Trends in the long-term impact of childhood visual impairment on health and social outcomes in the UK: a cross-cohort study across three decades of disability-related legislation and policy implementation Authors: Vasiliki Bountziouka, Lisanne Andra Horvat-Gitsels, Mario Cortina-Borja, Jugnoo Sangeeta Rahi Published: 2023 ABSTRACT AbstractBackgroundChildhood vision impairment (VI) can adversely impact health and social outcomes and limit life chances. We investigated whether its adverse impacts into adult life changed during a period in which legislation, policy and services to address inequalities relating to disability were implemented.MethodsCross-cohort study comprising 14 247 participants from the 1946, 1958 and 1970 British birth cohorts (BC). Participants dichotomized as VI at age 15/16 (distance visual acuity was 6/12 or worse in the better-seeing eye) or normally sighted. Associations of childhood VI with health, well-being, socioeconomic and social participation outcomes in mid-adult life were investigated using regression models adjusted for participants’ early life socioeconomic markers and sex. Change in adjusted odds ratios of >10% in the same direction in successive cohorts, or a > 20% difference between 1970BC and one older cohort were considered meaningful.ResultsTrends over time in impacts of childhood onset VI into mid-adult life were complex. This included worsening of odds of poorer physical health (odds ratio 1.47; 95% confidence interval 1.02–2.14), living in unsatisfactory (1.54; 1.03–2.29) or overcrowded (2.34; 1.26–4.06) households, being unemployed (2.19; 1.19–3.97) and not gaining additional educational qualifications during mid-adult life (1.61; 1.08–2.47). By contrast the odds of not participating in some social activities (e.g. seeing friends) improved over time. Associations with other outcomes were unchanged.ConclusionsMany adverse impacts of childhood VI do not appear ameliorated over time by legislation, policies and provision that would have been expected to reduce inequalities. Moreover, some were increased. Childhood VI continues to cast a life-long shadow. Introduction Over the past few decades disability-related public, health, social care and education interventions, policies and legislation have been implemented in UK.1–6 Through these acts and legislations UK government aimed to improve and protect the nation's health, through reducing health inequalities,1 to protect people from discrimination in the workplace and in wider society,2 to protect disabled people against any form of discrimination,3 to make provisions about housing conditions, especially in the case of multiple occupancy,4 to establish a new framework for the education of children requiring special educational provision,5 and to enable direct payments to persons for securing the provision of care services.6 Alongside this, the adoption of the World Health Organization’s holistic framework for conceptualizing disability7 has contributed to changing attitudes to issues affecting those living with disability.8–11 To our knowledge, there has been no published research on whether, how and the extent to which these changes in policy and legislation have improved the lives of those growing up with visual disability. We considered it reasonable to expect that changes in education provision5, disability anti-discrimination legislation including in relation to employment opportunities,2,3 governmental provision and financial assistance via disability living allowance6,12 and benefits of certification as sight impaired or severely sight impaired13 should have improved the lives of children with vision impairment (VI) over time. We hypothesized that this would be discernible in the reduced impact of VI on childhood on long-term health, well-being, social outcomes and life chances.14–16 Utilizing the 1946, 1958 and 1970 British birth cohorts (BC), we showed previously an increase in prevalence of impaired vision childhood VI over a period when social adversity in early life increased and health inequalities widened.17 Building on our previous work,17 we report a cross-cohort investigation across three key decades of societal change of the association between childhood VI and key outcomes in mid-adult life. Methods Timeline Figure 1 details the timeline of the key legislative and policy changes which would be expected to ameliorate the impact of childhood VI in adult life. For example, the landmark Warnock review18 of special educational needs in 1978 informed the subsequent Education Act 1981,5 to enforce appropriate provision for children with special education needs, by which time participants of the 1946BC were on their mid-30s, while participants of the 1970BC were age 10 and therefore transitioned into adolescence through period of implementation of the Education Act. A series of legislations specific to disability3,6,12 were introduced in the 1990s, namely the Disability Living Allowance (1992), the