Exploratory research on asthma exacerbation risk factors using the Japanese claims database and machine learning: A retrospective cohort study

Machine Learning-Based Asthma Attack Prediction Models From Routinely Collected Electronic Health Records: Systematic Scoping Review

BACKGROUND

An early warning tool to predict attacks could enhance asthma management and reduce the likelihood of serious consequences. Electronic health records (EHRs) providing access to historical data about patients with asthma coupled with machine learning (ML) provide an opportunity to develop such a tool. Several studies have developed ML-based tools to predict asthma attacks.

OBJECTIVE

This study aims to critically evaluate ML-based models derived using EHRs for the prediction of asthma attacks.

METHODS

We systematically searched PubMed and Scopus (the search period was between January 1, 2012, and January 31, 2023) for papers meeting the following inclusion criteria: (1) used EHR data as the main data source, (2) used asthma attack as the outcome, and (3) compared ML-based prediction models' performance. We excluded non-English papers and nonresearch papers, such as commentary and systematic review papers. In addition, we also excluded papers that did not provide any details about the respective ML approach and its result, including protocol papers. The selected studies were then summarized across multiple dimensions including data preprocessing methods, ML algorithms, model validation, model explainability, and model implementation.

RESULTS

Overall, 17 papers were included at the end of the selection process. There was considerable heterogeneity in how asthma attacks were defined. Of the 17 studies, 8 (47%) studies used routinely collected data both from primary care and secondary care practices together. Extreme imbalanced data was a notable issue in most studies (13/17, 76%), but only 38% (5/13) of them explicitly dealt with it in their data preprocessing pipeline. The gradient boosting-based method was the best ML method in 59% (10/17) of the studies. Of the 17 studies, 14 (82%) studies used a model explanation method to identify the most important predictors. None of the studies followed the standard reporting guidelines, and none were prospectively validated.

CONCLUSIONS

Our review indicates that this research field is still underdeveloped, given the limited body of evidence, heterogeneity of methods, lack of external validation, and suboptimally reported models. We highlighted several technical challenges (class imbalance, external validation, model explanation, and adherence to reporting guidelines to aid reproducibility) that need to be addressed to make progress toward clinical adoption.

The Link between Periodontal Disease and Asthma: How Do These Two Diseases Affect Each Other?

A growing body of evidence suggests that the effects of poor oral hygiene extend beyond the oral cavity and are associated with a variety of systemic diseases, including asthma. Asthma, which results in symptoms of cough, wheezing, and dyspnoea, and is characterized by airflow limitation with variability and (partial or complete) reversibility, is amongst the most prevalent respiratory diseases with approximately 262 million patients worldwide, and its prevalence and disease burden is on the increase. While asthma can occur at a young age, it can also develop later in life and affects a variety of age groups. Both of these diseases have a chronic course, and various researchers have suggested a link between the two. In this article, we aim to provide a literature review focusing on the association between the two diseases. The results demonstrate that medications (primarily, inhaler medicine), hypoxia induced by asthma, and the breathing behaviour of patients potentially trigger periodontal disease. In contrast, oral periodontopathogenic microorganisms and the inflammatory mediators produced by them may be involved in the onset and/or exacerbation of asthma. Common contributing factors, such as smoking, gastro-oesophageal reflux, and type-2 inflammation, should also be considered when evaluating the relationship between the two diseases.

The association between respiratory diseases and periodontitis: A systematic review and meta-analysis

AIM

To evaluate (1) whether periodontitis has an influence on the prevalence/incidence of respiratory diseases (chronic obstructive pulmonary disease [COPD], asthma, community-acquired pneumonia [CAP], obstructive sleep apnoea [OSA] and COVID-19), and (2) what is the impact of periodontal therapy on the onset or progression of respiratory diseases.

MATERIALS AND METHODS

An electronic search was performed on Pubmed, Cochrane Library and Scopus databases up to October 2021, to identify studies answering the PECOS and PICOS questions.

RESULTS

Seventy-five articles were selected. Meta-analyses identified statistically significant associations of periodontitis with COPD (nstudies  = 12, odds ratio [OR] = 1.28, 95% confidence interval [CI] [1.16; 1.42], p < .001), and OSA (ns  = 6, OR = 1.65, 95% CI [1.21; 2.25], p = .001), but not for asthma (ns  = 9, OR = 1.53, 95% CI [0.82; 2.86], p = .181). For acute conditions, two studies were found for CAP, while for COVID-19, significant associations were found for the need of assisted ventilation (ns  = 2, OR = 6.24, 95% CI [2.78; 13.99], p < .001) and COVID-related mortality (ns  = 3, OR = 2.26, 95% CI [1.36, 3.77], p = .002). Only four intervention studies were found, showing positive effects of periodontal treatment on COPD, asthma and CAP.

CONCLUSIONS

A positive association between periodontitis and COPD, OSA and COVID-19 complications has been found, while there is a lack of intervention studies.

Treatment response among asthmatic patients with and without reversible airflow limitations

Objectives

Asthma is a chronic airway disorder associated with variable airflow limitations, which are triggered by different stimuli. The reversibility of airflow limitations reflects patients' responses to the therapy with bronchodilators and improvements in airflow. This study aims to determine the treatment outcomes (improvements in forced expiratory volume in the first second (FEV1) and the number of asthma exacerbations) associated with the presence of airflow reversibility.

Methods

This retrospective cohort study included 154 adults (>18 years) who were diagnosed with asthma and had pulmonary function testing (PFT) at a tertiary care centre in KSA between January 1st, 2014 and May 31st, 2019. Smokers and patients with comorbidities or medications that could affect PFT were excluded from the analysis. Patients were classified as having a reversible airflow limitation when they exhibited a post-bronchodilator FEV1 increase of 12% and 200 mL. Exacerbations were defined as the need to use oral corticosteroids. Chi-square tests were used for comparative analyses.

Results

From our cohort, 42 patients exhibited reversibility. In contrast, 112 patients did not show any sign of reversibility. Asthmatics with baseline reversible airflow limitations experienced significant worsening of FEV1 during the follow-up period compared with those with no reversibility, showing a mean difference of 19.96 mL (p = 0.0206). There was no significant association between asthma reversibility and exacerbations (p = 0.23).

Conclusion

In our study, during the follow-up of patients with asthma, we found that the reversibility of airflow was associated with significantly worse FEV1, although this did not have a significant effect on exacerbations. Therefore, we recommend regular spirometry follow-ups, particularly for patients with significant airway reversibility.