Journal of Global Health最新文献

Background: Lower respiratory tract infections (LRTI) in children under five years of age remain a leading cause of mortality, especially in low- and middle-income countries (LMIC). Although many of these deaths are preventable, accurate diagnosis of bacterial pneumonia is essential to ensure appropriate treatment and to reduce unnecessary antibiotic use. Imaging technologies are essential for improving the specificity of diagnoses; however, chest x-rays are often unavailable in rural LMIC settings. In this context, point-of-care lung ultrasound (POCLUS) offers a promising alternative, but the limited training remains a significant barrier to its widespread adoption and long-term sustainability. This study proposes and evaluates a training methodology to address this challenge.

Methods: We conducted a pre-post observational study without a control group across 10 rural health districts in Alta Verapaz, Guatemala. We divided the training programme into four progressive phases: training in technical skills and basic image acquisition (phase 0); assessment of image acquisition skills (phase 1); evaluation of the ability to identify pathological findings (phase 2); and training in a diagnostic algorithm and evaluation in the clinical context (phase 3). In each phase, two specialists independently evaluated images, and a third resolved disagreements. We combined in-person and remote training activities and implemented them between April 2021 and December 2024.

Results: A total of 23 healthcare professionals (six physicians, 15 nurses, and two nursing assistants) participated in the training. Of these, 19 successfully completed phase 1, 18 completed phase 2, and 13 completed phase 3. Eight participants discontinued the programme, primarily due to changes in professional roles or location.

Conclusions: The blended learning model proposed in this study enabled physician and non-physician healthcare providers to use POCLUS to diagnose LRTI in children under five years of age in a rural primary care setting in a low-income country.

Background: Systemic inflammation plays a critical role in coronary artery disease (CAD), yet comprehensive profiling of inflammatory markers and their integration into predictive models remain incompletely characterised. We here sought to identify key CAD-related inflammation markers and construct an integrated inflammatory risk score (IRS) to enhance conventional cardiovascular risk prediction.

Methods: Associations between 18 complete blood count based inflammatory markers and incident CAD were assessed among 475 134 UK Biobank participants free of CAD at baseline. Weighted quantile sum (WQS) regression evaluated the relative directional contributions of individual markers, and mediation analysis further examined the role of inflammation in linking accelerated aging and unhealthy lifestyle factors to CAD. Predictive performance was assessed by comparing IRS-augmented models in terms of AUC (area under the receiver operating characteristic curve), net reclassification improvement (NRI), and decision curve analysis.

Results: All analysed inflammatory markers showed statistically significant associations with CAD risk, although effect sizes varied. Monocyte-to-HDL-C ratio (MHR), neutrophil-to-HDL-C ratio (NHR), and systemic inflammation response index (SIRI) exhibited the strongest positive associations (21-41% increased risk per SD), while platelet-to-lymphocyte ratio (PLR) and platelet-to-leukocyte ratio (PWR) demonstrated modest inverse associations (5-13% decreased risk). Several markers (e.g. MHR, PLR, PWR) displayed discernible group-level differences over a decade before CAD onset. WQS regression highlighted heterogeneous, direction-specific contributions of these markers to CAD risk. Mediation analysis suggested that a portion of the observed associations may operate through accelerated aging (mediation proportion = 7-43%) or via inflammation in the link between unhealthy lifestyle and CAD. The integrated IRS modestly improved CAD risk prediction, particularly within the short-term window of 0 - 5 years (absolute increase in AUC (ΔAUC) = 2.7%, NRI = 1.6%; net benefit = 5%).

Conclusions: Inflammatory markers captured by routine test were consistently associated with future CAD, suggesting that part of CAD risk is reflected in low-cost hematologic parameters. An integrated IRS showed modest but statistically significant improvement in risk discrimination, but external validation and clinical impact studies are needed before implementation.

Background: Tajikistan has accomplished reductions in maternal, newborn, and child mortality over the past decades through targeted policies and interventions. Challenges remain in providing quality healthcare due to limited resources, geographic barriers, and inadequate infrastructure. We aimed to evaluate the impact of a quality improvement (QI) initiative implemented in ten district hospitals from 2021 to 2024 to improve maternal, newborn, and childcare to accelerate progress towards achieving the Sustainable Development Goals.

Methods: A baseline assessment was conducted in 2021, with an endline assessment in 2023, using updated WHO quality assessment tools. A multidisciplinary team of national and international experts evaluated hospital performance across three domains: support systems, clinical management, and organisation of care. Data was collected through observations, interviews, and medical record reviews. Project interventions included working with hospital-level Quality Improvement committees, capacity building in effective perinatal care, and the use of the WHO Pocketbook of Hospital Care for Children. Regular supportive supervision and half-yearly collaborative quality improvement meetings were held among the ten hospitals. We graphically displayed, analysed, and summed the assessment scores.

