Deep learning-based prediction of major arrhythmic events in dilated cardiomyopathy: A proof of concept study.

Prediction of major arrhythmic events (MAEs) in dilated cardiomyopathy represents an unmet clinical goal. Computational models and artificial intelligence (AI) are new technological tools that could offer a significant improvement in our ability to predict MAEs. In this proof-of-concept study, we propose a deep learning (DL)-based model, which we termed Deep ARrhythmic Prevention in dilated cardiomyopathy (DARP-D), built using multidimensional cardiac magnetic resonance data (cine videos and hypervideos and LGE images and hyperimages) and clinical covariates, aimed at predicting and tracking an individual patient's risk curve of MAEs (including sudden cardiac death, cardiac arrest due to ventricular fibrillation, sustained ventricular tachycardia lasting ≥30 s or causing haemodynamic collapse in <30 s, appropriate implantable cardiac defibrillator intervention) over time. The model was trained and validated in 70% of a sample of 154 patients with dilated cardiomyopathy and tested in the remaining 30%. DARP-D achieved a 95% CI in Harrell's C concordance indices of 0.12-0.68 on the test set. We demonstrate that our DL approach is feasible and represents a novelty in the field of arrhythmic risk prediction in dilated cardiomyopathy, able to analyze cardiac motion, tissue characteristics, and baseline covariates to predict an individual patient's risk curve of major arrhythmic events. However, the low number of patients, MAEs and epoch of training make the model a promising prototype but not ready for clinical usage. Further research is needed to improve, stabilize and validate the performance of the DARP-D to convert it from an AI experiment to a daily used tool.

Fitting Early Phases of the COVID-19 Outbreak: A Comparison of the Performances of Used Models.

The COVID-19 outbreak involved a spread of prediction efforts, especially in the early pandemic phase. A better understanding of the epidemiological implications of the different models seems crucial for tailoring prevention policies. This study aims to explore the concordance and discrepancies in outbreak prediction produced by models implemented and used in the first wave of the epidemic. To evaluate the performance of the model, an analysis was carried out on Italian pandemic data from February 24, 2020. The epidemic models were fitted to data collected at 20, 30, 40, 50, 60, 70, 80, 90, and 98 days (the entire time series). At each time step, we made predictions until May 31, 2020. The Mean Absolute Error (MAE) and the Mean Absolute Percentage Error (MAPE) were calculated. The GAM model is the most suitable parameterization for predicting the number of new cases; exponential or Poisson models help predict the cumulative number of cases. When the goal is to predict the epidemic peak, GAM, ARIMA, or Bayesian models are preferable. However, the prediction of the pandemic peak could be made carefully during the early stages of the epidemic because the forecast is affected by high uncertainty and may very likely produce the wrong results.

Automatic Forecast of Intensive Care Unit Admissions: The Experience During the COVID-19 Pandemic in Italy.

The experience of the COVID-19 pandemic showed the importance of timely monitoring of admissions to the ICU admissions. The ability to promptly forecast the epidemic impact on the occupancy of beds in the ICU is a key issue for adequate management of the health care system.Despite this, most of the literature on predictive COVID-19 models in Italy has focused on predicting the number of infections, leaving trends in ordinary hospitalizations and ICU occupancies in the background.This work aims to present an ETS approach (Exponential Smoothing Time Series) time series forecasting tool for admissions to the ICU admissions based on ETS models. The results of the forecasting model are presented for the regions most affected by the epidemic, such as Veneto, Lombardy, Emilia-Romagna, and Piedmont.The mean absolute percentage errors (MAPE) between observed and predicted admissions to the ICU admissions remain lower than 11% for all considered geographical areas.In this epidemiological context, the proposed ETS forecasting model could be suitable to monitor, in a timely manner, the impact of COVID-19 disease on the health care system, not only during the early stages of the pandemic but also during the vaccination campaign, to quickly adapt possible preventive interventions.

