RJRhodri David Johnson

Rhodri David Johnson

*This is an algorithmically generated author record based on publicly available information.

Rhodri David Johnson has a total of 125 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Characterising residential mobility during childhood in the United Kingdom: a longitudinal study using linked electronic health records

    by Nicola Firman, Laura North, Jane Lyons, Marta Wilk, Michael Edwards, Rhodri D. Johnson, Rowena Bailey, Jo Davies, Gill Harper, Carol Dezateux, et al.

    Background: Childhood residential mobility (RM) has been associated with adverse health and social outcomes. We examined differences in RM by family size and household composition in two diverse, disadvantaged UK populations. Methods: Using electronic health records (EHRs) and anonymised addresses, we examined if household characteristics are associated with RM in north-east London (NEL) and Wales. We conducted a longitudinal study of 240,509 (51.0 per cent male) and 398,959 (51.3 per cent male) children aged 0–13 years in NEL and Wales respectively on 1 January 2015, with complete follow-up until 31 December 2019. We defined RM as at least one change in address during follow-up. We estimated odds ratios (OR) and 95 per cent confidence intervals (CI) for RM by interactions between age group (0–4, 5–9, 10–13 years), number of children in household and household composition, adjusting for demographic, household and area-level characteristics. Results: Overall, 20.6 per cent and 32.1 per cent of children experienced RM. RM was highest in the youngest age group: NEL: 25.3 per cent, 18.7 per cent, 15.6 per cent; Wales: 40.1 per cent, 29.4 per cent, 23.6 per cent. RM was more likely among 10–13-year-olds living with a single adult than two working-age adults (NEL OR: 1.15; 95 per cent CI: 1.07,1.25; Wales: 1.82; 1.73,1.91). Children in households with four or more children were more likely to experience RM than those with no other child among 10–13-year-olds in NEL (1.31; 1.18,1.46) and 5–9-year-olds and 10–13-year-olds in Wales (1.21; 1.14,1.29 and 1.68; 1.56,1.82). Conclusion: RM is more common in early childhood and varies by household composition, region and age group. EHRs can be used to estimate RM using real-world data and have the potential to examine its impact on health and other outcomes.

  • Open Access

    Exploring Loneliness and Living Alone Through Linked Administrative and Survey Data in Wales, UK

    by Rhodri David Johnson, Stuart Bedston, Rowena Bailey, Richard Fry, Lucy Griffiths

    BackgroundLoneliness is a public health priority, yet population-scale evidence that distinguishes types of loneliness and links with living alone circumstances is scarce. Population-wide linked administrative data within the Secure Anonymised Information Linkage (SAIL) Databank provide an opportunity to investigate these relationships across Wales. Objectives(1) Describe long-term living alone trends; (2) develop indicators for reason for living alone; and (3) examine associations between living alone (including derived indicators) and emotional, social and overall loneliness, alongside demographic, health and place-based factors. MethodsWe conducted a two-phase observational study in Wales. Phase 1 used annual, population level snapshots (2008--2024) to quantify living alone trends and correlates. Phase 2 linked individual level data in SAIL to National Survey for Wales loneliness data (De Jong Gierveld 6-item) to model adjusted associations. We inferred living alone reasons (partner loss, separation, unassigned) from life events data. Models adjusted for age and gender interaction, health, tenure, housing type, deprivation, and built environment indicators. ResultsLiving alone households increased from 27.0% (314,806) in 2008 to 29.6% (363,133) in 2024. In adjusted models, living alone was strongly associated with loneliness risk, especially emotional loneliness. These associations were modified by reason for living alone, with the largest effects observed following recent partner loss. Younger adults reported highest overall loneliness; men showed higher social loneliness around midlife; women showed higher emotional loneliness. Built environment measures showed little evidence of independent association after adjustment. ConclusionsPopulation-scale linkage can generate robust, policy-relevant indicators of living alone dynamics. Such indicators support monitoring and targeting within health and social care, complementing national statistics and enabling future linked data studies.