Results: We observed notable improvements in the quality of hospital care, with most facilities progressing from substandard to better-performing categories. Seven out of ten hospitals demonstrated advancements in their maternity and neonatal units, with improvements in clinical management and hospital support systems, including access to drugs and equipment. Challenges remained in paediatric care, with only two of ten hospitals showing improvements in infrastructure and laboratory services, and none improving drug availability. Improvements in infection prevention and control were minimal; however, four in ten hospitals managed to improve their practices despite challenges with resource availability, infrastructure, and current protocols.

Conclusions: Comprehensive QI interventions can raise standards of care in resource-limited settings like Tajikistan. Despite measurable progress, systemic barriers persist, with weak infrastructure, unstable workforce, and limited infection prevention and control, requiring targeted investment and political commitment. Sustained success depends on equitable resource allocation, robust monitoring systems, and the promotion of a non-punitive, systems-oriented culture. Scaling up this initiative nationwide is critical to achieving long-term improvements in health.

Background: Heart failure mortality has risen sharply after years of decline, highlighting the limitations of current risk assessment tools in accuracy, complexity, and cost, and the need for improved predictive models. To address this gap, we developed and validated a deep learning model to improve short-term mortality prediction in heart failure patients.

Methods: In this retrospective study, we leveraged the Medical Information Mart for Intensive Care IV database to develop HF-ECGNet, combining an EfficientNet neural network and a Transformer architecture. We also developed a composite model integrating electrocardiogram-based (ECG) predictions and clinical features. We evaluated model performance using the area under the curve (AUC) and other metrics, with gradient-weighted class activation mapping (Grad-CAM) and Shapley additive explanations (SHAP) analyses for interpretability. We conducted comparisons with N-terminal pro-B-type natriuretic peptide and sequential organ failure assessment (SOFA) scores.

Results: We analysed a total of 104 844 ECGs from 36 222 admissions. HF-ECGNet achieved an AUC of 0.664 for the first ECG during initial admission, improving to 0.721 for the last ECG. Incorporating three-day ECG data further enhanced performance, with AUCs of 0.691 (first admission) and 0.698 (last admission). HF-ECGNet outperformed NT-proBNP and SOFA. A composite model integrating ECG data and clinical features achieved the highest AUC of 0.725. Grad-CAM identified critical ECG patterns, while SHAP analysis highlighted ECG-derived features as the most influential predictors.

Conclusions: HF-ECGNet demonstrates potential as a powerful tool for predicting short-term mortality in heart failure patients. Its innovative architecture and integration of clinical data enable more accurate and interpretable risk stratification. Future multi-centre validation is the critical step to fully ascertain its clinical utility and generalisability.

Background: Child anaemia remains a major public health concern in low- and middle-income countries, particularly among children aged <2 years. Given Tanzania's slow progress towards addressing this challenge, despite the implementation of targeted national strategies, we aimed to examine the prevalence and associated factors of anaemia in children aged <2 years.

Methods: This cross-sectional study was based on secondary data from the Tanzania Demographic and Health Surveys (2004-2005, 2010, 2015-2016, and 2022). We applied individual sampling weights to ensure national representativeness and used descriptive analyses to estimate the prevalence of child anaemia, both overall and in specific age group distributions (6-23 and 24-59 months), by severity. Then, employed geographic information system mapping to visualise the regional anaemia prevalence among children aged <2 years. Finally, we conducted a multivariable logistic regression analysis to identify risk factors related to child anaemia and sensitivity analyses to check the robustness of our findings.

Results: We included 25 590 children aged <5 years with available haemoglobin data, of whom 10 166 were aged 6-23 months. In 2022, 59.9% of children aged <5 years and 73.1% of those aged 6-23 months were anaemic. Using spatial analysis, we found that 16 of 31 regions experienced increased prevalence of anaemia, while the remaining 15 saw slight to moderate declines since 2015. In the multivariable analysis, being male (adjusted odds ratio (aOR) = 1.380; 95% confidence interval (CI) = 1.133-1.681), having low birth weight (aOR = 1.713; 95% CI = 1.076-2.727), being underweight (aOR = 1.438; 95% CI = 1.032-2.005), and not having health insurance (aOR = 1.768; 95% CI = 1.247-2.508) were factors significantly associated with increased risk of anaemia.

