Implementing a Child Welfare Decision Aide in Douglas County: Methodology Report
TThis reports summarises the development and implementation of the Douglas County Decision Aide (DCDA), a child welfare predictive risk-modelling tool built and deployed by the Centre for Social Data Analytics. The DCDA is intended to support call screening of child maltreatment referrals made to the Douglas County Department of Human Services (CO, United States). The methodology outlines how Douglas County was triaging referrals prior to the deployment of the DCDA and how the DCDA has changed that, as well as details of how the DCDA tool was built and validated and how ethical concerns were identified and responded to.
Vaithianathan, R., Dinh, H., Kalisher, A., Kithulgoda, C. I., Kulick, E., Mayur, M., … Putnam-Hornstein, E. (December 2019). Implementing a Child Welfare Decision Aide in Douglas County: Methodology Report. Centre for Social Data Analytics.
Exploring the protective factors of children and families identified at highest risk of adverse childhood experiences by a predictive risk model: An analysis of the Growing Up in New Zealand cohort
The authors explore what protective factors might exist for the families of children identified by a predictive risk model as at high risk of experiencing adverse childhood experiences. Identifying protective factors is an important step in designing preventive services for families as well as helping social workers take a strengths-based approach to these families. The authors identify 56 factors associated with protective effects against adversities and find that a positive mother-partner relationship helps children at risk of adversities.
Walsh, M. C., Joyce, S., Maloney, T., & Vaithianathan, R. (2019). Exploring the Protective Factors of Children and Families Identified at Highest Risk of Adverse Childhood Experiences by a Predictive Risk Model: An Analysis of the Growing up in New Zealand Cohort. Children and Youth Services Review, 104556.
Indigenous voices on measuring and valuing health states
This article sets out to start a conversation around what an Indigenous measure of health might look like and how it might value key dimensions of health. Within a Kaupapa Māori theoretical paradigm, in-depth interviews were conducted with six Māori key informants who had cared for whānau (family) members through illness to give voice to dimensions of health and illness that Western economic measures of health fail to capture.
Willing, E., Paine, S.-J., Wyeth, E., Ao, B. T., Vaithianathan, R., & Reid, P. (2019). Indigenous voices on measuring and valuing health states. AlterNative: An International Journal of Indigenous Peoples, 117718011988541. doi: 10.1177/1177180119885418
Ethnic Disparities in Childhood Prevalence of Maltreatment: Evidence From a New Zealand Birth Cohort
This article aims to document ethnic disparities in childhood abuse and neglect among New Zealand children. By following the 1998 New Zealand birth cohort of 56,904 children through 2016, the authors identify substantial ethnic differences in child maltreatment and child protection involvement. They find that despite long-standing child support policies and reparation for breaches of Indigenous people’s rights, significant child maltreatment disparities persist.
Rouland, B., Vaithianathan, R., Wilson,
Adverse childhood experiences and school readiness outcomes: Results from the Growing up in New Zealand study
This report maps standard ‘adverse childhood experiences’ (ACEs) to the Growing Up in New Zealand (GUiNZ) study cohort to explore associations between adverse experiences in early childhood and measures of school readiness.
Walsh, M. C., Maloney, T., Vaithianathan, R., & Joyce, S. (2019). Adverse childhood experiences and school readiness outcomes Results from the Growing up in New Zealand study, Wellington: Ministry of Social Development.
Protective factors of children and families at highest risk of adverse childhood experiences: An analysis of children and families in the Growing up in New Zealand data who “beat the odds”
With increasing access to integrated administrative data, it is easy to identify infants who are likely to suffer childhood adversities. However, many infants who appear “at risk” end up thriving, experiencing few of the adversities that beset other children with similar risk factors. Understanding what helps children “beat the odds” is important for policy-makers and frontline services that want to help families at risk. The present report analyses the Growing Up in New Zealand (GUiNZ) birth cohort to identify protective factors for at-risk children who “beat the odds”.
Walsh, M. C., Maloney, T., Vaithianathan, R., & Joyce, S. (2019). Protective factors of children and families at highest risk of adverse childhood experiences: An analysis of children and families in the Growing up in New Zealand data who “beat the odds” Wellington: Ministry of Social Development.
Adverse childhood experiences and school readiness outcomes: Results from the Growing Up in New Zealand study
(N. Z. Med. J.)
