BMJ Paediatrics Open最新文献

Objective: Early-onset sepsis (EOS), which occurs within the first 72 hours of life, can often be fatal for neonates. Machine learning (ML) models demonstrate promise for timely diagnosis. However, current ML models primarily rely on data from high-income countries, which reduces their applicability to low-income and middle-income countries (LMICs) that have a higher burden and different disease profiles. We developed an ML model for the timely prediction of culture-proven EOS in LMICs.

Methods: We conducted a secondary analysis of the Delhi Neonatal Infection Study (2011-2014) carried out in three level-3 neonatal units in India. Data from inborn neonates suspected of having EOS were extracted, and cases of culture-negative sepsis were excluded. By implementing a dynamic 80:20 (train:test) data split, two feature selection methods were employed-Boruta and Lasso-across 64 variables, and five ML techniques were applied. The aim was to achieve 90% sensitivity to identify the optimal model based on performance metrics. The developed model was integrated into a web application and validated in an external cohort of neonates born between 2015 and 2021.

Results: Of 2924 neonates, 548 (18.7%) had culture-proven sepsis. The mean gestation and birth weight were 35.3 (±3.8) weeks and 2112 (±754) g, respectively. The Boruta and random forest classifier yielded the best model, which included 28 perinatal-neonatal variables. The sensitivity and specificity of the model were 90.3% and 40.6%, respectively. In external validation (n=147; 26 culture-proven sepsis cases), the model's sensitivity, specificity, positive predictive value and negative predictive value were 92.3%, 37.2%, 24.0% and 95.7%, respectively. The sensitivity was 100% in asymptomatic neonates with only perinatal risk factors for EOS. The use of the model could have reduced antibiotic usage from 74.8% to 55.7% (risk difference: -19.1%; 95% CI -8.3 to -29.7).

Conclusions: The ML model demonstrated high sensitivity and acceptable specificity in predicting EOS. This prediction model has the potential to assist in the timely and reliable identification of culture-positive sepsis and may serve as a bedside decision support tool in LMICs.

Study purpose: This study examines the implementation of the Screening Instrument for Child Abuse and Neglect (SCAN) and the Structured Tool for Evaluating Positive Screened cases (STEPS) in nine emergency departments (EDs). The study aimed to assess whether implementation could enhance compliance with SCAN and to evaluate the usability of SCAN&STEPS within an implementation-focused design, in response to the decline in child maltreatment recognition observed in Dutch EDs over the past decade.

Methods: SCAN&STEPS was embedded in ED workflow and electronic health records (EHRs), supported by a selection of implementation strategies, including e-learning, policy manuals and an awareness campaign. Effectiveness was assessed through compliance with SCAN, comparing preimplementation and postimplementation screening rates, with a ≥10% increase defined as clinically relevant. Usability of SCAN&STEPS was examined using a mixed-method design combining the Measurement Instrument for Determinants of Innovations and semistructured interviews. Subgroup analyses were conducted by hospital type, EHR, profession and years of working experience.

Results: After implementation, the average compliance rate increased from 57.5% to 70.5%, with 3 of 8 sites achieving ≥10% improvement. Rates varied by site, with university hospital EDs having the highest improvement. Compliance was influenced by EHR configurations. Usability analysis identified five facilitators (perception of responsibility, social support, self-efficacy, knowledge and formal management ratification) and one barrier (unsettled organisation). Users considered SCAN&STEPS user-friendly, though perceived support differed between nurses and physicians due to role-specific factors.

Relevance: SCAN&STEPS can improve compliance in recognising child maltreatment concerns, but tailored strategies are needed for further implementation in Dutch hospitals. The standardised approach enhances uniform data collection, enabling comparative analysis and interdisciplinary collaboration, advancing early detection of child maltreatment. Broader international adoption should account for policy and system-level differences and requires further validation to ensure applicability beyond the Dutch context.

Aim: This national survey aimed to evaluate the adoption and implementation of delivery room cuddles (DRC) for preterm infants (<32 weeks' gestation) in neonatal units across the UK. We sought to determine the extent of the practice, and understand the consistency and challenges associated with DRC, with the goal of enhancing care for preterm infants.

Methods: Between January and March 2024, all UK level 2 and 3 neonatal units were invited to complete a structured survey on DRC practices. We collected data on the existence, frequency and facilitation of DRC, as well as any institutional guidelines and barriers for its practice.

Results: We received complete responses from all 131 centres invited (100% response rate). Among these, 60 centres (45.8%) reported routinely practising DRC, 66 (50.4%) did so occasionally and 5 (3.8%) reported not practising DRC at all. Notably, 81 centres (61.8%) practised DRC without formal institutional guidelines. Main barriers to practice included equipment limitations, varying staff attitudes, and concerns about the clinical condition of preterm infants.