Conclusion: s Child anaemia in Tanzania has stagnated between 2015-2022, failing to meet national targets. The burden has remained disproportionately high among low birth weight children and those aged <2 years, with those residing in coastal regions and high food production areas being especially vulnerable. Targeted interventions during the first 1000 days of life should be prioritised to break the intergenerational cycle of anaemia.

Background: The continuity and quality of maternal and child health (MCH) services represent significant challenges in low- and middle-income countries (LMICs). Home-based records (HBRs), including the integrated Maternal and Child Health Handbook (MCHHB), may support consistent service usage and improved care quality. We aimed to evaluate the association between HBRs and continued quality MCH care, identifying differences in associations between MCHHBs and the continuum of care and those between other HBR types and the continuum of care.

Methods: We conducted a pooled analysis of Demographic and Health Surveys (DHS) from 18 LMICs, including women and their youngest eligible children for the health card module. We defined the continuum of quality MCH care using receipt of quality antenatal care (ANC), skilled birth attendance, and quality postnatal care (qPNC), with ANC and PNC measured as composite indicators. We based the HBR ownership on the DHS variable 'Has health card'. We used multivariable logistic regressions to examine associations between HBR ownership, HBR type, and the continuum of care, adjusting for residence, maternal education, and wealth.

Results: The final analysis included 89 902 samples. Ownership of HBR was associated with significantly higher odds of completing the continuum of quality MCH care compared with not owning HBR. However, no significant differences were observed between owners of MCHHB and those of other HBR types. Based on analysing the subgroups of MCH service components, the MCHHB may facilitate the provision of ANC services, such as urine and blood tests, more effectively than other HBR types.

Conclusions: Owning HBR was positively associated with greater use of quality MCH services. However, no significant differences were observed for MCHHBs, despite MCHHB ownership being significantly associated with improved uptakes of urine and blood tests. Further research is needed to explore the influence of actual HBR use, provider-related factors, and variations in HBR content and type.

Background: Understanding the rates and determinants of severe acute malnutrition (SAM) relapse is crucial for stakeholders in Somalia, where evidence is limited. This study aimed to assess SAM relapse rates and associated risk factors among children discharged from outpatient therapeutic programmes (OTP) in the Bay and Hiran regions of Somalia.

Methods: We conducted a prospective cohort study of 160 children aged 7-53 months discharged as recovered from OTP SAM treatment between August-September 2023. Children were followed monthly for six time points post-discharge. Anthropometric measurements, morbidity data, and household information were collected. Survival analysis was used to calculate cumulative incidence of SAM relapse, defined by weight-for-height z-score (WHZ)<-3 standard deviation (SD) or mid-upper arm circumference (MUAC)<11.5 cm or oedema. Cox proportional hazard models identified factors associated with relapse.

Results: Cumulative incidence of SAM relapse at Time 1 (T1) = 5.2% (95% confidence interval (CI) = 2.5, 10.6%), T2 = 14.3% (95% CI = 9.4, 21.5%) and T6 = 26.0% (95% CI = 19.3, 34.5%) by WHZ and 13.2% (95% CI = 8.8, 19.5%) by MUAC. The relapse rate for combined SAM and moderate acute malnutrition by WHZ at T1 = 26.9% (95% CI = 19.5, 36.3%), T2 = 36.2% (95% CI = 28.0, 46.1%) and T6 = 50.1% (95% CI = 41.0, 60.0%). Weight-for-height z-score (WHZ)-based relapse was higher in rural areas (31.4% vs. 22.7% urban, P = 0.285) and among children with WHZ<-3 SD at admission (37.4% vs. 21.2%, P = 0.029). Mid-upper arm circumference (MUAC)-based relapse was higher in urban areas (20.8% vs. 4.1% rural, P = 0.002), among younger children (19.7% vs. 5.5% > 2 years, P = 0.009), and internally displaced persons (21.8% vs. 5.8% non-internally displaced persons, P = 0.003). Factors significantly associated with increased relapse risk included WHZ<-3 SD at admission (adjusted hazard ratio (HR) = 2.22; 95% CI = 1.04, 4.72) and longer OTP stay (adjusted HR = 1.02 per day; 95% CI = 1.00, 1.04). Participation in a cash assistance programme was protective (adjusted HR = 0.44; 95% CI = 0.22, 0.90).

Conclusions: Severe acute malnutrition (SAM) relapse rates in Somalia are considerable, with varying patterns by anthropometric indicator, region, and demographic factors. Cash assistance programme offers a promising complementary intervention. These findings can inform targeted interventions and policy changes to reduce relapse and improve long-term outcomes for children recovering from SAM in Somalia and similar contexts.