Pediatric Injury Surveillance From Uncoded Emergency Department Admission Records in Italy: Machine Learning-Based Text-Mining Approach.

BACKGROUND: Unintentional injury is the leading cause of death in young children. Emergency department (ED) diagnoses are a useful source of information for injury epidemiological surveillance purposes. However, ED data collection systems often use free-text fields to report patient diagnoses. Machine learning techniques (MLTs) are powerful tools for automatic text classification. The MLT system is useful to improve injury surveillance by speeding up the manual free-text coding tasks of ED diagnoses. OBJECTIVE: This research aims to develop a tool for automatic free-text classification of ED diagnoses to automatically identify injury cases. The automatic classification system also serves for epidemiological purposes to identify the burden of pediatric injuries in Padua, a large province in the Veneto region in the Northeast Italy. METHODS: The study includes 283,468 pediatric admissions between 2007 and 2018 to the Padova University Hospital ED, a large referral center in Northern Italy. Each record reports a diagnosis by free text. The records are standard tools for reporting patient diagnoses. An expert pediatrician manually classified a randomly extracted sample of approximately 40,000 diagnoses. This study sample served as the gold standard to train an MLT classifier. After preprocessing, a document-term matrix was created. The machine learning classifiers, including decision tree, random forest, gradient boosting method (GBM), and support vector machine (SVM), were tuned by 4-fold cross-validation. The injury diagnoses were classified into 3 hierarchical classification tasks, as follows: injury versus noninjury (task A), intentional versus unintentional injury (task B), and type of unintentional injury (task C), according to the World Health Organization classification of injuries. RESULTS: The SVM classifier achieved the highest performance accuracy (94.14%) in classifying injury versus noninjury cases (task A). The GBM method produced the best results (92% accuracy) for the unintentional and intentional injury classification task (task B). The highest accuracy for the unintentional injury subclassification (task C) was achieved by the SVM classifier. The SVM, random forest, and GBM algorithms performed similarly against the gold standard across different tasks. CONCLUSIONS: This study shows that MLTs are promising techniques for improving epidemiological surveillance, allowing for the automatic classification of pediatric ED free-text diagnoses. The MLTs revealed a suitable classification performance, especially for general injuries and intentional injury classification. This automatic classification could facilitate the epidemiological surveillance of pediatric injuries by also reducing the health professionals' efforts in manually classifying diagnoses for research purposes.

[ChatGPT in scientific research: a guide to informed use]. / ChatGPT nella ricerca scientifica: guida all'uso consapevole.

Using ChatGPT in scientific research offers revolutionary opportunities thanks to its natural language interaction capabilities and production of coherent and sophisticated text.Artificial intelligence can automate activities such as information synthesis and schematization, improving scientific communication and computer code writing.However, the lack of a complete understanding of context, the risk of spreading misleading information, and the possibility of plagiarism represent some of the biggest limitations in the current use of this technology.The role of human experience remains fundamental for in-depth understanding of context, exercising critical thinking, and ensuring respect for the ethical principles of scientific research.A responsible and aware use of tools such as ChatGPT can offer great benefits to the scientific community, but it is essential to remember that these tools are only a support and cannot replace human judgment and experience.

Monitoring the Epidemiology of Otitis Using Free-Text Pediatric Medical Notes: A Deep Learning Approach.