Disability Rights Act (1995) and the Community Care Act (1998). Participants of the 1946BC were in their early-40s when these came into force, while the 1970BC were in on their early-20s and were more likely to experience major life events and transitions,19 such as leaving education, beginning employment and cohabiting, during that time. Provisions relating to housing conditions,4 promoting equality of opportunity,2 regulating health and adult social care services,1 and commitment of the local authorities for ongoing improvement of the economic, environmental and social wellbeing9 of the local populations areas came in force over a decade in 2000s, by which time the youngest cohort were in their 30s to 40s. Timeline of cohort sweeps and societal changes Cohorts We analyzed data from three UK national studies that sampled all births in a single week in England, Scotland and Wales. These comprise (i) the National Survey of Health and Development of 1946 (1946BC)20 with 5 362 singleton participating babies, (ii) the National Child Development Study of 1958 (1958BC)21 with 17 634 participating babies and (iii) the British Cohort Study of 1970 (1970BC)22 with participating 17 287 babies. They are particularly suitable to investigate our hypothesis because each has measures of visual acuity in childhood23 and rich harmonized data on health and social outcomes throughout life. Members of the 1946BC would have been young adults when they experienced key policy changes (see figure 1) implemented in the UK, for example, legislation relating to employment, while members of the 1970BC would have experienced the policy changes earlier in their lives, for example, changes in educational provision. Ethics approval for the cohort studies was given by the UK Medical Research Council, the National Health Service and the London Central Ethics Committees, respectively.20–22 All participants gave informed consent and could withdraw from the study at any point. Childhood visual impairment (exposure) The best-attained distance visual acuity (i.e. with correction if required) was tested using standard Snellen charts at a distance of 20 ft (i.e. 6 m) in each cohort at the ages of 15 or 16 years.23 By this age, 98% of childhood visual impairment would have presented itself.24 Utilizing harmonized data,23 cohort members were classified as either having normal vision or VI, which was defined as having an acuity of 6/12 (i.e. logMAR 0.3) or worse in the better-seeing eye, encompassing the full range of VI from mild VI (below the threshold for driving a car in the UK) to blindness25,26 and which has been shown to be associated with increased mortality27 (Supplementary table S1). Outcomes We investigated outcomes in mid-adult life as new onset VI between childhood and mid-adult life due to new onset eye disorders is uncommon. Therefore, it is reasonable to ascribe and assess the long-term functional impacts of childhood VI and assess any changes over time. We selected those outcomes available at the same age (mid-40s) and wherever possible in each cohort to cover four domains: (i) physical and mental health, (ii) general well-being/satisfaction with life, (iii) socioeconomic position and (iv) social outcomes. Harmonized data by the Cohort and Longitudinal Studies Enhancement Resources (CLOSER) consortium were used to ensure comparability across the three cohorts.28 The selected outcomes, in accordance with the relevant policies and the classification we followed are summarized in table 1. Table 1Coding of outcomes and hypotheses of cohort effect linked to new legislations and societal changesOutcome at mid-40sDescriptionCodingLinked legislation or major societal eventsSelf-reported health statusGeneral state of participant’s healthPoor vs. good2012 Health and Social Care ActLong-standing illnessWhether the participant suffers from any long-term health problem, long-standing illness, infirmity or disability of any kind lasting or expected to last 12 months or moreYes vs. noImproved health careAnxiety/depressionWhether the participant has seen a specialist, been to a hospital, or seen a doctor in the past ten years for feeling low, depressed, sad, or generally anxious or jittery and still has this problemYes vs. noGaining wider acceptance of mental health issues; increased mental health problemsJob satisfactionSatisfaction with current jobPoor vs. good1995 Disability Discrimination Act; 2010 Equality ActAccommodation satisfactionSatisfaction with current accommodationPoor vs. good2004 Housing ActGeneral life satisfactionSatisfaction with the way life has turned out so farPoor vs. good2005 Improving the life chances of disabled people; 2010 Equality ActEducational qualificationHighest NVQ level from an academic or vocational qualificationNone/statutory/extended schooling vs. higher/further education1981 Education ActGained additional higher qualification between 23 and 45 yearsIncrease in NVQ level from an academic or vocational qualification