Understanding COVID-19 Impacts on the Health Workforce: AI-Assisted Open-Source Media Content Analysis
Notice bibliographique
Résumé
BACKGROUND: To investigate the impacts of the COVID-19 pandemic on the health workforce, we aimed to develop a framework that synergizes natural language processing (NLP) techniques and human-generated analysis to reduce, organize, classify, and analyze a vast volume of publicly available news articles to complement scientific literature and support strategic policy dialogue, advocacy, and decision-making. OBJECTIVE: This study aimed to explore the possibility of systematically scanning intelligence from media that are usually not captured or best gathered through structured academic channels and inform on the impacts of the COVID-19 pandemic on the health workforce, contributing factors to the pervasiveness of the impacts, and policy responses, as depicted in publicly available news articles. Our focus was to investigate the impacts of the COVID-19 pandemic and, concurrently, assess the feasibility of gathering health workforce insights from open sources rapidly. METHODS: We conducted an NLP-assisted media content analysis of open-source news coverage on the COVID-19 pandemic published between January 2020 and June 2022. A data set of 3,299,158 English news articles on the COVID-19 pandemic was extracted from the World Health Organization Epidemic Intelligence through Open Sources (EIOS) system. The data preparation phase included developing rules-based classification, fine-tuning an NLP summarization model, and further data processing. Following relevancy evaluation, a deductive-inductive approach was used for the analysis of the summarizations. This included data extraction, inductive coding, and theme grouping. RESULTS: After processing and classifying the initial data set comprising 3,299,158 news articles and reports, a data set of 5131 articles with 3,007,693 words was devised. The NLP summarization model allowed for a reduction in the length of each article resulting in 496,209 words that facilitated agile analysis performed by humans. Media content analysis yielded results in 3 sections: areas of COVID-19 impacts and their pervasiveness, contributing factors to COVID-19-related impacts, and responses to the impacts. The results suggest that insufficient remuneration and compensation packages have been key disruptors for the health workforce during the COVID-19 pandemic, leading to industrial actions and mental health burdens. Shortages of personal protective equipment and occupational risks have increased infection and death risks, particularly at the pandemic's onset. Workload and staff shortages became a growing disruption as the pandemic progressed. CONCLUSIONS: This study demonstrates the capacity of artificial intelligence-assisted media content analysis applied to open-source news articles and reports concerning the health workforce. Adequate remuneration packages and personal protective equipment supplies should be prioritized as preventive measures to reduce the initial impact of future pandemics on the health workforce. Interventions aimed at lessening the emotional toll and workload need to be formulated as a part of reactive measures, enhancing the efficiency and maintainability of health delivery during a pandemic.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,013 | 0,058 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,000 |
| Méta-épidémiologie (sens large) | 0,001 | 0,001 |
| Bibliométrie | 0,027 | 0,011 |
| Études des sciences et des technologies | 0,001 | 0,002 |
| Communication savante | 0,006 | 0,006 |
| Science ouverte | 0,002 | 0,004 |
| Intégrité de la recherche | 0,001 | 0,002 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,003 | 0,001 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».