Objective Young people transitioning from out-of-home care (OHC) frequently experience poor mental health and resilience due to adverse childhood experiences (ACEs). However, there is limited understanding of the factors that mediate and moderate these outcomes. This is the first study to integrate linked administrative and longitudinal data to examine the mediation and moderation effects of placement stability, independent living skills (ILS), social inclusion, and self-determination when examining the association between ACEs and care status on mental health and resilience. Method We integrated retrospective linked datasets with a prospective self-report longitudinal study involving 122 young people aged 15 to 25 transitioning from OHC between April 2019 and May 2022. Path analysis was used to model complex relationships involving moderators and mediators. Results ILS moderated the association between ACEs and resilience, while social inclusion, self-determination, and mental health mediated the effects of care status and maltreatment exposure on resilience. Placement stability independently influenced mental health outcomes but did not mediate the ACEs-resilience relationship. Conclusions This study highlights the complex interplay of risk and protective factors in shaping resilience and mental health, emphasising the importance of skills development, social connectedness and autonomy among care leavers. Findings support early intervention, strength-based approaches, and trauma-informed interventions, including emotional regulation, therapeutic relationships, and stable support networks, to mitigate past trauma and enhance resilience and well-being.

Background: Researchers have examined sub-groups that may exist among young people transitioning from out-of-home care (OHC) using various theoretical models. However, this population group has not been examined for trajectories of homelessness risk. Objectives: To examine whether different subtypes of homelessness risk exist among young people transitioning from care and whether these trajectories of homelessness are associated with mental health and substance use disorders. Participants and setting: A retrospective population-based cohort study was conducted from a population of 1018 young people (aged 15-18 years) who transitioned from out-of-home in 2013 to 2014 in the state of Victoria, Australia, with follow-up to 2018. Methods: Latent Class Growth Analysis was conducted using linked data from homelessness data collections, child protection, mental health information systems, alcohol and drug use, and youth justice information systems. Results: Three sub-groups of young people were identified. The 'moving on' group (88 %) had the lowest levels of homelessness, with the slope of this trajectory remaining almost stable. The 'survivors' (7 %) group started off with a high risk of homelessness, followed by a sharp decrease in homelessness risk over time. The 'complex' (5 %) group started off with a low risk of homelessness but faced sharp increases in the risk of homelessness over time. Conclusions: Our study demonstrates that subgroups of young people transitioning from care exist with distinct longitudinal trajectories of homelessness, and these classes are associated with different risk factors. Early intervention and different approaches to tackling homelessness should be considered for these three distinct groups before transitioning from care and during the first few years after leaving care.

Background Young people who were in out-of-home care (OHC) face an accelerated transition to independent adulthood. Current evidence on outcomes for Australian care-leavers is scant. Objective This study aims to develop a better understanding of the outcomes for young people leaving care. Participants and setting A birth cohort of children and young people born in Western Australia (WA) from 1993 to 2008. Three groups were identified and compared: young people with care-experience (OHC Cohort), those with child protection involvement but not care experience (CP Contact Cohort), and peers in the general population (No Contact Cohort). Methods This is a retrospective, population-based study utilising de-identified, linked administrative records provided by the WA state government agencies. Data from the three cohorts were compared through descriptive statistics, independent samples t-tests, and logistic regression modelling. Results The birth cohort contained records for 414,266 individuals. The smallest comparison group in this study was the OHC Cohort (n = 6526), followed by the CP Contact Cohort (n = 78,095), and the No Contact Cohort (n = 329,645). Care-experienced young people in WA fared significantly worse than their peers across the domains of health (physical and mental), disability, education, social housing and criminal justice involvement. Conclusions Those who have had child protection involvement, but have not been placed in care, had better outcomes than those who had been in care. However, their outcomes were still poorer than the population cohort with no child protection contact.

