Social Science & Medicine

Background. Population-representative evidence on social inequalities in youth mental health from an intersectional perspective is limited. We investigated inequalities across multiple mental health outcomes among young adults and examined whether measures operated equivalently across social and intersectional groups. Methods. Cross-sectional analyses were conducted with 9,675 participants aged 23 years from the nationally representative UK Millennium Cohort Study (born 2000-2002). Self-reported outcomes included psychological distress (Kessler-6), probable serious psychological distress/'caseness' (Kessler-6 score [≥]13), psychological wellbeing (Short Warwick-Edinburgh Mental Well-Being Scale; SWEMWBS), past-year self-harm, and lifetime suicide attempts. Measurement invariance analyses assessed comparability of the Kessler-6 and SWEMWBS across groups. Intersectional MAIHDA examined inequalities by sex assigned at birth, ethnic identity/background, sexual identity, and childhood poverty. Young people with lived experience of mental ill-health and/or social discrimination contributed to study design, interpretation, and reporting. Findings. Large inequalities were observed across all outcomes, with the largest disparities by sexual identity, followed by sex assigned at birth, childhood poverty history, and ethnic identity/background. Across intersectional strata, predicted psychological distress ranged from -0.27 to 1.32 SDs, and psychological wellbeing from -0.54 to 0.11 SDs. Predicted prevalence varied 6.99-fold (9.81% to 68.55%) for psychological distress 'caseness', 6.84-fold (8.60% to 58.79%) for self-harm, and 22.10-fold (1.96% to 43.32%) for suicide attempts. Young people at the intersection of sexual minority status, female birth sex, childhood poverty, and some ethnic minority identities/backgrounds experienced particularly poor outcomes. Most inequalities showed cumulative patterns consistent with additive expectations, although some intersectional groups experienced better or worse outcomes than expected. Accounting for measurement differences revealed inequalities that were otherwise undetected. Interpretation. Findings highlight the large extent of mental health inequalities when multiple dimensions of social (dis)advantage are considered, and the need to address both the social conditions generating these inequalities and the methodological assumptions shaping how they are measured and understood.

Background: As medicine grows increasingly technological and scientific, biomedical researchers working to bring new knowledge to bear on clinical practice face a key question: when is a new intervention or treatment ready for clinical use? Because this is both a technical and ethical dilemma, it is crucial to examine the social history of how different approaches to this question emerge and the values or assumptions they embed. Methods: We examine the rise and proliferation of an increasingly common framework for assessing the value of new biomedical data or technology, "actionability," through a computational analysis of published scientific literature referencing this and related terms, including topic modeling and Medical Subject Headings (MeSH) term analysis of over 7000 scientific abstracts indexed in PubMed. Results: We find that actionability, as a term, began appearing more commonly in published literature in the mid-2000s, and proliferated throughout the 2010s and into the 20s. While originally used primarily in research on healthcare quality and implementation, the concept's rise in popularity is ultimately driven by uptake in the fields of clinical genetics and oncology. Conclusions: The adoption of actionability in these fields suggests that actionability as a conceptual framework may be most valuable to areas of translational medicine seeking to make sense of increasing amounts of data and technological innovation with differing levels of scientific validity and clinical utility. Recognizing this value, we also caution that actionability drives our attention primarily towards whether a test or piece of information can lead to action, not whether that action has proven benefits. As clinicians and researchers face difficult questions about how to sort through growing amounts of data to generate knowledge that can have a real impact on patient health, empirical bioethics should play a key role in analyzing the trade-offs and impacts of different approaches.

We analyse the effect of waiting time for elective musculoskeletal treatment on labour market outcomes in England using hospital records for working-age adults who received treatment between April 2015 and March 2023 to Census, tax and benefit records. To address clinical prioritisation and unobserved severity, we leveraged provider-level waiting time as an instrument for individual waiting time and estimated two-stage least squares models adjusted for baseline labour market outcomes, demographic and socioeconomic characteristics, and provider, referral-time and procedure fixed effects. We find that longer waits reduce annual employee earnings, with effects persisting for up to five years. For an additional four weeks of waiting, the largest estimated annual losses range from {pound}301 for osteoarthritis to {pound}487 for other joint disorders. There was no evidence of effects on employment status or benefit receipt, suggesting that delays primarily operate through reduced productivity or hours worked among those who remain employed. Reducing waiting times may not only improve health outcomes but help patients maintain earnings and support economic

