families, workplaces, societies and economies\u0026nbsp;(Doran and Kinchin, 2017) making mental health a prominent feature of health and social policy across nations.\u003c/p\u003e\n\u003cp\u003eYoung people involved with social services face an increased likelihood of mental ill-health (Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003).\u0026nbsp;In Wales, where the state places a child in foster care, a residential setting or with relatives under a legal order to ensure their safety and wellbeing, a child is said to be \u0026lsquo;looked after\u0026rsquo;. A broader category of \u0026lsquo;children receiving care and support\u0026rsquo; refers to all children receiving support from a local authority under a statutory care and support plan which can be due to issues such as disability or family difficulties. For young people looked after or receiving care and support, at age 18 their social care provision will change either to adult social services, leaving care support, or independent living. The Mental Health and Wellbeing Strategy (Welsh Government, 2025a) cites the relationship between social care and mental health services as critical and commits to assessing the mental health support available to transition age youth with complex needs. There remains little research about the mental health service transitions of young people who are also transitioning from and between social care services (Butterworth et al., 2017; Jones, 2012; Tarren-Sweeney, 2013; Fledderjohann et al., 2021). This research aims to address this gap.\u003c/p\u003e\n\u003cp\u003eSociodemographic factors are known to influence mental health and access to support.\u0026nbsp;Rates of probable mental disorder were found to be twice as high for young women than young men in England (Trethewey et al., 2023) and racially minoritised people in the UK face poorer mental health outcomes as well as complex barriers to accessing support (Alam et al., 2024).\u0026nbsp;There are well documented associations between poverty and mental ill-health\u0026nbsp;(Public Health England, 2019; Ridley et al., 2020; Zaneva et al., 2022). In Wales, more than twice as many people over 16 experienced mental health problems in the fifth most deprived areas than in the fifth least deprived areas\u0026nbsp;(Public Health Wales, 2020). Intergenerational impacts mean adolescents exposed to persistent poor parental mental health and poverty are at increased risk of poor mental health themselves\u0026nbsp;(Leijdesdorff et al., 2017; Reupert et al., 2022). Furthermore, the relationship between deprivation and social care need is complex and contested,\u0026nbsp;with debate over whether higher levels of social services involvement in deprived areas reflect greater underlying need, structural inequalities, or differential system responses\u0026nbsp;(Bywaters et al., 2015). There is also limited research as to how social services involvement may interact to influence mental health service use. In Wales, health and social care policy are devolved to the Welsh Government with health care delivered through seven health boards. Differences in localised service delivery and urban/rural geographies could be of importance regarding young peoples\u0026rsquo; transitions from CAMHS to AMHS.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eClinical severity indicators, including\u0026nbsp;a diagnosis of a severe mental illness, ongoing medication, and previous hospitalisation, have all been linked with greater likelihood of\u0026nbsp;transition from CAMHS to AMHS. However, evidence also suggests that older adolescents and those with either complex or lower-level needs may be at heightened risk of referral not resulting in service access (Crenna-Jennings and Hutchinson, 2020) indicating that clinical need alone does not guarantee successful transition.\u003c/p\u003e\n\u003cp\u003eThis study examined whether access to outpatient mental health services among young people in Wales at transition age (16\u0026ndash;20 years) differs according to their involvement with social services. Linked Welsh health and social care administrative datasets were used to construct a population-based cohort of young people born April 1993 to March 2001 and registered with a Welsh General Practice (GP) between ages 16 and 20. We hypothesised that young people who were concurrently involved with social serviecs would have different patterns of mental health service access compared with their peers not involved with social services.\u003c/p\u003e\n\u003ch2\u003eEthics\u003c/h2\u003e\n\u003cp\u003eThis research was approved by the Information Governance Review Panel at the Secure Anonymised Data Linkage (SAIL) Databank SAIL Databank which reviews proposals as ethical and in the public interest.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eThe routinely collected administrative data used were fully anonymised and accessed within the SAIL Trusted Research Environment. No identifiable human participants were involved and therefore informed consent was not required.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis was a retrospective national population-based cohort study. Table 1 shows the anonymised datasets accessed through SAIL (Jones et al., 2019; Lyons et al., 2009). More detail about these datasets is available on the Health Data Research Innovation Gateway (2023).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e: Datasets accessed through the SAIL Databank\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study sample of n=223,458 was derived from the Welsh Demographic Service Dataset (WDSD) and contained individuals born April 1993 to March 2001 who had lived within six of the seven health boards in Wales from age 16-to-20 (see Figure 1).\u003c/p\u003e\n\u003ch3\u003eOutcome variables\u003c/h3\u003e\n\u003cp\u003eThe primary outcome variable was transition from CAMHS to AMHS. Using the consultant specialism code in the Outpatient Dataset Wales (OPDW), the transition variable reflected whether a young person who had an appointment made with CAMHS 16-plus also had an appointment made with AMHS before age 21. Welsh Government recommendations highlight transition planning should be in place by age 16 (Welsh Government, 2022) and a cut-off of age 21 included three years of possible transition. For context, a variable for whether an individual had a CAMHS appointment 16-plus, or not, was also investigated. Both variables were based on appointments made, whether seen or missed, to better reflect individual need over service access.\u003c/p\u003e\n\u003ch3\u003ePrimary exposure\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe primary exposure was social services involvement 16-plus. This includes whether a young person was receiving care and support 16-plus (YPRCS) or was looked after 16-plus (YPLA) and was derived from the three social care census returns (Table 1). For clarity, young people identified in either the Children in Need Wales (CINW) or the Children Receiving Care and Support (CRCS) returns at age 16-plus are collectively referred to as the YPRCS 16-plus cohort. Those additionally recorded in the YPLA cohort were excluded from this cohort to ensure mutually exclusive exposure groups.