Purpose Despite the volume of accumulating knowledge from prospective Aboriginal cohort studies, longitudinal data describing developmental trajectories in health and well-being is limited. The linkage of child and carer cohorts from a historical cross-sectional survey with longitudinal health-service and social-service administrative data has created a unique and powerful data resource that underpins the Western Australian Aboriginal Child Health Survey (WAACHS) linked data study. This study aims to provide evidence-based information to Aboriginal communities across Western Australia, governments and non-government agencies on the heterogeneous life trajectories of Aboriginal children and families.Participants This study comprises data from a historical cross-sectional household study of 5289 Aboriginal children from the WAACHS (2000-2002) alongside their primary (N=2113) and other (N=1040) carers, and other householders. WAACHS data were linked with Western Australia (WA) government administrative datasets up to 2020 including health, education, child protection, police and justice system contacts. The study also includes two non-Aboriginal cohorts from WA, linked with the same administrative data sources allowing comparisons of outcomes across cohorts in addition to between-group comparisons within the Aboriginal population.Findings to date Linked data coverage rates are presented for all WAACHS participants. Child health outcomes for the WAACHS children (Cohort 1) are described from birth into adulthood along with other outcomes including child protection and juvenile justice involvement.Future plans Analysis of data from both the child and carer cohorts will seek to understand the contribution of individual, family (intergenerational) and community-level influences on Aboriginal children's developmental and health pathways, identify key developmental transitions or turning points where interventions may be most effective in improving outcomes, and compare service pathways for Aboriginal and non-Aboriginal children. All research is guided by Aboriginal governance processes and study outputs will be produced with Aboriginal leadership to guide culturally appropriate policy and practice for improving health, education and social outcomes.

Introduction Over the past decade there has been a marked growth in the use of linked population administrative data for child protection research. This is the first systematic review of studies to report on research design and statistical methods used where population-based administrative data is integrated with longitudinal data in child protection settings. Methods The systematic review was conducted according to Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) statement. The electronic databases Medline (Ovid), PsycINFO, Embase, ERIC, and CINAHL were systematically searched in November 2019 to identify all the relevant studies. The protocol for this review was registered and published with Open Science Framework (Registration DOI: 10.17605/OSF.IO/96PX8) Results The review identified 30 studies reporting on child maltreatment, mental health, drug and alcohol abuse and education. The quality of almost all studies was strong, however the studies rated poorly on the reporting of data linkage methods. The statistical analysis methods described failed to take into account mediating factors which may have an indirect effect on the outcomes of interest and there was lack of utilisation of multi-level analysis. Conclusion We recommend reporting of data linkage processes through following recommended and standardised data linkage processes, which can be achieved through greater co-ordination among data providers and researchers.

Introduction Organisations are increasingly aware of the risks and responsibilities of handling personally identifying information (PII). These factors not only influence internal data management practices but also impact on data linkage arrangements with other parties. Some linkage environments are particularly sensitive to the release or use of PII. Advances in privacy-preserving record linkage methods such as PPRL-using-Bloom make it possible to undertake highly accurate data linkage without release or disclosure of PII. Such methods play a role in enabling data linkage in risk-sensitive environments. Objectives and ApproachWe present and describe several Australian case studies where the PPRL-using-Bloom method has been used to enable data linkage between organisations. We report on the defining elements of each case, the associated risks and solutions, as well as quality and performance issues. We also reflect on challenges and opportunities for future improvement. ResultsAustralian use cases utilising privacy preserving linkage (PPRL-using-Bloom) include projects linking state-based datasets to Commonwealth datasets, some linking primary care data to state-based secondary healthcare data, and others linking healthcare data to non-health datasets such as police and criminal justice datasets. Conclusion / ImplicationsMethods such as PPRL-using-Bloom play a critical role in enabling data linkage in highly risk-sensitive environments. However, in an ever evolving world where risks and requirements are constantly changing, linkage methodologies and technologies must remain adaptable to meet evolving demands.

