iCoverT: A rich data source on the incidence of child maltreatment over time in England and Wales

Abstract

Background
Child maltreatment is a major public health problem affecting one quarter of children in England and Wales. Good epidemiological data are needed to establish how many and which children are most at risk, and to evaluate the impact of policies and interventions. However, a comprehensive data source on child maltreatment is currently lacking.

Aim
We aimed to create a rich data source on the incidence of Child maltreatment over Time (iCoverT) in England and Wales.

Methods
We developed systematic methods to search and identify administrative data sources that regularly measured child maltreatment. Data sources were investigated and assessed against pre-specified eligibility criteria and a bespoke quality assessment tool. Relevant data were extracted, digitalised, and harmonised over time. All data and their accompanying documentation were prepared to form an open access data source: the iCoverT (osf.io/cf7mv).

Results
We identified 13 unique sources of administrative data, six of which met our eligibility criteria: Child protection statistics, Children in care, Criminal statistics, Homicide index, Mortality statistics and NSPCC statistics. Data and documentation were prepared and combined to form the iCoverT, including 272 variables, over 43,500 data points, and spanning over 150 years. A subsequent time series analysis demonstrated the utility of the iCoverT; identifying large overall decreases in child maltreatment from 1858 to 2016 (e.g. 90% decrease in child homicides (2.7 per fewer per 100,000 children)) but worrying recent increases from 2000 to 2016.

Conclusion
We systematically developed a rich data source on child maltreatment in England and Wales. Our methodology overcomes practical obstacles and offers a new approach for harnessing administrative data for research. Our resulting data source is a valuable public health surveillance tool, which can be used to monitor national levels of child maltreatment and to evaluate the effectiveness of child protection initiatives.

Background

Child maltreatment is a major public health problem affecting one quarter of children in England and Wales. Good epidemiological data are needed to establish how many and which children are most at risk, and to evaluate the impact of policies and interventions. However, a comprehensive data source on child maltreatment is currently lacking.

Aim

We aimed to create a rich data source on the incidence of Child maltreatment over Time (iCoverT) in England and Wales.

Methods

We developed systematic methods to search and identify administrative data sources that regularly measured child maltreatment. Data sources were investigated and assessed against pre-specified eligibility criteria and a bespoke quality assessment tool. Relevant data were extracted, digitalised, and harmonised over time. All data and their accompanying documentation were prepared to form an open access data source: the iCoverT (osf.io/cf7mv).

Results

We identified 13 unique sources of administrative data, six of which met our eligibility criteria: Child protection statistics, Children in care, Criminal statistics, Homicide index, Mortality statistics and NSPCC statistics. Data and documentation were prepared and combined to form the iCoverT, including 272 variables, over 43,500 data points, and spanning over 150 years. A subsequent time series analysis demonstrated the utility of the iCoverT; identifying large overall decreases in child maltreatment from 1858 to 2016 (e.g. 90% decrease in child homicides (2.7 per fewer per 100,000 children)) but worrying recent increases from 2000 to 2016.

Conclusions

We systematically developed a rich data source on child maltreatment in England and Wales. Our methodology overcomes practical obstacles and offers a new approach for harnessing administrative data for research. Our resulting data source is a valuable public health surveillance tool, which can be used to monitor national levels of child maltreatment and to evaluate the effectiveness of child protection initiatives.