Identification of ICD codes suggestive of child maltreatment

Evaluating the real-world implementation of the Family Nurse Partnership in England: a data linkage study

Background/objectives

The Family Nurse Partnership is an intensive home visiting programme for adolescent mothers. We aimed to evaluate the effectiveness of the Family Nurse Partnership on outcomes up to age 7 using national administrative data.

Design

We created a linked cohort of all mothers aged 13-19 using data from health, educational and children's social care and defined mothers enrolled in the Family Nurse Partnership or not using Family Nurse Partnership system data. Propensity scores were used to create matched groups for analysis.

Setting

One hundred and thirty-six local authorities in England with active Family Nurse Partnership sites between 2010 and 2017.

Participants

Mothers aged 13-19 at last menstrual period with live births between April 2010 and March 2019, living in a Family Nurse Partnership catchment area and their firstborn child(ren).

Interventions

The Family Nurse Partnership includes up to 64 home visits by a family nurse from early pregnancy until the child's second birthday and is combined with usual health and social care. Controls received usual health and social care.

Main outcome measures

Indicators of child maltreatment (hospital admissions for injury/maltreatment, referral to social care services); child health and development (hospital utilisation and education) outcomes and maternal hospital utilisation and educational outcomes up to 7 years following birth.

Data sources

Family Nurse Partnership Information System, Hospital Episode Statistics, National Pupil Database.

Results

Of 110,520 eligible mothers, 25,680 (23.2%) were enrolled in the Family Nurse Partnership. Enrolment rates varied across 122 sites (range: 11-68%). Areas with more eligible mothers had lower enrolment rates. Enrolment was higher among mothers aged 13-15 (52%), than 18-19 year-olds (21%). Indicators of child maltreatment: we found no evidence of an association between the Family Nurse Partnership and indicators of child maltreatment, except for an increased rate of unplanned admissions for maltreatment/injury-related diagnoses up to age 2 for children born to Family Nurse Partnership mothers (6.6% vs. 5.7%, relative risk 1.15; 95% confidence interval 1.07 to 1.24). Child health and developmental outcomes: there was weak evidence that children born to Family Nurse Partnership mothers were more likely to achieve a Good Level of Development at age 5 (57.5% vs. 55.4%, relative risk 1.05; 95% confidence interval 1.00 to 1.09). Maternal outcomes: There was some evidence that Family Nurse Partnership mothers were less likely to have a subsequent delivery within 18 months of the index birth (8.4% vs. 9.3%, relative risk 0.92; 95% confidence interval 0.88 to 0.97). Younger and more vulnerable mothers received higher numbers of visits and were more likely to achieve fidelity targets. Meeting the fidelity targets was associated with some outcomes.

Limitations

Bias by indication and variation in the intervention and usual care over time and between areas may have limited our ability to detect effects. Multiple testing may have led to spurious, significant results.

Conclusions

This study supports findings from evaluations of the Family Nurse Partnership showing no evidence of benefit for maltreatment outcomes measured in administrative data. Amongst all the outcomes measured, we found weak evidence that the Family Nurse Partnership was associated with improvements in child development at school entry, a reduction in rapid repeat pregnancies and evidence of increased healthcare-seeking in the mother and child.

Future work

Future evaluations should capture better measures of Family Nurse Partnership interventions and usual care, more information on maternal risk factors and additional outcomes relating to maternal well-being.

Study registration

The study is registered as NIHR CRN Portfolio (42900).

Funding

This award was funded by the National Institute of Health and Care Research (NIHR) Health and Social Care Delivery Research programme (NIHR award ref: 17/99/19) and is published in full in Health and Social Care Delivery Research; Vol. 12, No. 11. See the NIHR Funding and Awards website for further award information.

Risk of Injury-Related Readmission Varies by Initial Presenting-Hospital Type Among Children at High Risk for Physical Abuse

OBJECTIVES

Children experiencing physical abuse may initially present to hospitals with underappreciated minor injuries, only to experience more severe injuries in the future. The objectives of this study were to 1) describe young children presenting with high-risk diagnoses for physical abuse, 2) characterize the hospitals to which they initially presented, and 3) evaluate associations of initial presenting-hospital type with subsequent admission for injury.

METHODS

Patients aged younger than 6 years from the 2009-2014 Florida Agency for Healthcare Administration database with high-risk diagnoses (codes previously associated with >70% risk of child physical abuse) were included. Patients were categorized by the hospital type to which they initially presented: community hospital, adult/combined trauma center, or pediatric trauma center. Primary outcome was subsequent injury-related hospital admission within 1 year. Association of initial presenting-hospital type with outcome was evaluated with multivariable logistic regression, adjusting for demographics, socioeconomic status, preexisting comorbidities, and injury severity.

RESULTS

A total of 8626 high-risk children met inclusion criteria. Sixty-eight percent of high-risk children initially presented to community hospitals. At 1 year, 3% of high-risk children had experienced subsequent injury-related admission. On multivariable analysis, initial presentation to a community hospital was associated with higher risk of subsequent injury-related admission (odds ratio, 4.03 vs level 1/pediatric trauma center; 95% confidence interval, 1.83-8.86). Initial presentation to a level 2 adult or combined adult/pediatric trauma center was also associated with higher risk for subsequent injury-related admission (odds ratio, 3.19; 95% confidence interval, 1.40-7.27).

CONCLUSIONS

Most children at high risk for physical abuse initially present to community hospitals, not dedicated trauma centers. Children initially evaluated in high-level pediatric trauma centers had lower risk of subsequent injury-related admission. This unexplained variability suggests stronger collaboration is needed between community hospitals and regional pediatric trauma centers at the time of initial presentation to recognize and protect vulnerable children.

Identifying Children at Risk for Maltreatment Using Emergency Medical Services' Data: An Exploratory Study

The objective of this study was to use natural language processing to query Emergency Medical Services (EMS) electronic health records (EHRs) to identify variables associated with child maltreatment. We hypothesized the variables identified would show an association between the Emergency Medical Services encounter and risk of a children maltreatment report. This study is a retrospective cohort study of children with an EMS encounter from 1/1/11-12/31/18. NLP of EMS EHRs was conducted to generate single words, bigrams and trigrams. Clinically plausible risk factors for child maltreatment were established, where presence of the word(s) indicated presence of the hypothesized risk factor. The EMS encounters were probabilistically linked to child maltreatment reports. Univariable associations were assessed, and a multivariable logistic regression was conducted to determine a final set of predictors. 11 variables showed an association in the multivariable modeling. Sexual, abuse, chronic condition, developmental delay, unconscious on arrival, criminal activity/police, ingestion/inhalation/exposure, and <2 years old showed positive associations with child maltreatment reports. Refusal and DOA/PEA/asystole held negative associations. This study demonstrated that through EMS EHRs, risk factors for child maltreatment can be identified. A future direction of this work include developing a tool that screens EMS EHRs for households at risk for maltreatment.

Association of maternal risk factors with infant maltreatment: an administrative data cohort study

OBJECTIVE

We aimed to evaluate the risk of infant maltreatment associated with commonly used criteria for home visiting programmes: young maternal age, maternal adversity (homelessness, substance abuse, intimate partner violence), newcomer status and mental health concerns in Ontario, Canada.

DESIGN

This retrospective cohort study included infants born in hospital in Ontario from 1 April 2005 to 31 March 2017 captured in linked health administrative and demographic databases. Infants were followed from newborn hospitalisation until 1 year of age for child maltreatment captured in healthcare or death records. The association between type and number of maternal risk factors, and maltreatment, was analysed using multivariable logistic regression modelling, controlling for infant characteristics and material deprivation. Further modelling explored the association of each year of maternal age with maltreatment.

RESULTS

Of 989 586 infants, 434 (0.04%) had recorded maltreatment. Maternal age <22 years conferred higher risk of infant maltreatment (adjusted OR (aOR) 5.5, 95% CI 4.5 to 6.8) compared with age ≥22 years. Maternal mental health diagnoses (aOR 2.0, 95% CI 1.6 to 2.5) were also associated with maltreatment, while refugee status appeared protective (aOR 0.6, 95% CI 0.4 to 1.0). The odds of maltreatment increased with higher numbers of maternal risk factors. Maternal age was associated with maltreatment until age 28 years.

CONCLUSION

Infants born to young mothers are at greater risk of infant maltreatment, and this association remained until age 28 years. These findings are important for ensuring public health interventions are supporting populations experiencing structural vulnerabilities with the aim of preventing maltreatment.

Follow-Up Care After Maltreatment: Sociodemographic Associations With Timeliness in a Southern State

Children who experience maltreatment are at elevated risk of developing mental health difficulties. Even so, they often do not receive timely, evidenced-based mental health treatment, which may exacerbate the risk of poor outcomes. This study aims to describe the receipt of timely follow-up care after maltreatment in a southern state with known treatment shortages and aims to identify factors associated with timely follow-up care. We utilized a retrospective cohort design using 2014 Mississippi Medicaid administrative claims data for youth 0-18 years. Prevalence estimates and associations with definite and probable maltreatment (based on recorded age/injury combinations) during inpatient and outpatient healthcare encounters were evaluated. Rates of 30-day maltreatment follow-up with any medical or behavioral health provider were also assessed. Prevalence estimates of definite and probable maltreatment in the eligible study population (N = 324,752) were 0.53% and 3.8%, respectively. Only one-third of identified children received 30-day follow-up. Black and older children as well as children diagnosed with anxiety or depression were more likely to receive 30-day follow-up than younger children, white children, and children without anxiety or depression. Low rates of timely follow-up indicate the need for intentional workflow practices to increase the likelihood of follow-up.