Background: Rising rates of young people Not in Education, Employment, or Training (NEET) in the UK have recently coincided with declining youth physical and mental health but no study has asked whether this reflects a growing proportion of young people with health problems (prevalence) or those with health problems becoming more likely to be NEET (penalty). Methods: Using 15 years of Understanding Society data (2009-23), we analysed 15,242 respondents aged 16-24 (66,160 observations). We employed three complementary approaches: descriptive trends, Blinder-Oaxaca-Kitagawa (BO) probit decomposition comparing 2009-2013 and 2019-2023 against a 2014-2018 reference period, and fixed-effects (FE) Poisson models with lagged health status. Exposures included self-reported health conditions or disability (SRHD), psychological distress , diagnosed conditions and socio-demographic factors. Findings: NEET rates were lowest in 2014-18 (10.5-11.5%) and higher in 2009-13 (12-15%) and 2019-23 (15-16%). Higher prevalence of SRHD, psychological distress, diagnosed depression and multimorbidity explained changes in NEET prevalence across both the 2009-13 to 2014-2018 and 2014-18 to 2019-23 periods. No change in penalty was observed for any health variable across periods, except for an increase in the penalty for SRHD between the 2009-13 to 2014-18 periods. Interpretation: Rising NEET rates among UK youth are driven largely by more young people having physical and psychological ill health. Whilst labour market and education accommodations remain important, reducing NEET rates will require reversing the decline in youth health, not just accommodating it.

Background: The decline of mental health among adolescents in the UK over the past two decades has occurred amidst many societal and technological changes making it difficult to identify drivers of these trends. We aimed to identify which aspect of life showed the most rapid deterioration in satisfaction in the UK since 2009 and which aspects have become increasingly correlated with mental health. Methods: Between 2009 and 2024, 15,829 adolescents aged 10-15 rated their satisfaction with life as well as five aspects of life (family, friends, school, schoolwork, appearance), via the UK Household Longitudinal Study. Population-weighted regressions were used to estimate annual summary statistics on these items (means, proportions) and the relative associations between satisfaction items and socioemotional difficulties, as measured by the Strengths and Difficulties Questionnaire (SDQ). Meta-regressions estimated linear changes over time, with interactions testing for heterogeneity across sexes, age-groups, and household income. Results: Mean satisfaction with life dropped over the 16 years of study (B=-0.028, 95% CI=-0.034, -0.022), as did satisfaction with all five aspects of life. Satisfaction with school declined most rapidly (B=-0.027, 95% CI=-0.034,-0.020), becoming accelerated in the most recent 5-year window (2020-24; B=-0.139, 95% CI=-0.178,-0.099). The proportion of dissatisfied individuals rose most rapidly for schoolwork, particularly for females. Satisfaction with appearance became increasingly associated with general life satisfaction (B=0.009, 95% CI=0.005, 0.013, p<.001) and socioemotional difficulties (B=-0.051, 95% CI=-0.073,-0.029, p<.001) over time. Conclusions: Our findings add to international evidence for academic pressure and school environments as plausible drivers of recent declines in life satisfaction, particularly among female adolescents. Future studies should aim to identify the specific aspects of school life that are influencing adolescent mental health and the impact of existing educational policies on student wellbeing should be evaluated.

Background.Respect for autonomy is central to biomedical ethics but is rarely assessed with a structured, reproducible measure. This gap matters as patients, families, and clinicians increasingly use general-purpose language models for health-related questions. We developed the Autonomy Index (AIx) to characterize autonomous agency in clinical-ethics vignettes and compared ratings from trained human reviewers with ratings from current language models. Methods.We reviewed ten decisional-capacity instruments and classified them by target construct into consent capacity, decision capacity, and functional autonomy, identifying a region of autonomous agency (values, deliberation, and enactment) that existing instruments measure least well. The resulting thirteen-item instrument scores four domains: Values Awareness, Factual Understanding, Rational Deliberation, and Intentional Action, on a five-point ordinal scale with an explicit not-applicable option; the External Constraint Index and Support Provided Index are reported separately. Nine models scored 89 clinical vignettes. We selected 50 vignettes by model-score tier and inter-model disagreement, then obtained 173 ratings from 30 clinical and bioethics-informed reviewers and 3,431 valid scores from eight models in July 2026. The prespecified comparison summarized each source within vignette and estimated the paired human-minus-model difference; mixed-effects, domain, agreement, and variance-component analyses were secondary or exploratory. Human data were collected under Harvard Longwood Campus IRB protocol IRB26-0146. Results.Across the selected 50-vignette comparison set, human and model composites were strongly associated (Pearson r = 0.781, P = 2.3 x 10-{superscript 1}{superscript 1}; Spearman {rho} = 0.775). The human mean was 50.2 and the model mean was 39.2, a paired human-minus-model difference of 11.0 index points (95% CI, 6.3 to 15.8; P = 2.6 x 10-; d_z = 0.66). Human means exceeded model means on 36 of 50 vignettes. Bland-Altman limits of agreement were wide (-21.9 to 43.9), indicating substantial case-level variation. Mean differences were positive in all four domains: Factual Understanding, 13.7; Intentional Action, 12.3; Values Awareness, 10.4; and Rational Deliberation, 7.9 points; the joint domain-by-source test did not detect heterogeneity (Wald {chi}{superscript 2}[3] = 2.27, P = 0.52). Vignette content accounted for 80.9% of score variance, whereas source accounted for 0.7%. Models never selected not applicable; human reviewers did so for 22.5% of item responses. Among 19 vignettes with at least four human ratings, 9 had human-minus-model differences greater than 15 points and none had a difference below -15 points. Conclusions.In this deliberately selected 50-vignette set, trained reviewers assigned higher autonomy scores than the tested general-purpose language models on average, while the two sources ranked cases similarly. Wide limits of agreement and variation across well-rated vignettes preclude treating the mean difference as a prediction for an individual case. Because the comparison set was selected using prior model scores, reviewers were a convenience sample, and human ratings are a reference standard rather than ground truth, the findings should be interpreted as evidence of systematic divergence in this sample, not proof that models underestimate patient capacity. External validation in prospectively sampled cases with denser human rating is required before clinical use.