\u003c/p\u003e\n\u003ch3\u003eCovariates\u003c/h3\u003e\n\u003cp\u003eCovariates included gender, ethnicity, area-level deprivation, health board of residence, and mental health related diagnosis codes in health records. The variable name gender is used in the WDSD (and is therefore used throughout this study). This variable reflects sex as recorded in GP records that can be different from sex at birth. Only male or female gender are recorded and therefore non-binary and self-identified genders were not available. Ethnicity was derived from the Education Wales (EDUW) dataset and recoded to a six-level categorisation of Asian, black, mixed, other, white and unknown. Ethnic identities in the EDUW dataset reflect the ethnic group with which the pupil identifies (Welsh Government, 2024). Because there were very small numbers of individuals from ethnic groups other than White in Wales, a binary variable was created for some analyses, categorising individuals as either \u0026ldquo;White\u0026rdquo; or \u0026ldquo;Asian, Black, Mixed, Other or Unknown\u0026rdquo;. This is for statistical purposes and recognises that in Wales these ethnicities are minoritized but cannot reflect the diverse health and social care experiences of ethnic minority young people in Wales.\u003c/p\u003e\n\u003cp\u003eThe deprivation variable was derived from the Welsh Index of Multiple Deprivation (WIMD) 2019 and reflected the lowest quintile of relative deprivation a young person had lived in age 16-to-17. The WIMD is based on eight domains of area-level deprivation, and attributes relative deprivation to 1896 Lower Super Output Areas (LSOAs) in Wales (Welsh Government, 2014). Due to frequencies under 10 for some analyses a binary variable comparing the 60% least deprived with the 40% most deprived areas was used. The health board area of residence variable was derived from the WDSD and whilst health board area was retained in models as a control, results are not reported due to frequencies under 10. One health board did not return outpatient mental health appointment data for the period of the study and young people who had lived in that health board age 16-to-20 were removed from the sample (see Figure 1).\u003c/p\u003e\n\u003cp\u003eDiagnosis code variables were derived from the three health datasets (Table 1) which contain \u0026lsquo;Read\u0026rsquo; codes and ICD-10 codes of the primary diagnosis for health events (prescriptions, consultations or inpatient episodes). Validated lists from the Adolescent Mental Health Data Platform (2026) were used to search health events age 11-to-17 for diagnosis codes of severe mental illness, anxiety, depression, eating disorders, alcohol use, drug use, self-harm, and neurodevelopmental conditions. The range of diagnosis codes investigated reflected data availability and diagnosis codes relevant to CAMHS and transition. To prevent disclosure of frequencies under 10, in some analyses diagnosis codes were grouped as: 1) severe mental illness / anxiety / depression / eating disorder, 2) alcohol / drug use, 3) self-harm and 4) neurodevelopmental conditions.\u003c/p\u003e\n\u003ch3\u003eData access and analysis\u003c/h3\u003e\n\u003cp\u003eDatasets were accessed within the secure SAIL platform and SQL in Eclipse IDE (Version 2024-12) and StataSE (Version 17) were used for analysis. Descriptive statistics, Chi-square tests and logistic regression models were used. Separate logistic regression models were used for the primary outcome of transition and the contextual outcome of CAMHS 16-plus, and odds ratios were used as the measures of association. Both models were stratified by the primary exposure of social care status age 16-plus. Diagnostic tests did not indicate specification errors or significant collinearity of reported predictor variables. To formally test differences in odds ratios between the primary outcome of transition and the contextual outcome of CAMHS 16-plus, pooled logistic regression models were fitted using a stacked dataset including both outcomes to establish relative odds ratios (RORs) of the interaction terms between outcome types.\u003c/p\u003e\n\u003ch2\u003eFindings\u003c/h2\u003e\n\u003ch3\u003eStudy sample characteristics by cohort\u003c/h3\u003e\n\u003cp\u003eOf the 223,458 young people in the study sample, 5,933 (2.7%) had received care and support 16-plus and 852 (0.4%) had been looked after 16-plus. Study sample characteristics by cohort are shown in Table 2. There were slightly more males in the sample than females. Higher percentages of YPRCS (39.2%) and YPLA (50.1%) had lived in areas within the quintile of greatest deprivation compared with young people with no social services involvement (SSI) (22.9%). There was also variation in the cohort proportions by different health board areas. YPRCS and YPLA were over-represented across all diagnosis codes investigated compared with young people with no SSI. For example, 8.3% of YPRCS and 17.0% of YPLA had a recorded self-harm code compared with only 2.1% of those with no SSI. In addition, higher proportions of YPRCS (18.0%) and YPLA (24.2%) had two or more of the eight diagnosis codes present than those with no SSI (8.9%).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003cstrong\u003e: Sociodemographic characteristics and diagnosis code presence for the study sample, by social services involvement\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e*frequencies \u0026lt; 10 supressed\u003c/p\u003e\n\u003ch3\u003eCAMHS and transition sub-sample characteristics by cohort\u003c/h3\u003e\n\u003cp\u003eFor 92.9% of CAMHS appointments in the study dataset the patient was under 18, and for 93.5% of AMHS appointments the patient was 18 or over reflecting the expected transition age of approximately 18 years. Characteristics of the sub-samples who had CAMHS 16-plus and who transitioned are shown in Table 3 and Table 4 respectively. Higher percentages of YPRCS (19.1%, n=1,136) and YPLA (29.2%, n=249) had CAMHS 16-plus than young people with no SSI (4.3%, n=9,389). The differences were smaller for transition where 35.7% (n=406) of YPRCS who had CAMHS 16-plus and 35.3% (n=88) of YPLA who had CAMHS 16-plus had transitioned compared to 28.2% (n=2,645) of young people with no SSI. Chi-square tests confirmed these differences were statistically significant (p \u0026lt; 0.05).