The Impact of COVID-19 on Infant Maltreatment Emergency Department and Inpatient Medical Encounters

OBJECTIVE

To assess the counts of infant maltreatment-related medical encounters at a large medical system during a 21-month span of the COVID-19 pandemic.

METHODS

Retrospective data for this study came from all inpatient and emergency department medical encounters for infants from January 1, 2016, through November 30, 2021, at a single children's hospital system in California. Distributions of medical encounters were tabulated and plotted over time. Interrupted time series models were used to evaluate changes in child maltreatment medical encounters.

RESULTS

Medical encounters for infants with child maltreatment diagnoses increased following the onset of COVID-19. Monthly counts of encounters with indicated maltreatment trended upward following the start of the pandemic. Interrupted time series models showed that the count of maltreatment encounters increased 64% with the onset of COVID-19.

CONCLUSIONS

We found an increase in infant maltreatment medical encounters during a 21-month period following the onset of COVID-19. These findings suggest that the pandemic may have adversely affected the safety of infants and ongoing work is needed to understand better the pandemic impacts on child maltreatment.

Child Maltreatment: A Comparison of Wisconsin Children Identified in Health Care and Child Protective Services (CPS) Records, 2018-2019

CONTEXT

Child Protective Services (CPS) reports and health records (hospital inpatient and emergency department visits) are the primary data sources to measure child maltreatment; yet, they are not linked at the state or national level. Linking provides novel insight into the demographic characteristics of the populations served by one or both agencies, thus informing opportunities for prevention and intervention.

OBJECTIVE

This study compares children identified as maltreated in health records (based on International Classification of Diseases, Tenth Revision, Clinical Modification [ ICD-10-CM ] codes indicating suspected or confirmed child maltreatment) and/or CPS reports.

DESIGN

Three mutually exclusive comparison groups were created: group 1- children with a nonfatal hospitalization and/or emergency department visit with a maltreatment ICD-10-CM code and an investigated CPS report; group 2- children with a maltreatment ICD-10-CM code in a health record without an investigated CPS report; and group 3- children with an investigated CPS report without a health record with a maltreatment ICD-10-CM code. Descriptive statistics and tests for statistically significant differences were conducted.

SETTING AND PARTICIPANTS

Wisconsin children with maltreatment ICD-10-CM code in health record and/or CPS-investigated report from 2018 to 2019.

MAIN OUTCOME MEASURE

Demographic and maltreatment type differences between groups.

RESULTS

Group 1 children were significantly younger than those in other groups ( P <. 001), and group 2 children were significantly more female (69.6%; P < .001). Black children were overrepresented in all groups and most disproportionate among group 1 (39.5%; P < .001). Maltreatment type also differed significantly across groups.

CONCLUSIONS

CPS staff and health providers encounter overlapping and nonoverlapping populations of children experiencing different types of maltreatment. Although interventions may be tailored toward the type of maltreatment and other relevant child characteristics, all populations could benefit from referrals and access to supportive social services. Agency contact provides an opportunity to intervene and support at-risk children and families.

Hospitalization for child maltreatment and other types of injury during the COVID-19 pandemic

BACKGROUND

The possibility that child maltreatment was misclassified as unintentional injury during the COVID-19 pandemic has not been evaluated.

OBJECTIVE

We assessed if child maltreatment hospitalizations changed during the pandemic, and if the change was accompanied by an increase in unintentional injuries.

PARTICIPANTS AND SETTING

This study included children aged 0-4 years who were admitted for maltreatment or unintentional injuries between April 2006 and March 2021 in hospitals of Quebec, Canada.

METHODS

We used interrupted time series regression to estimate the effect of the pandemic on hospitalization rates for maltreatment, compared with unintentional transport accidents, falls, and mechanical force injuries. We assessed if the change in maltreatment hospitalization was accompanied by an increase in specific types of unintentional injury.

RESULTS

Hospitalizations for child maltreatment decreased from 16.3 per 100,000 (95 % CI 9.1-23.4) the year before the pandemic to 13.2 per 100,000 (95 % CI 6.7-19.7) during the first lockdown. Hospitalizations for most types of unintentional injury also decreased, but injuries due to falls involving another person increased from 9.0 to 16.5 per 100,000. Hospitalization rates for maltreatment and unintentional injury remained low during the second lockdown, but mechanical force injuries involving another person increased from 3.8 to 8.1 per 100,000.

CONCLUSIONS

Hospitalizations for child maltreatment may have been misclassified as unintentional injuries involving another person during the pandemic. Children admitted for these types of unintentional injuries may benefit from closer assessment to rule out maltreatment.

An external validation of coding for childhood maltreatment in routinely collected primary and secondary care data

Validated methods of identifying childhood maltreatment (CM) in primary and secondary care data are needed. We aimed to create the first externally validated algorithm for identifying maltreatment using routinely collected healthcare data. Comprehensive code lists were created for use within GP and hospital admissions datasets in the SAIL Databank at Swansea University working with safeguarding clinicians and academics. These code lists build on and refine those previously published to include an exhaustive set of codes. Sensitivity, specificity and positive predictive value of previously published lists and the new algorithm were estimated against a clinically assessed cohort of CM cases from a child protection service secondary care-based setting-'the gold standard'. We conducted sensitivity analyses to examine the utility of wider codes indicating Possible CM. Trends over time from 2004 to 2020 were calculated using Poisson regression modelling. Our algorithm outperformed previously published lists identifying 43-72% of cases in primary care with a specificity ≥ 85%. Sensitivity of algorithms for identifying maltreatment in hospital admissions data was lower identifying between 9 and 28% of cases with high specificity (> 96%). Manual searching of records for those cases identified by the external dataset but not recorded in primary care suggest that this code list is exhaustive. Exploration of missed cases shows that hospital admissions data is often focused on the injury being treated rather than recording the presence of maltreatment. The absence of child protection or social care codes in hospital admissions data poses a limitation for identifying maltreatment in admissions data. Linking across GP and hospital admissions maximises the number of cases of maltreatment that can be accurately identified. Incidence of maltreatment in primary care using these code lists has increased over time. The updated algorithm has improved our ability to detect CM in routinely collected healthcare data. It is important to recognize the limitations of identifying maltreatment in individual healthcare datasets. The inclusion of child protection codes in primary care data makes this an important setting for identifying CM, whereas hospital admissions data is often focused on injuries with CM codes often absent. Implications and utility of algorithms for future research are discussed.

Economic Hardship and Violence: A Comparison of County-Level Economic Measures in the Prediction of Violence-Related Injury

Economic hardship may lead to a wide range of negative outcomes, including violence. However, existing literature on economic hardship and violence is limited by reliance on official reports of violence and conflation of different measures of economic hardship. The goals of this study are to measure how violence-related injuries are associated with five measures of county-level economic shocks: unemployment rate, male mass layoffs, female mass layoffs, foreclosure rate, and unemployment rate change, measured cross-sectionally and by a 1-year lag. This study measures three subtypes of violence outcomes (child abuse, elder abuse, and intimate partner violence). Yearly county-level data were obtained on violence-related injuries and economic measures from 2005 to 2012 for all 87 counties in Minnesota. Negative binomial models were run regressing the case counts of each violence outcome at the county-year level on each economic indicator modeled individually, with population denominator offsets to yield incidence rate ratios. Crude models were run first, then county-level socio-demographic variables and year were added to each model, and finally fully-adjusted models were run including all socio-demographic variables plus all economic indicators simultaneously. In the fully-adjusted models, a county's higher foreclosure rate is the strongest and most consistently associated with an increase in all violence subtypes. Unemployment rate is the second strongest and most consistent economic risk factor for all violence subtypes. Lastly, there appears to be an impact of gender specific to economic impacts on child abuse; specifically, male mass-lay-offs were associated with increased rates while female mass-lay-offs were associated with decreased rates. Understanding the associations of different types of economic hardship with a range of violence outcomes can aid in developing more holistic prevention and intervention efforts.

Validating Use of ICD-10 Diagnosis Codes in Identifying Physical Abuse Among Young Children

OBJECTIVE

Evaluate the positive predictive value of International Classification of Disease, 10th Revision, Clinical Modification (ICD-10-CM) codes in identifying young children diagnosed with physical abuse.

METHODS

We extracted 230 charts of children <24 months of age who had any emergency department, inpatient, or ambulatory care encounters between Oct 1, 2015 and Sept 30, 2020 coded using ICD-10-CM codes suggestive of physical abuse. Electronic health records were reviewed to determine if physical abuse was considered during the medical encounter and assess the level of diagnostic certainty for physical abuse. Positive predictive value of each ICD-10-CM code was assessed.