The Center for Disease Control’s (CDC) Adverse Childhood Experiences (ACEs) have been associated with adverse health consequences in adults and children, but less is known about any association between ACE and early learning skills. We investigated the relationship between ACEs and objective preschool measures of skills using the Growing up In New Zealand (GUiNZ) cohort study.
Walsh, M. C., Joyce, S., Maloney, T., & Vaithianathan, R. (2019). Adverse childhood experiences and school readiness outcomes: results from the Growing Up in New Zealand study. The New Zealand medical journal, 132(1493), 15‐24.
Toward Algorithmic Accountability in Public Services: A Qualitative Study of Affected Community Perspectives on Algorithmic Decision‐making in Child Welfare Services.
Researchers, policy experts, and civil rights groups have all voiced concerns that algorithmic decision-making systems are being deployed without adequate consideration of potential harms, disparate impacts, and public accountability practices. Yet little is known about the concerns of those most likely to be affected by these systems. The authors, including Rhema Vaithianathan, report and discuss the findings of workshops conducted to learn about the concerns of affected communities in the context of child welfare services.
Brown, A., Chouldechova, A., Putnam‐Hornstein, E., Tobin, A., & Vaithianathan, R. (2019, April). Toward Algorithmic Accountability in Public Services: A Qualitative Study of Affected Community Perspectives on Algorithmic Decision‐making in Child Welfare Services. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems (p. 41). ACM.
He awa whiria—braided rivers: Understanding the outcomes from Family Start for Māori
In Aotearoa New Zealand the “braided rivers—he awa whiria” metaphor is facilitating conversations between Māori (indigenous peoples) and non-Māori researchers about the integration of knowledge systems. Rhema Vaithianathan, Tim Maloney and their co-authors explore how an approach based on he awa whiria can work in practice in the examination of the efficacy for Māori whānau (families) of the government’s intensive home-visiting programme, Family Start. Published in Evaluation Matters—He Take Tō Te Aromatawai Online First.
Cram, F., Vette, M., Wilson, M., Vaithianathan, R., Maloney, T., & Baird, S. (2018). He awa whiria—braided rivers: Understanding the outcomes from Family Start for Maori. Evaluation Matters, 165-207.
Estimating the economic costs of ethnic health inequalities: protocol for a prevalence-based cost-of-illness study in New Zealand (2003-2014)
This paper is a protocol for a proposed study, by researchers including Rhema Vaithianathan, that will investigate inequities in health between the indigenous Māori and non-Māori adult population in New Zealand and estimate the economic costs associated with these differences.
Reid, P., Paine, S., Te Ao, B., Vaithianathan, R., Willing, E., & Wyeth, E. (2018, June 19). Estimating the Economic Costs of Ethnic Health Inequties: Protocol for a Prevalence-Based-Cost-of-Illness-Study in New Zealand (2003-2014). BMJ Open, 1-7.
Labour market effects of activating sick-listed workers
Bénédicte Rouland and her co-authors use data from a large-scale randomized controlled trial conducted in Danish job centres to investigate the effects of activating sick-listed workers on subsequent labour market outcomes. Comparing treated and controls, the authors find an overall unfavourable effect on subsequent labour market outcomes.
Rehwald, K., Rosholm, M., & Rouland, B. (2018). Labour market effects of activating sick-listed workers. Labour Economics.
Cumulative Prevalence of Maltreatment Among New Zealand Children, 1998–2015
Bénédicte Rouland and Rhema Vaithianathan explore the cumulative prevalence in New Zealand of notifications to child protective services, substantiated maltreatment cases and out-of-home placements. The study shows that 1 in 4 New Zealand children will be subject to at least 1 notification at age 17, an incidence of notification higher than that of medicated asthma among children.
Rouland, B., & Vaithianathan, R. (2018). Cumulative Prevalence Of Maltreatment Among New Zealand Children, 1998–2015 . American journal of public health, 108(4), 511-513.
The impact of Social Workers in Schools: A preliminary investigation using linked administrative data
This paper reports on a preliminary investigation of the impact of the government-funded Social Workers in Schools service (SWiS) which is available in selected primary and intermediate schools. The aim of SWiS is to see safe, healthy and socialised children with a strong sense of identity, who are fully engaged in school. The present analysis compares outcomes for children who attended SWiS schools with similar children who attended similar schools that at the time were not part of the SWiS programme. The study finds evidence of a reduction in non-enrollment days, a 5% increase for girls attaining NCEA level 1 by their 16th birthday, and a reduction in CYF youth justice referrals for the first three years of high school.