\u003c/p\u003e\n\u003cp\u003eFemales were overrepresented among those with a CAMHS appointment 16-plus, although gender differences in transition were minimal, meaning more males than females transitioned. Small but significant differences were observed in CAMHS 16-plus appointments by ethnicity, with minimal ethnic differences in transition. Across both the CAMHS 16‑plus and transition sub‑samples, young people who had lived in the most deprived areas were over‑represented, with the pattern most pronounced among YPRCS and YPLA. This was especially evident in transition, where 45.6% of YPRCS and 56.8% of YPLA had lived in the most deprived quintile compared with 33.2% of those with no SSI. Rates of CAMHS appointments 16-plus and transition also varied by health board. Chi-square tests indicated that all sociodemographic differences described were statistically significant.\u003c/p\u003e\n\u003cp\u003eComparing between sub‑samples, neurodevelopmental condition diagnosis codes were present at noticeably higher percentages in the transition sub-sample than in the CAMHS sub-sample for all cohorts. Between cohorts, YPRCS 16-plus and YPLA 16-plus consistently showed lower percentages of those with severe mental illness, anxiety, depression and eating disorder diagnosis codes, and higher percentages of those with alcohol or drug use, self‑harm and neurodevelopmental diagnosis codes when compared with young people with no SSI. Chi‑square tests indicated that these differences were statistically significant in both sub‑samples, except for anxiety, eating disorder and alcohol use codes in the transition sub‑sample\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003cstrong\u003e: Sociodemographic characteristics and diagnosis code presence for young people with a CAMHS appointment age 16+, by social services involvement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e*frequencies \u0026lt; 10 supressed\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e4\u003c/strong\u003e\u003cstrong\u003e\u003cspan id=\"_Toc214445682\"\u003e: Sociodemographic characteristics and diagnosis code presence for young people who transitioned from CAMHS to AMHS before age 21, by social services involvement\u003c/span\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e*frequencies \u0026lt; 10 supressed\u003c/p\u003e\n\u003ch3\u003eAssociation between social services involvement and mental health service use\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eLogistic regression models adjusted for gender, ethnicity, deprivation, health board and diagnosis codes in health records showed that the odds of a CAMHS appointment 16-plus for YPRCS were approximately two times higher (OR = 2.03, 95% CI [1.86, 2.21], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001), and for YPLA were nearly four times higher (OR = 3.74, 95% CI [3.07, 4.56], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001) compared with young people with no SSI. However, the odds of transition for YPRCS were only modestly higher (18%) (OR = 1.18, 95% CI [1.03, 1.35], \u003cem\u003ep\u003c/em\u003e = 0.020) and YPLA did not have significantly higher odds of transition compared to those with no SSI (OR = 1.27, 95% CI [0.96, 1.98], \u003cem\u003ep\u003c/em\u003e = 0.095). Stratified logistic regression models were used with both sub-samples to enable between-cohort comparison and the results are presented in Table 5 and Table 6 showing odds ratios (OR), P-values and 95% confidence intervals (CI).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e5\u003c/strong\u003e\u003cstrong\u003e: Odds ratios (OR) and 95% confidence intervals (CI) for having a CAMHS appointment 16+ by social services involvement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e* = \u003cem\u003ep\u003c/em\u003e\u0026lt;0.05\u003c/p\u003e\n\u003cp\u003e** =\u003cem\u003e\u0026nbsp;p\u003c/em\u003e\u0026lt;0.01\u003c/p\u003e\n\u003cp\u003e*** = \u003cem\u003ep\u003c/em\u003e\u0026lt;0.001\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003cstrong\u003e: Odds ratios (OR) and 95% confidence intervals (CI) for transitioning from CAMHS to AMHS by social services involvement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e*p\u0026lt;0.05\u003c/p\u003e\n\u003cp\u003e**p\u0026lt;0.01\u003c/p\u003e\n\u003cp\u003e***p\u0026lt;0.001\u003c/p\u003e\n\u003ch3\u003eSociodemographic covariate effects within cohorts\u003c/h3\u003e\n\u003cp\u003eTable 5 shows that gender was only associated with CAMHS 16-plus and transition for young people with no SSI, increasing odds of CAMHS for females by 7% and reducing the odds of transition for females by 15% (Table 6). Having lived in the 40% most deprived areas in Wales increased the odds of CAMHS 16-plus for young people with no SSI but was not associated with CAMHS 16-plus for YPRCS and YPLA. Deprivation was also not associated with transition for any cohort. Although descriptive statistics suggested young people who had CAMHS 16-plus and transitioned were disproportionately from the most deprived areas in Wales, logistic regression indicated only a modest deprivation effect on CAMHS 16-plus and only for those with no SSI (OR = 1.21, 95% CI [1.15, 1.27], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001).\u003c/p\u003e\n\u003ch3\u003eDiagnosis code effects within cohorts\u003c/h3\u003e\n\u003cp\u003eDiagnosis codes present in health records were the strongest predictors of CAMHS 16-plus but stratified models indicated that diagnosis codes increased the odds of CAMHS 16-plus most strongly for young people with no SSI. Differences between cohorts were minimal for alcohol or drug use diagnosis codes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eConsidering the ORs for transition, diagnosis codes had substantially weaker associations than for CAMHS 16-plus across all cohorts, with the exception of alcohol or drug use diagnosis codes for YPLA. For example, among young people with no SSI, a neurodevelopmental condition diagnosis code increased the odds of CAMHS 16-plus 49-fold but only fourfold for transition. For YPLA, a self-harm code increased CAMHS 16-plus odds by over sixfold but was not significantly associated with transition. The only diagnosis code category with a relatively stronger association with transition than CAMHS 16-plus was alcohol or drug use for YPLA (OR = 2.24, 95% CI [1.01, 5.00], \u003cem\u003ep\u003c/em\u003e = 0.048).\u003c/p\u003e\n\u003ch3\u003eFormal comparison of CAMHS vs transition effects\u003c/h3\u003e\n\u003cp\u003eTo test whether the effect of social services involvement differed between CAMHS 16-plus and transition outcomes relative odds ratios (RORs) were calculated. Two pooled logistic regression models were fitted using a stacked dataset including both outcomes. Interaction terms between outcome type and SSI cohort were significant for YPRCS (\u0026chi;\u0026sup2; = 36.30, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; coefficient = \u0026minus;0.63, 95% CI: \u0026minus;0.83 to \u0026minus;0.42) and YPLA (\u0026chi;\u0026sup2; = 31.14, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001; coefficient = \u0026minus;1.21, 95% CI: \u0026minus;1.63 to \u0026minus;0.78). Relative odds ratios (RORs) confirmed these differences showing that the odds of transition were 47% lower than the odds of CAMHS 16-plus for YPRCS (ROR = 0.53, 95% CI: 0.44\u0026ndash;0.66) and 70% lower for YPLA (ROR = 0.30, 95% CI: 0.20\u0026ndash;0.46).