RESULTS

Of 230 charts with ICD-10 codes concerning for physical abuse, 209 (91%) had documentation that a diagnosis of physical abuse was considered during an encounter. The majority of cases, 138 (60%), were rated as definitely or likely abuse, 36 cases (16%) were indeterminate, and 35 (15%) were likely or definitely accidental injury. Other forms of suspected maltreatment were discussed in 16 (7%) charts and 5 (2%) had no documented concerns for child maltreatment. The positive predictive values of the specific ICD-10 codes for encounters rated as definitely or likely abuse varied considerably, ranging from 0.89 (0.80-0.99) for T74.12 "Adult and child abuse, neglect, and other maltreatment, confirmed" to 0.24 (95% CI: 0.06-0.42) for Z04.72 "Encounter for examination and observation following alleged child physical abuse."

CONCLUSIONS

ICD-10-CM codes identify young children who experience physical abuse, but certain codes have a higher positive predictive value than others.

Leveraging Administrative Data to Better Understand and Address Child Maltreatment: A Scoping Review of Data Linkage Studies

BACKGROUND

This scoping review aimed to overview studies that used administrative data linkage in the context of child maltreatment to improve our understanding of the value that data linkage may confer for policy, practice, and research.

METHODS

We searched MEDLINE, Embase, PsycINFO, CINAHL, and ERIC electronic databases in June 2019 and May 2020 for studies that linked two or more datasets (at least one of which was administrative in nature) to study child maltreatment. We report findings with numerical and narrative summary.

RESULTS

We included 121 studies, mainly from the United States or Australia and published in the past decade. Data came primarily from social services and health sectors, and linkage processes and data quality were often not described in sufficient detail to align with current reporting guidelines. Most studies were descriptive in nature and research questions addressed fell under eight themes: descriptive epidemiology, risk factors, outcomes, intergenerational transmission, predictive modelling, intervention/service evaluation, multi-sector involvement, and methodological considerations/advancements.

CONCLUSIONS

Included studies demonstrated the wide variety of ways in which data linkage can contribute to the public health response to child maltreatment. However, how research using linked data can be translated into effective service development and monitoring, or targeting of interventions, is underexplored in terms of privacy protection, ethics and governance, data quality, and evidence of effectiveness.

Do children evaluated for maltreatment have higher subsequent emergency department and inpatient care utilization compared to a general pediatric sample?

BACKGROUND

Child maltreatment leads to substantial adverse health outcomes, but little is known about acute health care utilization patterns after children are evaluated for a concern of maltreatment at a child abuse and neglect medical evaluation clinic.

OBJECTIVE

To quantify the association of having a child maltreatment evaluation with subsequent acute health care utilization among children from birth to age three.

PARTICIPANTS AND SETTING

Children who received a maltreatment evaluation (N = 367) at a child abuse and neglect subspecialty clinic in an academic health system in the United States and the general pediatric population (N = 21,231).

METHODS

We conducted a retrospective cohort study that compared acute health care utilization over 18 months between the two samples using data from electronic health records. Outcomes were time to first emergency department (ED) visit or inpatient hospitalization, maltreatment-related ED use or inpatient hospitalization, and ED use or inpatient hospitalization for ambulatory care sensitive conditions (ACSCs). Multilevel survival analyses were performed.

RESULTS

Children who received a maltreatment evaluation had an increased hazard for a subsequent ED visit or inpatient hospitalization (hazard ratio [HR]: 1.3, 95 % confidence interval [CI]: 1.1, 1.5) and a maltreatment-related visit (HR: 4.4, 95 % CI: 2.3, 8.2) relative to the general pediatric population. A maltreatment evaluation was not associated with a higher hazard of health care use for ACSCs (HR: 1.0, 95 % CI: 0.7, 1.3).

CONCLUSION

This work can inform targeted anticipatory guidance to aid high-risk families in preventing future harm or minimizing complications from previous maltreatment.

Population-Level Trends in Emergency Department Encounters for Sexual Assault Preceding and During the COVID-19 Pandemic Across Ontario, Canada

IMPORTANCE

Lockdown measures and the stress of the COVID-19 pandemic are factors associated with increased risk of violence, yet there is limited information on trends in emergency department (ED) encounters for sexual assault.

OBJECTIVE

To compare changes in ED encounters for sexual assault during the COVID-19 pandemic vs prepandemic estimates.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective, population-based cohort study used linked health administrative data from 197 EDs across Ontario, Canada, representing more than 15 million residents. Participants included all patients who presented to an ED in Ontario from January 11, 2019, to September 10, 2021. Male and female individuals of all ages were included. Data analysis was performed from March to October 2022.

EXPOSURES

Sexual assault, defined through 27 International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, procedure and diagnoses codes.

MAIN OUTCOMES AND MEASURES

Ten bimonthly time periods were used to compare differences in the frequency and rates of ED encounters for sexual assault between 2020 to 2021 (during the pandemic) compared with baseline prepandemic rates in 2019. Rate differences (RDs) and age adjusted rate ratios (aRRs) and Wald 95% CIs were calculated using Poisson regression.

RESULTS

From January 11, 2019, to September 10, 2021, there were 14 476 656 ED encounters, including 10 523 for sexual assault (9304 [88.4%] among female individuals). The median (IQR) age was 23 (17-33) years for female individuals and 15 (4-29) years for male individuals. Two months before the pandemic, ED encounters increased for sexual assault among female individuals (8.4 vs 6.9 cases per 100 000; RD, 1.51 [95% CI, 1.06 to 1.96]; aRR, 1.22 [95% CI, 1.09 to 1.38]) and male individuals (1.2 vs 1.0 cases per 100 000; RD, 0.19 [95% CI, 0.05 to 0.36]; aRR, 1.19 [95% CI, 0.87 to 1.64]). During the first 2 months of the pandemic, the rates decreased for female individuals (4.2 vs 8.3 cases per 100 000; RD, -4.07 [95% CI, -4.48 to -3.67]; aRR, 0.51 [95% CI, 0.44 to 0.58]) and male individuals (0.5 vs 1.2 cases per 100 000; RD, -0.72 [95% CI, -0.86 to -0.57]; aRR, 0.39 [95% CI, 0.26 to 0.58]). For the remainder of the study period, the rates of sexual assault oscillated, returning to prepandemic levels during the summer months and between COVID-19 waves.

CONCLUSIONS AND RELEVANCE

These findings suggest that lockdown protocols should evaluate the impact of limited care for sexual assault. Survivors should still present to EDs, especially when clinical care or legal interventions are needed.

Violence in the Great Recession

Substantial evidence suggests that economic hardship causes violence. However, a large majority of this research relies on observational studies that use traditional violence surveillance systems that suffer from selection bias and over-represent vulnerable populations, such as people of color. To overcome limitations of prior work, we employed a quasi-experimental design to assess the impact of the Great Recession on explicit violence diagnoses (injuries identified to be caused by a violent event) and proxy violence diagnoses (injuries highly correlated with violence) for child maltreatment, intimate partner violence, elder abuse, and their combination. We used Minnesota hospital data (2004-2014), conducting a difference-in-differences analysis at the county level (n = 86) using linear regression to compare changes in violence rates from before the recession (2004-2007) to after the recession (2008-2014) in counties most affected by the recession, versus changes over the same time period in counties less affected by the recession. The findings suggested that the Great Recession had little or no impact on explicitly identified violence; however, it affected proxy-identified violence. Counties that were more highly affected by the Great Recession saw a greater increase in the average rate of proxy-identified child abuse, elder abuse, intimate partner violence, and combined violence when compared with less-affected counties.

The Association of Race, Ethnicity, and Poverty With Child Maltreatment Reporting

OBJECTIVES

To determine the role of race/ethnicity and poverty in the likelihood of children younger than age 3 years hospitalized because of child abuse and neglect-related injuries being reported to child protective services (CPS) and being assigned a specific maltreatment diagnostic code.

METHODS

We used population-based linked administrative data comprising of birth, hospitalization, and CPS records. Children were identified for maltreatment-related hospitalizations using standardized diagnostic codes. Regression models were used to compute crude and adjusted race/ethnicity estimates regarding the likelihood of being reported to CPS and assigned a specific maltreatment diagnostic code during the maltreatment-related hospitalization.

RESULTS

Of the 3907 children hospitalized because of child maltreatment, those with public health insurance were more likely than those with private insurance (relative risk [RR]: 1.29; 95% confidence interval [CI], 1.16-1.42) and those with Asian/Pacific Islander mothers were less likely than those with White mothers to be reported to CPS (RR: 0.78; 95% CI, 0.65-0.93). No differences were found for children with Black, Hispanic, and Native American mothers compared with those with White mothers for CPS reporting. However, children with Native American mothers (RR: 1.45; 95% CI, 1.11-1.90) and public health insurance (RR: 2.00; 95% CI, 1.63-2.45) were more likely to have a specific maltreatment diagnostic code, the second strongest predictor of a CPS report.

CONCLUSIONS

Race/ethnicity and poverty were factors for CPS reports during a child maltreatment-related hospitalization. It is necessary to implement programs and policies that mitigate implicit bias to prevent inequities in which children receive protective intervention.