Background: Girls are more likely than boys to experience psychological distress during adolescence. Drivers of this gender gap remain under-studied, despite its large impact on later life and outcomes, especially for women. Explanatory factors can be gendered in their prevalence (differential exposure) and/or in their impact (differential susceptibility). We developed a multi-domain framework of explanatory factors for distress and applied it in four low- and middle-income countries. Methods: We first proposed a framework of explanatory factors for psychological distress grouped into seven domains (biological, cognitive, psychological, behavioural, social, environmental and broader societal). We applied this framework to identify explanatory variables assessed in childhood and adolescence in the Young Lives longitudinal cohort studies in Ethiopia (N=908), India (N=952), Peru (N=634) and Vietnam (N=887). Distress symptoms at 18-19 years were measured with the Strengths and Difficulties Questionnaire Emotional Symptoms scale. Gendered prevalence of each explanatory factor was tested with Mann-Whitney tests, and gendered impact with country-specific linear regressions including a gender-by-factor interaction term. Exploratory boosted-tree machine-learning models, interpreted with SHAP values and permutation feature importance, examined the contribution of these factors to the gender gap in each country. Results: Distress was higher in girls than boys in every country, the gap ranging from 0.62 points in Ethiopia and Vietnam to 1.76 in India and 2.45 in Peru. Some explanatory factors were consistently gendered in prevalence (e.g. girls experienced greater role overload in each country), while others were gendered only in some countries (e.g. study amount was higher for girls in Peru and Vietnam, but not Ethiopia or India). Gendered impacts on distress displayed more variability across countries. For example, higher self-esteem was more protective for boys than girls in India, Peru and Vietnam, but more protective for girls in Ethiopia. The machine-learning models identified important predictors across all framework domains, though their salience varied by country. Conclusions: Our findings underscore the importance of considering multi-domain explanations for mental-health outcomes and observed inequalities in mental health. Large cross-country differences in the gender gap and in the prevalence and influence of explanatory factors caution against generalising drivers of mental-health outcomes and inequalities across contexts.