Jiang, N., Maloney, T., Staneva, A., Wilson, M., & Vaithianathan, R. (2018). The impact of Social Workers in Schools: A preliminary investigation using linked administrative data.
A Case Study of Algorithm-Assisted Decision Making in Child Maltreatment Hotline Screening Decisions
Diana Benavides-Prado, Oleksandr Fialko, Rhema Vaithianathan, Alexandra Chouldechova and Emily Putnam-Hornstein describe their work on developing, validating, fairness auditing, and deploying a risk prediction model in Allegheny County, PA, USA. They discuss the results of their analysis to-date and highlight critical problems and data bias issues that present challenges for model evaluation and deployment.
Chouldechova, A., Putnam-Hornstein, E., Benavides-Prado, D., Fialko, O., & Vaithianathan, R. (2018). A case study of algorithm-assisted decision making in child maltreatment hotline screening decisions. Proceedings of Machine Learning Research, 1-15.
Injury and Mortality Among Children Identified as at High Risk of Maltreatment
This study by Rhema Vaithianathan, Bénédicte Rouland and Emily Putnam-Hornstein explored a model which assigned risk scores for the risk of a substantiated finding of maltreatment to children born in New Zealand in 2010, to see whether children at the highest 10% and 20% of risk would have an elevated chance of injury or death in early childhood. The study found that children assessed as being “very high risk” (highest 10%) and “high-risk” (highest 20%) had 4.8 times and 4.2 times respectively higher post-neonatal mortality rates than other children.
Vaithianathan, R., Rouland, B., & Putnam-Hornstein, E. (2017). “Injury and Mortality Among Children Identified as at High Risk of Maltreatment”. Pediatrics 141(2).
Risk assessment and decision making in child protective services: Predictive risk modeling in context
The authors, including Rhema Vaithianathan, review the literature and provide a context for predictive risk modelling in the current risk assessment paradigm in child protective services. They describe how predictive analytics or predictive risk modelling using linked administrative data may provide a useful complement to current approaches.
Cuccaro-Alamin, S., Foust, R., Vaithianathan, R., & Putnam-Hornstein, E. (2017). “Risk assessment and decision making in child protective services: Predictive risk modelling in context”. Children and Youth Services Review 79
Decomposing ethnic differences in university academic achievement in New Zealand
Tim Maloney and Zhaoyi Cao use individual-level administrative data to examine the extent and potential explanations for the relatively poorer academic performance of three ethnic minority groups in their first year of study at a New Zealand university. Substantial differences in course completion rates and letter grades are found for Māori, Pasifika, and Asian students relative to their European counterparts. These large and significant gaps persist in the face of alternative definitions of ethnicity and sample restrictions.
Cao, Zhaoyi & Maloney, Tim. (2017). Decomposing ethnic differences in university academic achievement in New Zealand. Higher Education. 75. 10.1007/s10734-017-0157-6.
Family Start Impact Study: Selected Extensions
This report provides selected extensions to a 2016 quasi-experimental impact evaluation of the Family Start Home Visiting programme. The extensions explore the efficacy of Family Start for additional sub-populations of participant families, and when delivered by sub-populations of providers. Results indicate that Family Start was effective in reducing some measures of post-neonatal infant mortality across sub-groups studied, including teen and non-teen mothers, children in families with and without past contact with Child Youth and Family, and Maori children receiving Family Start from Maori and mainstream providers.
Vaithianathan, R., Wilson, M., Maloney, T. and Baird, S. (2016). The Impact of the Family Start Home Visiting Programme on Outcomes for Mothers and Children: A Quasi-Experimental Study. Wellington: Ministry of Social Development.
Developing Predictive Risk Models to Support Child Maltreatment Hotline Screening Decisions: Allegheny County Methodology and Implementation
This methodology document traces the development and implementation of the Allegheny Family Screening Tool (AFST), a child welfare predictive risk-modelling tool built by a research team led by Rhema Vaithianathan. The AFST is intended to support call screening of child maltreatment referrals by Allegheny County Department of Human Services. The methodology document includes an overview of existing practice, methodology and performance metrics of the model and details of the external validation and next steps for the model rebuild.