The Typology and Topography of Child Abuse and Neglect: The Experience of a Tertiary Children's Centre

Effective child protection systems and processes require reliable and accurate data. The aim of this study was to determine what data could be extracted from hospital records in a single site that reflected a child’s journey from admission with suspected abuse to the decisions regarding substantiation made by the multidisciplinary child protection team. A retrospective study of the case records of 452 children referred to a major UK children’s tertiary centre for suspected child maltreatment was undertaken. Child maltreatment was substantiated in 65% of referred cases, with the majority of referrals coming from children living in the most deprived neighbourhoods in the country. Domestic violence and abuse and the child’s previous involvement with statutory bodies were associated with case substantiation. Physical abuse predominated, with soft tissue injuries, including dog bites and burns, most frequent. Burns were related almost exclusively to supervisory neglect. There were also cases of medical neglect. Emotional abuse was associated with exposure to domestic violence and abuse and to self-harm. The strengths and limitations for single-centre data systems were explored, concluding with a recommendation to establish an agreed national and international minimum data set to protect children from maltreatment.

Child maltreatment-related children's emergency department visits before and during the COVID-19 pandemic in Connecticut

BACKGROUND

Previous studies of national emergency department (ED) data demonstrate a decrease in visits coded for physical abuse during the pandemic period. However, no study to date has examined the incidence of multiple child maltreatment types (physical abuse, sexual abuse, and neglect), within a single state while considering state-specific closure policies. Furthermore, no similar study has utilized detailed chart review to identify cases, nor compared hospital data to Child Protective Services (CPS) reports.

OBJECTIVE

To determine the incidence of child maltreatment-related ED visits before and during the COVID-19 pandemic, including characterizing the type of maltreatment, severity, and CPS reporting.

PARTICIPANTS AND SETTING

Children younger than 18 years old at two tertiary-care, academic children's hospitals in X state.

METHODS

Maltreatment-related ED visits were identified by ICD-10-CM codes and keywords in chief concerns and provider notes. We conducted a cross-sectional retrospective review of ED visits and child abuse consultations during the pre-COVID (1/1/2019-3/15/2020) and COVID (3/16/2020-8/31/2020) periods, as well as state-level CPS reports for suspected maltreatment.

RESULTS

Maltreatment-related ED visits decreased from 15.7/week in the matched pre-COVID period (n = 380 total) to 12.3/week (n = 296 total) in the COVID period (P < .01). However, ED visits (P < .05) and CPS reports (P < .001) for child neglect increased during this period. Provider notes identified 62.4% of child maltreatment ED visits, while ICD-10 codes identified only-CM captured 46.8%.

CONCLUSION

ED visits for physical and sexual abuse declined, but neglect cases increased during the COVID-19 pandemic in X state.

Comparison of crosswalk methods for translating ICD-9 to ICD-10 diagnosis codes for child maltreatment

BACKGROUND

A validated source of International Classification of Disease (1CD) 10th revision diagnostic codes to identify child maltreatment has not been developed. Such a reference would be essential for the practical purposes of administrative data-based research and public health surveillance.

OBJECTIVE

To evaluate the validity of diagnosis code classifications for child maltreatment following conversion from ICD 9th edition, clinical modification (ICD-9-CM) to 10th revision.

PARTICIPANTS AND SETTING

Children receiving inpatient or emergency medical care in the United States with ICD-9-CM/ICD-10-CM diagnosis codes for child maltreatment, identified using two large multicenter hospital-based datasets.

METHODS

We evaluated the performance of general equivalence mappings (GEMs) and network-based mappings for previously-validated ICD-9-CM diagnosis codes for child maltreatment in the 2013-2014 PHIS and 2012 KID and resulting ICD-10-CM codes in the 2018-2019 PHIS and 2016 KID datasets.

RESULTS

Of 56 previously-validated ICD-9-CM diagnoses, GEMs identified 15 with a similar proportion of diagnosed children in the KID ICD-9-CM and ICD-10 eras and 14 diagnoses in PHIS. Network-based mapping identified 18 diagnoses with similar proportions in the KID datasets, and 13 diagnoses in PHIS. For six diagnoses, the proportion of children identified in the ICD-10 era using network-based mapping was more than ten times the proportion identified in the ICD-9-CM era.

CONCLUSION

Neither crosswalk method provided consistently reliable conversions, due to both crosswalk methodology and changes introduced by the ICD 10th revision. These findings highlight the need for independent construction and validation of ICD-10-based definitions of child maltreatment as a precursor to administrative data-based research and public health surveillance.

Reports and Removals of Child Maltreatment-Related Hospitalizations: A Population-Based Study

Despite U.S. child protective services (CPS) agencies relying on mandated reporters to refer concerns of child maltreatment to them, there is little data regarding which children mandated reporters decide to report and not to report. This study addresses this gap by utilizing a population-based linked administrative dataset to identify which children who are hospitalized for maltreatment-related reasons are reported to CPS and which are removed by CPS. The dataset was comprised of all children born in Washington State between 1999 and 2013 (N = 1,271,416), all hospitalizations for children under the age of three, and all CPS records. We identified maltreatment-related hospitalizations using standardized diagnostic codes. We examined the records for children with maltreatment-related hospitalizations to identify hospitalization-related CPS reports and if the child was removed from their parents. We tested for differences in these system responses using multinomial regression. About two-thirds of children identified as experiencing a child maltreatment-related hospitalization were not reported to CPS. We found differences in responses by maltreatment subtype and the type of diagnostic code. Children whose hospitalizations were related to abuse and associated with a specific maltreatment code had increased odds of being both reported to CPS and subsequently removed by CPS.

National Trends in Emergency Department Visits for Child Maltreatment, 2007-2014

OBJECTIVE

To understand the prevalence of child maltreatment-related emergency department (ED) visits in the United States, we examined data from the 2007 to 2014 Nationwide Emergency Department Sample.

METHODS

Based on existing literature, International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9 CM) ED discharge codes for children less than 10 years of age were characterized as specified child maltreatment, defined as visits with an explicit maltreatment ICD-9 CM or external causes of injury codes. The prevalence of child maltreatment visits per 100,000 children in the United States (based on Center for Disease Control Wide-ranging ONline Data for Epidemiologic Research estimates) overall and by sociodemographic factors was examined, and tests for trends over time were evaluated with Cochran-Armitage tests. Analyses were conducted in 2019.

RESULTS

The prevalence of child maltreatment based in ICD-9 CM discharge codes ED visits dropped from 69.2 visits per 100,000 in 2007 to 65.9 visits per 100,000 in 2014; this trend was statistically significant. The prevalence was lowest in 2010 (60.1 visits per 100,000 children). There were increases observed for some demographic groups in this period. Throughout the 8-year period examined, the prevalence of child maltreatment visits was highest for physical abuse compared with other forms of maltreatment, higher for boys compared with girls; highest for children younger than 1 year, and higher for children living in neighborhoods with the lowest median income compared with children in higher-income neighborhoods.

CONCLUSIONS

The Nationwide Emergency Department Sample data set is a valuable surveillance tool for examining trends in child maltreatment. Future studies should explore what factors may explain variations in child maltreatment over time to best develop prevention strategies.

An Algorithm Using Administrative Data to Measure Adverse Childhood Experiences (ADM-ACE)

OBJECTIVE

To develop an algorithm using administrative data to measure adverse childhood experiences (ADM-ACE) within routinely collected health insurance claims and enrollment data.

DATA SOURCES

We used claims and enrollment data from Tennessee's Medicaid program (TennCare) in 2018.

STUDY DESIGN

We studied five types of ACEs: maltreatment and peer violence, foster care and family disruption, maternal mental illness, maternal substance use disorder, and abuse of the mother. We used diagnosis and procedure codes, prescription drug fills, and enrollment files to develop the ADM-ACE, which we applied to measure the prevalence of ACEs and to examine prevalence by demographic characteristics among our sample of children in TennCare. We compared ADM-ACE prevalence to child welfare records and survey results from Tennessee.

DATA COLLECTION/EXTRACTION METHODS

Our study sample included children aged 0-17 years who were linked to their mothers if also enrolled in TennCare in 2018 (N=763,836 children).

PRINCIPAL FINDINGS

Approximately 19.2% of children in TennCare had indicators for ADM-ACEs. The prevalence of ACEs was higher among children who were younger (p<0.001), non-Hispanic White or Black (compared to Hispanic) (p<0.001), and children residing in rural vs. urban counties (p<0.001). The prevalence of maltreatment identified through the ADM-ACE (1.6%) falls between the percent of children in Tennessee who were reported to child welfare authorities and the percent for whom reports of maltreatment were substantiated. Comparison with survey reports from Tennessee parents suggests an advantage in measuring maternal mental illness with the ADM-ACE using health insurance claims data.

CONCLUSIONS

The ADM-ACE can be applied to health encounter data to study and monitor the prevalence of certain ACEs, their association with health conditions, and the effects of policies on reducing exposure to ACEs or improving health outcomes for children with ACEs.

Maltreatment Related Hospitalizations Among Children Ages 17 Years and Younger: New York State, 2011-2013

BACKGROUND

Child maltreatment is an important societal and public health problem. However, there are limited data on the epidemiology of maltreatment related hospitalizations.