  • Open Access

    Residential mobility and receipt of measles, mumps and rubella vaccination: analysis of linked primary care electronic health records in a London region

    by Nicola Firman, Laura North, Milena Marszalek, Marta Wilk, Ana Gutierrez, Rhodri Johnson, Carol Dezateux

    BackgroundResidential mobility in early life may disrupt access to health care. We examined associations between residential mobility in the first two years of life and receipt of first measles, mumps and rubella (MMR) vaccination by 24 months of age. MethodsWe analysed electronic health records for children born between 01/01/2014 and 30/10/2019 and registered with primary care general practices (GPs) in north-east London (NEL). Primary outcome was receipt of first MMR vaccination between 12 and 24 months of age and residential mobility was defined by number of GP-recorded addresses by vaccination date or 24 months of age. We used logistic regression to estimate the odds ratio (OR) and 95% confidence intervals (CI) of receipt of MMR vaccination by residential mobility, adjusting for sex, ethnicity, number of children in the household, household composition, area-level deprivation, and local government area of residence. ResultsWe included 150,949 children (51.0% boys) of whom 127,958 (84.8%) had received a first MMR vaccination and 22.3% had more than one GP-recorded address. Compared to children with one GP-recorded address, children with multiple GP-recorded addresses were at higher risk of not receiving their first MMR vaccination by 24 months of age. Those with two GP-recorded addresses had a 54% increased likelihood (OR: 0.46; 95% CI: 0.44,0.48), and those with three or more GP-recorded addresses a 68% increased likelihood (0.32; 95% CI: 0.29,0.36), compared to those with one. ConclusionsChildren who change address in early life are less likely to be protected against measles and other infections. Measles outbreaks in the UK have been in urban areas with high residential mobility and low MMR vaccine uptake suggesting the need for review of the immunisation status of children newly registered with GPs.


  • Open Access

    The Desirable Health Indicator: A New Indicator of Population Health and Healthcare Utilisation

    by Harri Doel, Lucy J Griffiths, Rhodri David Johnson, Samantha Turner, Ronan A Lyons, Jane Lyons

    Background and ObjectiveHealthcare research faces challenges in developing metrics that resonate with the general public or policymakers. We created a Desirable Health Indicator (DHI) to address this gap, centred around New Year's wishes for survival and non-occurrence of undesired events in the following year, for the population of Wales, UK, following discussions with policymakers and members of the public. MethodsWe created retrospective, population-based individual-level cohorts from linked routinely collected anonymised, health and demographic data from the Secure Anonymised Information Linkage (SAIL) Databank (2015-2022). The DHI was calculated per person per year and quantified the distribution of the population who survive calendar years and do not use selected health services (not admitted to hospital; no emergency department attendance; and not prescribed medication used in infection, analgesics, or mental health drugs). Group and individual interviews were held with members of the public and policy makers seeking their views of the indicator. ResultsThe findings were understood and well received by members of the general public and policymakers. Between 2015 and 2019, the percentage of individuals meeting the DHI ranged between 39.6%-41.9%, increasing to 48.6% and 46.2% for2020 and 2021respectively, and reducing to 43.1% in 2022. Focussing on the year 2022, 1,154,630 (43.1%) met the DHI from a population of 2,677,829. The percentage of people with desirable health decreased significantly with age and with increasing socioeconomic deprivation. A higher proportion of males (49.2%) met the DHI compared to females (37.1%). Being male (aOR = 1.62 [95%CI 1.61,1.63]), 10-19 years of age (aOR = 1.69 [95%CI 1.68,1.71]), and living in the least deprived areas of Wales (aOR = 1.31 [95%CI 1.30,1.32]) were the characteristics associated with the highest odds of meeting the desirable health indicator. The most prevalent reasons for not meeting the indicator were GP prescriptions for drugs used in infections (29.5%), analgesics (22.8%) and mental health conditions (20.2%). ConclusionThe DHI provides an insightful and novel tool for monitoring aspects of population health and healthcare utilisation. The DHI's coverage of important topics, derived from routine data sources, makes it a reproducible, temporally flexible, and easily understood indicator, suitable for informing policy development and addressing aspects of health inequalities. As data linkage capabilities expand internationally there are opportunities for implementation to aid comparison and better understanding of how systems perform.

