Introduction: Over the years there have been several interventions targeted at the public to increase their knowledge and awareness about Antimicrobial Resistance (AMR). In this work, we updated a previously published review by Price et al. (2018), on effectiveness of interventions to improve the public’s antimicrobial resistance awareness and behaviours associated with prudent use of antimicrobials to identify which interventions work best in influencing public behaviour. Methods: Five databases—Medline (OVID), CINAHL (EBSCO), Embase, PsycINFO, and Cochrane Central Register of Controlled Trials (CENTRAL-OVID)—were searched for AMR interventions between 2017 and 2021 targeting the public. All studies which had a before and after assessment of the intervention were considered for inclusion. Results: In total, 17 studies were found to be eligible for inclusion in the review. Since there was a variety in the study interventions and in particular outcomes, a narrative synthesis approach was adopted for analysis. Whereas each study showed some impact on awareness and knowledge, none measured long-term impact on behaviours towards antibiotic use, awareness, or knowledge. Engagement was higher in interventions which included interactive elements such as games or videos. Social media was not used for recruitment of participants or as a mode of communication in any AMR interventions included in this review.

Introduction: The use of antibiotics leads to antibiotic resistance (ABR). Different methods have been used to predict and control ABR. In recent years, artificial intelligence (AI) has been explored to improve antibiotic (AB) prescribing, and thereby control and reduce ABR. This review explores whether the use of AI can improve antibiotic prescribing for human patients. Methods: Observational studies that use AI to improve antibiotic prescribing were retrieved for this review. There were no restrictions on the time, setting or language. References of the included studies were checked for additional eligible studies. Two independent authors screened the studies for inclusion and assessed the risk of bias of the included studies using the National Institute of Health (NIH) Quality Assessment Tool for observational cohort studies. Results: Out of 3692 records, fifteen studies were eligible for full-text screening. Five studies were included in this review, and a narrative synthesis was carried out to assess their findings. All of the studies used supervised machine learning (ML) models as a subfield of AI, such as logistic regression, random forest, gradient boosting decision trees, support vector machines and K-nearest neighbours. Each study showed a positive contribution of ML in improving antibiotic prescribing, either by reducing antibiotic prescriptions or predicting inappropriate prescriptions. However, none of the studies reported the engagement of AB prescribers in developing their ML models, nor their feedback on the user-friendliness and reliability of the models in different healthcare settings. Conclusion: The use of ML methods may improve antibiotic prescribing in both primary and secondary settings. None of the studies evaluated the implementation process of their models in clinical practices. Prospero Registration: (CRD42022329049).

Governments and healthcare organisations collect data on antibiotic prescribing (AP) for surveillance. This data can support tools for visualisations and feedback to GPs using dashboards that may prompt a change in prescribing behaviour. The objective of this systematic review was to assess the effectiveness of interactive dashboards to optimise AP in primary care. Six electronic databases were searched for relevant studies up to August 2022. A narrative synthesis of findings was conducted to evaluate the intervention processes and results. Two independent reviewers assessed the relevance, risk of bias and quality of the evidence. A total of ten studies were included (eight RCTs and two non-RCTs). Overall, seven studies showed a slight reduction in AP. However, this reduction in AP when offering a dashboard may not in itself result in reductions but only when combined with educational components, public commitment or behavioural strategies. Only one study recorded dashboard engagement and showed a difference of 10% (95% CI 5% to 15%) between intervention and control. None of the studies reported on the development, pilot or implementation of dashboards or the involvement of stakeholders in design and testing. Interactive dashboards may reduce AP in primary care but most likely only when combined with other educational or behavioural intervention strategies.