Objective To describe the use of privacy preserving linkage methods operationally in Australia, and to present insights and key learnings from their implementation. Methods Privacy preserving record linkage (PPRL) utilising Bloom filters provides a unique practical mechanism that allows linkage to occur without the release of personally identifiable information (PII), while still ensuring high accuracy. Results The methodology has received wide uptake within Australia, with four state linkage units with privacy preserving capability. It has enabled access to general practice and private pathology data amongst other, both much sought after datasets previous inaccessible for linkage. Conclusion The Australian experience suggests privacy preserving linkage is a practical solution for improving data access for policy, planning and population health research. It is hoped interest in this methodology internationally continues to grow.

This data note describes a new resource for crime-related research: the Avon Longitudinal Study of Parents and Children (ALSPAC) linked to regional police records. The police data were provided by Avon & Somerset Police (A&SP), whose area of responsibility contains the ALSPAC recruitment area. In total, ALSPAC had permission to link to crime records for 12,662 of the ‘study children’ (now adults, who were born in the early 1990s). The linkage took place in two stages: Stage 1 involved the ALSPAC Data Linkage Team establishing the linkage using personal identifiers common to both the ALSPAC participant database and A&SP records using deterministic and probabilistic methods. Stage 2 involved A&SP extracting attribute data on the matched individuals, removing personal identifiers and securely sharing the de-identified records with ALSPAC. The police data extraction took place in July 2021, when the participants were in their late 20s/early 30s. This data note contains details on the resulting linked police records available. In brief, electronic police records were available from 2007 onwards. In total, 1757 participants (14%) linked to at least one police record for a charge, offence ‘taken into consideration’, caution, or another out of court disposal. Linked participants had a total of 6413 records relating to 6283 offences. Almost three quarters of the linked participants were male. The most common offence types were violence against the person (22% of records), drug offences (19%), theft (17%) and public order offences (11%). This data note also details important issues that researchers using the local police data should be aware of, including the importance of defining an appropriate denominator, completeness, and biases affecting police records.

OBJECTIVES:To examine the relationship between structural characteristics of Australian residential aged care facilities (RACFs) and breaches of the aged care quality standards.METHODS:Facility-level analysis of audits, sanctions and non-compliance notices of all accredited Australian RACFs between 2015/16 and 2018/19. Structural factors of interest included RACF size, remoteness, ownership type and jurisdiction. Two government data sources were joined. Each outcome was analysed to calculate time trends, unadjusted rates and relative risks.RESULTS:Non-compliance notices were imposed on 369 RACFs (13%) and 83 sanctions on 75 RACFs (3%). Compared with New South Wales (NSW), non-compliance notices were less likely in Victoria, Queensland and the Northern Territory (NT), more likely in South Australia (SA), and comparable in Western Australia (WA), Tasmania and the Australian Capital Territory (ACT). RACFs with more than 100 beds and RACFs located in remote and outer regional areas (vs. major cities) also increased the likelihood of non-compliance notices. Compared with NSW, sanctions were less likely in Victoria, Queensland, NT and WA and comparable in SA, Tasmania and ACT. Additionally, the likelihood of sanctions was higher for RACFs with more than 40 beds. For both non-compliance notices and sanctions, no significant relationship was found with RACF ownership type.CONCLUSIONS:We partially confirmed other Australian findings about the relationship between RACF structural characteristics and regulatory sanctions and reported new findings about non-compliance notices. Routine and standardised public reporting of RACF performance is needed to build trust that Australia's latest aged care reforms have led to sustained quality improvements.

