Millennium Cohort Study Research Articles

School connectedness as a protective factor between childhood adversity and adolescent mental health outcomes.

School connectedness may offset mental health risks associated with childhood adversity. The present study examined the potential protective effects of school connectedness against childhood adversity when predicting adolescent mental health outcomes in 9,964 individuals (51% female, 81% white) from the Millennium Cohort Study. Structural equation models were fitted to examine the longitudinal relationships between childhood adversity, school connectedness, and adolescent mental health. Childhood adversity was a risk factor, predicting greater internalizing and externalizing problems and lower levels of positive mental health. School connectedness was a promotive factor as it predicted fewer mental health problems and greater positive mental health. Furthermore, school connectedness at age 11 was protective against childhood adversity when predicting internalizing and externalizing problems at age 14. That is, students with a history of adversity who felt more connected to school were less likely to exhibit internalizing and externalizing symptoms than those who felt less connected to school. Only school connectedness at age 11 was protective against childhood adversity, indicating that feeling connected to school at younger ages may disrupt processes linking childhood adversity to adolescent mental health. Schools should foster students' feelings of connectedness to protect vulnerable individuals and benefit all pupils' mental health.

Understanding young caring in the UK pre- and post-COVID-19: Prevalence, correlates, and insights from three UK longitudinal surveys

BackgroundDespite increasing awareness of young carers in recent years, there remains a significant gap in our understanding of both the prevalence and the characteristics of young carers. Our study aims to address this gap by examining the impact of the COVID-19 pandemic on the prevalence and characteristics of young carers. MethodsThis research utilised data from three UK longitudinal surveys: the UK Household Longitudinal Study (UKHLS), the COVID Social Mobility and Opportunities (COSMO) study, and the Millennium Cohort Study (MCS). We focused on adolescents aged 16–18, and examined two pre-COVID (UKHLS and MCS) and two post-COVID (UKHLS and COSMO) samples. ResultsThe prevalence of young carers increased from 8.0 % pre-COVID to 9.8–11.9 % since COVID. Young carers were more commonly found in single-parent and socioeconomically disadvantaged households, with a higher prevalence of young carers in homes where parents were out of paid employment or held lower educational qualifications. Young carers were also more likely to reside in deprived areas. Most young carers engaged in low-intensity caring (< 10 h/week), but post-COVID there was an increase in high-intensity caring (10 + hours/week), predominantly assumed by young female carers. The primary recipients of care were parents, followed by grandparents and siblings, with no change in the care recipient type since COVID. ConclusionThis study showed an increase in the prevalence of young carers, particularly those providing high-intensity care, since the onset of the COVID pandemic. Further, young carers were more likely to come from socioeconomically disadvantaged households and areas. Given the potential impacts that young caring can have on young peoples’ lives, it is imperative that support for young carers is increased, particularly for those facing multiple disadvantages. In tandem, services that support adult health and social care need to play a key role in identifying young carers.

Longitudinal Associations Between Physical Health Conditions in Childhood and Attention-Deficit/Hyperactivity Disorder Symptoms at Age 17

ADHD is associated with lower birth weight, but also with obesity in childhood. Findings on the direction of this association are mixed. This study investigated the relationship between weight and ADHD from birth across development. We used data from the Millennium Cohort Study (MCS), collected at seven time points between 9 months and 17 years. ADHD diagnosis status and scores on the Strength and Difficulties Questionnaire (SDQ) were used to create an ADHD group and a control group. Random intercept cross-lagged panel models were conducted in females (N=4051) and males (N=3857) to examine bidirectional associations between BMI z-scores and SDQ scores between ages three and 17. Analyses were adjusted for common risk factors for ADHD and obesity, such as sex assigned at birth, multiple births, and ADHD medication status. Children in the ADHD group were significantly lighter at birth than the control group (t(5674)=2.65, 95% CI [0.02, 0.14] p=.008) and significantly more likely to have obesity at age five onwards (OR range 1.57-2.46, RR range 0.98-2.29). Path analyses conducted separately for males and females showed that higher ADHD symptoms in females at ages 7, 11 and 14 significantly predicted higher BMI z-score at ages 11, 14 and 17 respectively. In males, this association was only seen between age 11 and 14 (β= 0.07; 95% CI, 0.04-0.10, p < .001). Results suggest that interventions for children with ADHD, and their parents, should begin as soon as possible, prenatally if possible. Developmental sex differences should be considered.