Background Under-five mortality remains a major public health challenge in Zambia despite substantial reductions over recent decades. Demographic and Health Survey (DHS) data are commonly analysed using Cox proportional hazards models; however, violations of the proportional hazards assumption and unobserved cluster-level heterogeneity may compromise inference. This study compared advanced survival modelling approaches to determine the most appropriate framework for analysing under-five mortality using pooled Zambia DHS data. Methods We conducted a pooled secondary analysis of the 2018 and 2024 Zambia Demographic and Health Survey child recode datasets. The analysis included 9,982 children, among whom 266 died before their fifth birthday. A survey-weighted Cox proportional hazards model was fitted as the reference model and assessed using Schoenfeld residual diagnostics. To address identified methodological limitations, an extended Cox model, a shared-frailty model, and a Bayesian survival model were subsequently fitted. Model performance, assumptions, and the consistency of adjusted hazard ratio estimates were compared across modelling approaches. Results The global Schoenfeld test indicated violation of the proportional hazards assumption ({chi}2=51.26, p=0.002). In the adjusted extended Cox model, male children had higher mortality risk than females (aHR 1.95, 95% CI 1.40-2.73), while children with gestation [&ge;]9 months had lower mortality risk than those with shorter gestation (aHR 0.21, 95% CI 0.14-0.32). Average and large birth size were associated with lower mortality compared with small birth size (aHR 0.47, 95% CI 0.34-0.66 and aHR 0.57, 95% CI 0.39-0.84, respectively). Newborn health checks before discharge were associated with lower mortality (aHR 0.50, 95% CI 0.38-0.66), as was any antenatal care (aHR 0.25, 95% CI 0.09-0.75). A high-risk fertility pattern involving fourth or higher births with an interval [&ge;]24 months was also associated with lower mortality (aHR 0.57, 95% CI 0.41-0.80). The shared-frailty model produced similar estimates, including increased mortality among male children (aHR 1.39, 95% CI 1.08-1.78) and twins or multiple births (aHR 2.74, 95% CI 1.36-5.53). Bayesian estimates were broadly consistent with the extended Cox model. Conclusions Advanced survival models provided a more robust analytical framework than the conventional Cox model for evaluating under-five mortality in nationally representative DHS data. While the extended Cox model accommodated time-varying effects, the shared-frailty model captured unobserved cluster-level heterogeneity, and the Bayesian model confirmed the stability of parameter estimates. Incorporating these complementary approaches can improve the validity of survival analyses and strengthen evidence for child survival research and policy in low- and middle-income settings.

Mental disorders typically emerge before age 25, yet little is known about how childhood human capital and large economic shocks jointly shape mental health into adulthood in low- and middle-income countries (LMICs). We use seven rounds of the Young Lives study (2002-2024), which followed two birth cohorts in Ethiopia, India, Peru, and Vietnam, to examine (i) the long reach of early human capital, proxied by the height-for-age z-score in infancy; (ii) the medium-term association between household food insecurity during the COVID-19 pandemic and adult depressive (PHQ-8) and anxiety (GAD-7) symptoms measured at ages 22 and 29 in 2023-2024; and (iii) whether early human capital buffered the pandemic shock. We estimate value-added, latent-factor, dynamic panel, and mediation models with cluster-robust inference. Early height-for-age strongly predicts adult height (1.6 cm per standard deviation, p < 0.01) but not adult mental health or formal employment. Pandemic food insecurity predicts higher depressive (0.41 points, 0.12 standard deviations) and anxiety (0.41 points, 0.11 standard deviations) symptoms three to four years later (both p < 0.01); per step, the probability of clinically relevant anxiety (GAD-7 of 10 or more; base 7.6 percent) rises by 1.9 percentage points (p = 0.001) and of clinically relevant depression (PHQ-8 of 10 or more; base 5.6 percent) by 1.0 percentage point (p = 0.07). The gradient travels with the persistence of distress and of material hardship, is robust to attrition corrections and bounds, and its causal reading is explicitly bounded, since mental health was first measured during the pandemic itself. Exploratory analyses suggest the gradient differs across countries and may be larger in the older cohort, exposed at about age 26 rather than 19; these contrasts rest on few clusters and are not robust to multiple-comparison adjustment. Early human capital neither predicts adult mental health--a precise null of at most {+/-}0.03 standard deviations--nor moderates the gradient. Protecting household food security during aggregate shocks, especially for young adults, appears central to safeguarding mental health in LMICs.

Background: High rates of young people who are not in education, employment or training (NEET) are a major societal concern in the UK. Whilst other studies have highlighted that adolescent health can predict NEET status in young adulthood, robust and recent longitudinal evidence remains limited. Methods: This study used data from the Millennium Cohort Study, a longitudinal study of people born in the UK in the early 2000s, to estimate the extent to which mental health conditions, physical health conditions and health behaviours during adolescence predict NEET status in early adulthood (median age: 23). Co-occurrence of exposures was also considered and population attributable fractions were calculated to account for differences in exposure prevalence. Results: Among 8,374 young people, 12.5% were NEET at age 23; approximately two thirds were seeking work and one third were economically inactive. Estimates adjusted for demographic factors indicated that multiple health exposures increased risk of being NEET at age 23, with mental health conditions predicting greater risk than physical health conditions and health behaviours. For instance, a longstanding mental health condition more than doubled the risk of being NEET (adjusted relative risk [aRR] = 2.39, 95% CIs = 1.85, 3.09), while autism (aRR = 3.60, 95% CIs = 2.69, 4.83) and ADHD (aRR = 3.25, 95% CIs = 2.38, 4.44) more than tripled the risk. A greater number of reported adolescent mental health conditions was associated with greater risk of being NEET in young adulthood. Obesity predicted being NEET at age 23 (aRR = 1.54, 95% CIs = 1.18, 2.01) and obesity accompanied by a mental health condition further increased risk (aRR = 2.01, 95% CIs = 1.38, 2.93). Follow-up analyses indicated that associations between adolescent mental health and young adult NEET status were more pronounced for females than males and for the economically inactive than those seeking work. Conclusions: Findings indicate that adolescent health, especially mental health, strongly predicts being NEET in early adulthood. Early, integrated health and education interventions may help reduce later educational and labour market disengagement.