Free-text information represents a valuable resource for epidemiological surveillance. Its unstructured nature, however, presents significant challenges in the extraction of meaningful information. This study presents a deep learning model for classifying otitis using pediatric medical records. We analyzed the Pedianet database, which includes data from January 2004 to August 2017. The model categorizes narratives from clinical record diagnoses into six types: no otitis, non-media otitis, non-acute otitis media (OM), acute OM (AOM), AOM with perforation, and recurrent AOM. Utilizing deep learning architectures, including an ensemble model, this study addressed the challenges associated with the manual classification of extensive narrative data. The performance of the model was evaluated according to a gold standard classification made by three expert clinicians. The ensemble model achieved values of 97.03, 93.97, 96.59, and 95.48 for balanced precision, balanced recall, accuracy, and balanced F1 measure, respectively. These results underscore the efficacy of using automated systems for medical diagnoses, especially in pediatric care. Our findings demonstrate the potential of deep learning in interpreting complex medical records, enhancing epidemiological surveillance and research. This approach offers significant improvements in handling large-scale medical data, ensuring accuracy and minimizing human error. The methodology is adaptable to other medical contexts, promising a new horizon in healthcare analytics.

COVID-19 hospitalizations and patients' age at admission: The neglected importance of data variability for containment policies.

Introduction: An excess in the daily fluctuation of COVID-19 in hospital admissions could cause uncertainty and delays in the implementation of care interventions. This study aims to characterize a possible source of extravariability in the number of hospitalizations for COVID-19 by considering age at admission as a potential explanatory factor. Age at hospitalization provides a clear idea of the epidemiological impact of the disease, as the elderly population is more at risk of severe COVID-19 outcomes. Administrative data for the Veneto region, Northern Italy from February 1, 2020, to November 20, 2021, were considered. Methods: An inferential approach based on quasi-likelihood estimates through the generalized estimation equation (GEE) Poisson link function was used to quantify the overdispersion. The daily variation in the number of hospitalizations in the Veneto region that lagged at 3, 7, 10, and 15 days was associated with the number of news items retrieved from Global Database of Events, Language, and Tone (GDELT) regarding containment interventions to determine whether the magnitude of the past variation in daily hospitalizations could impact the number of preventive policies. Results: This study demonstrated a significant increase in the pattern of hospitalizations for COVID-19 in Veneto beginning in December 2020. Age at admission affected the excess variability in the number of admissions. This effect increased as age increased. Specifically, the dispersion was significantly lower in people under 30 years of age. From an epidemiological point of view, controlling the overdispersion of hospitalizations and the variables characterizing this phenomenon is crucial. In this context, the policies should prevent the spread of the virus in particular in the elderly, as the uncontrolled diffusion in this age group would result in an extra variability in daily hospitalizations. Discussion: This study demonstrated that the overdispersion, together with the increase in hospitalizations, results in a lagged inflation of the containment policies. However, all these interventions represent strategies designed to contain a mechanism that has already been triggered. Further efforts should be directed toward preventive policies aimed at protecting the most fragile subjects, such as the elderly. Therefore, it is essential to implement containment strategies before the occurrence of potentially out-of-control situations, resulting in congestion in hospitals and health services.

A topic trend analysis on COVID-19 literature.

Objective: In the past 2 years, the number of scientific publications has grown exponentially. The COVID-19 outbreak hugely contributed to this dramatic increase in the volume of published research. Currently, text mining of the volume of SARS-CoV-2 and COVID-19 publications is limited to the first months of the outbreak. We aim to identify the major topics in COVID-19 literature collected from several citational sources and analyze the temporal trend from November 2019 to December 2021. Methods: We performed an extensive literature search on SARS-Cov-2 and COVID-19 publications on PubMed, Scopus, and Web of Science (WoS) and a structural topic modelling on the retrieved abstracts. The temporal trend of the recognized topics was analyzed. Furthermore, a comparison between our corpus and the COVID-19 Open Research Dataset (CORD-19) repository was performed. Results: We collected 269,186 publications and identified 10 topics. The most popular topic was related to the clinical pictures of the COVID-19 outbreak, which has a constant trend, and the least popular includes studies on COVID-19 literature and databases. "Telemedicine", "Vaccine development", and "Epidemiology" were popular topics in the early phase of the pandemic; increasing topics in the last period are "COVID-19 impact on mental health", "Forecasting", and "Molecular Biology". "Education" was the second most popular topic, which emerged in September 2020. Conclusions: We identified 10 topics for classifying COVID-19 research publications and estimated a nonlinear temporal trend that gives an overview of their unfolding over time. Several citational databases must be searched to retrieve a complete set of studies despite the efforts to build repositories for COVID-19 literature. Our collected data can help build a more focused literature search between November 2019 and December 2021 when carrying out systematic and rapid reviews and our findings can give a complete picture on the topic.