Utilizing machine learning to predict participant response to follow-up health surveys in the Millennium Cohort Study

The Millennium Cohort Study is a longitudinal study which collects self-reported data from surveys to examine the long-term effects of military service. Participant nonresponse to follow-up surveys presents a potential threat to the validity and generalizability of study findings. In recent years, predictive analytics has emerged as a promising tool to identify predictors of nonresponse. Here, we develop a high-skill classifier using machine learning techniques to predict participant response to follow-up surveys of the Millennium Cohort Study. Six supervised algorithms were employed to predict response to the 2021 follow-up survey. Using latent class analysis (LCA), we classified participants based on historical survey response and compared prediction performance with and without this variable. Feature analysis was subsequently conducted on the best-performing model. Including the LCA variable in the machine learning analysis, all six algorithms performed comparably. Without the LCA variable, random forest outperformed the benchmark regression model, however overall prediction performance decreased. Feature analysis showed the LCA variable as the most important predictor. Our findings highlight the importance of historical response to improve prediction performance of participant response to follow-up surveys. Machine learning algorithms can be especially valuable when historical data are not available. Implementing these methods in longitudinal studies can enhance outreach efforts by strategically targeting participants, ultimately boosting survey response rates and mitigating nonresponse.

Challenges in using data on fathers/partners to study prenatal exposures and offspring health.

Paternal exposures (and other non-maternal factors) around pregnancy could have important effects on offspring health. One challenge is that data on partners are usually from a subgroup of mothers with data, potentially introducing selection bias, limiting generalisability of findings. We aimed to investigate the potential for selection bias in studies using partner data.We characterise availability of data on father/partner and mother health behaviours (smoking, alcohol, caffeine and physical activity) around pregnancy from three UK cohort studies: the Avon Longitudinal Study of Parents and Children (ALSPAC), Born in Bradford and the Millennium Cohort Study. We assess the extent of sample selection by comparing characteristics of families where fathers/partners do and do not participate. Using the association of parental smoking during pregnancy and child birthweight as an example, we perform simulations to investigate the extent to which missing father/partner data may induce bias in analyses conducted only in families with participating fathers/partners.In all cohorts, father/partner data were less detailed and collected at fewer timepoints than mothers. Partners with a lower socio-economic position were less likely to participate. In simulations based on ALSPAC data, there was little evidence of selection bias in associations of maternal smoking with birthweight, and bias for father/partner smoking was relatively small. Missing partner data can induce selection bias. In our example analyses of the effect of parental smoking on offspring birthweight, the bias had a relatively small impact. In practice, the impact of selection bias will depend on both the analysis model and the selection mechanism.

How can health practitioners support children and young people in social care?

Abstract Health practitioners are responsible for safeguarding, delivering preventive and responsive care for children and young people in social care settings. However, drivers and determinants that place children in care settings lie outside of the doctor’s office. In fact, primary care health practitioners often deal with children and young people who face or are at risk of severe neglect, abuse or vulnerability outside the care system. Recent findings from the UK’s Millennium cohort study have repeatedly demonstrated that social disadvantage in the early years accounts for more than 2/3 of health inequalities in adolescence. Professor Saxena is a practising GP with 25 years of experience of inner-city primary care and global expert on primary care of children. She will present dilemmas posed in clinical settings for health practitioners and professionals working in a broad range of areas including public health, education settings and social care. What responsibility do health professionals have when they encounter children living in social care settings? What should they be especially alert to? How can they make a difference to reduce health inequality faced by young people growing up in the care system.