  • Open Access

    Administrative data linkage to Census 2021 in Wales, UK: A cross-sectional study examining completeness and representativeness for population analytics

    by Jane Lyons, Rhodri Johnson, Michael Edwards, Samantha Turner, Richard Fry, Lucy J Griffiths, Ronan Lyons

    IntroductionMeasuring population representativeness is an important methodological step in public health and epidemiological studies. ObjectivesTo explore the representativeness of Census 2021 data linkage when compared with the Welsh Demographic Service Dataset (WDSD) within the Secure Anonymised Information Linkage (SAIL) Databank for research on the population of Wales, UK. To understand the characteristics of individuals linked and not linked and which subgroups of the population are disproportionately represented in data linkage population-wide studies. MethodsAn observational, population-wide cross-sectional comparison study, utilising administrative demographic data and decennial survey data held in SAIL. Two data sources, the WDSD and Census 2021, were used to create and compare two cohorts of the resident population of Wales, UK, on 21st March 2021. The two cohorts were linked to understand how many individuals from Census 2021 can be successfully linked within SAIL, in WDSD and not in Census 2021, and found across both sources. Logistic regression models analysed the variation in the linkability of the survey data within SAIL by various demographic and household characteristics. ResultsThe central analytical cohort contained 2,440,191 individuals present in both data sources. WDSD contained 3,090,976 individuals with 2,965,196 individuals in Census data. With a positively classed outcome indicating non-linkage from WDS to Census the characteristics associated with the highest odds of individuals being registered in WDS but not linked to Census (in SAIL) are male (aOR = 1.28 [95%CI 1.28,1.32]), 75+ years of age (aOR = 1.27 [95%CI 1.25,1.29]), of Asian ethnicity (aOR = 1.27 [95%CI 1.24,1.30]), a more recent migrant (arriving to UK after 2000) (aOR = 1.30 [95%CI 1.28,1.32]), a member of the LGBTQ+ community (aOR = 1.29 [95%CI 1.25,1.29]) or not disclosing LGBTQ+ status (aOR = 1.41 [95%CI 1.39,1.43]), being separated, divorced or widowed (aOR = 1.28 [95%CI 1.27,1.29]), or living in rental accommodation (aOR = 1.47 [95%CI 1.45,1.48]). ConclusionsResults show that certain personal characteristics and sub-groups of the population of Wales are disproportionately represented when combining population estimates and utilising Census data in data linkage population-wide studies in SAIL.

  • Open Access

    Residential mobility amongst children and young people in Wales: A longitudinal study using linked administrative records

    by Jo Davies, Rowena Bailey, Amy Mizen, Theodora Pouliou, Richard Fry, Rebecca Pedrick-Case, Gareth Stratton, Rhodri Johnson, Hayley Christian, Ronan Lyons, et al.

    BackgroundChild poverty remains a major global concern and a child's experience of deprivation is heavily shaped by where they live and the stability of their local neighbourhood. This study examines frequencies and patterns of residential mobility in children and young people (CYP) at a population level using novel geospatial techniques to assess how often their physical environment changes and to identify geographical variations in social mobility.MethodsWe used routinely collected administrative records held in the Secure Anonymised Information Linkage (SAIL) Databank for CYP aged under 18 years living in Wales between 2012 and 2022. We calculated the Moran's I statistic to assess the magnitude of Lower layer Super Output Area (LSOA)-level geographic variation in residential mobility and used the Local Indicator of Spatial Association (LISA) to identify clusters of LSOAs where there are higher rates of residential mobility.ResultsThis study included 923,531 CYP, with 58% having moved at least once during the study period. A total number of 1,209,102 house moves were recorded, 59% of which occurred between the ages of 0 and 5 years. Almost 10% of the cohort resided in five or more dwellings before the age of 18 years. In terms of area-level (LSOA) deprivation, 75% of house moves were to areas with the same or higher levels of deprivation, leaving only 25% of house moves that achieved upward social mobility. Clustering of residential mobility was identified predominantly in areas of high deprivation.ConclusionThe findings of this study show that residential mobility is linked with socio-economic circumstances and is experienced by over half of CYP in Wales. Understanding where CYP live, their mobility patterns and which areas have high levels of influx and efflux is crucial for policymakers to generate well-informed, targeted and effective child-focused interventions.


