Identifying adverse childhood experiences with electronic health records of linked mothers and children in England: a multistage development and validation study

Preconception indicators and associations with health outcomes reported in UK routine primary care data: a systematic review

BACKGROUND

Routine primary care data may be a valuable resource for preconception health research and to inform the provision of preconception care.

AIM

To review how primary care data could provide information on the prevalence of preconception indicators and examine associations with maternal and offspring health outcomes.

DESIGN AND SETTING

Systematic review of observational studies using UK routine primary care data.

METHOD

Literature searches were conducted in March 2023 using five databases to identify observational studies that used national primary care data from individuals aged 15-49 years. Preconception indicators were defined as medical, behavioural, and social factors that may impact future pregnancies; health outcomes included those that may occur during and after pregnancy.

RESULTS

From 5259 screened records, 42 articles were included. The prevalence of 37 preconception indicator measures was described for female patients, ranging from 0.01% for sickle cell disease to >20% for each of advanced maternal age, previous caesarean section (among those with a recorded pregnancy), overweight, obesity, smoking, depression, and anxiety (irrespective of pregnancy). Few studies reported indicators for male patients (n = 3) or associations with outcomes (n = 5). Most studies had a low risk of bias, but missing data may limit generalisability of the findings.

CONCLUSION

The findings demonstrated that routinely collected UK primary care data could be used to identify patients' preconception care needs. Linking primary care data with health outcomes collected in other datasets is underutilised, but could help to quantify how optimising preconception health and care could reduce adverse outcomes for mothers and children.

Large-scale deep learning analysis to identify adult patients at risk for combined and common variable immunodeficiencies

BACKGROUND

Primary immunodeficiency (PI) is a group of heterogeneous disorders resulting from immune system defects. Over 70% of PI is undiagnosed, leading to increased mortality, co-morbidity and healthcare costs. Among PI disorders, combined immunodeficiencies (CID) are characterized by complex immune defects. Common variable immunodeficiency (CVID) is among the most common types of PI. In light of available treatments, it is critical to identify adult patients at risk for CID and CVID, before the development of serious morbidity and mortality.

METHODS

We developed a deep learning-based method (named "TabMLPNet") to analyze clinical history from nationally representative medical claims from electronic health records (Optum® data, covering all US), evaluated in the setting of identifying CID/CVID in adults. Further, we revealed the most important CID/CVID-associated antecedent phenotype combinations. Four large cohorts were generated: a total of 47,660 PI cases and (1:1 matched) controls.

RESULTS

The sensitivity/specificity of TabMLPNet modeling ranges from 0.82-0.88/0.82-0.85 across cohorts. Distinctive combinations of antecedent phenotypes associated with CID/CVID are identified, consisting of respiratory infections/conditions, genetic anomalies, cardiac defects, autoimmune diseases, blood disorders and malignancies, which can possibly be useful to systematize the identification of CID and CVID.

CONCLUSIONS

We demonstrated an accurate method in terms of CID and CVID detection evaluated on large-scale medical claims data. Our predictive scheme can potentially lead to the development of new clinical insights and expanded guidelines for identification of adult patients at risk for CID and CVID as well as be used to improve patient outcomes on population level.

Family adversity and health characteristics associated with intimate partner violence in children and parents presenting to health care: a population-based birth cohort study in England

BACKGROUND

Little is known about the clinical characteristics of children and parents affected by intimate partner violence (IPV) presenting in health-care settings. We examined the associations between family adversities, health characteristics, and IPV in children and parents using linked electronic health records (EHRs) from primary and secondary care between 1 year before and 2 years after birth (the first 1000 days). We compared parental health problems in in children and parents with and without recorded IPV.

METHODS

We developed a population-based birth cohort of children and parents (aged 14-60 years) in England, comprising linked EHRs from mother-child pairs (with no identified father) and mother-father-child triads. We followed the cohort across general practices (Clinical Practice Research Datalink GOLD), emergency departments, outpatient visits, hospital admissions, and mortality records. Family adversities included 33 clinical indicators of parental mental health problems, parental substance misuse, adverse family environments, and high-risk child maltreatment-related presentations. Parental health problems included 12 common comorbidities, ranging from diabetes and cardiovascular diseases to chronic pain or digestive diseases. We used adjusted and weighted logistic-regression models to estimate the probability of IPV (per 100 children and parents) associated with each adversity, and period prevalences of parental health problems associated with IPV.