Conclusion: This survey demonstrates an encouraging high uptake of DRC in recent years, typically offered after initial stabilisation (within 30-60 min post-birth). Despite the recognised benefits of DRC to parents and in enhancing neonatal outcomes, considerable variability remains in its implementation underscoring the need for comprehensive guidelines and professional training. These could standardise DRC practices across the UK, promoting more consistent and effective care for preterm infants.

Background: The purpose of the study was to explore parents' experiences, perceptions, challenges and coping mechanisms after discharge from the neonatal intensive care unit.

Methods: The study adopted a qualitative descriptive design. Semi-structured interviews with 15 families (including 15 mothers and 5 fathers) from private and public hospitals in Lebanon. We analysed data following the six-step framework of inductive thematic analysis.

Results: 15 families participated in the interviews. The emerging themes reflected that the post-discharge period is mentally, physically and emotionally overwhelming to parents, who perceived their discharge preparation as insufficient. Parents faced challenges with infant care and struggled with financial strain. However, they used positive coping strategies, such as gratitude and joy, as well as problem-focused coping mechanisms, including seeking help from healthcare professionals, family and friends, online resources and digital platforms. Parents also reflected on insights for a better post-discharge experience.

Conclusion: Transitioning from the neonatal intensive care unit to home, parents of preterm infants experienced conflicting feelings of excitement, anxiety and fear. In the context of Lebanon, their needs for emotional, social and financial support were often unmet. Understanding their challenges is essential to developing effective strategies for support in their caregiving responsibilities.

Background: Adolescent substance use is a growing public health concern in Sri Lanka, yet little is known about how youth are first exposed to and continue using illicit substances.

Methods: This qualitative study explored the initiation and continuation of substance use among adolescents aged 10-17 years with a history of drug abuse. 48 adolescents were interviewed at the Youth Drug Addiction Prevention, Treatment & Rehabilitation Centre in Peradeniya, Sri Lanka, over a 1-year period. Data were collected through in-depth, semistructured interviews and analysed using reflexive thematic analysis with manual coding.

Results: Five major themes were identified: (1) curiosity-driven and self-initiated use; (2) peer modelling and social apprenticeship; (3) coercion and manipulated consent; (4) emotional triggers including grief and relationship stress and (5) sustained use driven by coping needs, pleasure and physical dependence. Initiation often occurred in familiar environments such as schools, temples and homes, with many adolescents describing use as functional rather than recreational.

Conclusions: Adolescent substance use in Sri Lanka is shaped by complex psychosocial and environmental influences. Prevention and intervention strategies must address not only individual behaviour but also social networks, emotional vulnerability and structural access to drugs in community settings.

Background: Maternal smoking has been associated with adverse neonatal outcomes; specifically, outcomes of preterm-born after maternal smoking have not been studied.

Objective: To compare the short-term hospital outcomes of preterm infants born to mothers who 'ever smoked' with matched preterm infants born to mothers who 'never smoked'.

Methods: This was a matched retrospective cohort study including all preterm infants <30 weeks between March 2019 and March 2023. Matching was performed, creating propensity scores with 'Mother ever smoked versus Mother never smoked' as dependent outcomes with birth gestational age, baby gender and Apgar score at 1 min as covariates. The standardised mean difference (SMD) was used to assess any potential imbalance. Appropriate statistical tests were applied.

Results: We had a total of 179 preterm infants born <30 weeks. After excluding 26 deaths, we had a total of 153 infants available for analysis before matching. After matching, we had 46 infants in each group (table 1). There was no imbalance between the baseline demographic and clinical factors, with all SMDs ≤0.1, except for the Apgar score at 10 min. Infants in the smoking group were significantly less likely to need surfactant (OR: 0.33) and were highly likely to be discharged on home oxygen (OR: 2.7). There was no difference in death rates between the two groups (16% in the non-smoking group vs 12% in the smoking group).

Conclusions: In this propensity score matching study, infants whose mothers ever smoked were more likely to be discharged on home oxygen.

Background: Children with disabilities were unlikely to be included in response strategies during the COVID-19 pandemic in the poorest countries that prioritise medical models. We aimed to understand the impact of COVID-19 on these children and their families in South Africa, through the lens of the WHO International Classification of Functioning, Disability and Health (ICF) framework, to advance service delivery that could make a difference to their lives.

Methods: Caregivers of children with developmental disabilities from an online parent network (n=400), initiated by Shonaquip Social Enterprise during the COVID-19 pandemic, were invited to participate in an online survey in October 2021, to understand their experiences. A quantitative analysis was conducted using Stata Statistical Software. NVivo 14, QSR International, enabled an iterative thematic analysis, using the ICF Framework.