Background: Multiple observational studies have reported associations between greater parity and increased CVD risk. Whether these associations reflect causal effects or are confounded by socioeconomic factors remains unclear. Methods: We investigated associations between number of children ever born (NEB) and 16 cardiometabolic traits in up to 172,122 females and 138,390 males in the UK Biobank, and an independent sample of 53,237 UK Biobank spousal pairs. We additionally conducted sex-stratified two-sample Mendelian randomization (MR) and applied a novel spousal MR framework, in which an individual's spouse's genotype was used as the instrumental variable to estimate the causal effect of NEB on cardiometabolic health outcomes, as an approach to minimize bias from horizontal pleiotropy. Results: NEB was associated with multiple cardiometabolic traits in the multivariable regression, even after adjustment for socioeconomic status, with differences in the strength of association observed between males and females. Traditional MR provided evidence that higher NEB causally increases type 2 diabetes risk in females, body mass index (BMI) in both sexes, female basal metabolic rate (BMR) and male body fat percentage but decreases female blood pressure. Spousal MR corroborated positive effects on female BMI and BMR and additionally suggested inverse causal effects on female HDL cholesterol and ApoA1 and male blood glucose. Conclusion: These findings indicate a possible causal relationship between NEB and long-term cardiometabolic health, although causal effects are likely to be small.

Background: Young people who are Not in Education, Employment, or Training (NEET) represent a major public health and societal challenge. Existing evidence has linked adolescent mental health problems and health risk behaviours to NEET but has largely treated NEET as a static, rather than longitudinal outcome and overlooked the combined effects of multiple health conditions. Methods: Using data from 5,262 participants born between 1993 and 2000 in the UK Household Longitudinal Study, this study examined the independent and combined associations of adolescent mental health problems (emotional symptoms, conduct problems, hyperactivity) and health risk behaviours (regular smoking, drug use, alcohol use, and high social media use) with ever-NEET status, NEET chronicity, and NEET trajectories from ages 16 to 24, using modified Poisson, proportional odds, and multilevel logistic regression models, respectively. Findings: All mental health problems were associated with ever-NEET status (RRs 1.24-1.27) and NEET chronicity (ORs 1.41-1.98); emotional symptoms showed a widening disadvantage with age, while the disadvantages associated with conduct problems and hyperactivity remained stable. Among health risk behaviours, regular smoking showed the strongest and most persistent relationships with NEET (ever-NEET RR 1.54; chronicity OR 1.64); drug use was related to ever-NEET status (RR 1.37) and an increasing disadvantage after age 21-22, while alcohol use and social media use showed limited associations. NEET risk generally increased with the number of co-occurring conditions, but for recurrent NEET (three or more occasions), risk was only elevated at three or more co-occurring conditions. Interpretation: Adolescent health exposures were associated with NEET risk during ages 16-24, but the strength and pattern varied by exposure and outcome, offering potential insights into the timing and emphasis of any interventions.

Background Child marriage/cohabitation before 18 years of age disproportionately affects girls worldwide, but its impacts on mental health are poorly understood. Existing research is mostly cross-sectional, hence unable to estimate causal impacts or longer-term outcomes of child marriage. Methods We analysed longitudinal cohort data across Ethiopia, India, Peru, and Vietnam from Young Lives to investigate the relationship between child marriage/cohabitation and life satisfaction and emotional symptoms. We used linear regression, difference-in-differences estimation (DiD), and propensity score matching (PSM) to examine short and longer-term impacts from age 19 to 29. Findings Across countries 5% (Peru) to 20% (India) of girls were child brides. Child marriage/cohabitation predicted lower life satisfaction -- pooled estimates at age 19 (-0.25 [95%CI=0.40,-0.10]), across 19-29 (-0.22 [95%CI=-0.32,-0.13]), and heterogeneity estimates suggest consistently negative impacts across countries. However, child brides have lower life satisfaction at age 8 even before marriage occurs, likely due to pre-existing disadvantages such as poverty that increase child marriage likelihood. DiD and PSM analyses confirm that most of the differences are due to pre-existing vulnerability and there is a small additional impact of child marriage on worse life satisfaction at age 19 (pooled matched estimate -0.09 [95%CI=-0.19, 0.00]). No differences in emotional symptoms were observed, nor any differences by age of marriage within the child-married sample. Interpretation Child brides have lower life satisfaction from childhood through to early adulthood, which is mostly explained by prior disadvantages and risk factors, with a small additional impact of child marriage itself.