Study design Cross-sectional survey. Objectives To describe design and methods of Australian arm of International Spinal Cord Injury (Aus-InSCI) community survey, reporting on participation rates, potential non-response bias and cohort characteristics. Setting Survey of community-dwelling people with SCI at least 12 months post-injury, recruited between March 2018 and January 2019, from state-wide SCI services, a government insurance agency and not-for-profit consumer organisations across four Australian states. Methods The Aus-InSCI survey combined data for people with SCI from nine custodians, using secure data-linkage processes, to create a population-based, anonymised dataset. The Aus-InSCI questionnaire comprised 193 questions. Eligibility, response status and participation rates were calculated. Descriptive statistics depict participant characteristics. Logistic regression models were developed for probability of participation, and inverse probability weights generated to assess potential non-response bias. Results 1579 adults with SCI were recruited, a cooperation rate of 29.4%. Participants were predominantly male (73%), with 50% married. Mean age was 57 years (range 19–94) and average time post-injury 17 years (range 1–73). Paraplegia (61%) and incomplete lesions (68%) were most common. Males were more likely than females to have traumatic injuries ( p < 0.0001) and complete lesions ( p = 0.0002), and younger age-groups were more likely to have traumatic injuries and tetraplegia ( p < 0.0001). Potential non-response bias evaluated using selected outcomes was found to be negligible in the Aus-InSCI cohort. Conclusions The Aus-InSCI survey made efforts to maximise coverage, avoid recruitment bias and address non-response bias. The distributed, linked and coded (re-identifiable at each custodian level) ‘virtual quasi-registry’ data model supports systematic cross-sectional and longitudinal research.

Background Privacy preserving record linkage (PPRL) methods using Bloom filters have shown promise for use in operational linkage settings. However real-world evaluations are required to confirm their suitability in practice. Methods An extract of records from the Western Australian (WA) Hospital Morbidity Data Collection 2011–2015 and WA Death Registrations 2011–2015 were encoded to Bloom filters, and then linked using privacy-preserving methods. Results were compared to a traditional, un-encoded linkage of the same datasets using the same blocking criteria to enable direct investigation of the comparison step. The encoded linkage was carried out in a blinded setting, where there was no access to un-encoded data or a ‘truth set’. Results The PPRL method using Bloom filters provided similar linkage quality to the traditional un-encoded linkage, with 99.3% of ‘groupings’ identical between privacy preserving and clear-text linkage. Conclusion The Bloom filter method appears suitable for use in situations where clear-text identifiers cannot be provided for linkage.

Objectives The study examined the relationship between mental health, homelessness and housing instability among young people aged 15-18 years old who transitioned from out-of-home in 2013 to 2014 in the state of Victoria, Australia with follow-up to 2018. We determined the various mental health disorders and other predictors that were associated with different levels of homelessness risk, including identifying the impact of dual diagnosis of mental health and substance use disorder on homelessness. Methodology Using retrospective de-identified linked administrative data from various government departments we identified various dimensions of homelessness which were mapped from the European Topology of Homelessness (ETHOS) framework and associated mental health variables which were determined from the WHO ICD-10 codes. We used ordered logistic regression and Poisson regression analysis to estimate the impact of homelessness and housing instability respectively. Results A total homelessness prevalence of 60% was determined in the care-leaving population. After adjustment, high risk of homelessness was associated with dual diagnosis of mental health and substance use disorder, intentional self-harm, anxiety, psychotic disorders, assault and maltreatment, history of involvement with the justice system, substance use prior to leaving care, residential and home-based OHC placement and a history of staying in public housing. Conclusions There is clearly a need for policy makers and service providers to work together to find effective housing pathways and integrated health services for this heterogeneous group of vulnerable young people with complex health and social needs. Future research should determine longitudinally the bidirectional relationship between mental health disorders and homelessness.