  • Open Access

    Risk Factors Associated With Primary Care–Reported Domestic Violence for Women Involved in Family Law Care Proceedings: Data Linkage Observational Study

    by Rhodri D Johnson, Lucy J Griffiths, Laura E Cowley, Karen Broadhurst, Rowena Bailey

    Background: Domestic violence and abuse (DVA) has a detrimental impact on the health and well-being of children and families but is commonly underreported, with an estimated prevalence of 5.5% in England and Wales in 2020. DVA is more common in groups considered vulnerable, including those involved in public law family court proceedings; however, there is a lack of evidence regarding risk factors for DVA among those involved in the family justice system. Objective: This study examines risk factors for DVA within a cohort of mothers involved in public law family court proceedings in Wales and a matched general population comparison group. Methods: We linked family justice data from the Children and Family Court Advisory and Support Service (Cafcass Cymru [Wales]) to demographic and electronic health records within the Secure Anonymised Information Linkage (SAIL) Databank. We constructed 2 study cohorts: mothers involved in public law family court proceedings (2011-2019) and a general population group of mothers not involved in public law family court proceedings, matched on key demographics (age and deprivation). We used published clinical codes to identify mothers with exposure to DVA documented in their primary care records and who therefore reported DVA to their general practitioner. Multiple logistic regression analyses were used to examine risk factors for primary care–recorded DVA. Results: Mothers involved in public law family court proceedings were 8 times more likely to have had exposure to DVA documented in their primary care records than the general population group (adjusted odds ratio [AOR] 8.0, 95% CI 6.6-9.7). Within the cohort of mothers involved in public law family court proceedings, risk factors for DVA with the greatest effect sizes included living in sparsely populated areas (AOR 3.9, 95% CI 2.8-5.5), assault-related emergency department attendances (AOR 2.2, 95% CI 1.5-3.1), and mental health conditions (AOR 1.7, 95% CI 1.3-2.2). An 8-fold increased risk of DVA emphasizes increased vulnerabilities for individuals involved in public law family court proceedings. Conclusions: Previously reported DVA risk factors do not necessarily apply to this group of women. The additional risk factors identified in this study could be considered for inclusion in national guidelines. The evidence that living in sparsely populated areas and assault-related emergency department attendances are associated with increased risk of DVA could be used to inform policy and practice interventions targeting prevention as well as tailored support services for those with exposure to DVA. However, further work should also explore other sources of DVA, such as that recorded in secondary health care, family, and criminal justice records, to understand the true scale of the problem. Trial Registration:


  • Open Access

    Household Composition and Inequalities in COVID-19 Vaccination in Wales, UK

    by Alex Lench, Malorie Perry, Rhodri D. Johnson, Richard Fry, Gill Richardson, Ronan A. Lyons, Ashley Akbari, Adrian Edwards, Brendan Collins, Natalie Joseph-Williams, et al.

    The uptake of COVID-19 vaccination in Wales is high at a population level but many inequalities exist. Household composition may be an important factor in COVID-19 vaccination uptake due to the practical, social, and psychological implications associated with different living arrangements. In this study, the role of household composition in the uptake of COVID-19 vaccination in Wales was examined with the aim of identifying areas for intervention to address inequalities. Records within the Wales Immunisation System (WIS) COVID-19 vaccination register were linked to the Welsh Demographic Service Dataset (WDSD; a population register for Wales) held within the Secure Anonymised Information Linkage (SAIL) databank. Eight household types were defined based on household size, the presence or absence of children, and the presence of single or multiple generations. Uptake of the second dose of any COVID-19 vaccine was analysed using logistic regression. Gender, age group, health board, rural/urban residential classification, ethnic group, and deprivation quintile were included as covariates for multivariable regression. Compared to two-adult households, all other household types were associated with lower uptake. The most significantly reduced uptake was observed for large, multigenerational, adult group households (aOR 0.45, 95%CI 0.43–0.46). Comparing multivariable regression with and without incorporation of household composition as a variable produced significant differences in odds of vaccination for health board, age group, and ethnic group categories. These results indicate that household composition is an important factor for the uptake of COVID-19 vaccination and consideration of differences in household composition is necessary to mitigate vaccination inequalities.


  • Open Access

    Built Environments And Child Health in WalEs and AuStralia (BEACHES): a study protocol

    by Rebecca Pedrick-Case, Rowena Bailey, Ben Beck, Bridget Beesley, Bryan Boruff, Sinead Brophy, Donna Cross, Gursimran Dhamrait, John Duncan, Peter Gething, et al.