OBJECTIVE

The objective of this study was to describe maltreatment related hospitalizations among children ages 17 and younger in New York State (NYS).

METHODS

Using 2011-2013 statewide planning and research cooperative system (SPARCS) inpatient hospital discharge data, maltreatment related hospitalizations among children ages 17 years and younger were identified using international classification of diseases, ninth revision, clinical modification codes for diagnoses and external cause of injury. Distributions of demographic and inpatient care characteristics were compared between hospitalizations for maltreatment and those for other causes, and between different types of maltreatment, using chi-square tests (for categorical variables) and t-tests (for continuous variables).

RESULTS

During 2011-2013, a total of 853 maltreatment related hospitalizations among 836 children ages 17 years and younger were documented in NYS SPARCS. Infants (children < 1) had the highest rates of hospitalization. Overall, physical abuse was the most prevalent maltreatment type reported.

CONCLUSIONS

This is the first study in NYS to describe the epidemiology of child maltreatment hospitalizations; it establishes a statewide baseline for this public health and societal issue.

Measuring the hidden burden of violence: use of explicit and proxy codes in Minnesota injury hospitalizations, 2004-2014

PURPOSE

Commonly-used violence surveillance systems are biased towards certain populations due to overreporting or over-scrutinized. Hospital discharge data may offer a more representative view of violence, through use of proxy codes, i.e. diagnosis of injuries correlated with violence. The goals of this paper are to compare the trends in violence in Minnesota, and associations of county-level demographic characteristics with violence rates, measured through explicitly diagnosed violence and proxy codes. It is an exploration of how certain sub-populations are overrepresented in traditional surveillance systems.

METHODS

Using Minnesota hospital discharge data linked with census data from 2004 to 2014, this study examined the distribution and time trends of explicit, proxy, and combined (proxy and explicit) codes for child abuse, intimate partner violence (IPV), and elder abuse. The associations between county-level risk factors (e.g., poverty) and county violence rates were estimated using negative binomial regression models with generalized estimation equations to account for clustering over time.

RESULTS

The main finding was that the patterns of county-level violence differed depending on whether one used explicit or proxy codes. In particular, explicit codes suggested that child abuse and IPV trends were flat or decreased slightly from 2004 to 2014, while proxy codes suggested the opposite. Elder abuse increased during this timeframe for both explicit and proxy codes, but more dramatically when using proxy codes. In regard to the associations between county level characteristics and each violence subtype, previously identified county-level risk factors were more strongly related to explicitly-identified violence than to proxy-identified violence. Given the larger number of proxy-identified cases as compared with explicit-identified violence cases, the trends and associations of combined codes align more closely with proxy codes, especially for elder abuse and IPV.

CONCLUSIONS

Violence surveillance utilizing hospital discharge data, and particularly proxy codes, may add important information that traditional surveillance misses. Most importantly, explicit and proxy codes indicate different associations with county sociodemographic characteristics. Future research should examine hospital discharge data for violence identification to validate proxy codes that can be utilized to help to identify the hidden burden of violence.

A Quasi-Experimental Study of the Effects of Pre-Kindergarten Education on Pediatric Asthma

Ensuring access to pre-kindergarten (Pre-K) education remains a pressing policy issue in the United States. Prior research has shown the positive effects that Pre-K has on children’s cognitive development. However, studies on its effects on children’s health outcomes are scarce. This study aimed to investigate the effects of the Pre-K program on pediatric asthma. Children’s individual data from existing research conducted in North Carolina were linked with state Medicaid claims data from 2011–2017. There were 51,408 observations (person-month unit) of 279 children enrolled in Pre-K and 333 unenrolled children. Asthma was identified using the ICD 9/10 codes. A difference-in-differences model was adopted using a panel analysis with three time periods: before, during, and after Pre-K. The explanatory variables were interaction terms between Pre-K enrollment and (a) before vs. during period and (b) during vs. after period. The results indicated that children enrolled in Pre-K had a greater risk of asthma diagnosis during Pre-K (b = 0.0145, p = 0.058). Conversely, in the post-intervention period, the enrolled children had a lower of receiving an asthma diagnosis (b = −0.0216, p = 0.002). These findings indicate that Pre-K may increase the use of asthma-related health services in the short term and decrease the service use after participants leave the program.

Characteristics of child welfare investigations reported by healthcare professionals in Ontario: secondary analysis of a regional database

OBJECTIVES

This study examines the characteristics and outcomes of child welfare investigations reported by hospital-based and community-based healthcare professionals.

METHODS

A sample of 7590 child maltreatment-related investigations from the Ontario Incidence Study of Reported Child Abuse and Neglect-2018, a cross-sectional study, was analysed. Bivariate analyses compared characteristics of hospital and community healthcare-reported investigations. Chi-square automatic interaction detector analyses were used to predict the most influential factors in the decision to provide a family with services following a child welfare investigation from each referral source.

RESULTS

Community healthcare-reported investigations were more likely to have a primary concern of physical abuse while hospital-reported investigations were more likely to be focused on assessing risk of future maltreatment. Hospital-reported investigations were more likely to involve noted primary caregiver (eg, mental health issues, alcohol/drug abuse, victim of intimate partner violence (IPV)) and household risk factors. The most significant predictor of service provision following an investigation was having a caregiver who was identified as a victim of IPV in hospital-reported investigations (χ²=30.237, df=1, adj. p<0.001) and having a caregiver for whom few social supports was noted in community healthcare-reported investigations (χ²=18.892, df=1, adj. p<0.001).

CONCLUSION

Healthcare professionals likely interact with children who are at high risk for maltreatment. This study's findings highlight the important role that healthcare professionals play in child maltreatment identification, which may differ across hospital-based and community-based settings and has implications for future collaborations between the healthcare and child welfare systems.

Predictive value of indicators for identifying child maltreatment and intimate partner violence in coded electronic health records: a systematic review and meta-analysis

OBJECTIVE

Electronic health records (EHRs) are routinely used to identify family violence, yet reliable evidence of their validity remains limited. We conducted a systematic review and meta-analysis to evaluate the positive predictive values (PPVs) of coded indicators in EHRs for identifying intimate partner violence (IPV) and child maltreatment (CM), including prenatal neglect.

METHODS

We searched 18 electronic databases between January 1980 and May 2020 for studies comparing any coded indicator of IPV or CM including prenatal neglect defined as neonatal abstinence syndrome (NAS) or fetal alcohol syndrome (FAS), against an independent reference standard. We pooled PPVs for each indicator using random effects meta-analyses.

RESULTS

We included 88 studies (3 875 183 individuals) involving 15 indicators for identifying CM in the prenatal period and childhood (0-18 years) and five indicators for IPV among women of reproductive age (12-50 years). Based on the International Classification of Disease system, the pooled PPV was over 80% for NAS (16 studies) but lower for FAS (<40%; seven studies). For young children, primary diagnoses of CM, specific injury presentations (eg, rib fractures and retinal haemorrhages) and assaults showed a high PPV for CM (pooled PPVs: 55.9%-87.8%). Indicators of IPV in women had a high PPV, with primary diagnoses correctly identifying IPV in >85% of cases.

CONCLUSIONS

Coded indicators in EHRs have a high likelihood of correctly classifying types of CM and IPV across the life course, providing a useful tool for assessment, support and monitoring of high-risk groups in health services and research.

The Incidence of Child Maltreatment Resulting in Hospitalizations for Children Under Age 3 Years

OBJECTIVE

To assess the incidence of child maltreatment-related hospitalizations for children under 3 years for the population of Washington State.

STUDY DESIGN

A population-based study using retrospective linked administrative data for all children born in Washington State from 2000 through 2013 (n = 1 191 802). The dataset was composed of linked birth and hospitalization records for the entire state. Child maltreatment-related hospitalizations were identified using diagnostic codes, both specifically attributed to and suggestive of maltreatment. Incidence were calculated for the population, by birth year, by sex, and by maltreatment subtype.

RESULTS

A total of 3885 hospitalizations related to child maltreatment were identified for an incidence of 10.87 per 10 000 person-years. Hospitalizations related to child maltreatment accounted for 2.1% of all hospitalizations for children under the age of 3 years. This percentage doubled over time, reaching a high in 2012 (3.6%). More than one-half of all hospitalizations were related to neglect. Maltreatment-related hospitalizations occurred most frequently in the first year of life for all subtypes except for neglect, which occurred the most between 1 and 2 years of age. Male children had higher incidence than female children in general (11.97 vs 9.70 per 10 000 person-years) and across all subtypes.

CONCLUSIONS

Hospitalizations can be a useful source of population-based child maltreatment surveillance. The identification of neglect-related hospitalizations, likely the result of supervisory neglect, because the most common subtype is an important finding for the development of prevention programming.