How does family support mitigate the impact of childhood adversity on adolescent mental health?

Abstract Background Childhood adversity is associated with increased risk of developing mental health problems over the life course. We investigated how family emotional support in adolescence may mitigate the impact of childhood adversity on adolescent mental health. Methods We used longitudinal data from the UK Millennium Cohort Study on 9,269 children followed to age 17 years. Individuals were assigned to exposure trajectories characterised using group-based trajectory models that included measures of poverty and family adversities (parental mental ill health, parental alcohol misuse and domestic violence) up to age 14 years. Using counterfactual causal mediation analysis and four-way decomposition modelling approach, we evaluated how perceived family emotional support at age 14 (measured using the three-item Short Social Provisions Scale) influences the association of childhood adversity trajectories on mental health at ages 14 and 17, assessing the relative contributions of mediation and interaction simultaneously. Results Compared with children experiencing low adversity and poverty, those exposed to childhood adversity were almost three times more likely to experience poor mental health (RR 2·99, 95% CI 2 ·41 to 3·57) at age 14 and age 17 (RR 2·58, 95% CI 2·09 to 3·06). Perceived emotional support mitigates up to 18% (95% CI: 9% to 26%) of this effect at age 14, and 13% (95% CI: 5% to 22%) at age 17, mainly due to interactive mechanisms. Conclusions Childhood adversity has a strong effect on mental health, which is partially mitigated by perceived family support in adolescence. Policies that support positive family functioning may be particularly beneficial for children who have experienced adversity. Key messages • Socio-economic – associated mental health problems in adolescence differed based on levels of family support. • Approximately 20% of the social inequality in mental health outcomes in adolescence may be mitigated or reduced by improving levels of family support.

Exploring the association between low birthweight and adolescent mental health in the United Kingdom

Abstract Low birthweight (LBW) is associated with a range of medical conditions including behavioural and emotional problems over the life course. This study aimed to explore the association between LBW and internalising and externalising mental health problems (MHP) in adolescents born in the United Kingdom between 2000 and 2002. Data were analysed for 10,485, 9,228, and 8,402 Millennium Cohort Study participants at ages 11, 14 and 17, respectively. Internalising and externalising MHP were measured using the Strengths and Difficulties Questionnaire (SDQ). LBW was subcategorised by size for gestational age and prematurity. Crude and adjusted associations between LBW subgroup and internalising and externalising MHP were estimated using binary logistic regression. Internalising MHP were more prevalent in LBW than NBW participants at ages 11 (14.2% vs 10.3%, p &amp;lt; 0.001), 14 (19.4% vs 13.7%, p &amp;lt; 0.001), and 17 (parent-report: 20.3% vs 14.8%, p &amp;lt; 0.001; self-report: 25.7% vs 20.1%, p &amp;lt; 0.001), respectively. Externalising MHP were more common amongst LBW participants at ages 11 (9.4% vs 7.3%, p = 0.02) and 14 (9.0% vs 6.6%, p = 0.01). Compared with NBW peers, the odds of internalising (males: aOR 2.18, 95% CI 1.21-3.73, p = 0.007; females: aOR 2.17, 95% CI 1.22-3.66, p = 0.005) and externalising MHP (males: aOR 2.01, 95% CI 1.05-3.60, p = 0.014; females: aOR 2.58, 95% CI 1.14-5.24, p = 0.016) were significantly increased in male and female premature small for gestational age (SGA) participants at age 11. Externalising MHP were significantly increased amongst premature SGA females aged 14 (aOR: 2.53, 95% CI 1.01-5.66, p = 0.016). Internalising MHP were significantly increased in premature SGA females at age 17 based on parent (aOR: 2.41, 95% CI 1.38-4.22, p = 0.005) and self-report (aOR: 1.95, 95% CI 1.14-3.31, p = 0.014). Amongst LBW individuals, premature SGA females have the greatest risk of internalising and externalising MHP during adolescence, thus identifying an important target group for intervention. Key messages • This study provides evidence for the longitudinal association between low birth weight and mental health in adolescents born in the United Kingdom in the 21st century. • Stratification of low birth weight individuals by size for gestational age and prematurity enabled the identification of a subgroup of adolescents for mental health promotion.