Rural-urban fertility differences may reflect unequal distributions of education, household resources, union timing, and regional characteristics rather than an independent effect of residence. This study examined whether the rural-urban gap in near-completed fertility among Nigerian women aged 40-49 persisted after accounting for these compositional factors. Data were drawn from 7,370 women in the 2024 Nigeria Demographic and Health Survey. Children ever born were analyzed using weighted descriptive statistics, residence-specific comparisons, and four survey-weighted Poisson generalized linear models with a log link and cluster-robust standard errors. The weighted mean number of children ever born was 5.53, ranging from 4.68 among urban women to 6.39 among rural women. In the age-adjusted model, rural women had 37% more children than urban women (IRR = 1.37, p < .001), but residence was no longer statistically significant after adjustment for education and household wealth. In the fully adjusted model, women aged 45-49 had more children than those aged 40-44 (IRR = 1.09, p < .001). Secondary and higher education were associated with lower fertility (IRR = 0.92 and 0.82, respectively; p < .001), as was residence in the richest households (IRR = 0.84, p < .001). First cohabitation at ages 20-24 and 25 or older was also associated with lower fertility (IRR = 0.82 and 0.64; p < .001). The rural-urban fertility gap largely reflected socioeconomic, marital-timing, and regional inequalities.

Importance: Preschool children in care are at high risk for psychosocial health concerns. Population-based evidence is limited. Objective: Estimate prevalence of psychosocial health concerns in children in care and not in care, and assess care-status differences stratified by deprivation. Design: Population-based cross-sectional study using 27-30 Month Health Review data from April 2013 to March 2023. Setting: Universal health review program in Scotland. Participants: 7887 children in care and 445 547 children not in care. Exposures: Care status at review, classified as in care or not. Main Outcomes and Measures: Four outcome categories (emotional, behavioral, and/or attentional; personal and/or social; speech, language, and/or communication; and other developmental concerns) plus an aggregate indicator of any of the four. We estimated adjusted odds ratios between children in care and not in care, including variation with deprivation. Models adjusted for sex, age, ethnicity, and deprivation. Results: Psychosocial health concerns were more common in children in care (2290; 29.0%) than children not in care (77 836; 17.5%; relative risk 1.66). Concerns were more common in children in care across all outcomes. The adjusted odds ratio comparing children in care with children not in care for any recorded concern was 1.86 (95% CI, 1.77-1.96). Adjusted odds ratios varied by outcome from 1.57 (95% CI, 1.49-1.66) for speech, language, and/or communication concerns to 2.49 (95% CI, 2.34-2.66) for emotional, behavioral, and/or attentional concerns. Relative inequities between children in care and not in care decreased with increasing deprivation from aOR of 1.58 (95% CI, 1.45-1.72) in the most deprived fifth of areas to 2.61 (95% CI, 2.25-3.03) in the least deprived fifth. Prevalence of any recorded concern increased with deprivation in both care groups. The relative risk comparing the most deprived with least deprived fifth of areas was 1.46 (95% CI, 1.29-1.66) among children in care and higher at 2.34 (95% CI, 2.29-2.40) among children not in care. Conclusions and Relevance: Psychosocial health inequities are evident at an early age between children in care and not in care, and vary with deprivation. Support for children in care and children living in more deprived areas should be prioritized.