\u003c/p\u003e\n\u003cp\u003eTo examine whether the effect of diagnosis code presence differed by social services involvement further interaction tests were carried out. These showed no significant differences in transition odds for those with severe mental illness, anxiety, depression, eating disorders, substance use, or self‑harm diagnosis codes. In contrast, neurodevelopmental conditions diagnoses showed a significant interaction (\u0026chi;\u003csup\u003e2\u003c/sup\u003e = 13.84, p = 0.001). The ratio of odds ratios (ROR) for neurodevelopmental conditions was 0.55 (95% CI: 0.40-0.77) for YPRCS indicating approximately 45% lower odds of transition compared to those with no SSI. The ROR for neurodevelopmental conditions did not differ significantly between YPLA and those with no SSI (ROR = 0.59, 95% CI: 0.29-1.21). Overall, diagnosis codes strongly predicted CAMHS 16-plus but had weaker and less consistent associations with transition, except for neurodevelopmental condition diagnosis codes among YPRCS.\u003c/p\u003e\n\u003ch3\u003eSub analysis of sample with no diagnosis codes present\u003c/h3\u003e\n\u003cp\u003eGiven the known increased mental health needs of YPRCS and YPLA\u0026nbsp;(Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003), it could be hypothesised that \u0026nbsp;these cohorts would be more likely to transition to AMHS even when diagnosis codes were not present. To examine this, the logistic regression models were repeated on a sub-sample of young people with none of the investigated diagnosis codes recorded in health records age 11-to-17. Whilst the odds of CAMHS 16-plus were six times higher for YPRCS (OR = 5.66, 95% CI [4.87, 6.58], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001) and 11 times higher for YPLA (OR = 11.12, 95% CI [8.05, 15.36], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001), the odds of transition were not significantly different for YPRCS (OR = 1.37, 95% CI [0.97, 1.94], \u003cem\u003ep\u003c/em\u003e = 0.074) or YPLA (OR = 1.33, 95% CI [0.65, 2.73], \u003cem\u003ep\u003c/em\u003e = 0.441) compared with young people with no SSI. This indicates that YPRCS and YPLA were not more likely to transition when diagnosis codes were not present.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe transition from Child and Adolescent Mental Health Services (CAMHS) to Adult Mental Health Services (AMHS) is widely recognized as a high-risk period for service disengagement. The findings from this study show that this risk may be especially great for young people involved with social services, raising pressing questions about equity and continuity of care. The results indicated that, compared to young people with no social services involvement 16-plus (SSI), higher proportions of young people receiving care and support (YPRCS) and looked after (YPLA) 16-plus in Wales had mental health related diagnosis codes recorded in their health records age 11-to-17. In addition, YPRCS and YPLA 16-plus had higher prevalence of multiple (two or more) diagnosis codes investigated present in their health records. This finding aligns with longstanding evidence of the increased mental health needs and complex support requirements of young people involved with social services, (Bronsard et al., 2016; Dubois-Comtois et al., 2021; Engler et al., 2022; Ford et al., 2018; Meltzer et al., 2003). It is therefore unsurprising that in this study YPRCS and YPLA 16-plus in Wales were over-represented in CAMHS and had significantly higher odds of having a CAMHS appointment 16-plus after adjusting for sociodemographic factors and diagnosis codes present in health records.\u003c/p\u003e \u003cp\u003eHowever, among young people who had a CAMHS appointment 16-plus, those involved with social services 16-plus were far less represented in transition to AMHS. Transitions from childhood to adulthood, from living in care to leaving care, and from CAMHS to AMHS are well-established points of instability that may heighten the risk of worsening mental health as well as disengagement from services (Dunn, 2017; Singh et al., 2010). At a time when their support needs may be greatest, concurrently being in receipt of local authority care and support had only a modest effect of increasing the likelihood of transitioning to AMHS, and being looked after had no significant effect when compared with those not involved with social services. Overall, only 29% of young people who had a CAMHS appointment 16-plus also had an AMHS appointment by age 21 years. These results suggest that the well-documented reduction in mental health support at transition age may be most pronounced for young people concurrently involved with social services. However, given that social services involvement encompasses a diverse range of circumstances and levels of need, the factors influencing transition are unlikely to be uniform and future research should seek to identify the specific mechanisms operating within this broad category.\u003c/p\u003e \u003cp\u003eDescriptive statistics suggested that higher proportions of males and young people who had lived in the most deprived areas in Wales had transitioned to AMHS. These kinds of proportional representations may reflect what practitioners experience as they support young people at transition age, and thus feed into narratives around which young people do or do not transition. However, stratified logistic regressions showed that gender and deprivation did not independently influence the likelihood of transition when adjusting for social care status, ethnicity, health board area of residence and diagnosis codes in health records. This demonstrates the importance of moving beyond descriptive patterns to understand the drivers of transition outcomes\u003c/p\u003e \u003cp\u003ePrevious research has shown that having mental health related diagnoses, prescription medications, or periods of inpatient care are all strongly associated with mental health service usage (Appleton et al., 2023; Singh et al., 2010). In this study, diagnosis codes in health records captured these clinical needs. Relative odds ratios showed that diagnosis codes in health records had a much weaker effect on the likelihood of transition than on the likelihood of CAMHS 16-plus across all cohorts. Diagnosis codes did not have statistically significantly different effects on the odds of transition between cohorts, except for neurodevelopmental condition diagnosis codes which had a weaker association with the likelihood of transition for YPRCS than for young people with no SSI. Given the rising prevalence of ADHD and ASD diagnoses among young people in the UK (Department of Health and Social Care, 2025; McKechnie et al., 2023) it is concerning to see that whilst these diagnoses increase the likelihood of transition to adult mental health services nearly fourfold for young people not concurrently involved with social services, the same types of diagnosis codes only increase the likelihood of transition for young people concurrently in receipt of care and support from their local authority less than twofold. To plan effective support for young people transitioning from CAMHS, further research is needed to explore how different diagnoses interact with social services involvement to shape continuity of mental health care.