Punishment and Reward Sensitivity in Risk-Taking as Potential Mechanisms Explaining the Relationships Between Childhood Callous-Unemotional Traits and Adolescent Substance Use in a Longitudinal Cohort Study Sample.

Childhood callous-unemotional (CU) traits are associated with a neurocognitive response style of high reward and low punishment sensitivity, which may make these children particularly vulnerable to substance misuse. However, the mechanisms explaining the link between CU traits and substance use are poorly understood. This study investigated the mediating influences of reward and punishment sensitivity on the association between childhood CU traits and adolescent substance use. Using data from the UK Millennium Cohort Study, mediation analyses were conducted to investigate the potential indirect effects of age 14 reward and punishment sensitivity in risk-taking on the relationships between age 11 CU traits and alcohol, cannabis, and other illicit drug use at age 17. No direct effects of CU traits on substance use were found when accounting for gender, baseline alcohol use, poverty, emotional symptoms, conduct problems, hyperactivity, and verbal ability at age 11. Indirect effects of increased reward sensitivity on the relationship between CU traits and increased use were seen for alcohol, cannabis, and other drugs. There was a significant indirect effect of reduced punishment sensitivity on the relationship between CU traits and increased alcohol use, but not cannabis or other substance use. Findings suggest that reward and punishment sensitivity may have independent effects on decision-making processes contributing to adolescent substance use. Prevention and early intervention for substance use should consider modifying intervention strategies to fit the needs of adolescents with a callous interpersonal style and a neurocognitive profile characterized by a high drive for rewards and low risk aversion.

Progressing "Positive epidemiology": A cross-national analysis of adolescents' positive mental health and outcomes during the COVID-19 pandemic.

"Positive epidemiology" emphasizes strengths and assets that protect the health of populations. Positive mental health refers to a range of social and emotional capabilities that may support adaptation to challenging circumstances. We examine the role of positive mental health in promoting adolescent health during the crisis phase of the COVID-19 pandemic. We used four long-running Australian and UK longitudinal cohorts: Childhood to Adolescence Transition Study (CATS; analyzed N=809; Australia); Longitudinal Study of Australian Children (LSAC) - Baby (analyzed N=1,534) and Kindergarten (analyzed N=1,300) cohorts; Millennium Cohort Study (MCS; analyzed N=2,490; UK). Measures included: (Pre-pandemic exposure): Positive mental health (parent-reported, 13-15 years) including regulating emotions, interacting well with peers, and caring for others; and pandemic outcomes: psychological distress, life satisfaction, and sleep and alcohol use outside of recommendations (16-21 years; 2020). We used two-stage meta-analysis to estimate associations between positive mental health and outcomes across cohorts, accounting for potential confounders. Estimates suggest meaningful effects of positive mental health on psychosocial outcomes during the pandemic, including lower risk of psychological distress (Risk Ratio [RR]=0.83 95%CI=0.71, 0.97) and higher life satisfaction (RR=1.1, 95%CI=1.0, 1.2). The estimated effects for health behaviors were smaller in magnitude (sleep: RR=0.95, 95%CI=0.86, 1.1; alcohol use: RR=0.97, 95%CI=0.85, 1.1). Our results are consistent with the hypothesis that adolescents' positive mental health supports better psychosocial outcomes during challenges such as the COVID-19 pandemic, but relevance for health behaviors is less clear. These findings reinforce the value of extending evidence to include positive health states and assets.

Patterns of early childhood adversity and neighborhood deprivation predict unique challenges in adolescence: A UK birth cohort study.