    Introduction Childhood obesity and physical inactivity are two of the most significant modifiable risk factors for the prevention of non-communicable diseases (NCDs). Yet, a third of children in Wales and Australia are overweight or obese, and only 20% of UK and Australian children are sufficiently active. The purpose of the Built Environments And Child Health in WalEs and AuStralia (BEACHES) study is to identify and understand how complex and interacting factors in the built environment influence modifiable risk factors for NCDs across childhood. Methods and analysis This is an observational study using data from five established cohorts from Wales and Australia: (1) Wales Electronic Cohort for Children; (2) Millennium Cohort Study; (3) PLAY Spaces and Environments for Children’s Physical Activity study; (4) The ORIGINS Project; and (5) Growing Up in Australia: the Longitudinal Study of Australian Children. The study will incorporate a comprehensive suite of longitudinal quantitative data (surveys, anthropometry, accelerometry, and Geographic Information Systems data) to understand how the built environment influences children’s modifiable risk factors for NCDs (body mass index, physical activity, sedentary behaviour and diet). Ethics and dissemination This study has received the following approvals: University of Western Australia Human Research Ethics Committee (2020/ET000353), Ramsay Human Research Ethics Committee (under review) and Swansea University Information Governance Review Panel (Project ID: 1001). Findings will be reported to the following: (1) funding bodies, research institutes and hospitals supporting the BEACHES project; (2) parents and children; (3) school management teams; (4) existing and new industry partner networks; (5) federal, state and local governments to inform policy; as well as (6) presented at local, national and international conferences; and (7) disseminated by peer-reviewed publications.


  • Open Access

    Risk Factors Associated With Primary Care–Reported Domestic Violence for Women Involved in Family Law Care Proceedings: Data Linkage Observational Study (Preprint)

    by Rhodri D Johnson, Lucy J Griffiths, Laura E Cowley, Karen Broadhurst, Rowena Bailey

    BACKGROUND Domestic violence and abuse (DVA) has a detrimental impact on the health and well-being of children and families but is commonly underreported, with an estimated prevalence of 5.5% in England and Wales in 2020. DVA is more common in groups considered vulnerable, including those involved in public law family court proceedings; however, there is a lack of evidence regarding risk factors for DVA among those involved in the family justice system. OBJECTIVE This study examines risk factors for DVA within a cohort of mothers involved in public law family court proceedings in Wales and a matched general population comparison group. METHODS We linked family justice data from the Children and Family Court Advisory and Support Service (Cafcass Cymru [Wales]) to demographic and electronic health records within the Secure Anonymised Information Linkage (SAIL) Databank. We constructed 2 study cohorts: mothers involved in public law family court proceedings (2011-2019) and a general population group of mothers not involved in public law family court proceedings, matched on key demographics (age and deprivation). We used published clinical codes to identify mothers with exposure to DVA documented in their primary care records and who therefore reported DVA to their general practitioner. Multiple logistic regression analyses were used to examine risk factors for primary care–recorded DVA. RESULTS Mothers involved in public law family court proceedings were 8 times more likely to have had exposure to DVA documented in their primary care records than the general population group (adjusted odds ratio [AOR] 8.0, 95% CI 6.6-9.7). Within the cohort of mothers involved in public law family court proceedings, risk factors for DVA with the greatest effect sizes included living in sparsely populated areas (AOR 3.9, 95% CI 2.8-5.5), assault-related emergency department attendances (AOR 2.2, 95% CI 1.5-3.1), and mental health conditions (AOR 1.7, 95% CI 1.3-2.2). An 8-fold increased risk of DVA emphasizes increased vulnerabilities for individuals involved in public law family court proceedings. CONCLUSIONS Previously reported DVA risk factors do not necessarily apply to this group of women. The additional risk factors identified in this study could be considered for inclusion in national guidelines. The evidence that living in sparsely populated areas and assault-related emergency department attendances are associated with increased risk of DVA could be used to inform policy and practice interventions targeting prevention as well as tailored support services for those with exposure to DVA. However, further work should also explore other sources of DVA, such as that recorded in secondary health care, family, and criminal justice records, to understand the true scale of the problem. CLINICALTRIAL