\u003c/p\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eStrengths and limitations\u003c/h2\u003e \u003cp\u003eTo our knowledge this is the first study to explore how the effects of sociodemographic factors and recorded diagnosis codes on mental health service usage may differ for young people who are concurrently involved with social care services as they transition to adulthood. Using population level administrative data enables more comprehensive mapping of young people\u0026rsquo;s mental health service access nationally than using localised service or survey data. However, administrative data are subject to restrictions and biases, including how information is captured, how variables are conceptualised, and errors in recording. Large sample numbers add strength to statistical testing and generalisability, but in this study the sample is substantially reduced by mental health usage sub-samples, social services involvement cohorts, and low numbers of some predictor variables such as minority ethnic categories, health board areas of residence, and specific diagnosis codes.\u003c/p\u003e \u003cp\u003eAdapting existing data to research questions, however, can be considered a more ethical approach compared to using resources to gather or re-gather study specific data. This is particularly important when considering the data of vulnerable populations and those who may be over-researched.\u003c/p\u003e \u003cp\u003eThe WDSD uses a gender code reflecting sex as GP-registered, which can be different than sex at birth, but does not reflect non-binary or subjective gender identities. It was not possible to maintain six categories of ethnic identities for all the analyses, and even where six categories were used, these cannot reflect heterogeneity within each group.\u003c/p\u003e \u003cp\u003eThe read code lists used for types of diagnosis have been validated through a rigorous process combining expert clinical input with systematic review of coding standards. Relevant diagnostic, medication, and treatment codes identified are cross-checked against established classification systems and tested for accuracy using real-world data to ensure validity and completeness. Clinical diagnoses provide more complete case ascertainment than surveys or cohort studies, which can be susceptible to selection bias, low recruitment and high attrition. However, routinely collected data commonly underestimate prevalence as not all individuals in need will access services, and some conditions might not be recognised or recorded. Conversely, including GP records and using a simple binary variable for the presence or absence of diagnosis codes means that individuals presenting once in relation to a particular condition will be identified with presence of that diagnosis code, even with no subsequent contacts or diagnoses. Thus, whilst in many instances diagnosis code records will reflect a formal diagnosis, in some cases they do not. Diagnoses may therefore be both over-represented and under-represented when using diagnosis codes as a proxy. Finally, there are no validated measures of the clinical severity in these data.\u003c/p\u003e \u003cp\u003eThe study dataset sub-sample contains a wide range of mental health service access experiences from young people with a single CAMHS appointment that was never attended, through to those with frequent and ongoing contact with CAMHS and AMHS across the five-year period. Further research is needed to examine different patterns or typologies of mental health service access and how these relate to the predictor and outcome variables of interest.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study of Welsh administrative health data concurs with previous research showing that young people involved with social services have an increased risk of mental ill-health and complex mental health needs. Whilst these groups are over-represented in CAMHS 16-plus, the same is not true of their transition to AMHS. Across all cohorts, most young people accessing CAMHS 16-plus did not transition to AMHS, highlighting a significant gap in continuity of care. The \u0026lsquo;cliff edge\u0026rsquo; of formal mental health support at transition age in Wales is therefore of considerable concern, particularly given the rising levels of mental ill-health among children, and the financial pressures facing public services.\u003c/p\u003e \u003cp\u003eFurthermore, the finding that young people concurrently involved with social services have distinct patterns of mental health service use at transition suggests that they may experience the greatest reduction in support at this critical point. This study also indicates that established predictors of mental health transitions such as diagnoses, medication and inpatient care, may have different impacts on transition for young people concurrently involved with social services. Greater attention from mental health services to the social care contexts of young people is essential to ensure their needs are identified, continuity of care is maintained, and they are supported to flourish through their transition to adulthood and beyond.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eAMHS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eAdult mental health services\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCAMHS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eChild and adolescent mental health services\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eSSI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eSocial services involvement\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eYPRCS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eYoung people receiving care and support\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eYPLA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eYoung people looked after\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate: \u003c/strong\u003eThis research was approved by the Information Governance Review Panel (IGRP) at the Secure Anonymised Data Linkage (SAIL) Databank which reviews proposals as ethical and in the public interest. Approval was granted under SAIL project 1418. Once IGRP approval is in place further institutional ethical approval is not required. The routinely collected administrative data used were fully anonymised and accessed within the SAIL Trusted Research Environment. No identifiable human participants were involved and therefore informed consent was not required. All procedures were conducted in accordance with the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication:\u003c/strong\u003e N/A\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials:\u003c/strong\u003e The data that support the findings of this study are available from the Secure Anonymised Data LinkageDatabank. The data are accessible via a two-stage application process consisting of scoping and governance review to assess user and project approvals Apply to work with the data - SAIL Databank.