between ages 23/26 and 42/43 yearsNo vs. yes1981 Education ActUnemployedCurrent economic activity is unemployed and seeking workYes vs. no1981 Education Act; 1995 Disability Discrimination Act; 2010 Equality ActOccupational class given current/previous employmentSocial class of current or last job based on Registrar-General's Social Classes (RGSC)Unskilled/partially skilled (class IV or V) vs. Skilled (class IIINM or IIIM) or professional (class I or II)1981 Education Act; 1995 Disability Discrimination Act; 2010 Equality ActWorsened occupational class compared with childhoodDecrease in RGSC class from childhood to adulthood. Childhood’s social class is derived from father’s current or last job when cohort member was 10/11 years old, based on RGSCYes vs. no1981 Education Act; 1995 Disability Discrimination Act; 2010 Equality ActDisability allowanceReceiving disability-related allowances (including employment and support allowance (ESA))No vs. yes1992 Disability Living Allowance; 1996 Community Care (Direct Payments) ActHome ownershipOwnership of homeNo vs. yes2004 Housing act; Improvement of occupational and economic opportunitiesOvercrowded householdMore people in household than rooms at homeYes vs. no2004 Housing ActMarried/partnerCurrent marital statusNo vs. yesIncrease in acceptance of disability (from medical to social model)Seeing friends at least weeklyHow often the participant meets up with friends or relatives sociallyNo vs. yesIncrease in acceptance of disability (from medical to social model)Doing sports at least weeklyFrequency of regular exercise, that is at least once a month for most of the yearNo vs. yes2012 Olympics & Paralympics in UKGoing to pubs, clubs, social activities at least weeklyFrequency of taking part in organizations, attending meetings or events, and/or going for a drink at a pub or clubNo vs. yes1995 Disability Discrimination Act; 2005 Improving the life chances of disabled people; 2010 Equality Act Coding of outcomes and hypotheses of cohort effect linked to new legislations and societal changes Physical and mental health Participants (i) self-reported their general health status (poor vs. good), (ii) whether experiencing a long-standing illness (yes vs. no) and (iii) having seen a doctor/specialist (GP) for being anxious or depressed in the past and still feeling that way now (yes vs. no). General well-being/satisfaction with life Participants reported their overall satisfaction (poor vs. good) with their (i) job, (ii) accommodation and (iii) the way their life had turned out so far. Socioeconomic position Participants reported their (i) highest educational qualification (none/statutory/extended schooling vs. higher/further education), (ii) current unemployment (yes vs. no), (iii) receiving disability allowance (no vs. yes), (iv) home ownership (no vs. yes) and (v) overcrowded household (yes vs. no). Based on the highest educational qualifications obtained by age 23/26 vs. 42/43 years, we derived whether (vi) participants gained additional higher qualifications later in life (no vs. yes). Participants also reported their occupation in current or previous employment, from which (vii) social class (unskilled/partially skilled vs. skilled/professional) was derived using the Registrar-General’s Social Classes classification.29 Based on paternal social class when aged 10/11 years and own social class when aged 42/43 years, we derived whether (viii) participants had a downward social mobility (yes vs. no) since childhood. Social outcomes Participants reported (i) marital status (married/partnership vs. other) and frequency (less than weekly vs. weekly or more) of (ii) friend visits, (iii) doing sports and (iv) going to pubs, clubs or engaging with other social activities. Confounders/covariates Based on prior work by us17,30,31 and others14–16 we included as important confounders key early life markers of social position: maternal education at birth (none/statutory schooling vs. extended/further/higher education) and paternal social class at age 10/11 years (classified as unskilled/semiskilled vs. skilled/professional), and cohort member’s sex (Supplementary table S1). Statistical analysis Chi-squared tests assessed differences in baseline characteristics (sex, maternal education at birth and paternal social class at age 10/11 years) by childhood VI and birth cohort. A logistic regression model for each outcome was fitted to assess its association with VI stratified by cohort and adjusted for sex, childhood socioeconomic position (maternal education and paternal social class). Outcomes ‘marital status’ and ‘long-standing illness’ were also treated as potential confounders and were included in the regression models. Results are shown as adjusted odds ratios (aORs) alongside their 95% confidence intervals (95% CIs). We considered a > 10% difference in same direction (i.e. reduction or increase) in the aOR point estimate in successive cohorts, or a > 20% difference when