Antimicrobial resistance (AMR) is a global public health challenge. Due to overuse of antibiotics, bacteria are becoming resistant to antibiotics, which creates infections that can't be treated by modern medicine. To combat AMR, the public need to both know that AMR is an issue and know what actions they can take to address AMR.

When patients are prescribed antibiotics, the information provided (and how this is provided) is likely to be influential on whether they are taken correctly, or if behaviours that contribute to antimicrobial resistance (such as not finishing the course) predominate.

This study, a large online experiment involving 7,000 people across the UK, aims explore the influence of what information and how it is provided. The study looks at different ways of presenting information on antibiotic packets-from clear instructions to warnings and even scannable QR codes with extra details.

The main goals are to see if these changes can help participants understand how to take antibiotics exactly as prescribed. The investigators also want to see if it helps participants understand what antibiotic resistance is and their role in preventing it.

By understanding what works best, the investigators hope to find simple, effective ways to help everyone use antibiotics wisely, protecting these vital medicines for the future.

Antimicrobial resistance (AMR), or antibiotic resistance, is a serious global health problem. It happens when bacteria evolve and become strong enough to resist the very antibiotics designed to kill them. This makes common infections much harder, or even impossible, to treat. A major reason for this growing threat is the overuse and misuse of antibiotics.

To fight AMR, it's crucial that the public not only understand that antibiotic resistance is a problem, but also know what steps can be taken to help. A key action is to take antibiotics exactly as prescribed. However, the investigators know that many people aren't entirely sure what "exactly as prescribed" really means. This can involve things like remembering to take the full course, taking it at the right times, and simply taking the medication at all.

To better understand these challenges, the Fleming Initiative conducted focus groups with the UK public in spring 2025. These discussions helped identify the main reasons people might not take antibiotics as directed, such as not understanding what "exactly as prescribed" means, or not knowing about AMR and how individual actions might contribute to it.

Using these insights, the Fleming Initiative then held co-design workshops with public participants. Together, several new ways to make it easier for individuals to take antibiotics correctly were developed.

This study is a seven-armed online behavioral experiment conducted in collaboration between the Fleming Initiative at Imperial College London, and the Behavioural Insights Team. It aims to understand how different presentations of information on antibiotic packaging about desired target behaviors and antimicrobial resistance (AMR) influence the public's correct antibiotic use, attitudes towards AMR, and future behavioral intent.

Primary Objective:

To measure the change in behavioral intent to "take antibiotics exactly as prescribed," using a composite measure of four individual behaviors related to antimicrobial stewardship (AMS).

Secondary Objectives:

To assess comprehension of how to take antibiotics. To assess comprehension of AMR.

Exploratory Objectives:

To evaluate changes in behavioral intent for individual behaviors related to "taking antibiotics exactly as prescribed" (e.g., finishing the course, number per day).

To understand if participants recognise that not taking antibiotics exactly as prescribed contributes to AMR.

To examine the impact of message framing and the mechanism of intervention delivery on behavioral intent to take antibiotics exactly as prescribed.

To assess the impact on perceived personal responsibility regarding AMR.

To determine the proportion of individuals who interacted with a QR code (in relevant arms).

To gauge the impact of interventions on sentiment towards information on the package itself.

To evaluate participants' opinions on whether text message reminders would help participants take antibiotics exactly as prescribed.

Study Design and Participant Recruitment:

The study will recruit 7,000 UK participants, aged 18 and over, through market research panels via the Predictiv online platform, an online survey platform created by the Behavioural Insights Team.

Participants will be randomised 1:1:1:1:1:1:1 into one of seven trial arms: a control arm with a standard antibiotic packet, and six intervention arms. Interventions include QR codes linking to additional information (Arms 2-4) and warning messages on the packaging (Arms 5-7).

Recruitment will use quotas for age, gender, region, and ethnicity to ensure a nationally representative sample.

The participant journey involves accessing the survey, an introduction slide, an attention check, randomisation, baseline measures of AMR understanding, a standardized medical scenario, exposure to their assigned antibiotic packet image (intervention), a series of survey questions, and finally, demographic questions.

Participants will be anonymous, with no personally identifying data transferred to the research team. Participants will receive a small financial incentive (£0.70) managed by the market research panels. The total study duration is anticipated to be around 3 weeks.

Sub-group Analysis:

The primary analysis will be repeated for subgroups based on age, health confidence, and baseline comprehension of AMR to identify differential impacts of the intervention.

Inclusion Criteria:

• Adults aged 18 and over
• United Kingdom residents
• Have passed the attention check at the start of the online survey.

Exclusion Criteria:

• Individuals who are not signed up a market research panel and therefore will not have access to the online survey.
• Those resident outside the UK.
• Those who fail the attention check at the start of the online survey. Individuals aged under 18.

Anonymised participant data will be shared at the time of study publication on an openly available database