Effects of Home Visiting Program Implementation on Preventive Health Care Access and Utilization: Results from a Randomized Trial of Healthy Families Oregon

Home visiting programs are an increasingly popular mechanism for providing a broad set of early prevention supports to high-risk families. A key intended outcome for these programs is to support maternal and child health by helping families increase access to and use of preventive health care services. For many community-based home visiting programs, however, there is less evidence of positive outcomes in the health care domain. The current study used a randomized trial conducted in a statewide early childhood home visiting program, Healthy Families Oregon (HFO), to examine program impacts on families' use of preventive health care services. The study recruited a large sample of participants (n = 1438 HFO families and n = 1289 controls) and utilized state agency health insurance and medical records as the primary data source. There were challenges in providing services in alignment with an intent-to-treat research design, leading to the need to take alternative approaches to analyzing effects of service receipt on outcomes. Results found that while there were no significant differences in health care access or utilization in the intent-to-treat models, positive outcomes were found when propensity score matching was used to limit the program sample to those who actually received services. Further, within the program group, children who were enrolled for longer had fewer gaps in health insurance coverage and received more well-baby visits and immunizations compared to those with less service. The role of the home visitor in helping families navigate the complexities of publicly funded health care is discussed. Investments in professional development strategies that can increase staff expertise in this area and improve family retention may be needed to more effectively achieve intended health outcomes.

Dissemination of child abuse clinical decision support: Moving beyond a single electronic health record

BACKGROUND

Child maltreatment is a leading cause of pediatric morbidity and mortality. We previously reported on development and implementation of a child abuse clinical decision support system (CA-CDSS) in the Cerner electronic health record (EHR). Our objective was to develop a CA-CDSS in two different EHRs.

METHODS

Using the CA-CDSS in Cerner as a template, CA-CDSSs were developed for use in four hospitals in the Northwell Health system who use Allscripts and two hospitals in the University of Wisconsin health system who use Epic. Each system had a combination of triggers, alerts and child abuse-specific order sets. Usability evaluation was done prior to launch of the CA-CDSS.

RESULTS

Over an 18-month period, a CA-CDSS was embedded into Epic and Allscripts at two hospital systems. The CA-CDSSs vary significantly from each other in terms of the type of triggers which were able to be used, the type of alert, the ability of the alert to link directly to child abuse-specific order sets and the order sets themselves.

CONCLUSIONS

Dissemination of CA-CDSS from one EHR into the EHR in other health care systems is possible but time-consuming and needs to be adapted to the strengths and limitations of the specific EHR. Site-specific usability evaluation, buy-in of multiple stakeholder groups and significant information technology support are needed. These barriers limit scalability and widespread dissemination of CA-CDSS.

Protocol for a randomized controlled trial evaluating the impact of the Nurse-Family Partnership's home visiting program in South Carolina on maternal and child health outcomes

Background

Policy-makers are increasingly seeking rigorous evidence on the impact of programs that go beyond typical health care settings to improve outcomes for low-income families during the critical period around the transition to parenthood and through early childhood.

Methods

This study is a randomized controlled trial evaluating the impact of the Nurse-Family Partnership’s expansion in South Carolina. The scientific trial was made possible by a “Pay for Success” program embedded within a 1915(b) Waiver from Medicaid secured by the South Carolina Department of Health and Human Services. This protocol describes study procedures and defines primary and secondary health-related outcomes that can be observed during the intervention period (including pregnancy through the child’s first 2 years of life). Primary study outcomes include (1) a composite indicator for adverse birth outcomes including being born small for gestational age, low birth weight (less than 2500 g), preterm birth (less than 37 weeks’ gestation), or perinatal mortality (fetal death at or after 20 weeks of gestation or mortality in the first 7 days of life), (2) a composite outcome indicating health care utilization or mortality associated with major injury or concern for abuse or neglect occurring during the child’s first 24 months of life, and (3) an indicator for an inter-birth interval of < 21 months. Secondary outcomes are defined similarly in three domains: (1) improving pregnancy and birth outcomes, (2) improving child health and development, and (3) altering the maternal life course through changes in family planning.

Discussion

Evidence from this trial on the impact of home visiting services delivered at scale as part of a Medicaid benefit can provide policy-makers and stakeholders with crucial information about the effectiveness of home visiting programs in improving health and well-being for low-income mothers and children and about novel financing mechanisms for cross-silo interventions.

Trial registration

The trial was registered prospectively on the American Economic Association Trial Registry (the primary registry for academic economists doing policy trials) on 16 February 2016 (AEARCTR-0001039). ClinicalTrials.gov NCT03360539. Registered on 28 November 2017.

Prevalence of Births and Interactions with Child Protective Services of Children Born to Mothers Diagnosed with an Intellectual and/or Developmental Disability

OBJECTIVES

Concerns have been raised that parents with intellectual and/or developmental disabilities (IDD) interact with child protective services (CPS) at disproportionate rates than the general population as a result of bias and discrimination. However, there has been little empirical evidence to ascertain if these concerns are grounded. This study's objectives were to identify (a) the prevalence and sociodemographic characteristics of children born to mothers diagnosed with IDD diagnoses, (b) how many of these children interact with CPS (reports and removals) and (c) when these CPS interactions are occurring.

METHODS

The dataset was comprised of linked administrative birth, hospital discharge, and CPS records for all children born in one U.S. state between 1999 and 2013 (N = 1,271,419). CPS records were available through the first quarter of 2018 and CPS reports and removals at the child's first and fourth birthdays were identified. We conducted chi-square tests and multivariate survival Cox regression models.

RESULTS

A total of 567 children were identified as born to mothers with IDD diagnoses, which is 4.5 per 10,000 births. Of these children, 21.7% were the subject of a CPS report within 1 year and 35.8% within 4 years. In terms of removals, 6.5% experienced removals by 1 year and 8.6% by 4 years.

CONCLUSIONS FOR PRACTICE

This study provides population-based knowledge about how and when the children born to mothers diagnosed with IDD interact with CPS. These children have higher rates of CPS interactions than the general population, but these rates are not as high as previously reported.

Intracranial Injury Among Children with Abuse-Related Long Bone Fractures

Background

Objectives

Methods

Results

Conclusions

Do socioeconomic and birth order gradients in child maltreatment differ by immigrant status?

Background

While literature has documented strong gradients in child maltreatment (CM) by socioeconomic status and family composition in the general population, how these patterns extend to immigrants remain inconclusive. Using population-based administrative data, we examined, for the first time, whether gradients in CM by neighbourhood income and childbirth order vary by immigrant status.

Methods

We used linked hospitalisation, emergency department visits, small-area income, birth and death records with an official Canadian immigration database to create a retrospective cohort of all 1 240 874 children born from 2002 to 2012 in Ontario, Canada, followed from 0 to 5 years. We estimated rate ratios of CM among immigrants and non-immigrants using modified Poisson regression.

Results

CM rates were 1.6 per 100 children among non-immigrants and 1.0 among immigrants. CM was positively associated with neighbourhood deprivation. The adjusted rate ratio (ARR) of CM in the lowest neighbourhood income quintile versus the highest quintile was 1.57 (95% CI 1.49 to 1.66) for non-immigrants and 1.33 (95% CI 1.15 to 1.54) for immigrants. The socioeconomic gradient disappeared when restricted to children of immigrant mothers arrived at 25+ years and in analyses excluding emergency department visits. Compared to a first child, the ARR of CM for a fourth or higher-order child was 1.75 (95% CI 1.63 to 1.89) among non-immigrants and 0.57 (95% CI 0.44 to 0.74) among immigrants.

Conclusions

Immigrants exhibited lower CM rates than non-immigrants across neighbourhood income quintiles and differences were greatest in more deprived neighbourhoods. The contrasting birth order gradients between immigrants and non-immigrants require further investigation.

Evaluating the real-world implementation of the Family Nurse Partnership in England: protocol for a data linkage study

INTRODUCTION

Almost 20 000 babies are born to teenage mothers each year in England, with poorer outcomes for mothers and babies than among older mothers. A nurse home visitation programme in the USA was found to improve a wide range of outcomes for young mothers and their children. However, a randomised controlled trial in England found no effect on short-term primary outcomes, although cognitive development up to age 2 showed improvement. Our study will use linked routinely collected health, education and social care data to evaluate the real-world effects of the Family Nurse Partnership (FNP) on child outcomes up to age 7, with a focus on identifying whether the FNP works better for particular groups of families, thereby informing programme targeting and resource allocation.

METHODS AND ANALYSIS

We will construct a retrospective cohort of all women aged 13-24 years giving birth in English NHS hospitals between 2010 and 2017, linking information on mothers and children from FNP programme data, Hospital Episodes Statistics and the National Pupil Database. To assess the effectiveness of FNP, we will compare outcomes for eligible mothers ever and never enrolled in FNP, and their children, using two analysis strategies to adjust for measured confounding: propensity score matching and analyses adjusting for maternal characteristics up to enrolment/28 weeks gestation. Outcomes of interest include early childhood development, childhood unplanned hospital admissions for injury or maltreatment-related diagnoses and children in care. Subgroup analyses will determine whether the effect of FNP varied according to maternal characteristics (eg, age and education).

ETHICS AND DISSEMINATION

The Nottingham Research Ethics Committee approved this study. Mothers participating in FNP were supportive of our planned research. Results will inform policy-makers for targeting home visiting programmes. Methodological findings on the accuracy and reliability of cross-sectoral data linkage will be of interest to researchers.

