ARCH was a multidisciplinary, four Work Package, project involving key stakeholders to improve understanding, and ultimately practice, around infection detection/management and antibiotic use in care homes for older people. WP1 used statistical analysis of anonymised quantitative data, finding wide variation in antibiotic prescribing rates, and informing selection of care homes for WP2/3. WP1 data are held by Health Informatics Centre (HIC), University of Dundee (https://www.dundee.ac.uk/hic/). WP2 and WP3 used anthropology, sociology and behavioural science to investigate individual, socio-cultural and contextual factors influencing antibiotic use, conducting ethnographic observations (61 periods, ~315hrs), interviews (n=101) and surveys (n=76) with care home managers, nurses, carers, senior carers, GPs, advanced nurse practitioners, pharmacists, residents and their relatives, across 7 care homes (survey broader). Observations and interviews highlighted variation in how suspected infections were identified and managed. Identified factors influencing antibiotic prescribing included: limited training around infections; lack of confidence in distinguishing infections from other conditions, and in ‘watchful waiting’; habitual/routine use of urine dip testing; the importance of communication internally and externally; limited handover documentation; duty of care linked to worry about ‘missing something’; antibiotic resistance a distant problem; pressure from family and residents, and; antibiotic stewardship not seen as care home staff’s role. WP4: Behavioural science intervention development frameworks were applied to integrated WP1-3 findings to identify candidate intervention strategies, prioritised by a co-design workshop (20 care home sector participants) into an intervention including: training (videos and online), appointment of Antibiotic Champions, an Assessment Flowchart, a Monitoring Tool, an SBAR (Situation-Background-Assessment-Recommendation) form for communication between care homes and GPs/ANPs, and reminder stickers. The feasibility trial was impacted by COVID-19 but some in-situ testing and online focus groups found that participants liked the intervention and were keen to engage. Implementation challenges included understaffing, protecting time for training, and balance between standardisation and individual contextual adaptation.Bacteria that cause common infections are becoming increasingly resistant to currently available antibiotics. Antibiotics are essential in modern healthcare but their use drives the development of resistance as bacteria develop ways of surviving their effects. Previous research has found that antibiotics are often used when they are not needed, which increases the development of resistance in bacteria. Older people living in care homes are prescribed many more antibiotics than average and as a result often get antibiotic resistant infections later, which are then harder to treat. The amount of antibiotics used in different care homes varies a lot but we don't really know why. Most research on finding ways to safely reduce antibiotic use has been carried out in hospitals or GP surgeries rather than care homes. There is general agreement that antibiotic use in care homes could and should be safely reduced. However, to design effective approaches to reducing antibiotic prescribing for care home residents we need to understand more about how, when and why they get prescribed, from the perspectives of nurses, carers, GPs, and residents and their relatives. Researchers with different areas of expertise will work together in a four-stage project: 1. Measuring patterns of antibiotic use (epidemiology): We will analyse information on antibiotic prescribing and antibiotic resistance for all care home residents in two Scottish health board regions. As well examining the link between antibiotic prescribing and later antibiotic resistance, this will help us better understand how common antibiotic use is, which residents are prescribed, and variation in prescribing between care homes. This information will also be used to invite care homes with different patterns of antibiotic use to participate in the next stages of the project. 2. Understanding how and why antibiotic prescribing happens in different care homes (sociology and social anthropology): We will work with staff, GPs, residents and relatives in eight care homes. We will observe how staff and GPs work together and react to residents being unwell, and how this leads to antibiotic prescribing. We will also interview staff, GPs, residents and relatives to understand how the way the care home and general practices are organised influences the care different residents receive. This will help us identify new approaches to improving antibiotic prescribing decisions. 3. Identifying staff behaviours that could be changed to reduce antibiotic prescribing (health psychology): We will also carry out more focussed interviews and questionnaires to find out specific reasons why carers, nurses and GPs perform certain actions, and investigate what might encourage or discourage these actions. For example, what causes a nurse to phone a GP to request a prescription for a resident? Why might the GP sometimes write a prescription without reviewing the patient first? This will help us identify specific changes in behaviour which could improve antibiotic prescribing decisions. 4. Creating and testing new approaches to changing antibiotic prescribing (intervention development and testing): We will use all the information from the first three phases, and evidence from other situations (e.g. hospital or GP antibiotic studies) to create an intervention that we will try out in a small number of care homes to see if it is acceptable to staff and residents, and whether it is feasible to do alongside normal work. The intervention will include different elements, such as education for care home staff and GPs, and feedback about how many antibiotics are being used in each care home. At the end of the study, we will have created new and useful knowledge about antibiotic prescribing and its consequences in care homes, but we will also have pilot-tested a new intervention. The next stage of our research programme will be testing it in a large-scale trial to see if it 'works'.

Deposited data include: transcripts of interviews, focus groups and a co-design workshop, and; repsonses to questionnaire surveys. Documents supporting data collection are also deposited along with co-designed intervention materials. Participants were care home managers, staff, residents and resident's relatives, plus healthcare professionals (General Practitioners (GPs) and Advanced Nurse Practitioners (ANPs) and pharmacists) involved in antibiotic use in care homes. The co-design workshop in Work Package 4 (WP4) also included representatives from wider organisations that support and/or manage care homes. Care homes for invitation to participate in WP2, WP3 and WP4 were purposively sampled based on data analysis from WP1. Within recruited care homes, individual participants were invited to paricipate through a combination of purposive (representing different staff groups/levels) and opportunistic (who was available within the overall purposive framing). Questionnaire surveys in WP3 were sent to all care homes and general practices in the study regions, not restricted to recruited care homes.