A SuperLearner Approach to Predict Run-In Selection in Clinical Trials.

A critical early step in a clinical trial is defining the study sample that appropriately represents the target population from which the sample will be drawn. Envisaging a "run-in" process in study design may accomplish this task; however, the traditional run-in requires additional patients, increasing times, and costs. The possible use of the available a-priori data could skip the run-in period. In this regard, ML (machine learning) techniques, which have recently shown considerable promising usage in clinical research, can be used to construct individual predictions of therapy response probability conditional on patient characteristics. An ensemble model of ML techniques was trained and validated on twin randomized clinical trials to mimic a run-in process within this framework. An ensemble ML model composed of 26 algorithms was trained on the twin clinical trials. SuperLearner (SL) performance for the Verum (Treatment) arm is above 70% sensitivity. The Positive Predictive Value (PPP) achieves a value of 80%. Results show good performance in the direction of being useful in the simulation of the run-in period; the trials conducted in similar settings can train an optimal patient selection algorithm minimizing the run-in time and costs of conduction.

A Deep Learning Approach to Estimate the Incidence of Infectious Disease Cases for Routinely Collected Ambulatory Records: The Example of Varicella-Zoster.

The burden of infectious diseases is crucial for both epidemiological surveillance and prompt public health response. A variety of data, including textual sources, can be fruitfully exploited. Dealing with unstructured data necessitates the use of methods for automatic data-driven variable construction and machine learning techniques (MLT) show promising results. In this framework, varicella-zoster virus (VZV) infection was chosen to perform an automatic case identification with MLT. Pedianet, an Italian pediatric primary care database, was used to train a series of models to identify whether a child was diagnosed with VZV infection between 2004 and 2014 in the Veneto region, starting from free text fields. Given the nature of the task, a recurrent neural network (RNN) with bidirectional gated recurrent units (GRUs) was chosen; the same models were then used to predict the children's status for the following years. A gold standard produced by manual extraction for the same interval was available for comparison. RNN-GRU improved its performance over time, reaching the maximum value of area under the ROC curve (AUC-ROC) of 95.30% at the end of the period. The absolute bias in estimates of VZV infection was below 1.5% in the last five years analyzed. The findings in this study could assist the large-scale use of EHRs for clinical outcome predictive modeling and help establish high-performance systems in other medical domains.

A Web-Based Application to Monitor and Inform about the COVID-19 Outbreak in Italy: The {COVID-19ita} Initiative.

The pandemic outbreak of COVID-19 has posed several questions about public health emergency risk communication. Due to the effort required for the population to adopt appropriate behaviors in response to the emergency, it is essential to inform the public of the epidemic situation with transparent data sources. The COVID-19ita project aimed to develop a public open-source tool to provide timely, updated information on the pandemic's evolution in Italy. It is a web-based application, the front end for the eponymously named R package freely available on GitHub, deployed both in English and Italian. The web application pulls the data from the official repository of the Italian COVID-19 outbreak at the national, regional, and provincial levels. The app allows the user to select information to visualize data in an interactive environment and compare epidemic situations over time and across different Italian regions. At the same time, it provides insights about the outbreak that are explained and commented upon to yield reasoned, focused, timely, and updated information about the outbreak evolution.

Prediction of treatment outcome in clinical trials under a personalized medicine perspective.