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests: \u003c/strong\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding: \u003c/strong\u003eThis PhD study was funded by the Economic and Social Research Council and supported by the CASCADE Partnership which receives funding from Health and Care Research Wales.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor’s contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLR conceptualised the study, developed the methods, conducted the data analyses and wrote the manuscript. SW, DW and KS supported the conceptualisation of the study, the development of methods, the data analysis and read and approved the final manuscript.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eALAM, S., O\u0026rsquo;HALLORAN, S. \u0026amp; FOWKE, A. 2024. What are the barriers to mental health support for racially-minoritised people within the UK? A systematic review and thematic synthesis. \u003cem\u003eThe Cognitive Behaviour Therapist,\u003c/em\u003e 17.\u003c/li\u003e\n\u003cli\u003eAPPLETON, R., CANAWAY, A., TUOMAINEN, H., DIELEMAN, G. C., GERRITSEN, S., OVERBEEK, M., MARAS, A., VAN BODEGOM, L., FRANIĆ, T., DE GIROLAMO, G., MADAN, J., MCNICHOLAS, F., PURPER-OUAKIL, D., SCHULZE, U., TREMMERY, S. \u0026amp; SINGH, S. 2023. Predictors of transitioning to adult mental health services and associated costs: a cross-country comparison. \u003cem\u003eBMJ Mental Health,\u003c/em\u003e 26\u003cstrong\u003e,\u003c/strong\u003e article e300814.\u003c/li\u003e\n\u003cli\u003eAPPLETON, R., ELAHI, F., TUOMAINEN, H., CANAWAY, A. \u0026amp; SINGH, S. P. 2020. \u0026quot;I\u0026apos;m just a long history of people rejecting referrals\u0026quot; experiences of young people who fell through the gap between child and adult mental health services. \u003cem\u003eEuropean child \u0026amp; adolescent psychiatry,\u003c/em\u003e 3\u003cstrong\u003e,\u003c/strong\u003e 401\u0026ndash;413.\u003c/li\u003e\n\u003cli\u003eARNETT, J. J. 2000. Emerging adulthood: A theory of development from the late teens through the twenties. \u003cem\u003eAmerican Psychologist,\u003c/em\u003e 55\u003cstrong\u003e,\u003c/strong\u003e 469.\u003c/li\u003e\n\u003cli\u003eBRONSARD, G., AUQUIER, P. \u0026amp; BOYER, L. 2016. Links between early child maltreatment, mental disorders, and cortisol secretion anomalies. \u003cem\u003eJournal of Physiology-Paris,\u003c/em\u003e 110\u003cstrong\u003e,\u003c/strong\u003e 448\u0026ndash;452.\u003c/li\u003e\n\u003cli\u003eBUTTERWORTH, S., SINGH, S. P., BIRCHWOOD, M., ISLAM, Z., MUNRO, E. R., VOSTANIS, P., PAUL, M., KHAN, A. \u0026amp; SIMKISS, D. 2017. \u0026lt;em\u0026gt;Child and Adolescent Mental Health\u0026lt;/em\u0026gt; | ACAMH Journal | Wiley Online Library. \u003cem\u003eChild and Adolescent Mental Health,\u003c/em\u003e 22.\u003c/li\u003e\n\u003cli\u003eBYWATERS, P., BRADY, G., SPARKS, T., BOS, E., BUNTING, L., DANIEL, B., FEATHERSTONE, B., MORRIS, K. \u0026amp; SCOURFIELD, J. 2015. Exploring inequities in child welfare and child protection services: Explaining the \u0026lsquo;inverse intervention law\u0026rsquo;. \u003cem\u003eChildren and Youth Services Review,\u003c/em\u003e 57\u003cstrong\u003e,\u003c/strong\u003e 98\u0026ndash;105.\u003c/li\u003e\n\u003cli\u003eCOPELAND, W. E., WOLKE, D., SHANAHAN, L. \u0026amp; COSTELLO, E. J. 2015. Adult Functional Outcomes of Common Childhood Psychiatric Problems. \u003cem\u003eJAMA Psychiatry,\u003c/em\u003e 72.\u003c/li\u003e\n\u003cli\u003eCRENNA-JENNINGS, W. \u0026amp; HUTCHINSON, J. 2020. Access to child and adolescent mental health services in 2019. Education Policy Institute.\u003c/li\u003e\n\u003cli\u003eDALSGAARD, S., MCGRATH, J., \u0026Oslash;STERGAARD, S. D., WRAY, N. R., PEDERSEN, C. B., MORTENSEN, P. B. \u0026amp; PETERSEN, L. 2020. Association of Mental Disorder With Subsequent Educational Achievement. \u003cem\u003eJAMA Psychiatry,\u003c/em\u003e 77.\u003c/li\u003e\n\u003cli\u003eDEPARTMENT OF HEALTH AND SOCIAL CARE 2025. Independent review into mental health conditions, ADHD and autism: terms of reference. Available at https://www.gov.uk/government/publications/independent-review-into-mental-health-conditions-adhd-and-autism-terms-of-reference. Accessed [9 Mar 2026].\u003c/li\u003e\n\u003cli\u003eDORAN, C. M. \u0026amp; KINCHIN, I. 2017. A review of the economic impact of mental illness. \u003cem\u003eAustralian Health Review,\u003c/em\u003e 43\u003cstrong\u003e,\u003c/strong\u003e 43\u0026ndash;48.\u003c/li\u003e\n\u003cli\u003eDUBOIS-COMTOIS, K., BUSSI\u0026Egrave;RES, E.-L., CYR, C., ST-ONGE, J., BAUDRY, C., MILOT, T. \u0026amp; LABB\u0026Eacute;, A.-P. 2021. Are children and adolescents in foster care at greater risk of mental health problems than their counterparts? A meta-analysis. \u003cem\u003eChildren and Youth Services Review,\u003c/em\u003e 127.\u003c/li\u003e\n\u003cli\u003eDUNN, V. 2017. Young people, mental health practitioners and researchers co produce a Transition Preparation Programme to improve outcomes and experience for young people leaving Child and Adolescent Mental Health Services (CAMHS. \u003cem\u003eBMC Health Services Research,\u003c/em\u003e 17\u003cstrong\u003e,\u003c/strong\u003e 12.\u003c/li\u003e\n\u003cli\u003eENGLER, A. D., SARPONG, K. O., VAN HORNE, B. S., GREELEY, C. S. \u0026amp; KEEFE, R. J. 2022. A systematic review of mental health disorders of children in foster care. \u003cem\u003eTrauma, Violence, \u0026amp; Abuse,\u003c/em\u003e 23\u003cstrong\u003e,\u003c/strong\u003e 255\u0026ndash;264.\u003c/li\u003e\n\u003cli\u003eFLEDDERJOHANN, J., ERLAM, J., KNOWLES, B. \u0026amp; BROADHURST, K. 2021. Mental health and care needs of British children and young people aged 6\u0026ndash;17. \u003cem\u003eChildren and Youth Services Review,\u003c/em\u003e 126\u003cstrong\u003e,\u003c/strong\u003e 106033.\u003c/li\u003e\n\u003cli\u003eFORD, T., VOSTANIS, P., MELTZER, H. \u0026amp; GOODMAN, R. 2018. Psychiatric disorder among British children looked after by local authorities: Comparison with children living in private households. \u003cem\u003eBritish Journal of Psychiatry,\u003c/em\u003e 190\u003cstrong\u003e,\u003c/strong\u003e 319\u0026ndash;325.\u003c/li\u003e\n\u003cli\u003eH\u0026Auml;GGMAN-LAITILA, A., SALOKEKKIL\u0026Auml;, P. \u0026amp; KARKI, S. 2018. Transition to adult life of young people leaving foster care: A qualitative systematic review. \u003cem\u003eChildren and Youth Services Review,\u003c/em\u003e 95.\u003c/li\u003e\n\u003cli\u003eHALE, D. R., BEVILACQUA, L. \u0026amp; VINER, R. M. 2015. Adolescent health and adult education and employment: a systematic review. \u003cem\u003ePediatrics,\u003c/em\u003e 136\u003cstrong\u003e,\u003c/strong\u003e 128\u0026ndash;140.