This study applies a comprehensive bioecological perspective to address a significant gap in the childhood adversity literature by employing latent profile analysis to examine the impact of diverse combinations of early childhood adversities and protective factors on adolescent psychosocial and behavioral outcomes. Drawing from the United Kingdom's Millennium Cohort Study (N = 19,444), we identified eight unique profiles of early childhood adversity and protective factors. These profiles provide a nuanced understanding of adversity combinations and allow for differentiation between groups with similar profiles. Latent profile membership was a significant predictor of all adolescent outcome variables, indicating that profiles differed significantly from one another on psychosocial and behavioral outcomes (Wald values ranged from 10.10-623.22; p < .001). Some findings support the cumulative risk model, indicating that exposure to multiple early adversities increases the likelihood of adverse outcomes. However, we also found that specific adversities, such as parental psychopathology, parental alcohol use, and neighborhood deprivation, uniquely impact adolescent outcomes. This study highlights the necessity for tailored interventions and policies to support children with distinct early life experiences, emphasizing the importance of addressing both cumulative and specific adversities at multiple levels to prevent psychosocial and behavioral problems in adolescence.

More Than Just Treats? Effects of Grandparental Support for Children Growing up in Adversity.

This study examined whether grandparental support is a protective factor for children's socio-emotional development in the context of adversity. Using longitudinal data from the Millennium Cohort Study, we investigated the effects of grandparental support across development in children with and without adverse childhood experiences (ACEs). Socio-emotional development was assessed with the Strengths and Difficulties Questionnaire when children were aged 3 years (N=10,186), 5 years (N=10,412) and 7 years (N=10,551). Parent-reported grandparental childcare, coresidence and financial help were assessed and parents reported on the occurrence of five ACEs: physical and emotional abuse assessed with the Straus' Conflict Tactics Scale, parental mental illness assessed with the Kessler scale, domestic violence and parental separation. We found that children with relatively higher levels of ACEs showed more prosocial behaviour and less externalizing problems when they received grandparental care compared to non-grandparental (in)formal care, but only at age 3. By age 7, children with higher levels of ACEs receiving grandparental care showed less prosocial behaviour and more externalizing problems. In addition, grandparental financial support at age 3 was related to more externalizing problems. Post-hoc analyses showed that internalizing and externalizing behaviours at age 5 were related to an increased probability of grandparental childcare at age 7, indicating that children's socio-emotional problems trigger grandparental support. Our findings point to a protective effect of grandparental care on children's socio-emotional development at age 3. Our results highlight the importance of going beyond the nuclear family towards the impact of the wider family network when examining children's socio-emotional development. RESEARCH HIGHLIGHTS: Three-year-old children with high levels of adverse childhood experiences (ACEs) show more prosocial behaviour and less externalizing behaviour when they receive grandparental care. Grandparental care has therefore protective effects on young children's socio-emotional development in the context of family adversity. Grandparents respond to children's socio-emotional problems and family adversity by increasing financial support and involvement in care. These findings underscore the importance of going beyond the nuclear family towards the impact of the wider family network when examining children's socio-emotional development.

Measurement error in pre-school children’s emerging knowledge and skills. A case study from the UK

ABSTRACT Many studies attempt to measure pre-school children’s emerging knowledge and skills. While such data are of great value to researchers, it also brings significant measurement challenges. Unfortunately, in most secondary analyses, measurement error in these data is ignored. This paper consequently presents a case study documenting the magnitude and structure of measurement errors in three-year-olds Bracken School Readiness test scores within the UK’s Millennium Cohort Study. The analysis illustrates how measurement error in three-year-olds test scores is substantial, non-classical and varies by factors such as children’s behaviour during the test and the survey fieldworker assigned. It is then shown how ignoring these measurement errors could lead to appreciable changes to one’s understanding of pre-school children’s knowledge and skills. The paper concludes by making recommendations for groups collecting and analysing pre-school assessment data.