BACKGROUND:Trajectory analysis has been used to study long-term offending patterns and identify offender subgroups, but few such studies have included people with psychotic disorders (PDs) and these have been restricted to adult offenders.AIMS:To compare offending trajectories among 10-26-year-olds with PDs with those with other mental disorders (OMDs) or none (NMD) and identify associated risk factors.METHODS:This is a record-linkage study of 184,147 people born in Western Australia (WA) 1983-1991, drawing on data from WA mental health information system, WA corrective services and other state-wide registers. Group-based trajectory modelling was used to identify offending trajectories.RESULTS:Four offender groups were identified in each mental health status group: G1-no/negligible offending; G2-early onset, adolescent, desisting by age 18; G3-early onset, low rate, offending into early adulthood; and G4-very early onset, high rate, peaking at age 17, continuing into early adulthood. The PDs group had the lowest proportion of individuals with no or negligible offending histories-84% compared with 88.5% in the OMDs group and 96.6% in the no mental disorder group. Within mental health status offender groups, the PDs group was characterised by early or very early onset offending persisting into adulthood, accounting for 5.4% and 3.7% of the group respectively (OMD: 3.8%, 1.5%; NMD: 1.0%, 0.5%). Gender, indigenous status, substance use problems, childhood abuse and parental offending were generally associated with trajectory group membership, although among those with PDs childhood abuse and parental offending were only significant in the early onset-life-course-persistent group.CONCLUSIONS:While most people with PDs never offend, some are disproportionately vulnerable from a particularly early age. If the offending subgroup is to be helped away from criminal justice involvement, interventions must be considered in childhood.

Trajectory analysis has been used to study long-term patterns of offending and identify offending groups. Only few studies have explored patterns in people with psychotic illness and these were restricted to adult offenders. This study examines offending trajectories, and identifies risk factors, for people aged 10-26 with psychotic illness (PI) and other mental disorders (OMD) compared to those with no mental disorders (NMD). This is a whole-population record-linkage study of 184,147 people born in Western Australia (WA) 1983-1991 using data from WA psychiatric case register, WA corrective services and other state-wide registers. Group-based trajectory modelling was used to identify offending trajectories. Four offender groups were identified for each mental health status (MHS) group: MHS groups had similar offending patterns, however PI had a lower proportion of individuals in the G1 group and later offending onset in the G3 group. Gender, indigenous status, substance use, childhood victimisation and parental offending were risk factors associated with group membership; for PI, childhood victimisation and parental offending were only significant in the G4 group. Overall offending patterns and risk factors were similar for all MHS groups, however, some differences were observed for PI. To reduce offending in this population, interventions need to occur at an early age.

Introduction While the quantity and type of datasets used by data linkage projects is growing, there remain some datasets that are ‘not available’ or ‘hard to access’ by researchers and linkers, either due to legal/regulatory constraints restricting the release of personally identifying information or because of privacy or reputational concerns. Advances in privacy-preserving record linkage methods (e.g. PPRL-Bloom) have made it possible to overcome this impasse. These techniques aim to provide strong privacy protection while still maintaining high linkage quality. PPRL-Bloom methods are being used in practice. The Centre for Data Linkage (CDL) at Curtin University has been involved in several PPRL linkage and evaluation projects using real-world data. As the methods are relatively new, published information on achievable linkage quality in real-world scenarios is limited. Objectives and Approach We present and describe several real-world applications of privacy preserving record linkage (PPRL-Bloom) where the quality of the linkage could be ascertained. In each case, data was linked ‘blind’; that is, without linkers having access to the original personal identifiers at any stage, or having any additional information about the records. Evaluations include a linkage of state-based morbidity and mortality records, a linkage of a number of general practice datasets to morbidity and emergency records, and a linkage of a range of state-based non-health administrative data, including education, police, housing, birth and child protection records. Results The privacy preserving record linkage performed admirably, with very high-quality results across all evaluations. Conclusion / Implications Privacy preserving linkage is a useful and innovative methodology that is currently being used in real world projects. The results of these evaluation suggest it can be an appropriate linkage tool when legal or other constraints block release of personally identifying information to third party linkage units.