Incidence and Risk Factors for Abusive Head Trauma: A Population-Based Study

Previous studies on Abusive Head Trauma (AHT) suggest incidence may vary by geographic location and there is limited information regarding population-based risk factors on this form of child maltreatment. This study provides new knowledge regarding these two aspects using the population of the US state of Washington born between 1999 and 2013. We used a linked administrative dataset comprised of birth, hospital discharge, child protective services (CPS) and death records to identify the scale and risk factors for AHT for the state population using quantitative survival methods. We identified AHT using diagnostic codes in hospital discharge records defined by the US Centers for Disease Control. A total of 354 AHT hospitalisations were identified and the incidence for the state was 22.8 per 100 000 children under the age of one. Over 11 per cent of these children died. Risk factors included a teenaged mother at the time of birth, births paid for using public insurance, child's low birth weight, and maternal Native American race. The strongest risk factor was a prior CPS allegation, a similar finding to a California study on injury mortality. The practice and policy implications of these findings are discussed.

Child maltreatment surveillance following the ICD-10-CM transition, 2016-2018

BACKGROUND

Child maltreatment is poorly documented in clinical data. The International Classification of Diseases and Related Health Problems, 10th Revision, Clinical Modification (ICD-10-CM) represents the first time that confirmed and suspected child maltreatment can be distinguished in medical coding. The utility of this distinction in practice remains unknown. This study aims to evaluate the application of these codes by patient demographic characteristics and injury type.

METHODS

We conducted secondary data analysis of emergency department (ED) discharge records of children under 18 years with an ICD-10-CM code for confirmed (T74) or suspected (T76) child maltreatment. Child age, sex, race/ethnicity, insurance status and co-occurring injuries (S00-T88) were compared by maltreatment type (confirmed or suspected).

RESULTS

From 2016 to 2018, child maltreatment was documented in 1650 unique ED visits, or 21.7 per 10 000 child ED visits. Suspected maltreatment was documented most frequently (58%). Half of all maltreatment-related visits involved sexual abuse, most often in females and individuals of non-Hispanic white race. Physical abuse was coded in 36% of visits; injuries to the head were predominant. Non-Hispanic black children were more frequently documented with confirmed physical abuse than suspected (38.7% vs 23.7%, p<0.01). The rate of co-occurring injuries documented with confirmed and suspected maltreatment differed by 30% (9.2 vs 12.5 per 10 000 ED visits, respectively).

CONCLUSIONS

The ability to discriminate confirmed and suspected maltreatment may help mitigate clinical barriers to maltreatment surveillance associated with delayed diagnosis and subsequent intervention. Racial disparities in suspected and confirmed cases were identified which may indicate biased diagnostic behaviours in the ED.

Cluster analysis of child homicide in South Korea

BACKGROUND

There has been an insufficient in-depth analysis of the nature and prevalence of the typologies of child homicide in Asia, particularly in South Korea.

OBJECTIVE

In the current study, we sought to determine the prevalence and identify the heterogeneity of the child homicide phenomenon in South Korea.

PARTICIPANT AND SETTINGS

All 341 original case files (i.e., hospital, police, and autopsy reports) of homicide incidents involving children aged 0-18 in 2016 were obtained from the forensic autopsy archives of the National Forensic Service (NFS), which handles 100 % of the medico-legal autopsies in South Korea. These were examined and reclassified based on our definition.

METHOD

A cluster analysis using Gower's distance was applied, which has rarely been utilized in this field of research. By performing a qualitative analysis, we first extracted 70 (numerical, logical, categorical) crime, victim, perpetrator, and household relevant variables, which were later utilized in the cluster analysis.

RESULTS

Among the 341 cases from 2016, 95 were judged to be at least suspicious child homicide cases. When applying the cluster analysis, eight sub-clusters were extracted: child torture, maternal filicide, neonaticide, death not related to previous abuse, paternal filicide, paternal infanticide, maternal infanticide, and psychotic killings.

CONCLUSIONS

The commonality and the unique aspect of the child homicide phenomenon in South Korea, in comparison with the results from previous research from other countries, are discussed.

Risk Factors for Child Maltreatment Fatalities in a National Pediatric Inpatient Database

BACKGROUND AND OBJECTIVES

Child maltreatment (CM) is recognized as a major public health concern, and an important number of children suffer injuries related to abuse and neglect that result in death. We sought to identify risk factors for CM fatalities among hospitalized children that can provide clinicians with information to recognize at-risk children and reduce further death.

METHODS

In this study, we included cases from the 2012 Kids' Inpatient Database with diagnosis codes related to CM who were <5 years of age and were not transferred to another facility. Potential demographic and clinical risk factors were identified and compared to child fatality in the hospital by using bivariate and multivariate analyses. To assess how cases coded specifically for maltreatment differed from similar cases that only suggested maltreatment, a reduced-model multivariable logistic regression for fatality was created.

RESULTS

We found 10 825 children <5 years who had inpatient diagnoses coded in their medical record for CM. Most demographic variables (age, race, and sex) were not significantly associated with fatality, whereas clinical variables (transferring in, drowning, ingestions, and burns) were significantly associated with fatality. There were regional differences on the basis of hospital location as well as significantly more chronic conditions, procedure charges, and longer lengths of stay among children who died. Controlling for significant risk factors, those with diagnoses specific for physical abuse had ∼3 times the odds of dying (odds ratio = 2.797; 95% confidence interval: 1.941-4.031).

CONCLUSIONS

In this study, although infancy and decreased income were associated with increased risk for fatality, more important factors were the types of injuries the child endured and whether the inpatient clinician had identified specific injuries indicating physical abuse.

Differences in Early Childhood Maltreatment by Maternal Birthplace and Child Sex

OBJECTIVE

To identify patterns of health system-identified early childhood maltreatment by maternal birthplace and child sex, within a multicultural society with universal access to healthcare.

STUDY DESIGN

This retrospective population-based cohort study included 1240946 children born in Ontario, Canada, between 2002 and 2012, and followed from birth to age 5 years using administrative data. Modified Poisson regression was used to estimate adjusted rate ratios for maltreatment-physical abuse or neglect-among the children of immigrant vs nonimmigrant mothers. Conditional logistic regression was used to estimate further the odds of maltreatment comparing a daughter vs son of the same mother.

RESULTS

Maltreatment rates were 36% lower (adjusted rate ratio, 0.64; 95% CI, 0.61-0.66) among children of immigrant mothers (10 per 1000) than those of nonimmigrant mothers (16 per 1000). Maltreatment rates were 27%-48% lower among children of maternal immigrant groups relative to that among Canadian-born mothers, except children of Caribbean-born mothers (16 per 1000). No significant differences were seen between daughters and sons in the odds of early childhood health system-identified maltreatment by maternal birthplace.

CONCLUSIONS

Health system-identified maltreatment in early childhood is highest among children of Canadian- and Caribbean-born mothers. Maltreatment did not differ between daughters and sons of the same mother. These data may inform strategies aimed at decreasing maltreatment among vulnerable groups.

What We Can Learn From Failure: An EHR-Based Child Protection Alert System

This study aimed to evaluate the efficacy of a newly implemented Child Protection Alert System (CPAS) that utilizes triggering diagnoses to identify children who have been confirmed/strongly suspected as maltreated. We retrospectively reviewed electronic health records (EHRs) of 666 patients evaluated by our institution's child protection team between 2009 and 2014. We examined each EHR for the presence of a pop-up alert, a persistent text-based visual alert, and diagnoses denoting child maltreatment. Diagnostic accuracy of the CPAS for child maltreatment identification was assessed. Of 323 patients for whom child maltreatment was confirmed/strongly suspected, 21.7% (70/323) had a qualifying longitudinal diagnosis listed. The pop-up alert fired in 14% of cases (45/323) with a sensitivity and specificity of 13.9% (95% CI [10.4%, 18.2%]) and 100% (95% CI [98.9%, 100.0%]), respectively. The text-based visual alert displayed in 44 of 45 cases. The CPAS is a novel simple way to support clinical decision-making to identify and protect children at risk of (re)abuse. This study highlights multiple barriers that must be overcome to effectively design and implement a CPAS to protect at-risk children.

Pediatric Fingertip Injuries: Association With Child Abuse

Purpose

Methods

Results

Conclusions

Type of study/level of evidence

Trends in US physician diagnosis of child physical abuse and neglect injuries, 2006-2014

OBJECTIVE

To determine if US child physical abuse and neglect injury rates changed from 2006 to 2014, whether definitive diagnoses of physical abuse and neglect were used more often over time, and what patient factors influenced definitive physical maltreatment diagnoses.

METHODS

Nationally estimated rates of definitive and suggestive physical abuse and neglect injuries for children <10 years were generated using the Nationwide Emergency Department Sample, the National Inpatient Sample, and census estimates. Trends over time were evaluated, including the trend in the proportion of definitive diagnoses to all diagnoses (definitive plus suggestive). Logistic regression was used to evaluate whether patient characteristics and hospital patient volumes were associated with definitive versus suggestive diagnoses.