Results: Out of 400 parents surveyed, 68 (17%) responded. They reported declines in child health (n=11, 16%), contracting COVID-19 infections (n=6, 9%) and demise (n=2, 3%) alongside food insecurity (n=42, 62%), income losses (n=17, 26%) and restricted educational access (n=22, 32%) during the pandemic. Caregiving was described as 'very stressful' (n=33, 49%) and some needed substances to cope (n=5, 7%). Parents expressed that services for children with disabilities were not prioritised during the pandemic. Although the parent network empowered resilience, strong themes of isolation were equivalent to their pre-pandemic experiences.

Conclusion: By collaborating with community organisations, and leveraging technology, we reached vulnerable populations during the pandemic in South Africa. By contextualising the ICF framework and 'F-Words', we exposed the challenges of families which may be compared with a crisis in low- and middle-income countries, beyond a pandemic. Whereas COVID-19 forced us to reimagine addressing these needs, this is a call to relisten and rebuild systems around children with disabilities. Bridging research to practice gaps and strengthening family resilience should be 'just a normal day'.

Background: Limited studies in Iran have assessed child health-friendly neighbourhood domains or examined the extent to which neighbourhoods support the health of children under 6 years of age. This study aimed to assess the status of child health-friendly neighbourhood domains in selected districts of Tehran using a previously designed and locally validated tool, with a focus on identifying neighbourhood strengths and weaknesses.

Methods: This pilot study was conducted between April and July 2023 in Tehran. 10 neighbourhoods from five districts were selected using convenience random sampling. Neighbourhoods were assessed using the Iranian Child Health-Friendly Neighbourhood checklist. Six dimensions-green spaces and parks, recreational centres, cultural centres, passages and streets, healthcare centres and kindergartens-were evaluated across five domains: facilities and services, accessibility, security, safety and aesthetics and view. Descriptive statistics, including means and frequencies, were used for data analysis.

Results: Accessibility achieved the highest overall score (94%), particularly for kindergartens and cultural centres. The lowest scores were observed in green spaces and parks and in the domain of aesthetics and view (both 57.2%). Neighbourhood safety and security were scored 73.5% and 66.8%, respectively. The aesthetics and view of green spaces and parks (32.2%) and facilities and services in kindergartens (38%) were identified as the lowest-scoring components. Based on the design of the tool, the aim was to identify neighbourhood strengths and weaknesses; therefore, no predefined criteria were applied to classify neighbourhoods as child-friendly or not.

Conclusions: Despite some strengths, the overall status of the assessed neighbourhoods indicates a need for targeted improvements to better support and promote children's health. The integration of field observation, a locally validated assessment tool and a focus on children under 6 years of age provides practical evidence to inform urban planning and health promotion policies aimed at creating safer and more child-friendly neighbourhoods.

Background: Seizure-like events present a diagnostic challenge for paediatricians, particularly in third-tier cities in southwestern China. Although government initiatives have improved the accuracy of epilepsy diagnosis, substantial disparities persist between southwestern and eastern China. The aims of this study were to investigate clinical characteristics and prognosis of children with seizure-like events in resource-limited cities in southwestern China and to identify factors associated with future epilepsy.

Methods: This was a retrospective cohort study. We retrospectively reviewed the medical records of 519 children aged 23 days to 13.5 years who experienced seizure-like events. Demographic information, seizure characteristics, electroencephalograph (EEG) and brain MRI were collected. Epilepsy was diagnosed according to the International League Against Epilepsy criteria. Multivariate logistic regression was performed to determine independent predictors of epilepsy.

Results: We enrolled 519 children in our hospital. Forty-five (45/263, 17.1%) patients were first diagnosed with seizures. At the first seizure presentation, 263 patients (50.7%) were diagnosed with epilepsy. The aetiology of these patients with epilepsy was more commonly unknown (166/263, 63.2%), followed by a genetic aetiology (65/263, 24.6%). Normal MRI and EEG findings were observed in 218 (42.0%) and 169 (32.6%) children, respectively. Seizure frequency, positive genetic test results, abnormal brain MRI and abnormal video-EEG were predictive of recurrent febrile seizures. Multivariate analysis revealed that high seizure frequency, abnormal video-EEG findings and structural brain abnormalities are key predictors of subsequent epilepsy in children.

Conclusions: Although we aimed to identify risk factors applicable to resource-limited areas, we found that only high seizure frequency, abnormal video-EEG findings and structural brain MRI abnormalities independently predicted a subsequent diagnosis of epilepsy. These findings underscore the need for targeted paediatrician training, as well as increased financial investment in resource-limited rural areas to improve hospital infrastructure, in order to help reduce the epilepsy treatment gap.