Introduction During 2019, the Western Australian (WA) government and Curtin University’s Centre for Data Linkage (CDL) created a large, de-identified researchable database – the Social Investment Data Resource (SIDR) – to support government in delivering targeted early interventions to young offenders and their families to reduce the likelihood of re-offending (the Target 120 program). Objectives and Approach SIDR brings together administrative data from health, education, justice, child protection, disability and housing sectors. The linked, de-identified data provides an invaluable resource for actuarial assessment and social investment analytics to assess long-term costs and benefits of the Target 120 program. SIDR also provides an invaluable tool for academic research. SIDR adopted a distributed linkage model where linkage workload was shared between the Department of Health Data Linkage Branch who create and maintain the WA Data Linkage System (WADLS) and the CDL. Design elements of the model included a common spine (embedded into the infrastructure of both groups), methods for leveraging quality from WADLS, and inclusion of family relationships data from the WA Family Connections database. The linkage model within SIDR uses a combination of traditional and privacy-preserving record linkage (PPRL) methods. PPRL does not require release of personal identifiers; instead, data is irreversibly hashed prior to release for probabilistic linkage. Results Through cooperation (distributed linkage) and innovation (a mix of traditional and PPRL linkage), the project has delivered a large, linked, cross-sectoral data resource for policymakers and researchers. Sharing of the linkage workload maximised the capacity and unique capabilities of each linkage unit. PPRL enabled ‘hard to get’ datasets from justice to be included. SIDR is being updated in 2020. Conclusion / Implications SIDR provides a resource for whole-of-government policy development, service evaluation, academic research and social investment analytics for T120 and beyond. The SIDR linkage model has potential for adaptation and use elsewhere.

IntroductionPrivacy-preserving Record Linkage (PPRL) is a record linkage technique that can increase the security of personal information. PPRL uses techniques of either hashing identifiers (where exact matches are required) or Blooming identifiers (where partial matches are of interest before they are provided for linkage. Objectives and ApproachWe use LinXmart software to evaluate performance of PPRL linkage compared to linkage using clear text identifiers. The test linkage dataset is one that is routinely linked (N=2,672,257) at our linkage centre. The population spine (N=8,440,442) includes a record for every person who has resided in British Columbia, Canada over the past 30 years. Weights were determined using LinXmart’s implementation of the Expectation Maximization (EM) algorithm. For both linkages, accepted links were the highest-weighted candidate link with a weight above the threshold suggested by EM estimation. We compare linkage rates and quality and differences in weight and threshold estimations between clear-text and PPRL linkages ResultsClear-text and PPRL methods resulted in 97% and 90% linkage rates, respectively. Approximately 67% of records in the linked datasets contained a nominally unique ID. Records with a unique ID linked at higher rates (>99% for both clear-text and PPRL) while the linkage rate for records missing the ID differed substantially (92% /70% for clear-text/PPRL). Comparing PPRL linkage to the clear-text linkage, we obtain F-measures of 0.99 and 0.80 for records with and without the unique ID, respectively. Conclusion / ImplicationsLinkage performance may be attributable to differences in comparison operators between the two methods. Bloomed fields compared with Dice coefficient allow for partial matching but may not be as sensitive as clear-text string comparisons. Numerical comparisons in PPRL are exact matches while clear-text comparisons allow for more sophisticated matching. Further refinements in PPRL are being explored to improve these results.

IntroductionNotwithstanding the growth in the number and type of datasets that are being included in data linkage projects, some datasets remain ‘hard to include’ in operational linkage systems. Legal or regulatory constraints often restrict the release of personally identifying information from some datasets; alternatively, it may be privacy or reputational risks that prevent data release. Advances in privacy-preserving record linkage (PPRL) methods have made it possible to overcome this impasse. Objectives and ApproachWe present and describe a number of recent Australian ‘use cases’ where the PPRL-Bloom method has been used. For each, we report and reflect on the following: a) The nature of problem or ‘impasse’ being solvedb) The linkage model adoptedc) Quality, performance, privacy and automationd) Challenges, insights and opportunities for improvement ResultsAustralian projects utilising privacy preserving linkage (PPRL-Bloom) include several linking state-based datasets to Commonwealth datasets, some linking primary care data to state-based hospital and other health collections, and others linking state-based non-health datasets such as education, police and justice datasets. Conclusion / ImplicationsPPRL is a useful and innovative methodology for providing access to some ‘hard to get’ datasets. It has already enabled a number of research projects that for regulatory or other privacy related reasons would not have occurred. The use of PPRL in Australia appears likely to grow.