RESULTS

The population rates of child physical maltreatment medically treated injuries were unchanged from 2006 to 2014; the trends were not statistically significant for ED or hospitalized patients. Over time, physician definitive diagnoses as a proportion of all physical maltreatment diagnoses (definitive plus suggestive) increased in admitted children from 17.6% in 2006 to 22.0% in 2014 (p = 0.02). Older age, white race, lower income by zip code, and public insurance as well as larger patient volumes increased the odds of definitive rather than suggestive diagnoses of physical abuse and neglect injuries.

CONCLUSIONS

Definitive diagnoses of physical abuse and neglect increased over the study period and were associated with hospital volume and patient characteristics which may reflect provider experience and possible bias. The use of electronic medical records may have influenced the coding of definitive diagnoses.

Randomized controlled trial of Family Connects: Effects on child emergency medical care from birth to 24 months

One of Tom Dishion's most significant contributions to prevention science was the development of affordable, ecologically valid interventions, such as the Family Check-Up, that screen for child and family risk factors broadly, but concentrate family-specific interventions on those with greatest potential for population impact. In the spirit of this approach, investigators examined effects of a brief, universal postnatal home visiting program on child emergency medical care and billing costs from birth to age 24 months. Family Connects is a community-wide public health intervention that combines identification and alignment of community services and resources with brief, postpartum nurse home visits designed to assess risk, provide supportive guidance, and connect families with identified risk to community resources. Over 18 months, families of all 4,777 resident Durham County, North Carolina, births were randomly assigned based on even or odd birth date to receive a postnatal nurse home visiting intervention or services as usual (control). Independently, 549 of these families were randomly selected and participated in an impact evaluation study. Families, blind to study goals, provided written consent to access hospital administrative records. Results indicate that children randomly assigned to Family Connects had significantly less total emergency medical care (by 37%) through age 24 months, with results observed across almost all subgroups. Examination of billing records indicate a $3.17 decrease in total billing costs for each $1 in program costs. Overall, results suggest that community-wide postpartum support program can significantly reduce population rates of child emergency medical care through age 24 months while being cost-beneficial to communities.

Identification of Child Maltreatment-Related Emergency Department Visits in Connecticut, 2011 to 2014

Child maltreatment identified by medical professionals is poorly represented in records of child protection. International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) codes may better represent the burden of maltreatment treated in clinical settings. Using emergency department (ED) discharge data from 2011 to 2014, we enumerated presentations of maltreatment treated in Connecticut EDs for children under 10 years. Of 790 080 discharges, child maltreatment was explicitly documented in 265 (0.03%) unique ED visits, consistent with prior studies. Sexual maltreatment was most prevalent. A total of 3634 visits included an ICD-9-CM code suggestive of maltreatment. Children with these codes were significantly younger, more likely to be of white race, and use private insurance. Use of ICD codes in child maltreatment surveillance may elucidate characteristics of maltreatment not captured by child welfare data. Combining ICD codes for explicit and suggestive maltreatment will aid in understanding the extent of this problem.

A 15-Year Population-Based Investigation of Sexual Assault Cases Across the Province of Ontario, Canada, 2002-2016

Objectives. To estimate the population-level frequencies and standardized rates of sexual assault cases in the province of Ontario, Canada.Methods. We conducted a 15-year retrospective analysis (2002-2016) of sexual assault cases by linking 5 provincial administrative health databases. We defined sexual assault by an algorithm of 23 International Classification of Diseases, 10th Revision, and physician billing codes. We calculated age- and sex-stratified standardized rates per 100 000 census population, and we used age- and sex-stratified Poisson regressions to determine annual rate ratios.Results. Between 2002 and 2016, there were 52 780 incident cases of sexual assault in Ontario at a rate of 27.38 per 100 000 population. The highest rates were found among females aged 15 to 19 years (187 per 100 000) and 20 to 24 years (127 per 100 000). Among males, the highest rates were observed among children aged 0 to 4 years (41 per 100 000) and 5 to 9 years (29 per 10 000). Among males and females, the annual rate ratio increased among those aged 15 years and older and decreased among those aged 14 years and younger.Conclusions. Sexual assault was documented across all age groups and sexes, from children to elders, with high standardized rates among adolescents and children.

Child Sexual Abuse identified in Emergency Departments Using ICD-9-CM, 2011 to 2014

BACKGROUND

Child sexual abuse (CSA) is poorly identified due to its hidden nature and difficulty surrounding disclosure. Surveillance using emergency department (ED) data may identify victims and provide information on their demographic profile.

OBJECTIVES

Study aims were to calculate the prevalence of visits assigned an explicit or suggestive medical diagnosis code (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM]) for CSA and compare the demographic profile of ED visits by coding type.

METHODS

This study examined ED data for children < 10 years of age in Connecticut from 2011 to 2014. Cases involving CSA were identified using explicit and suggestive ICD-9-CM codes and age qualifiers previously established in the literature, and compared across visit characteristics (age, race/ethnicity, sex, and primary insurance payer, and town group).

RESULTS

ICD-9-CM codes for explicit CSA were identified in 110 ED visits, or 1.7 per 10,000 total ED visits. Inclusion of ICD-9-CM codes for suggestive CSA identified an additional 630 visits (9.7 per 10,000 visits). Suggestive codes identified proportionally more visits of younger (50% vs. 38%) and male (35% vs. 22%) children, compared with the explicit code (p < 0.05).

CONCLUSIONS

This study demonstrates one method for identifying CSA cases, which has the potential to increase surveillance of victims in the ED. Results imply that explicit codes alone may overlook most cases, whereas use of suggestive codes may identify additional cases, and proportionally more young and male victims. As the health consequences of CSA are severe, innovative forms of surveillance must be explored to detect a higher number of cases and improve the clinical care of patients.

Childhood adversity, mental health and suicide (CHASE): a methods protocol for a longitudinal case-control linked data study

Introduction

Suicide is a tragic outcome with devastating consequences. In 2018, Scotland experienced a 15% increase in suicide from 680 to 784 deaths. This was marked among young people, with an increase of 53% in those aged 15-24, the highest since 2007. Early intervention in those most at risk is key, but identification of individuals at risk is complex, and efforts remain largely targeted towards universal suicide prevention strategies with little evidence of effectiveness.

Recent evidence suggests childhood adversity is a predictor of subsequent poor social and health outcomes, including suicide. This protocol reports on methodology for harmonising lifespan hospital contacts for childhood adversity, mental health, and suicidal behaviour. This will inform where to 1) focus interventions, 2) prioritise trauma-informed approaches, and 3) adapt support avenues earlier in life for those most at risk.

Methods

This study will follow a case-control design. Scottish hospital data (physical health SMR01; mental health SMR04; maternity/birth record SMR02; mother’s linked data SMR01, SMR04, death records) from 1981 to as recent as available will be extracted for people who died by suicide aged 10-34, and linked on Community Health Index unique identifier. A randomly selected control population matched on age and geography at death will be extracted in a 1:10 ratio. International Classification of Disease (ICD) codes will be harmonised between ICD9-CM, ICD9, ICD10-CM and ICD10 for childhood adversity, mental health, and suicidal behaviour.

Results

ICD codes for childhood adversity from four key studies are reported in two categories, 1) Maltreatment or violence-related codes, and 2) Codes suggestive of maltreatment. ‘Clinical Classifications Software’ ICD codes to operationalise mental health codes are also reported. Harmonised lifespan ICD categories were achieved semi-automatically, but required labour-intensive supplementary manual coding. Cross-mapped codes are reported.

Conclusion

There is a dearth of evidence about touchpoints prior to suicide. This study reports methods and harmonised ICD codes along the lifespan to understand hospital contact patterns for childhood adversity, which come to the attention of hospital practitioners.

Key words

Childhood Adversity, Adverse Childhood Experiences, Mental Health, Self-harm, Suicide, Suicidality, Violence, Hospital episodes, Routine Data, Data Linkage, Study Protocol

Prematurity Does Not Increase Early Childhood Fracture Risk

OBJECTIVE

To evaluate the impact of prematurity on fracture by age 5, controlling for medications and comorbidities of prematurity.

STUDY DESIGN

We performed a retrospective cohort study of infants born in Military Treatment Facilities in 2009-2010 with ≥5 years of follow-up care. Gestational age, low birth weight, comorbidities of prematurity (osteopenia, necrotizing enterocolitis, chronic lung disease, and cholestasis) and fractures were identified by International Classification of Disease, 9th Edition, codes. Pharmaceutical records identified treatment with caffeine, diuretics, postnatal corticosteroids, and antacids. Poisson regression analysis determined fracture rate by 5 years of life.

RESULTS

There were 65 938 infants born in 2009-2010 who received care in the military health system for ≥5 years, including 3589 born preterm; 165 born at ≤28^6/7 weeks of gestation, 380 born at 29-31^6/7 weeks of gestation, and 3044 born at 32-36^6/7 weeks of gestation. Preterm birth at any gestational age was not associated with fracture rate in adjusted models. The fracture rate was increased with cholestasis, proton pump inhibitor exposure, and male sex.

CONCLUSIONS

Prematurity was not associated with fracture rate. Neonatal cholestasis and proton pump inhibitor treatment were associated with increased fractures by age 5.