Factors associated with mental healthcare utilization among United States military personnel with posttraumatic stress disorder or depression symptoms

Introduction: Although posttraumatic stress disorder (PTSD) and depression are prominent mental health conditions affecting United States service members, only a subset of individuals with these conditions utilize mental healthcare services. Identifying factors associated with mental healthcare utilization may elucidate military subgroups with unmet mental healthcare needs. Methods: Cross-sectional survey data from the 2019–2021 Millennium Cohort Study assessment were used to examine correlates of unmet mental healthcare needs among military personnel who screened positive for PTSD or depression symptoms (n=18,420) using modified Poisson regression models. Data analyses for this study were conducted between 2023 and 2024. Results: Approximately 32–43% of service members reported receiving any mental healthcare in the past 12 months. Hispanic and Asian or Pacific Islander personnel, and those with certain service characteristics (higher pay grade, recent deployment, experienced discrimination) had a lower likelihood of mental healthcare utilization. Female sex, greater symptom severity, experiencing bullying, and other psychosocial factors were associated with greater likelihood of mental healthcare utilization. Conclusions: One-third of service members with PTSD or depression symptoms reported any mental healthcare use, highlighting the need to identify factors that may impede or delay treatment. Racial and ethnic disparities in treatment utilization persist as do differences in utilization by military characteristics. Further research and initiatives are necessary to identify potential service-specific or cultural barriers and provide equitable quality and access to needed mental health services within the Military Health System.

The Relationship between PM2.5, Greenness, and Road Noise Exposures and Children’s Cognitive Performance in England: The Millennium Cohort Study

The Relationship between PM2.5, Greenness, and Road Noise Exposures and Children’s Cognitive Performance in England: The Millennium Cohort Study

Family poverty, neighbourhood greenspace and perinatal outcomes.

The relationship between low income and adverse perinatal outcomes, such as low birth weight and developmental delays, is well established making the search for protective factors important. One such factor may be neighbourhood greenspace. This study elucidates the role of urban neighbourhood greenspace in the relationship between income and perinatal outcomes in a nationally representative birth cohort from the UK. Data on 14 050 infants participating in the initial wave at age 9 months of the Millennium Cohort Study were used (51% male, 20% non-white, 52% living in disadvantaged areas). We tested whether the association between income and perinatal outcomes is moderated by urban greenspace (measured in deciles) before and after adjustments for confounding. The perinatal outcomes included birth weight, gestational age (in days), communication and motor delays. The models were adjusted for the infant's sex and ethnicity, mother's age, education, substance use and mental health as well as area disadvantage and air pollution. Neighbourhood greenspace moderated the association between income and gestational age, even after adjustment for all confounders, b=-0.11, 95% CI (-0.215, -0.004). For births in low-income households, in particular, it was associated with an increase in gestational age by an average of approximately 3 days. However, after adjustment, greenspace was not found to influence birth weight, communication or motor delays at age 9 months. The biophilic design of urban environments is a modifiable factor for improving perinatal outcomes in the UK as urban greenspaces appear to be mitigating the risk of preterm birth associated with family poverty.

Stimulation beliefs, parental reading involvement, and social inequalities in children's language development

Objective: This study investigates whether greater identification of mothers and fathers in different-sex couples with the stimulation dimension of intensive parenting promoted both parents' involvement in reading and benefited children's language development between ages 3 and 5 in the UK. In addition, we explore social class variations. Background: Our study tests the frequently assumed relationship of parenting beliefs about stimulation with language development, and contributes to our understanding of parenting practices and their implications for child development. Method: We draw on a large representative sample of young children from the UK Millennium Cohort Study (N = 8,071) and apply path analyses in the framework of structural equation models. Results: For mothers and fathers, stimulation beliefs partially mediated the relation between parental education and reading frequency. Mothers' and fathers' stimulation beliefs had positive effects on children’s language abilities, but their impact was small compared to the direct associations with parental education. Conclusion: While parental education emerges as a key determinant of children's language development, our study reveals the nuanced role of beliefs about stimulation within intensive parenting, prompting further investigation into the multifaceted nature of parental involvement.