FINDINGS

We included 129 948 children and parents, comprising 95 290 (73·3%) mother-father-child triads and 34 658 (26·7%) mother-child pairs only between April 1, 2007, and Jan 29, 2020. An estimated 2689 (2·1%) of 129 948 children and parents (95% CI 2·0-2·3) had recorded IPV and 54 758 (41·2%; 41·5-42·2) had any family adversity between 1 year before and 2 years after birth. All family adversities were significantly associated with IPV. Most parents and children with IPV had recorded adversities (1612 [60·0%] of 2689) before their first IPV recording. The probability of IPV was 0·6 per 100 children and parents (95% CI 0·5-0·6) with no adversity, increasing to 4·4 per 100 children per parents (4·2-4·7) with one adversity, and up to 15·1 per 100 parents and children (13·6-16·5) with three of more adversities. Mothers with IPV had a significantly higher prevalence of both physical (73·4% vs 63·1%, odds ratio [OR] 1·6, 95% CI 1·4-1·8) and mental health problems (58·4% vs 22·2%, OR 4·9, 4·4-5·5) than mothers without IPV. Fathers with IPV had a higher prevalence of mental health problems (17·8% vs 7·1%, OR 2·8, 2·4-3·2) and similar prevalences of physical health problems than those without IPV (29·6% vs 32·4%, OR 0·9, 0·8-1·0).

INTERPRETATION

Two in five of the children and parents presenting to health care had recorded parental mental health problems, parental substance misuse, adverse family environments, or high-risk presentations of maltreatment in the first 1000 days. One in 22 children and parents with family adversity also had recorded IPV before age 2 years. Primary and secondary care staff should safely ask about IPV when parents or children present with family adversity or health problems associated with IPV, and respond appropriately.

FUNDING

NIHR Policy Research Programme.

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.

Association of Interparental Violence and Maternal Depression With Depression Among Adolescents at the Population and Individual Level

Importance

Parental intimate partner violence (IPV) and maternal depression are associated with increased risk of depression in children at the population level. However, it is not known whether having information about these experiences can accurately identify individual children at higher risk of depression.

Objective

To examine the extent to which experiencing parental IPV and/or maternal depression before age 12 years is associated with depression at age 18 years at the population and individual level.

Design, Setting, and Participants

This cohort study used data from the Avon Longitudinal Study of Parents and Children, a UK population-based birth cohort, which initially recruited pregnant mothers with estimated due dates in 1991 and 1992. Data used in this study were collected from 1991 to 2009. Data analysis was performed from February to March 2022.

Exposures

Mother-reported parental IPV was assessed on 8 occasions (child age, 1-11 years). Maternal depression was assessed via the Edinburgh Postnatal Depression Scale or by the mother taking medication for depression, as reported by the mother on 8 occasions (child age, 2-12 years).

Main Outcomes and Measures

Depressive symptoms were measured with the Short Mood and Feelings Questionnaire (SMFQ) and Clinical Interview Schedule-Revised (CIS-R) when the child was aged 18 years. Binary indicators of a case of depression were derived the cutoff point of 11 points or above for the SMFQ and 12 points or above for the CIS-R.

Results

The study included 5029 children (2862 girls [56.9%]; 2167 boys [43.1%]) with a measure of depressive symptoms at age 18 years. IPV only was associated with a 24% (adjusted risk ratio, 1.24; 95% CI, 0.97-1.59) higher risk of depression at age 18 years, exposure to maternal depression only was associated with a 35% (adjusted risk ratio, 1.35; 95% CI, 1.11-1.64) higher risk, and exposure to both IPV and maternal depression was associated with a 68% (adjusted risk ratio, 1.68; 95% CI, 1.34-2.10) higher risk. At the individual level, the area under the receiver operating characteristic curve was 0.58 (95% CI, 0.55-0.60) for depression according to the SMFQ and 0.59 (95% CI, 0.55-0.62) for the CIS-R, indicating a 58% to 59% probability (ie, 8%-9% above chance) that a random participant with depression at age 18 years had been exposed to IPV and/or maternal depression compared with a random participant who did not have depression.

Conclusions and Relevance

In this cohort study, parental IPV and maternal depression were associated with depression in adolescence at the population level. However, estimation of an individual developing depression in adolescence based only on information about IPV or maternal depression is poor. Screening children for maternal depression and IPV to target interventions to prevent adolescent depression will fail to identify many children who might benefit and may unnecessarily target many others who do not develop depression.