Background: Adults with disability tend to need more healthcare and to have less capacity to pay for it. Kenya replaced the National Hospital Insurance Fund (NHIF) with the Social Health Authority (SHA) in October 2024, so KDHS 2022 is the last pre-transition national survey containing these measures. We examined differences by disability status in health insurance coverage, healthcare use, and outpatient payment before the transition, including how insurance coverage varied across disability severity and what could be observed about insurer contributions at the point of care. Methods: Cross-sectional analysis of the KDHS 2022 person recode, covering adults aged 18 and above in the long-questionnaire subsample that carried the disability, insurance and service-use modules. Disability used the Washington Group Short Set (WG-SS). Survey-weighted quasi-Poisson models gave prevalence ratios under three adjustment sets with marginal standardised differences; severity categories were contrasted directly; amounts and payer sources were analysed among those who reported paying, with bounds where payer source is unrecorded. Results: Of 40,197 adults, 6.7% met the WG threshold. Uninsurance was higher among them unadjusted (PR 1.10) and after adjustment for sex, age group and residence (PR 1.10, 95% CI 1.06 to 1.13; +6.7 percentage points), and was attenuated after adding wealth and education (PR 1.01, 95% CI 0.98 to 1.04; 0.6 percentage points, 95% CI -1.3 to 2.5). Outpatient use and hospitalisation were higher in every specification (fully adjusted PR 1.63, 14.0 percentage points, 95% CI 11.7 to 16.3; and PR 2.15, 7.7 percentage points, 95% CI 5.9 to 9.6; both p < 0.001). Among outpatient users with disability, 90.5% reported paying, insured or not. Among respondents who reported paying, insurance was associated with a lower probability that any part of the payment was made in cash (PR 0.89, p < 0.001), while among those who did pay cash the conditional cash amount was higher. Among insured outpatient users with disability who reported paying and had a recorded payer split, 15.1% had a recorded insurer contribution. Because payer information was not collected when no payment was reported, the corresponding proportion among all insured outpatient contacts was bounded between 11.9% and 32.9%. Conclusions: Adults with disability had higher crude uninsurance, but this difference was attenuated to the null after adjustment for wealth and education. They also had substantially higher outpatient use and hospitalisation, which persisted under every adjustment set. Most outpatient users reported making a payment at their last visit regardless of insurance status, while the insurer contribution could only be partially observed because payer information was not collected when no payment was reported. These estimates provide a pre-transition benchmark, against which SHA-era measurement should assess payment at the point of care alongside insurance coverage.

Background: Health visiting in England is a universal service designed to identify family need and offer more support where needed (proportionate universalism). We know little about how local areas deliver services for families facing adversity or their associated outcomes. Methods: Using linked Community Services Data Set and Hospital Episode Statistics data (2018-20) for 57/152 areas, we (a) described health visiting service delivery for children whose mother had a previous hospital admission with mental health issues, alcohol/substance misuse or violence/abuse recorded (study definition of adversity), (b) identified common service delivery patterns using latent class analysis; and (c) tested whether patterns were associated with child and maternal outcomes, overall and in terms of inequalities for families facing adversity. We conducted four experts-by-experience workshops to aid interpretation. Results: Three service delivery patterns for families facing adversity emerged, distinguished primarily by the intensity and age-distribution of contacts. In areas where health visiting was delivered with a high intensity across the whole 0-5 years age range (rather than targeted primarily at babies), children were more likely to meet expected levels of development at age 2-2.5 years (ORadj: 0.79, 95% CI:0.63-0.99) and child inequalities in injury-related hospital admissions and some developmental domains were reduced. Conclusions: We found that a pattern of health visiting aligned with proportionate universalism was associated with mitigating disadvantage for families facing adversity. With limited outcomes and forms of adversity captured in our study, we cannot say which service delivery pattern is best, but our findings evidence the potential of health visiting to reduce inequalities.

ABSTRACT Introduction: Trial evidence in older adults with hypertension suggests that meditation may lower mortality, but this association has not been tested in a nationally representative study with linked-mortality data. This study estimated meditation's association with mortality among US adults, examined variation by family income, and synthesized trial and survey evidence for the population matched to the trial evidence base. Methods: We pooled four National Health Interview Survey cohorts (2002 to 2017) with mortality follow-up through 2019 (N=90,041; 8,402 [9.3%] reported meditation practice in the past 12 months). Doubly robust survey-weighted Cox models combined stabilized inverse-probability weights with covariate adjustment, overall and by income stratum; 15 year risks were standardized using g computation, and probabilistic bias analysis addressed exposure misclassification, including a scenario by income strata. For non Hispanic Black adults [&ge;]55 years with hypertension (n=2,786), Bayesian synthesis combined subgroup estimates with a pooled randomized trial prior. Results: Over 818,248 person years of follow up, 10,027 all cause and 3,040 cardiovascular deaths occurred. Doubly robust hazard ratios (HR) were 1.02 (95% CI 0.90, 1.16) for all cause and 0.96 (95% CI 0.79, 1.16) for cardiovascular mortality. Among individuals below 200% of the federal poverty level, HRs were 0.85 (All cause: 95% CI 0.71, 1.03) and 0.80 (Cardiovascular mortality: 95% CI 0.57, 1.10), compared with 1.10 (95% CI 0.93, 1.30) and 1.02 (95% CI 0.80, 1.30) among those at/above that threshold (interaction p=0.19 and p=0.62). In the subgroup analysis, posteriors under a 50%-discounted trial prior were 0.84 (All-cause mortality: Pr[HR<1]=0.89) and 0.70 (Cardiovascular mortality: Pr[HR<1]=0.96). Conclusions: Meditation practice showed no overall association with mortality after doubly robust adjustment. However, estimates suggesting a protective effect were concentrated among adults below 200% of the federal poverty level, and trial-informed synthesis indicated probable benefit in cardiovascular mortality in the high-risk subgroup most comparable to trial participants. These hypothesis-generating findings motivate further study of effect modification using more detailed exposure measurement.