Early emotion regulation developmental trajectories and ADHD, internalizing, and conduct problems symptoms in childhood.

Emotion dysregulation is considered a transdiagnostic factor with importance for a range of neurodevelopmental and mental health issues, including attention deficit hyperactivity disorder (ADHD) symptoms, internalizing problems, and conduct problems. Emotion regulation skills are acquired from early in life and are thought to strengthen gradually over childhood. Children, however, acquire these skills at different rates and slower acquisition may serve as a marker for neurodevelopmental and mental health issues. The current study uses the UK Millennium Cohort Study, a large longitudinal study to evaluate whether developmental trajectories of emotion regulation across ages 3, 5, and 7 predict levels of ADHD symptoms, internalizing problems, and conduct problems at age 7. Both higher initial levels of and slower reductions in emotion dysregulation across ages 3, 5, and 7 predicted higher ADHD symptoms, conduct problems, and internalizing problems at age 7 in both male and female children. Our findings suggest that monitoring trajectories of emotion regulation over development could help flag at-risk children. Additionally, supporting the acquisition of emotion regulation skills in this critical period could be a promising transdiagnostic preventive intervention.

Using linked cohort data to help address residual confounding in analyses of population administrative data

ObjectivesAnalyses of population administrative data can often only be minimally adjusted due to the unavailability of a full set of control variables, leading to bias due to residual confounding. Cohort studies will often contain rich information on potential confounders but may not be sufficiently powered to meaningfully address the research question of interest. We aimed to use linked cohort data to help address residual confounding in analyses of population administrative data. ApproachWe propose a multiple imputation-based approach, introduced through application to simulated data. We then apply this approach to a real-world problem – examining the association between pupil mobility (changing schools at non-standard times) and Key Stage 2 attainment using data from the UK National Pupil Database (NPD). The limited control variables available in the NPD are supplemented by multiple measures of socioeconomic deprivation captured in linked Millennium Cohort Study (MCS) data. ResultsThe proposed approach is observed to perform well when using simulated data. The association between pupil mobility and Key Stage 2 attainment was attenuated after supplementing the NPD analysis with information from linked MCS data, though with a decrease in precision. ConclusionWe have demonstrated the potential of the proposed approach, but more work is required. ImplicationsThe principles underlying this innovative approach are widely applicable: any analysis of administrative data where confounder control is limited by the availability of information could potentially be strengthened by linking a subset of individuals into richer cohort data and leveraging the additional information to inform population-level analyses.

The unequal impacts of the COVID-19 pandemic on young adults' mental health. Predictors of vulnerability and resilience using longitudinal birth cohort data in the UK.

Previous studies have demonstrated deteriorations in young adult mental health during the COVID-19 pandemic, but evidence suggests heterogeneity in the mental health impacts of the pandemic. We sought to identify factors which may predict changes in psychological distress and wellbeing during the COVID-19 pandemic in UK young adults. A total of 2607 young adults from the Millennium Cohort Study were included. Psychological distress and mental wellbeing were measured using the Kessler-6 and Short Warwick-Edinburgh Mental Wellbeing Scale, respectively. Assessment occurred at three timepoints between the ages of 17-19: 2018/19 (pre-COVID Baseline), May 2020 (COVID Wave 1) and September/October 2020 (COVID Wave 2). Latent change score models were used to study change in distress and wellbeing across the study period, as well as the impact of sex, relative family poverty, parental education, preexisting mental health difficulties and perceived social support on these changes. The latent change score models suggested both distress and wellbeing tended to increase across the study period. Being female and in relative poverty predicted greater increases in distress and/or poorer wellbeing. Higher levels of parental education and greater perceived social support were protective against increased distress and associated with improved wellbeing. The impact of the COVID-19 pandemic on UK young adult mental health is complex. We provide further evidence for a distinction between symptoms of poor mental health and wellbeing. Research is urgently needed to assess the long-term impacts of the COVID-19 pandemic on the mental health and wellbeing of young people, particularly in more vulnerable groups.