comparing 1970BC with only one older cohort to represent a meaningful cohort effect. In the absence of any published work in this area that could have provided effects sizes for understanding the impact of policy or changes in societal attitudes, we pragmatically chose these thresholds as appropriate for providing results robust enough to be meaningful for assessing changes in policy or provision. Attrition and missingness were investigated by fitting logistic regression models with them as responses to understand the potential selection bias in carrying out a complete case analysis. Outcomes with more than 10% missing data were excluded from the analysis. All tests were two-sided and the significance level was set at 5%. All analyses were performed in R version 4.1.0.32 Results Study population Altogether 14 247 cohort members (2 296 of 1946BC, 7 752 of 1958BC and 4 199 of 1970BC) had complete visual acuity data at age 15/16 years, childhood social position data and outcomes of interest (figure 2). Flow of participants in the study Some outcomes were only collected in the 1958BC and 1970BC. Baseline characteristics and distribution of outcomes by VI for each cohort are shown in Supplementary tables S3 and S4 respectively. Attrition was higher in men and those with lower childhood social position (Supplementary table S2), who were underrepresented in the complete case analysis. Although missing data for the baseline characteristics and outcomes were in general minimal, 65% of data on social activities were missing in 1958BC, precluding analysis. Impact of childhood VI on outcomes in adulthood over time The adjusted associations (aOR) of childhood VI with the adverse health, well-being, socioeconomic position and social outcomes over time in mid-adult life are shown in table 2. While only some of the associations with adverse outcomes and cohort effects observed were significant at the 5% significance level, broad trends were discernible. Table 2Associations of childhood visual impairment with adverse health and social outcomes at early-40s over the three birth cohortsDomain1946BC aOR (95% CI)1958BC aOR (95% CI)1970BC aOR (95% CI)Association in the youngest cohort vs. either/both older cohorts (i.e. cohort effect) Adverse physical/mental health Poor self-reported health status1.10 (0.75–1.59)1.47 (1.02–2.14)Increased adverse impact Long-standing illness1.37 (1.20–1.56)0.83 (0.57–1.21)Reduced adverse impact Seeing a doctor for anxiety/depression0.74 (0.33–1.49)0.67 (0.39–1.09)1.18 (0.79–1.77)Increased adverse impact Adverse general well-being Not satisfied with job1.48 (0.99–2.20)1.05 (0.70–1.52)0.78 (0.46–1.27)Reduced adverse impact Not satisfied with accommodation1.06 (0.58–1.78)1.54 (1.03–2.29)Increased adverse impact Not satisfied with life in general1.03 (0.67–1.53)0.90 (0.55–1.42)No change Adverse socioeconomic position Completing statutory (compulsory) education only vs. university/college higher/further education1.06 (0.50–2.50)0.82 (0.60–1.12)1.00 (0.70–1.44)No change No educational qualifications vs. additional qualifications between mid-20s and mid-40s0.86 (0.42–1.99)0.70 (0.48–1.03)1.61 (1.08–2.41)Increased adverse impact Unemployed0.95 (0.36–2.07)1.88 (1.13–3.10)2.19 (1.19–3.97)Increased adverse impact Unskilled/partially skilled occupational class using current/previous employment0.80 (0.42–1.52)1.24 (0.81–1.86)0.81 (0.42–1.45)No change Downgrade social class since childhood1.25 (0.66–2.38)1.21 (0.89–1.67)1.01 (0.68–1.49)Reduced adverse impact No disability allowance0.92 (0.54–1.69)0.65 (0.35–0.98)Reduced adverse impact No home ownership0.80 (0.32–1.76)1.13 (0.74–1.70)0.99 (0.62–1.55)No change Living in overcrowded households0.89 (0.56–1.33)2.34 (1.26–4.06)Increased adverse impact Adverse social participation Not married/no partner1.07 (0.53–1.99)0.94 (0.63–1.36)1.03 (0.70–1.48)No change Seeing friends less than weekly1.98 (1.09–3.47)1.08 (0.75–1.55)Reduced adverse impact Doing sports less than weekly1.19 (0.68–2.13)1.18 (0.87–1.60)1.29 (0.87–1.88)No change Going to pubs, clubs, social activities less than weekly1.53 (1.02–2.82)0.78 (0.50–1.25)Reduced adverse impactaOR for sex, childhood socioeconomic position (maternal education and paternal social class), marital status, and illness. We considered a >10% difference in same direction (i.e. reduction or increase) in the aOR point estimate in successive cohorts or, a >20% difference when comparing 1970BC with only one older cohort to represent a meaningful cohort effect. Empty cells indicate that the outcome was not measured in the relevant cohort. Associations of childhood visual impairment with adverse health and social outcomes at early-40s over the three birth cohorts aOR for sex, childhood socioeconomic position (maternal education and paternal social class), marital status, and illness. We considered a >10% difference in same direction (i.e. reduction or increase) in the aOR point estimate in successive cohorts or, a >20% difference when