Background: Social media can be used to quickly disseminate focused public health messages, increasing message reach and interaction with the public. Social media can also be an indicator of people’s emotions and concerns. Social media data text mining can be used for disease forecasting and understanding public awareness of health-related concerns. Limited studies explore the impact of type, sentiment and source of tweets on engagement. Thus, it is crucial to research how the general public reacts to various kinds of messages from different sources. Objective: The objective of this paper was to determine the association between message type, user (source) and sentiment of tweets and public engagement during the COVID-19 pandemic. Methods: For this study, 867,485 tweets were extracted from January 1, 2020 to March 31, 2022 from Ireland and the United Kingdom. A 4-step analytical process was undertaken, encompassing sentiment analysis, bio-classification (user), message classification and statistical analysis. A combination of manual content analysis with abductive coding and machine learning models were used to categorize sentiment, user category and message type for every tweet. A zero-inflated negative binomial model was applied to explore the most engaging content mix. Results: Our analysis resulted in 12 user categories, 6 message categories, and 3 sentiment classes. Personal stories and positive messages have the most engagement, even though not for every user group; known persons and influencers have the most engagement with humorous tweets. Health professionals receive more engagement with advocacy, personal stories/statements and humor-based tweets. Health institutes observe higher engagement with advocacy, personal stories/statements, and tweets with a positive sentiment. Personal stories/statements are not the most often tweeted category (22%) but have the highest engagement (27%). Messages centered on shock/disgust/fear-based (32%) have a 21% engagement. The frequency of informative/educational communications is high (33%) and their engagement is 16%. Advocacy message (8%) receive 9% engagement. Humor and opportunistic messages have engagements of 4% and 0.5% and low frequenciesof 5% and 1%, respectively. This study suggests the optimum mix of message type and sentiment that each user category should use to get more engagement. Conclusions: This study provides comprehensive insight into Twitter (rebranded as X in 2023) users’ responses toward various message type and sources. Our study shows that audience engages with personal stories and positive messages the most. Our findings provide valuable guidance for social media-based public health campaigns in developing messages for maximum engagement.

Objective: Collaborate, Analyse, Research and Audit (CARA) project set out to provide an infrastructure to enable Irish general practitioners (GPs) to use their routinely collected patient management software (PMS) data to better understand their patient population, disease management and prescribing through data dashboards. This paper explains the design and development of the CARA infrastructure. Methods: The first exemplar dashboard was developed with GPs and focused on antibiotic prescribing to develop and showcase the proposed infrastructure. The data integration process involved extracting, loading and transforming de-identified patient data into data models which connect to the interactive dashboards for GPs to visualise, compare and audit their data. Results: The architecture of the CARA infrastructure includes two main sections: extract, load and transform process (ELT, de-identified patient data into data models) and a Representational State Transfer Application Programming Interface (REST API) (which provides the security barrier between the data models and their visualisation on the CARA dashboard). CARAconnect was created to facilitate the extraction and de-identification of patient data from the practice database. Discussion: The CARA infrastructure allows seamless connectivity with and compatibility with the main PMS in Irish general practice and provides a reproducible template to access and visualise patient data. CARA includes two dashboards, a practice overview and a topic-specific dashboard (example focused on antibiotic prescribing), which includes an audit tool, filters (within practice) and between-practice comparisons. Conclusion: CARA supports evidence-based decision-making by providing GPs with valuable insights through interactive data dashboards to optimise patient care, identify potential areas for improvement and benchmark their performance against other practices. Supplementary file 1. Graphical abstract

Background/objectives: Antimicrobial resistance (AMR) is a major worldwide concern with severe implications for public health, contributing to almost 5 million deaths in 2019. One of the main causes of AMR is overuse and misuse of antibiotics, which can be addressed by increasing awareness and educating the public on this issue. Studies have demonstrated the potential of social media to educate the public and influence behaviour. Instagram's unique features, such as its visual nature and user-friendly interface, make it well-suited for exploring health behaviours and disseminating information on various health topics. Research shows 65.4% of young adults (18–36 years old) use Instagram as their main source of information.Methods: This study examined AMR posts from Instagram from January 1, 2017 to July 15, 2023. In total, 4,606 images and videos were initially extracted which corresponded to 3,261 Instagram posts. After data cleaning, a final dataset of 574 posts were categorised into 6 message categories which were humour, shock/disgust/fear, personal stories/statements, educational/informative, opportunistic and advocacy.Results: The most common post category was educational/informative (78%) and humour was the least common (2%). We also looked at the average engagement (likes) with these posts, the educational/informative category received the most likes per post (mean of 30). The fear/shock/disgust category received 25 likes per post, humour and personal stories/statements 18 and 21 respectively.Conclusions: Our study shows Instagram has hardly been used for AMR interventions. An important population group of young adults who use Instagram as their main source of information, is missed in public health messaging on AMR.