Registration: The cluster-RCT associated with this cohort study is registered at ClinicalTrials.gov, ID: NCT06642012.

Pakistan's neonatal and under-five mortality rates remain among the highest in South Asia. This challenge is exacerbated by a critical shortage of paediatric intensive care staff, with <30 trained specialists serving >80 million children. Recent national data indicate that only 70% of accredited paediatric hospitals have functional paediatric intensive care units, and nearly half lack adequate nurse-to-patient ratios. These shortages are most severe in rural regions, where delayed access to life-saving interventions contributes to preventable deaths. Emerging solutions such as telemedicine, neonatal care units within general hospitals, and public-private partnerships demonstrate potential for scalable reform. Strengthening paediatric critical care is essential to achieving Sustainable Development Goal 3.2 and ending preventable child deaths by 2030.

Background: This meta-analysis aims to determine the clinical manifestations, prevalence, and risk factors of asthenopia across diverse populations.

Methods: We systematically searched PubMed up to April 2024 for studies published within the last five years on asthenopia, without language or design restrictions. Reference lists were also reviewed. The study quality was evaluated using the Newcastle-Ottawa Scale. A random-effects meta-analysis was conducted to calculate proportions, prevalence rates, odds ratios (ORs) and their 95% confidence intervals (CIs).

Results: Overall, 63 studies were included. The pooled prevalence of asthenopia detected via questionnaires or symptom report was 51% (95% CI = 50%, 52%). Subgroup analyses showed high prevalence among digital device users (90%) and computer workers (77%). During the COVID-19 pandemic, prevalence rose among adults (39%-45%), university students (36%-57%), and school-aged children (45%-64%). The most frequent ocular symptoms were eye tiredness (65%, 95% CI = 46%, 84%), eye strain (47%, 95% CI = 37%, 58%), and burning/irritation (43%, 95% CI = 35%, 51%). Musculoskeletal symptoms, including neck pain (45%, 95% CI = 28%, 62%) and shoulder pain (30%, 95% CI = 12%, 48%) were also prevalent. Neuropsychological symptoms included headache (50%, 95% CI = 41%, 59%) and difficulty concentrating (44%, 95% CI = 32%, 56%). Risk factors included short sleep duration (OR = 1.28; 95% CI = 1.04, 1.57), prior eye disease (OR = 2.59; 95% CI = 1.43, 4.69), prolonged screen time (OR = 1.15; 95% CI = 1.09, 1.21), and ambient conditions like air conditioning use (OR = 23.02; 95% CI = 4.94, 107.18). Protective measures included anti-glare filters (OR = 0.34; 95% CI = 0.19, 0.64), regular breaks (OR = 0.21; 95% CI = 0.09, 0.51), and computer use knowledge (OR = 0.20; 95% CI = 0.13, 0.30).

Conclusions: Asthenopia is prevalent across diverse populations, characterised by a wide range of symptoms and influenced by modifiable risk factors. Our findings support a unified definition to improve clinical recognition and offer preliminary evidence to help shape future research on preventive strategies.

Registration: PROSPERO: CRD42024536841.

Background: Astigmatism is a common refractive error in children and adolescents. This study aimed to investigate the prevalence, severity, types, correction status, and associated risk factors of astigmatism among children and adolescents in Shaanxi Province, China.

Methods: A cross-sectional study was conducted involving 236 397 children and adolescents aged 3-20 years from Shaanxi Province, selected through multistage stratified cluster random sampling. Demographic information was gathered via structured questionnaires, and refractive status was assessed using an autorefractor without cycloplegia. Astigmatism was defined as cylindrical refractive error ≥0.50 dioptres.

Results: The overall prevalence of astigmatism was 73.81% among the studied population. Age showed a clear positive association with prevalence, which was significantly higher in adolescents compared to younger children. Similar patterns were observed across advancing educational levels. Males showed slightly higher prevalence than females. Geographic distribution revealed a north-to-south gradient, with northern regions having the highest prevalence. Regarding astigmatism types, with-the-rule astigmatism and compound myopic astigmatism were the predominant forms. Only 30.73% of affected subjects had received refractive correction, with correction rates markedly increasing with age and educational level. Multivariate analyses confirmed that older age, higher educational attainment, male gender, northern residence, and non-Han ethnicity were independent risk factors for astigmatism.

Conclusions: Astigmatism prevalence among children in Shaanxi Province is considerably higher than previously reported in other Chinese regions, with significant variations across demographic factors. The low correction rate, particularly among younger children, highlights the need for enhanced early screening and timely intervention programmes to prevent visual function impairment.