comparing 1970BC with only one older cohort to represent a meaningful cohort effect. Empty cells indicate that the outcome was not measured in the relevant cohort. There was evidence of increasing odds over time of the following adverse outcomes for those with childhood VI: worse general physical health, more likely to see a doctor for anxiety or depression, being dissatisfied with accommodation, living in overcrowded accommodation, unemployment and not gaining any educational qualifications in adult life increasing over time. For example, the odds of VI participants of the 1946BC being unemployed were 0.95 (95% CI 0.36–2.07), rising to 1.88 (1.13–3.10) for the 1958BC and further increasing to 2.19 (1.19–3.97) for the 1970BC. By contrast, as shown in table 2, those in the youngest cohort had lower odds than older cohorts of having a long-standing illness, being unsatisfied with their job (where employed), experiencing downward social mobility, seeing friends less than weekly and having social activities less than weekly. They were also less likely to not receive (i.e. more likely to receive) Governmental financial support (disability living allowance) to which they were entitled for their disability [0.65 (0.35–0.98) for the 1970BC vs. 0.92 (0.54–1.69) for the 1958BC]. Using our thresholds for changes over time in effect sizes, there were no changes in associations between having childhood onset VI and the following: general dissatisfaction with life, not gaining any further/higher educational qualifications after statutory schooling, being unskilled/partially skilled, not owning a home, not being married/having a partner and doing sports less than weekly (table 2). Discussion This cross-cohort investigation spans key decades of legislative, policy and societal changes intended to improve outcomes for people living with disabilities. The findings show a complex picture of trends in adverse impacts of childhood onset VI into mid-adult life on health, socioeconomic participation outcomes. Many adverse impacts do not appear to have been ameliorated by policies and provisions intended to address them. Moreover, some adverse impacts increased over time such as the likelihood of being unemployed, having worse physical health, experiencing anxiety or depression, living in overcrowded accommodation and not pursuing education into mid-adult life. To our knowledge there are no equivalent studies with which we can compare our findings directly. We analyzed harmonized rich data collected longitudinally from three directly comparable UK birth cohorts, assessing outcomes at comparable ages to minimize confounding alongside multivariable analysis that included a priori potential confounders. Nevertheless, the possibility of residual confounding exists, as in any observational study. VI in childhood is uncommon,24,33 which limited the power of the multivariable analysis and thus true associations may have been missed as they either did not reach ‘statistical significance’ or resulted in overlapping corresponding 95% CIs. Therefore, we focused on the magnitude of effect sizes to be indicative of the change that policymakers would consider important. The small selection bias in our study with respect to male sex and lower socioeconomic class, suggests that the point estimates would be larger in their absence. We did not implement multiple imputation procedures as suggested in the CLS’ Guidelines34 because, as shown in figure 2, most missing data occurred in the exposure and the outcome variables. Also, Supplementary table S2 shows that missing data in the covariates are not completely random. So instead, we followed the literature35,36 that shows a complete case analysis is valid if, as in our study, data are not missing completely at random and covariates related to missingness are included in the model. While key legislation and policy changes were implemented during the period covered by our study, we are not able to assess the direct experiences of individual cohort members of the changes in the provision or access resulting from these policies. For example, we are unable to disentangle whether the increase in odds over the decades covered by our study of seeing a doctor for anxiety or depression represents increasing risk of adverse mental health, improved societal acceptability of seeking help for mental health problems, increased access to mental health provision, or a combination of all three. Despite this, some coherent cohort effects are discernible. Our findings of improved odds of some social participation activities are consistent with some improvement in societal acceptance and engagement with disability. It is encouraging to see findings that suggest effective implementation and uptake of some specific measures such as Governmental financial assistance (Disability Living Allowance) for those living with disability. Our findings relating to no change in educational attainment during childhood and worsening odds