Ting Fung Ma

Rationale: Chronic obstructive pulmonary disease (COPD) is known to be associated with substantial health‐care utilization and economic costs. Since the COVID‐19 pandemic, the epidemiology of various diseases has undergone significant changes, it is important to evaluate the health‐care impact and economic burden of acute exacerbations of COPD (AECOPD) in the post‐pandemic period, along with factors influencing this burden, to inform effective management strategies and health‐care planning. Objectives: To assess the health‐care and economic burden of severe AECOPD from 2022 to 2024, and to identify factors associated with direct health‐care costs. Methods: A territory‐wide retrospective cohort study was conducted. Adult patients with AECOPD admitted to acute public hospitals in Hong Kong were included. The primary outcomes were the total number of AECOPD admissions and the associated direct health‐care costs. Measurements and Main Results: From 2022 to 2024, a total of 11,465 patients were admitted to acute public hospitals for severe AECOPD, resulting in 25,053 hospitalization episodes and 150,717 bed days. The estimated direct health‐care costs totaled 1.259 billion Hong Kong dollars (HKD). AECOPD admission numbers increased significantly over the 3 years. Patients with eosinophilic and non‐eosinophilic phenotypes had similar annual direct health‐care costs related to severe AECOPD. Patients with co‐existing bronchiectasis had a significantly higher annual number of severe AECOPD episodes and direct health‐care costs from severe AECOPD. Conclusions: Severe AECOPD imposes a substantial burden on the health‐care system. The presence of bronchiectasis and a greater frequency of severe AECOPD episodes were identified as possible factors linked to increased direct health‐care costs.

Background: Co-existing asthma and diabetes mellitus (DM) are common, and there has been a debate on whether glucagon-like peptide-1 (GLP-1)-based therapies, such as GLP-1 receptor agonists (GLP-1RA) and inhibitors of dipeptidyl peptidase-4 (DPP-4i), confer benefits in these patients due to their anti-inflammatory effects in chronic inflammatory diseases. Methods: A territory-wide retrospective cohort study was conducted in Hong Kong among adult patients with co-existing asthma and DM to examine the impact of add-on GLP-1RA or DPP-4i on asthma exacerbations, ranging from mild exacerbations managed in outpatient settings to severe exacerbations requiring hospitalization. Adult patients were prescribed DPP-4i or GLP-1RA in year 2018 and for at least 6 months in Hospital Authority of Hong Kong were included. These patients were followed till 31st December 31, 2023. Propensity score matching was performed using optimal full matching, which is a sub-classification-based approach in which treated and control subjects are assigned to matched subclasses. Results: A total of 3295 patients with both asthma and DM, 3193 treated with DPP-4i and 102 treated with GLP-1RA, were included in the study. There were 1191 (36.1%) male patients with a mean age of 68.0 ± 14.2 years. There were 2531 (76.8%) patients with moderate-to-severe asthma according to Global Initiative for Asthma (GINA) Steps 3 to 5. GLP-1RA-treated patients had significantly lower risks of hospitalized asthma exacerbation (Average Treatment Effect on the Treated (ATT) Mean Ratio of 0.513, p < 0.001), ad hoc outpatient visits with OCS (ATT Mean Ratio of 0.343, p = 0.005), emergency visits (ATT Mean Ratio of 0.378, p = 0.009), and all asthma exacerbations (ATT Mean Ratio of 0.419, p < 0.001) compared with the DPP-4i group. Conclusion: GLP-1RA, compared with DPP-4i, as an add-on treatment for patients with co-existing DM and asthma was associated with a lower risk of asthma exacerbation. The use of GLP-1RA among diabetic patients with comorbid asthma may be considered though validation in clinical trials is needed.

Used car pricing is always complex and involves various factors. It is difficult for consumers to obtain full transparency into used car pricing, which sellers often obscure to protect margins. Even if the pricing model for one specific used car can be established, the wide variety of used cars and features further complicates the problem. Motivated by a real-world application in insurance premium pricing for used cars, our study presents a transparently explainable solution to predict various used car prices with elaborate feature engineering. We propose a novel three-phase approach, with each phase dedicated to prediction models, explainable machine learning, and the examination of marginal effects of binary variables. We conduct a comparative analysis to assess the performance of different predicting models with real-world data for robustness. The consistent results suggest that ensemble tree algorithms perform the best, with a mean absolute percentage error of 0.269 (out-of-sample), compared with 0.957 from the linear benchmark. The performance of deep neural networks with numerical features is salient when dealing with more homogeneous and specific data. By examining the individual feature attributes, the car’s age, make (i.e., brand), and engine capacity are the top three determinants of the price with strong monotonicity. Finally, the marginal effects of significant binary variables and their interactions on prices are quantified. Our analysis reveals the nonlinearity and complexity in effect sizes. The proposed three-phase approach provides a transparent and reliable pricing pipeline for the used car market and can essentially assist stakeholders in making decisions.