A central problem in most data-driven personalized medicine scenarios is the estimation of heterogeneous treatment effects to stratify individuals into subpopulations that differ in their susceptibility to a particular disease or response to a specific treatment. In this work, with an illustrative example on type 2 diabetes we showed how the increasing ability to access and analyzed open data from randomized clinical trials (RCTs) allows to build Machine Learning applications in a framework of personalized medicine. An ensemble machine learning predictive model is first developed and then applied to estimate the expected treatment response according to the medication that would be prescribed. Machine learning is quickly becoming indispensable to bridge science and clinical practice, but it is not sufficient on its own. A collaborative effort is requested to clinicians, statisticians, and computer scientists to strengthen tools built on machine learning to take advantage of this evidence flow.

Oversampling and replacement strategies in propensity score matching: a critical review focused on small sample size in clinical settings.

BACKGROUND: Propensity score matching is a statistical method that is often used to make inferences on the treatment effects in observational studies. In recent years, there has been widespread use of the technique in the cardiothoracic surgery literature to evaluate to potential benefits of new surgical therapies or procedures. However, the small sample size and the strong dependence of the treatment assignment on the baseline covariates that often characterize these studies make such an evaluation challenging from a statistical point of view. In such settings, the use of propensity score matching in combination with oversampling and replacement may provide a solution to these issues by increasing the initial sample size of the study and thus improving the statistical power that is needed to detect the effect of interest. In this study, we review the use of propensity score matching in combination with oversampling and replacement in small sample size settings. METHODS: We performed a series of Monte Carlo simulations to evaluate how the sample size, the proportion of treated, and the assignment mechanism affect the performances of the proposed approaches. We assessed the performances with overall balance, relative bias, root mean squared error and nominal coverage. Moreover, we illustrate the methods using a real case study from the cardiac surgery literature. RESULTS: Matching without replacement produced estimates with lower bias and better nominal coverage than matching with replacement when 1:1 matching was considered. In contrast to that, matching with replacement showed better balance, relative bias, and root mean squared error than matching without replacement for increasing levels of oversampling. The best nominal coverage was obtained by using the estimator that accounts for uncertainty in the matching procedure on sets of units obtained after matching with replacement. CONCLUSIONS: The use of replacement provides the most reliable treatment effect estimates and that no more than 1 or 2 units from the control group should be matched to each treated observation. Moreover, the variance estimator that accounts for the uncertainty in the matching procedure should be used to estimate the treatment effect.

Improving Healthy Aging by Monitoring Patients' Lifestyle through a Wearable Device: Results of a Feasibility Study.

Population aging is related to a huge growth in healthcare and welfare costs. Therefore, wearable devices could be strategic for minimizing years of disability in old age and monitoring patients' lifestyles and health. The purpose of this study was to assess the feasibility of using smart devices to monitor patients' physical activity in a primary care setting. To assess the acceptance of this novel technology from the point of view of both patients and healthcare professionals, two questionnaires (one paper-based and one ex-novo developed) were administered to 11 patients with type 2 diabetes mellitus and a non-compliant behavior towards the therapeutic indications of their general practitioner (GP). Seven participants would continue to use a wearable activity tracker to monitor their health. We observed that 75% of patients reported a device's characteristics satisfaction level of over 80% of the total score assigned to this dimension. No differences were observed in the questionnaire's scores between the two professionals categories (GPs and nurses). Three dimensions (equipment characteristics, subjective norm, perceived risks, perceived ease-of-use and facilitating conditions) correlated > 0.5 with the device's acceptability level. Some weak correlations were observed between healthcare professionals' perception and patients' parameters, particularly between the dimensions of collaboration and web interface ease-of-use and patients' median number of steps and hours of sleep. In conclusion, despite the limited number of subjects involved, a good acceptance level towards these non-medical devices was observed, according to both patients' and healthcare professionals' impressions.

Adjustment for Baseline Covariates to Increase Efficiency in RCTs with Binary Endpoint: A Comparison of Bayesian and Frequentist Approaches.