\u003c/li\u003e\n\u003cli\u003eISLAM, Z., FORD, T., KRAMER, T., PAUL, M., HARLEY, K., WEAVER, T., MCLAREN, S. \u0026amp; SINGH, S. P. 2016. Mind how you cross the gap! Outcomes for young people who failed to make the transition from child to adult services: The TRACK study. \u003cem\u003eBJPsych Bulletin,\u003c/em\u003e 40\u003cstrong\u003e,\u003c/strong\u003e 142\u0026ndash;148.\u003c/li\u003e\n\u003cli\u003eJOHNSON, D., DUPUIS, G., PICHE, J., CLAYBORNE, Z. \u0026amp; COLMAN, I. 2018. Adult mental health outcomes of adolescent depression: a systematic review. \u003cem\u003eDepression and Anxiety,\u003c/em\u003e 35\u003cstrong\u003e,\u003c/strong\u003e 700\u0026ndash;716.\u003c/li\u003e\n\u003cli\u003eJONES, K. H., FORD, D. V., THOMPSON, S. \u0026amp; LYONS, R. A. 2019. A profile of the SAIL databank on the UK secure research platform. \u003cem\u003eInternational journal of population data science,\u003c/em\u003e 4.\u003c/li\u003e\n\u003cli\u003eJONES, R. 2012. The effectiveness of interventions aimed at improving access to health and mental health services for looked after children and young people: A systematic review. \u003cem\u003eFamilies, Relationships and Societies,\u003c/em\u003e 1\u003cstrong\u003e,\u003c/strong\u003e 71\u0026ndash;85.\u003c/li\u003e\n\u003cli\u003eKESSLER, R. C., BERGLUND, P., DEMLER, O., JIN, R., MERIKANGAS, K. R. \u0026amp; WALTERS, E. E. 2007. Age of onset of mental disorders: a review of recent literature. \u003cem\u003eCurrent Opinion in Psychiatry,\u003c/em\u003e 20\u003cstrong\u003e,\u003c/strong\u003e 359\u0026ndash;364.\u003c/li\u003e\n\u003cli\u003eLEIJDESDORFF, S., VAN DOESUM, K. T. M., POPMA, A., KLAASSEN, R. \u0026amp; VAN AMELSVOORT, T. 2017. The impact of parental mental illness on children and adolescents: a systematic review. \u003cem\u003eEuropean Child \u0026amp; Adolescent Psychiatry,\u003c/em\u003e 26\u003cstrong\u003e,\u003c/strong\u003e 869\u0026ndash;879.\u003c/li\u003e\n\u003cli\u003eLYONS, R. A., JONES, K. H., JOHN, G., BROOKS, C. J., VERPLANCKE, J. P., FORD, D. V., BROWN, G. \u0026amp; LEAKE, K. 2009. The SAIL databank: linking multiple health and social care datasets. \u003cem\u003eBMC Medical Informatics and Decision Making,\u003c/em\u003e 9.\u003c/li\u003e\n\u003cli\u003eMCKECHNIE, D., O\u0026rsquo;NIONS, E., DUNSMUIR, S. \u0026amp; PETERSEN, I. 2023. Attention-deficit hyperactivity disorder diagnoses and prescriptions in UK primary care, 2000\u0026ndash;2018: population-based cohort study. \u003cem\u003eBJPsych Open,\u003c/em\u003e 9(4).\u003c/li\u003e\n\u003cli\u003eMCKENNA, S., DONNELLY, M., ONYEKA, I. N., O\u0026rsquo;REILLY, D. \u0026amp; MAGUIRE, A. 2021. Experience of child welfare services and long-term adult mental health outcomes: a scoping review. \u003cem\u003eSocial psychiatry and psychiatric epidemiology,\u003c/em\u003e 56\u003cstrong\u003e,\u003c/strong\u003e 1115\u0026ndash;1145.\u003c/li\u003e\n\u003cli\u003eMCLAREN, S., BELLING, R., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., HOVISH, K., ISLAM, Z., WHITE, S., SINGH, S. P., MCLAREN, S., BELLING, R., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., HOVISH, K., ISLAM, Z., WHITE, S. \u0026amp; SINGH, S. P. 2013. \u0026lsquo;Talking a different language\u0026rsquo;: an exploration of the influence of organizational cultures and working practices on transition from child to adult mental health services. \u003cem\u003eBMC Health Services Research 2013 13:1,\u003c/em\u003e 13.\u003c/li\u003e\n\u003cli\u003eMELEIS, A. I., SAWYER, L. M., IM, E., MESSIAS, D. K. H. \u0026amp; SCHUMACHER, K. 2010. Transition theory. \u003cem\u003eTransitions theory: middle-range and situation specific theories in nursing research and practice. New York: Springer Publishing Company\u003c/em\u003e\u003cstrong\u003e,\u003c/strong\u003e 52\u0026ndash;83.\u003c/li\u003e\n\u003cli\u003eMELTZER, H., GATWARD, R., CORBIN, T., GOODMAN, R. \u0026amp; FORD, T. 2003. \u003cem\u003eThe mental health of young people looked after by local authorities in England\u003c/em\u003e, London: The Stationery Office.\u003c/li\u003e\n\u003cli\u003ePITCHFORTH, J., FAHY, K., FORD, T., WOLPERT, M., VINER, R. M. \u0026amp; HARGREAVES, D. S. 2019. Mental health and well-being trends among children and young people in the UK, 1995\u0026ndash;2014: analysis of repeated cross-sectional national health surveys. \u003cem\u003ePsychological medicine,\u003c/em\u003e 49\u003cstrong\u003e,\u003c/strong\u003e 1275\u0026ndash;1285.\u003c/li\u003e\n\u003cli\u003ePRINCE, M., PATEL, V., SAXENA, S., MAJ, M., MASELKO, J., PHILLIPS, M. R. \u0026amp; RAHMAN, A. 2007. No health without mental health. \u003cem\u003eThe Lancet,\u003c/em\u003e 370.\u003c/li\u003e\n\u003cli\u003ePUBLIC HEALTH ENGLAND 2019. Health Profile for England 2019. Available at https://www.gov.uk/government/publications/health-profile-for-england-2019. Accessed [9 Mar 2026]. \u003cem\u003eIn:\u003c/em\u003e ENGLAND., P. H. (ed.). London.\u003c/li\u003e\n\u003cli\u003ePUBLIC HEALTH WALES 2020. Mental Wellbeing in Wales 2020. Available at: https://phw.nhs.wales/services-and-teams/observatory/data-and-analysis/mental-wellbeing-in-wales-2020/. [Accessed 9 Mar 2026]. \u003cem\u003eIn:\u003c/em\u003e WALES, P. H. (ed.). Cardiff.\u003c/li\u003e\n\u003cli\u003eREUPERT, A., MAYBERY, D., NICHOLSON, J., GOPFERT, M. \u0026amp; SEEMAN, M. V. 2022. Children of parents with a mental illness: a systematic review of interventions. \u003cem\u003eInternational Journal of Mental Health Nursing,\u003c/em\u003e 31\u003cstrong\u003e,\u003c/strong\u003e 6\u0026ndash;25.\u003c/li\u003e\n\u003cli\u003eRIDLEY, M., RAO, G., SCHILBACH, F. \u0026amp; PATEL, V. 2020. Poverty, depression, and anxiety: causal evidence and mechanisms. \u003cem\u003eScience,\u003c/em\u003e 370.\u003c/li\u003e\n\u003cli\u003eSINGH, S. P., PAUL, M., FORD, T., KRAMER, T., WEAVER, T., MCLAREN, S., HOVISH, K., ISLAM, Z., BELLING, R. \u0026amp; WHITE, S. 2010. Process, outcome and experience of transition from child to adult mental healthcare: Multiperspective study. \u003cem\u003eThe British Journal of Psychiatry,\u003c/em\u003e 197\u003cstrong\u003e,\u003c/strong\u003e 305\u0026ndash;312.\u003c/li\u003e\n\u003cli\u003eSINGH, S. P. \u0026amp; TUOMAINEN, H. 2015. Transition from child to adult mental health services: Needs, barriers, experiences and new models of care. \u003cem\u003eWorld Psychiatry,\u003c/em\u003e 14\u003cstrong\u003e,\u003c/strong\u003e 358\u0026ndash;361.\u003c/li\u003e\n\u003cli\u003eTARREN-SWEENEY, M. A. V., A. EDS., 2013. \u003cem\u003eMental health services for vulnerable children and young people: Supporting children who are, or have been, in foster care\u003c/em\u003e, Routledge.\u003c/li\u003e\n\u003cli\u003eTRETHEWEY, S. P., MATHEWS, F., RUSSELL, A. \u0026amp; NEWLOVE-DELGADO, T. 2023. Mental health of children and young people aged 5-16 in England: socio-demographic and clinical characteristics associated with support and service contact. \u003cem\u003eEuropean Psychiatry,\u003c/em\u003e 66\u003cstrong\u003e,\u003c/strong\u003e S582\u0026ndash;S582.