of gaining additional educational qualifications into mid-adult life for those with childhood VI are notable in light of the significant changes that have taken place in education policy in the UK. Although special educational provision for children and young people with disabilities dates from the late 1800s, it was the Warnock Committee Report18 in 1978 which recognized the needs of children and young people with ‘disabilities of body or mind’ which was the foundation for the Education Act 1981 setting out the statutory requirement to provide an appropriate education for children with special educational needs. Thus, while the schooling of members of the 1946BC and 1958BC would not have been affected, more appropriate secondary schooling for the 1970BC should have been available. In the decades since participants in our study were children, statutory provision via the formal process of agreeing on an Education and Healthcare Plan plan has been implemented and understanding of the specific needs of children and young people with VI has improved.37,38 Despite these changes via the 2014 Education Act, by 2019, compared with those without a disability, a much lower proportion of adults living with VI in the UK had a university degree (38% vs. 15%, respectively) and a much higher proportion had no educational qualifications (6% vs. 11%).39 We recently showed that the education attainment gap for children and young people living with VI was not explained by differences in their ambitions or intentions for further education after school.40 The recent UK Government review of Special Educational Needs and Disabilities (SEND) has resulted in its SEND and Alternative Provision Improvement Plan,41 which is ambitious but has been criticized by the UK Children’s Commissioner for ‘not going far enough swiftly enough’. While education in childhood is intricately linked to subsequent employment opportunities, our finding of a striking worsening of odds of being unemployed over time—such that members of the 1970BC who had childhood VI were more than twice as likely as those without to be unemployed in their mid-40s—cannot be solely attributed to lower educational attainment. This disparity persists today, with 57% of adults living with VI being in paid employment vs. 81% of those without a disability.42 Interestingly, the odds of downward ‘social mobility’ between childhood and adulthood (based on parental and own occupational social class) improved over time. These positive trends may partly reflect the benefits experienced by the 1970BC from the implementation of legislation and provisions that should have addressed some inequities arising from disability including the Disability Rights Act (1995), Community Care Act (1998), Disability Living Allowance (1992), Housing Act (2004) and Equality Act (2010). Nevertheless, the stark trend of worsening odds over time for those with childhood VI of finding themselves unemployed and living in overcrowded accommodation in their mid-40s demonstrates how far the UK is from realizing the ambition of the ‘Improving the Life Chances of Disabled People’ UK Government strategy (2005) that ‘disabled people should have full opportunities and choices to improve their quality of life’. The improved odds of some social participation activities over time in our study may reflect some improvement in societal attitudes, acceptance and engagement with disability,7 (including the reframing through adoption of the World Health Organization’s International Classification of Functioning Disability and Health in 20017) alongside the implementation of legislation to prevent discrimination and improve equitable provision. The 1995 World Summit for Social Development,43 saw an international commitment to ‘society for all’ followed by initiatives to promote inclusivity and opportunity that may have benefited the 1970BC. The evidence from our study of worse physical and mental health in mid-adult life demonstrates the long shadow that those with childhood onset VI casts over the lives of affected people, despite the reduction in odds of other long-standing illnesses which will likely be due to improved healthcare during this timespan. It also demonstrates that these adverse outcomes are not new, although rightly receiving greater attention,44,45 especially to escalation due to the COVID-19 pandemic.46 Our study shows that much remains to be done to tackle the inequities in health, social position, life chances and participation that can result from childhood onset VI. However, childhood VI tends to be overshadowed by the more prevalent adult-onset VI in initiatives relating to eye health or blindness47 and is rarely considered specifically within those relating to child health.48 We suggest that for progress to be made, an explicit focus is needed on the adverse impacts of childhood vision across the life course within initiatives to address inequities relating to disability. Supplementary Material