BACKGROUND: In a randomized controlled trial (RCT) with binary outcome the estimate of the marginal treatment effect can be biased by prognostic baseline covariates adjustment. Methods that target the marginal odds ratio, allowing for improved precision and power, have been developed. METHODS: The performance of different estimators for the treatment effect in the frequentist (targeted maximum likelihood estimator, inverse-probability-of-treatment weighting, parametric G-computation, and the semiparametric locally efficient estimator) and Bayesian (model averaging), adjustment for confounding, and generalized Bayesian causal effect estimation frameworks are assessed and compared in a simulation study under different scenarios. The use of these estimators is illustrated on an RCT in type II diabetes. RESULTS: Model mis-specification does not increase the bias. The approaches that are not doubly robust have increased standard error (SE) under the scenario of mis-specification of the treatment model. The Bayesian estimators showed a higher type II error than frequentist estimators if noisy covariates are included in the treatment model. CONCLUSIONS: Adjusting for prognostic baseline covariates in the analysis of RCTs can have more power than intention-to-treat based tests. However, for some classes of model, when the regression model is mis-specified, inflated type I error and potential bias on treatment effect estimate may arise.

The Surplus Transplant Lung Allocation System in Italy: An Evaluation of the Allocation Process via Stochastic Modeling.

BACKGROUND: Lung transplantation is a specialized procedure used to treat chronic end-stage respiratory diseases. Due to the scarcity of lung donors, constructing fair and equitable lung transplant allocation methods is an issue that has been addressed with different strategies worldwide. This work aims to describe how Italy's "national protocol for the management of surplus organs in all transplant programs" functions through an online app to allocate lung transplants. We have developed two probability models to describe the allocation process among the various transplant centers. An online app was then created. The first model considers conditional probabilities based on a protocol flowchart to compute the probability for each area and transplant center to receive each n-th organ in the period considered. The second probability model is based on the generalization of the binomial distribution to correlated binary variables, which is based on Bahadur's representation, to compute the cumulative probability for each transplant center to receive at least nth organs. Our results show that the impact of the allocation of a surplus organ depends mostly on the region where the organ was donated. The discrepancies shown by our model may be explained by a discrepancy between the northern and southern regions in relation to the number of organs donated.

Monitoring Patients Reported Outcomes after Valve Replacement Using Wearable Devices: Insights on Feasibility and Capability Study: Feasibility Results.

Wearable devices (WDs) can objectively assess patient-reported outcomes (PROMs) in clinical trials. In this study, the feasibility and acceptability of using commercial WDs in elderly patients undergoing transcatheter aortic valve replacement (TAVR) or surgical aortic valve replacement (SAVR) will be explored. This is a prospective observational study. Participants were trained to use a WD and a smartphone to collect data on their physical activity, rest heart rate and number of hours of sleep. Validated questionnaires were also used to evaluate these outcomes. A technology acceptance questionnaire was used at the end of the follow up. In our participants an overall good compliance in wearing the device (75.1% vs. 79.8%, SAVR vs. TAVR) was assessed. Half of the patients were willing to continue using the device. Perceived ease of use is one of the domains that scored higher in the technology acceptance questionnaire. In this study we observed that the use of a WD is accepted in our frail population for an extended period. Even though commercial WDs are not tailored for clinical research, they can produce useful information on patient behavior, especially when coordinated with intervention tailored to the single patient.

Psychological Effects of Lockdown Measures for the COVID-19 Outbreak in Patients with Systemic Lupus Erythematosus.