\u003c/li\u003e\n\u003cli\u003eWELSH GOVERNMENT. 2014. \u003cem\u003eWelsh Index of Multiple Deprivation 2011 to 2019 \u003c/em\u003e[Online]. Available: https://www.gov.wales/welsh-index-multiple-deprivation-2011-2019 [Accessed].\u003c/li\u003e\n\u003cli\u003eWELSH GOVERNMENT 2022. Transition and handover from children\u0026rsquo;s to adult health services: Guidance. Available at: https://www.gov.wales/transition-and-handover-childrens-adult-health-services/ [Accessed 9 Mar 2026]. Cardiff: Welsh Government.\u003c/li\u003e\n\u003cli\u003eWELSH GOVERNMENT 2024. Pupil Level Annual School Census (PLASC) Census day: 20 January 2026. Available at: https://www.gov.wales/sites/default/files/publications/2026-01/technical-completion-notes-pupil-annual-school-census-jan-2026.pdf. [Accessed 9 Mar 2026].\u003c/li\u003e\n\u003cli\u003eWELSH GOVERNMENT. 2025a. \u003cem\u003eMental health and wellbeing strategy 2025 to 2035 | GOV.WALES \u003c/em\u003e[Online]. Available: https://www.gov.wales/mental-health-and-wellbeing-strategy-2025-2035 [Accessed].\u003c/li\u003e\n\u003cli\u003eWELSH GOVERNMENT. 2025b. \u003cem\u003eMental health statistics: interactive dashboard \u003c/em\u003e[Online]. Available: https://www.gov.wales/mental-health-statistics-interactive-dashboard [Accessed 09 Mar 2026].\u003c/li\u003e\n\u003cli\u003eZANEVA, M., GUZMAN‑HOLST, C., R., A. \u0026amp; BOWES, L. 2022. The impact of monetary poverty alleviation programs on children\u0026rsquo;s and adolescents\u0026rsquo; mental health: a systematic review and meta‑analysis. \u003cem\u003eJournal of Adolescent Health,\u003c/em\u003e 71\u003cstrong\u003e,\u003c/strong\u003e 147\u0026ndash;156.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 6 are available in the supplementary files section\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-health-services-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bhsr","sideBox":"Learn more about [BMC Health Services Research](http://bmchealthservres.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/BHSR/default.aspx","title":"BMC Health Services Research","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Transition, mental health, CAMHS, AMHS, young people, social services involvement, administrative data","lastPublishedDoi":"10.21203/rs.3.rs-9105839/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9105839/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBackground\u003c/p\u003e\n\u003cp\u003eTransitions from Child and Adolescent Mental Health Services (CAMHS) to adult services (AMHS) are often complex and negative experiences for young people and those supporting them. Young people involved with social services are at greater risk of mental ill-health but little is known about their mental health service transitions.\u003c/p\u003e\n\u003cp\u003eMethods\u003c/p\u003e\n\u003cp\u003eThis retrospective population-based cohort study used Secure Anonymised Information Linkage (SAIL) data to examine whether young people in Wales with CAMHS contact transitioned to AMHS. Social care status was the primary exposure. Gender, ethnicity, deprivation, health board, and mental health related diagnosis codes in health records were included as covariates. Stratified logistic regression compared odds of CAMHS with odds of transition, and estimated differences in transition by social services involvement.\u003c/p\u003e\n\u003cp\u003eResults\u003c/p\u003e\n\u003cp\u003eResults showed young people receiving care and support (YPRCS) and young people looked after (YPLA) had higher prevalence of diagnosis codes in health records than those with no social services involvement. YPRCS and YPLA were more than twice and nearly four times as likely to access CAMHS (OR = 2.03, 95% CI [1.86, 2.21], \u003cem\u003ep\u003c/em\u003e\u0026lt; .001; OR = 3.74, 95% CI [3.07, 4.56], \u003cem\u003ep\u003c/em\u003e \u0026lt; .001). However, likelihood of transition was much smaller for YPRCS (OR = 1.18, 95% CI [1.03, 1.35], \u003cem\u003ep\u003c/em\u003e = 0.020) and not statistically significantly different for YPLA (OR = 1.27, 95% CI [0.96, 1.98], \u003cem\u003ep\u003c/em\u003e = 0.095) when compared with their peers with no social services involvement. Recorded diagnosis codes were more strongly associated with CAMHS access than transition to AMHS. Associations between diagnosis codes and CAMHS were weaker among those involved with social services. Neurodevelopmental condition diagnosis codes influenced transition differently according to social services involvement.\u003c/p\u003e\n\u003cp\u003eConclusions\u003c/p\u003e\n\u003cp\u003eDespite higher levels of diagnosis codes in health records and greater CAMHS access, young people involved with social services were only marginally or no more likely to transition to AMHS. This suggests potential discontinuity in care. The findings highlight the importance of considering how social services involvement interacts with diagnostic profiles in shaping transition outcomes. Further research is needed to explore mechanisms underlying these differences and to inform targeted support strategies.\u003c/p\u003e","manuscriptTitle":"Transitions from child and adolescent to adult mental health services for young people involved with social care services in Wales: A national population-based retrospective cohort study using linked administrative data","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-22 08:26:28","doi":"10.21203/rs.3.rs-9105839/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-24T23:32:36+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-24T09:14:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"30832231355514201779843485843673910780","date":"2026-04-20T06:52:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"221489106746612373841877200116252082337","date":"2026-04-15T17:15:47+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-15T06:20:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-13T08:32:33+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-24T11:39:19+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-20T07:47:12+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Health Services Research","date":"2026-03-20T07:41:04+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-health-services-research","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bhsr","sideBox":"Learn more about [BMC Health Services Research](http://bmchealthservres.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/BHSR/default.aspx","title":"BMC Health Services Research","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7ae25b6d-42ac-40ac-8878-0e49bd9ccea8","owner":[],"postedDate":"April 22nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-22T08:26:30+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-22 08:26:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9105839","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9105839","identity":"rs-9105839","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}