Background: BMI and height are usually related through mean-based models, which can mask how this relationship varies across the height distribution and across age. Whether BMI relates to height differently at different quantiles, and whether this differs by sex, is not well studied in Southeast Asian children. Methods: We analyzed longitudinal school health data from 32,006 Vietnamese children (16,489 boys, 15,517 girls) aged 5 to 18 with at least 3 annual visits from 2018 to 2025. Using sex-stratified linear quantile mixed models, we estimated the association between BMI-for-age Z-score (BMIZ) and height-for-age Z-score (HAZ) at 5 quantiles, adjusting for a flexible age spline and study site, with a child-level cluster bootstrap for inference, plus a pooled model with a formal sex interaction test. Results: A 1SD increase in BMIZ was linked to higher HAZ at every quantile in both sexes (boys 0.078 to 0.100, girls 0.055 to 0.089, all p < .01), with no significant sex interaction (p = .35), contrasting with a small negative association from a linear mixed model (-0.023). This association did not differ significantly across quantiles within either sex, pointing to a fairly uniform effect across the height distribution itself, but it did vary strongly by age, strongest in early and mid-childhood and reversing to negative by late adolescence in both sexes, so children with overweight or obesity were taller for age through most of childhood but shorter than leaner peers by the oldest ages studied. Within-group variability in HAZ narrowed with age in both sexes, and boys with thinness showed a consistently wider height spread than boys in the other BMI categories through most of childhood, a pattern that was less pronounced in girls. Conclusions: BMI's association with linear growth in Vietnamese children is positive through most of childhood but weakens and reverses by late adolescence, a pattern mean-based models obscure. BMI and height should be interpreted together in light of a child's age, not as one fixed relationship.

Background. Medical imaging is central to modern diagnosis and treatment, yet its contribution to population level survival has rarely been quantified, and the few existing estimates have been interpreted as evidence that imaging directly extends life. Whether national imaging utilisation reflects a distinct clinical mechanism or a broader process of health system development has not been formally tested. Methods. Longitudinal ecological panel study of 38 OECD countries, 2000 to 2024 (950 possible country years). CT and MRI availability and utilisation were linked to life expectancy at birth, treatable and preventable mortality, and seven cause specific mortality outcomes. Two-way fixed-effects models with country and year fixed effects and country clustered standard errors were adjusted sequentially for log GDP per capita, log health expenditure per capita, population aged [&ge;]65 years, physician density and smoking prevalence. Three falsification families were specified: lead-exposure (placebo) models regressing current life expectancy on future imaging utilisation; negative control exposures with no diagnostic-imaging pathway to survival; and wild cluster bootstrap inference. Results. In the principal adjusted model, CT utilisation at a three year lag was associated with higher life expectancy ({beta} = 0.85 years per 100 additional examinations per 1,000 inhabitants, 95% CI 0.16 to 1.54; cluster robust p = 0.015; bootstrap p = 0.034; n = 339, 29 countries). Scanner availability showed weaker, generally non-significant associations. All three falsification analyses failed to support a modality-specific effect. Future CT utilisation predicted current life expectancy as strongly as past utilisation (three year lead {beta} = 0.86, 95% CI 0.15 to 1.57, p = 0.017), consistent with high within-country autocorrelation of the exposure (r = 0.64 at three years). Radiotherapy equipment density, a therapeutic rather than diagnostic technology, showed an equal and more precise association ({beta} = 0.85 per 10 units per million, 95% CI 0.32 to 1.38, p = 0.0017), as did PET scanner density and health expenditure, whereas physician density showed none. Associations were stronger for preventable ({beta} = -43.3 per 100,000, p = 0.0002) than treatable mortality ({beta} = -13.5, p = 0.19), and all were attenuated to the null under country specific linear trends. Conclusions. National CT utilisation is a robust ecological correlate of life expectancy but does not behave as a causal determinant under falsification. It is best understood as a marker of health system capital development. Claims that imaging adds years of life are not supportable on ecological evidence of this kind.