OBJECTIVE: To compare the psychological impact of the lockdown measures contrasting the COVID-19 outbreak between systemic lupus erythematosus (SLE) and general population. PATIENTS AND METHODS: From July 15th to August 15th 2020, a retrospective survey referring to the period March 9th to May 18th 2020 was administered to SLE patients and the results of the survey, called LEPRE (Lupus Erythematosus PREsto) study, were compared with those from the PRESTO (imPact of quaRantine mEasures againST cOvid19) project, the same survey provided to the general population. Consecutive patients >18 years old affected by SLE and regularly followed in a single rheumatologic centre were involved. Primary outcome was to compare the scores of the Impact of Events Scale-Revised (IES-R), the General Health Questionnaire 12 (GHQ-12) and the Center for Epidemiological Depression Scale (CES-D) between patients and general population. RESULTS: A total of 64 patients completed the survey. After a propensity score matching, they were compared to 128 people from PRESTO project. The median age among patients was 43 years (I-III interquartile range 35-54.5), 88% were female and 100% Caucasian. IES-R [(score>23: 57% (34) vs 49% (58)], GHQ-12 [(score>13: 85% (52) vs 88% (106)], and CES-D [(score>15: 45% (28) vs 40% (46)] scores were not statistically different between patients and controls (p>0.05). CONCLUSION: Restrictive measures for COVID-19 pandemic had no greater impact on patients with SLE than in the general population. Strategy for coping to the SLE might be useful during lockdown measures and may be helpful for other chronic conditions.

Excess of all-cause mortality is only partially explained by COVID-19 in Veneto (Italy) during spring outbreak.

BACKGROUND: Italy has been the first European country to be affected by the COVID-19 epidemic which started out at the end of February. In this report, we focus our attention on the Veneto Region, in the North-East of Italy, which is one of the areas that were first affected by the rapid spread of SARS-CoV-2. We aim to evaluate the trend of all-cause mortality and to give a description of the characteristics of the studied population. METHODS: Data used in the analyses were released by the majority of municipalities and cover the 93% of the total population living in the Veneto Region. We evaluated the trend of overall mortality from Jan.01 to Jun.30. 2020. Moreover we compared the COVID-19-related deaths to the overall deaths. RESULTS: From March 2020, the overall mortality rate increased exponentially, affecting males and people aged > 76 the most. The confirmed COVID-19-related death rate in the Veneto region between Mar.01 and Apr.302020 is 30 per 100,000 inhabitants. In contrast, the all-cause mortality increase registered in the same months in the municipalities included in the study is 219 per 100,000 inhabitants. CONCLUSIONS: COVID-19 has a primary role in the increase in mortality but does not entirely explain such a high number of deaths. Strategies need to be developed to reduce this gap in case of future waves of the pandemic.

Assessing school-based intervention strategies to foster the prevention of choking injuries in children: The results of the CHOP (CHOking Prevention) trial.

There are few public health programs aimed at reducing choking injuries, even though choking is one of the leading causes of death among unintentional injuries in young children. We present the results from the CHOP (CHOking Prevention) project community intervention trial, which aimed to compare three different school-based intervention strategies for food choking prevention. The trial enrolled 41 educational facilities, which were randomized to one of three different intervention strategies to inform about prevention of food choking, or to serve as the control group. In strategy A, education was delivered directly to families, whereas in strategy B, education was delivered first to teachers and by them to families, and in strategy C, education was delivered only to healthcare coordinators in each school and by them to teachers and families. All educational interventions were delivered in the schools by experts and certified trainers. The participants were asked about sociodemographic information and completed questionnaires (pre-, post- and follow-up of intervention). Information from the postintervention and follow-up questionnaires was synthesized into four indicators to evaluate the effectiveness of each intervention strategy. Of the 1,426 participants, 298 were involved in strategy A, 474 in strategy B, 491 in strategy C and 163 in the control group. At postintervention, the scores of the indicators in each strategy significantly outperformed those of the control group, with adjusted p < 0.05. At follow-up, the distribution of the indicator scores of each strategy was found to be not significantly different compared to those of the control group (p > 0.05). The results of this study suggest that a sustainable school-based public health intervention mediated by teachers is effective as direct training for families in improving knowledge about food choking injury prevention. Nevertheless, further improvements could be made to increase long-term information retention.