Inter‐seeding red clover ( Trifolium pratense L.) provides an alternative method to incorporate cover crops into continuous corn ( Zea mays L.) in the Upper US Midwest. Red clover is a leguminous cover crop that can grow in low‐radiation environments and is winter hardy. Systems with red clover have demonstrated improved corn yield and a fertilizer N equivalence but understanding these effects with inter‐seeding warrants further investigation. The objectives of this study were to determine the effect of inter‐seeding red clover on (i) plant‐available N during and after red clover decomposition, (ii) optimum N rates for corn, and (iii) corn yields. The experimental design was a randomized, complete block‐split plot design, with cover crop as the main plot factor (treatments with and without inter‐seeded red clover) and N‐rate as a split‐plot factor (N‐rates between 0 and 315 kg‐N ha ⁻¹ in 45 kg‐N ha ⁻¹ intervals). Quadratic plateau response curves with a bootstrapping technique were used to determine differences in optimum N rate among treatments. Corn yields were evaluated with or without red clover inter‐seeded following the first inter‐seeding year. Red clover accumulated biomass values of 50 kg ha ⁻¹ up to 300 kg ha ⁻¹ pre‐termination when inter‐seeded with corn at the V4–V5 growth stage without detriment to yield. Corn yield was improved with clover treatments in one of four site years tested, resulting in a (15–17 kg ha ⁻¹ ) lower N requirement that year. Overall, our results indicate that inter‐seeding red clover into continuous corn did not provide an agronomically meaningful nitrogen fertilizer equivalence to the cropping system.

Background/Objectives: Vaccination is a key preventive measure against pneumococcal disease, but uptake rates remain low in high-risk populations. Limited information exists on pneumococcal vaccine uptake in individuals with a history of pneumococcal disease. This study aims to assess pneumococcal vaccine uptake and the factors associated with it in patients hospitalized for pneumococcal disease, before and after hospitalization, across time periods before, during, and after the COVID-19 pandemic. Methods: Data for patients aged ≥18 years who were hospitalized for pneumococcal disease between 2015 and 2024 were extracted from the Hospital Authority’s territory-wide electronic medical record database. The uptake of pneumococcal vaccines in subgroups aged 18–64 years and ≥65 years, with and without risk conditions, both before and after hospitalization for pneumococcal disease, was assessed, followed by multivariate analyses of the factors associated with vaccination uptake by logistic regression models. Results: This study included 5517 patients hospitalized for pneumococcal disease. Prior to hospitalization, the vaccination uptake among the eligible patients was 20.5%, with only 8.1% fully vaccinated, despite the majority (87.9%) having previous hospitalizations (subgroup medians 3–9 times) or outpatient clinic visits (subgroup median 61–107 times). After discharge, during a median follow-up of 1.85 years, almost all the eligible patients (98.4%) received subsequent inpatient (subgroup medians 3–4 times) and outpatient (subgroup medians 21–28 times) care, but only 32.2% of the eligible patients received the vaccine. Factors associated with increased vaccine uptake post-discharge included age ≥75 years (OR 1.6), ≥10 subsequent hospitalizations (OR 2.1), and ≥10 subsequent clinic visits (OR 55.9). Vaccination rates within 12 months post-discharge were significantly lower in the patients hospitalized during the COVID-19 pandemic (3.5%) compared to the baseline (11.6%) and post-COVID-19 (6.6%) periods. Conclusions: The uptake of the pneumococcal vaccine before hospitalization for pneumococcal disease was low and continued to be suboptimal post-discharge. Numerous vaccination opportunities were missed in both the inpatient and outpatient settings. These findings indicate a need to improve vaccination strategies.

Technology permeates daily life. During adolescence, screen time may limit time that could be spent on activities from sports to sleep. Studies often examine screen time in isolation, overlooking that time use is relative—more screen time results in less time available elsewhere. Drawing on a representative birth cohort of 2,539 U.K. adolescents, we examined how tradeoffs in screen time versus other activities are linked to adolescent well-being. Replacing activities like active time, physical activity, or sleep with screen time was associated with worse outcomes, and some effects held three years later. However, screen time was beneficial compared to passive time. Our approach accounts for the inherent tradeoffs of screen time, pinpointing when screen time can be harmful, helpful, or neutral for adolescents to lead a satisfying life.