Vaithianathan, R., Jiang, N., Maloney, T., Nand, P., & Putnam-Hornstein, E. (2017). Developing Predictive Risk Models to Support Child Maltreatment Hotline Screening Decisions: Allegheny County Methodology. Centre for Social Data Analytics. Auckland: Centre for Social Data Analytics.
Black–White Differences in Child Maltreatment Reports and Foster Care Placements: A Statistical Decomposition Using Linked Administrative Data
The age and marital status of parents explains racial disparities in Child Protective Service (CPS) involvement, according to a study co-authored by Tim Maloney and Rhema Vaithianathan. In the study, birth records for all children born in Allegheny County, Pennsylvania, between 2008 and 2010 were linked to administrative service records.
Maloney, T., Jiang, N., Putnam-Hornstein, E., Dalton, E., & Vaithianathan, R. (2017). Black–White differences in child maltreatment reports and foster care placements: A statistical decomposition using linked administrative data. Maternal and child health journal, 21(3), 414-420.
Impact of school-based support on educational outcomes of teen-mothers: Evidence from New Zealand's "Teen Parent Units"
Researchers found New Zealand teenage mothers with access to Teen Parent Units (TPUs) had better educational outcomes than those with no access. This research was conducted through the linkage of social sector administrative data.
Vaithianathan, R., Maloney, T., Wilson, M., Staneva, A., & Jiang, N. (2017) Impact of school-based support on educational outcomes of teen-mothers: Evidence from New Zealand's "Teen Parent Units" Working Paper.
Impact of the Family Start Home Visiting Programme on Outcomes for Mothers and Children: A Quasi-Experimental Study
Family Start workers make regular home visits and, using a structured program, seek to improve parenting capability and practice. These resources from the Ministry of Social Development evaluate the impact of the Family Start programme in improving outcomes for participating families
Vaithianathan, R., Wilson, M., Maloney, T., & Baird, S. (2016). The Impact of the Family Start Home Visiting Programme on Outcomes for Mothers and Children: A Quasi-Experimental Study. Ministry of Social Development.
Using predictive modelling to identify students at risk of poor university outcomes
This paper outlines the use of predictive risk modeling tools to identify vulnerable students and understand factors that place these students at risk of non-completion and non-retention early in their university careers. Administrative data from the enrollment process is used to identify the factors contributing to these adverse outcomes.
Jia, P., & Maloney, T. (2015).Using predictive modelling to identify students at risk of poor university outcomes. Higher Education, 70(1), 127-149.
Demand in New Zealand hospitals: Expect the unexpected?
This article presents a model to predict patient demand at the hospital facility level, which is established with data from a national database of hospital admissions.The paper also constructs two indicators of demand shocks, at hospital and disease chapter levels, including an assessment of the consequent impact on patient outcomes.
Jiang, N., & Pacheco, G. (2014). Demand in New Zealand hospitals: expect the unexpected? Applied Economics, 46(36), 4475-4489.
Addressing child maltreatment in New Zealand: Is poverty reduction enough?
This publication discusses the use of predictive risk models to target early and appropriate intervention to reduce child maltreatment and suggests potential issues that may arise from this approach to child protection.\
Dare, T., Vaithianathan, R., & De Haan, I. (2014). Addressing child maltreatment in New Zealand: is poverty reduction enough?. Educational philosophy and theory, 46(9), 989-994.
Integrating care for high risk patients in England using the virtual ward model
This paper provides a discussion of the extent to which integrated care was achieved in the analysis of virtual wards in Croydon, Devon and Wandsworth. Key success factors and challenges are identified from the results of the case studies.
Lewis, G., Vaithianathan, R., Wright, L., Brice, M., Lovell, P., Rankin, S., & Bardsley, M. (2013). Integrating care for high-risk patients in England using the virtual ward model: lessons in the process of care integration from three case sites. International journal of integrated care, 13(4).
Children in the public benefit system at risk of maltreatment
This paper outlines the method of using an automated predictive risk model to identify children at high risk of maltreatment, as well as discussing the strategic targeting of prevention activities toward these individuals.
Vaithianathan, R., Maloney, T., Putnam-Hornstein, E., & Jiang, N. (2013). Children in the public benefit system at risk of maltreatment: Identification via predictive modeling. American journal of preventive medicine, 45(3), 354-359.