Click here for additional data file. Acknowledgements We would like to thank the study participants, the field investigators responsible for recruiting and the data management staff of all British birth cohorts. We are also grateful to the Centre for Longitudinal Studies (CLS), UCL Social Research Institute, for the use of these data and to the UK Data Service for making them available. However, neither CLS nor the UK Data Service bears any responsibility for the analysis or interpretation of these data. Contributor Information Vasiliki Bountziouka, Population, Policy and Practice Research and Teaching Department, Great Ormond Street Institute of Child Health, University College London, London, UK; Computer Simulations, Genomics and Data Analysis Lab, Department of Food Science and Nutrition, University of the Aegean, Lemnos, Greece; Department of Cardiovascular Science, Cardiovascular Research Centre, University of Leicester, Leicester, UK. Lisanne Andra Horvat-Gitsels, Population, Policy and Practice Research and Teaching Department, Great Ormond Street Institute of Child Health, University College London, London, UK; Ulverscroft Vision Research Group, Great Ormond Street Institute of Child Health, University College London, London, UK. Mario Cortina-Borja, Population, Policy and Practice Research and Teaching Department, Great Ormond Street Institute of Child Health, University College London, London, UK. Jugnoo Sangeeta Rahi, Population, Policy and Practice Research and Teaching Department, Great Ormond Street Institute of Child Health, University College London, London, UK; Ulverscroft Vision Research Group, Great Ormond Street Institute of Child Health, University College London, London, UK; Ophthalmology Department, Great Ormond Street Hospital for Children, NHS Foundation Trust, London, UK; Institute of Ophthalmology, University College London, London, UK; NIHR Moorfields Biomedical Research Centre London, London, U, K. Supplementary data Supplementary data are available at EURPUB online. Funding LAH-G was supported by the Ulverscroft Vision Research Group (no award number). JSR is a National Institute for Health and Care Research (NIHR) Senior Investigator and was supported by the NIHR Biomedical Research Centre at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology (no award numbers). All research at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health is made possible by the NIHR Great Ormond Street Hospital Biomedical Research Centre. Conflicts of interest: None declared. Author contributions VB contributed to the design of the study, data analysis and interpretation, prepared the initial draft of the manuscript, and gave final manuscript approval. LAH-G contributed to the design of the study, performed the data analysis, contributed to the interpretation of the findings and the preparation of the manuscript and gave final manuscript approval. MC-B contributed to the data analysis, interpretation and critical revision of the manuscript, and gave final manuscript approval. JSR was accountable for the design of the study, data interpretation and critical revision of the manuscript, gave final manuscript approval and was the guarantor. Childhood onset vision impairment adversely impacts health and social outcomes and limit life chances. Childhood onset vision impairment adversely impacts health and social outcomes and limit life chances. Many adverse impacts of childhood VI do not appear to have been ameliorated by legislation, policies and provision in the UK spanning three decades that were intended to address them. Many adverse impacts of childhood VI do not appear to have been ameliorated by legislation, policies and provision in the UK spanning three decades that were intended to address them. A specific focus on the adverse impacts of childhood vision across the life course is needed within initiatives to address inequities relating to disability. A specific focus on the adverse impacts of childhood vision across the life course is needed within initiatives to address inequities relating to disability. Data availability The dataset used in this research uses harmonized data from three British national birth cohort studies. The original and the harmonized 1946 birth cohort data are made available to researchers who submit data requests at https://skylark.ucl.ac.uk/NSHD/. The original data for the 1958 and 1970 birth cohorts are available from the UK Data Service (https://ukdataservice.ac.uk/). References Associated Data Supplementary Materials Click here for additional data file. Data Availability Statement The dataset used in this research uses harmonized data from three British national birth cohort studies. The original and the harmonized 1946 birth cohort data are made available to researchers who submit data requests at https://skylark.ucl.ac.uk/NSHD/. The original data for the 1958 and 1970 birth cohorts are available from the UK Data Service (https://ukdataservice.ac.uk/).