Background: The incidence of pneumococcal pneumonia in the context of the Coronavirus Disease 2019 (COVID-19) pandemic, along with the real-world data on the ratio of non-invasive to invasive pneumococcal pneumonia, is an area that has not been thoroughly studied. The outcomes associated with coinfection of influenza and COVID-19 remain unknown. This study examined the incidence, demographics, coinfection with influenza and/or COVID-19, and clinical outcomes of pneumococcal hospitalizations in Hong Kong during the baseline, pandemic, and post-pandemic periods. Methods: Hospitalization records of individuals aged 18 years and above with pneumococcal disease from January 2015 to August 2024 were extracted from the territory-wide electronic medical record database. Pneumococcal disease was categorized into invasive pneumococcal pneumonia (IPP), invasive pneumococcal disease without pneumonia (IPDWP), and non-invasive pneumococcal pneumonia (NIPP), followed by univariate and multivariate analyses. Effects of coinfection with influenza and COVID-19 were analyzed. Results: The incidence of pneumococcal disease decreased during the COVID-19 pandemic but rebounded in the post-pandemic period. There were no significant changes in the distribution of pneumococcal serotypes across the three periods. The study revealed a strong positive correlation between monthly pneumococcal hospitalizations and the indicator of influenza activity, while the correlation with the COVID-19 indicator was weak. Additionally, strong positive correlations were observed between the indicator of influenza activity and influenza coinfections, as well as between the indicator of COVID-19 activity and COVID-19 coinfections. Multivariate analyses identified male gender, a higher comorbidity index, older age, invasive pneumococcal disease (IPP and IPDWP), coinfection with influenza and COVID-19, and hospitalization during the pandemic period as factors associated with adverse outcomes. Conclusions: The study showcases changes in the epidemiology of pneumococcal disease during and after the COVID-19 pandemic. It highlights the adverse effects of influenza and COVID-19 coinfections on the outcomes of patients with pneumococcal disease and emphasizes the need to vaccinate vulnerable populations against these infections.

Introduction Chronic obstructive pulmonary disease (COPD) has been reported to have significant healthcare and economic burdens worldwide. Since the COVID-19 pandemic, there have been changes in the epidemiology of various chronic diseases. The healthcare and economic burden of severe acute exacerbation COPD (AECOPD) and the factors affecting the burden should be assessed in the post-pandemic era. Methods A territory-wide study was conducted in Hong Kong to study the healthcare and economic burden of severe AECOPD from 2022 to 2024, and identify factors associated with an increase in the direct healthcare costs related to severe AECOPD. Adult patients with severe AECOPD admitted to the acute medical unit managed by the Hong Kong Hospital Authority from 2022 to 2024 were included. The primary outcome was the total headcount and admission numbers related to severe AECOPD, as well as the direct healthcare costs associated with severe AECOPD. The secondary outcomes included factors associated with increased direct healthcare costs resulting from severe AECOPD. Results From 2022 to 2024, there were 11,465 COPD patients admitted to acute medical hospitals for severe AECOPD, contributing to 25,053 hospital admissions and 150,717 bed days. The total direct healthcare cost was HKD $1.259 billion for these severe AECOPD. There was a significant increase in the severe AECOPD admission numbers over the 3 years. While the phenotype by baseline BEC did not affect the annual total healthcare costs related to severe AECOPD, patients with co-existing bronchiectasis had significantly higher annual number of severe AECOPD and annual direct healthcare costs. The annual number of severe AECOPD significantly correlates with total direct healthcare costs and annual direct healthcare costs related to severe AECOPD. Conclusions Severe AECOPD carries a significant burden towards healthcare system. Patients with bronchiectasis and more frequent severe AECOPD were factors associated with a higher number of severe AECOPD episodes and the associated direct healthcare costs.

Spatiotemporal causal inference methods are needed to detect the effect of interventions on indirectly measured epidemiological outcomes that go beyond studying spatiotemporal correlations. Chronic wasting disease (CWD) causes neurological degeneration and eventual death to white‐tailed deer (Odocoileus virginianus) in Wisconsin. Targeted culling involves removing deer after traditional hunting seasons in areas with high CWD prevalence. The evaluation of the causal effects of targeted culling in the spread and growth of CWD is an important unresolved research and CWD management question that can guide surveillance efforts. Reaction–diffusion partial differential equations (PDEs) can be used to mechanistically model the underlying spatiotemporal dynamics of wildlife diseases, like CWD, allowing researchers to make inference about unobserved epidemiological quantities. These models indirectly regress spatiotemporal covariates on diffusion and growth rates parameterizing such PDEs, obtaining associational conclusions. In this work we develop an innovative method to obtain causal estimators for the effect of targeted culling interventions on CWD epidemiological processes using an inverse‐probability‐of‐treatment‐weighted technique by means of marginal structural models embedded in the PDE fitting process. Additionally we establish a novel scheme for sensitivity analysis under unmeasured confounder for testing the hypothesis of a significant causal effect in the indirectly measured epidemiological outcomes. Our methods can be broadly used to study the impact of spatiotemporal interventions and treatment exposures in the epidemiological evolution of infectious diseases that can help to inform future efforts to mitigate public health implications and wildlife disease burden.