Posters Abstracts 14-23

Persistent delirium in the months following discharge from acute settings has been associated with long-term cognitive impairment. This in turn requires community-based services and support for the individuals and their families. The experience of people who live with persistent delirium and their families is poorly understood. This project aims to explore the experience of individuals with persistent delirium post discharge, including their information and support needs (formal and informal). And examine the health and social care needs of this population and their families, including third sector provisions.

This is a mixed method study gathering qualitative data on the experiences and quantitative data on the outcomes of individuals with persistent delirium. Qualitative data collection from individuals with persistent delirium may require a considerate interviewing approach as communication might be a challenge within this group. Also, thoughtful consideration will be given when interviewing families as this might be a sensitive subject for families to discuss. Quantitative data on the service needs and outcomes of these individuals focusing on hospital re-admission, care home admission and mortality will be collected. Interdisciplinary perspectives from health care, social care and third sector will help give a background when designing the study. This will be explored further during the presentation.

The project aims to make a positive impact on services received by individuals and families with persistent delirium post discharge from acute hospital by creating awareness about the subject amongst clinical and academic colleagues. Also, suggesting and advising staff to improve skills and information so that individuals and families receive the right support when discharged. Understanding the experiences and outcomes will enable policy makers to invest in resources needed to improve care received by individuals and families with persistent delirium. Over all this project aims to make a wider impact on clinical, policy and economic aspects of health and social care services with regards to persistent delirium.

As our population continues to age, it is essential that older adults are provided with high-quality care. But what does this mean? Many argue that because of demographic, economic and institutional pressures, there is a demand for current care practices to be reformed. One way in which this can be done is with the introduction of “care-bots”. The role of these care-bots would be to either assist or replace human care-givers. If we are to introduce robots into environments where they will be interacting with vulnerable older adults, we have a moral duty to ensure they are ethical. But what does this entail?

This study endeavours to formulate an ethical framework which can be used at each step of the process by stakeholders involved in the creation and use of robots in care. This includes but is not limited to: developers; designers; computer scientists; care providers, carers and care-recipients.

This ethical framework will attempt to predict key ethical challenges which may arise when robots are used to assist older people in the community. It will explore and attempt to answer important questions such as:

• Will the introduction of robots help facilitate autonomy and independence or take it away?
• How will older people feel about the introduction of robots in their homes? Will this interaction foster trust or will older adults feel their privacy is being diminished?
• Will interactions with robots add or deduct value from the lives of older adults?

Answering the above questions requires an understanding of the lived experience of older people. How they view the world and their perceptions of both care and robots, provides invaluable insight into the considerations needed to construct an ethical framework for human-robot interaction in care environments.

As such, this study plans to draw from the disciplines of Philosophy and Ethics, Anthropology and Design Informatics to gain further insight into the ethical challenges relating to human-robot interaction. The methodology for this study will use qualitative research methods as well as an ethnography, which will involve carrying out interviews and observational studies with older people. These methods will not only allow myself to gain a deeper understanding of the lived experience of older people and their perceptions of robots, but they will also be essential as part of an ethnography, to allow myself as the researcher, to reflect upon and question my own preconceptions and assumptions regarding ethical interactions between older people and robots.

Where we live significantly impacts our health. Neighbourhoods are both physical places and socially constructed environments, shaped by socioeconomic factors, built environments, and social dynamics. These factors can profoundly influence health outcomes and care needs for older people. Existing research indicates that neighbourhood characteristics, such as deprivation, access to greenspace, community participation, and demographic composition, may independently influence health outcomes, but less is available on how these factors interact.

In order to conduct research which attempts to draw together a more holistic view of neighbourhood characteristics and their influence on health outcomes, an interdisciplinary approach is required. How a neighbourhood is defined, and thus what characteristics it has, is necessarily informed by perspective and positionality. For example, where a neighbourhood may be examined by using census data relating to ‘lower layer super output areas’, the lived experience of local residents may not recognise the same boundaries. Moreover, when viewed from a longitudinal perspective, neighbourhoods are fluid; both their characteristics and how they may be defined change over time.

My proposed research project will seek to take account of these multiple perspectives. Using an interdisciplinary approach, the proposed research will triangulate quantitative and qualitative data using a Bayesian modelling framework to analyse and ultimately predict how neighbourhood characteristics influence health outcomes and associated care demands in later life.

Using mortality statistics and census data on health indicators, this project will first identify neighbourhoods in Scotland and the NE of England with health outcomes among older residents that are concordant or in conflict with model predictions based on neighbourhood characteristics; specifically, where inequalities are higher or lower than might be expected given the socio-demographic profile of the area. It will then use a qualitative design involving non-participant observation and focus groups with residents of the selected communities, to explore local intersectionalities that might explain concordance with or deviance from statistical models. Contextual reasons for concordance or conflict will be qualitatively explored with local older residents. Finally, Bayesian methods and longitudinal survey data will be used to accommodate quantitative and qualitative insights to improve model estimates of health outcomes and associated care needs.

By investigating the complex interplay between neighbourhood characteristics and health outcomes, this project has the potential to inform targeted interventions and policy, design healthier neighbourhoods, and ultimately improve the lives of older people navigating the challenges of later life.

Delirium is a clinical syndrome describing an acute change in mental ability marked by disturbed attention, awareness, and cognition. In hospitals, up to half of older adults experience delirium. Despite being mostly temporary and reversible, delirium is a severe mental impairment associated with more difficulties in recovery after hospital discharge, as well as a decline in mental well-being and social functioning.

Recent studies suggest a bi-directional relationship between delirium and sleep disturbances, indicating a potential neurocognitive similarity between these conditions mediated by shared underlying mechanisms. People with delirium can also exhibit disrupted circadian rhythms, which otherwise help regulate the body's natural sleep-wake cycle. Nevertheless, the extent to which intrinsic factors versus environmental factors influence the dynamic interplay between poor sleep and delirium remains uncertain.

This project will adopt an interdisciplinary approach, drawing insights from psychology, medicine, and engineering, to better understand the multifaceted nature of delirium and sleep disturbances in older people. The first aim is to analyse sleep patterns and circadian rhythms in older adults at risk for delirium or experiencing delirium using self-reported and novel non-invasive objective methods. Participant observation will be conducted in a hospital and a care home to evaluate sleep-wake cycles in populations with varying medical complexity and environmental influences.

Further, the second aim is to test the efficacy, safety, and acceptability of an intervention combining bright light therapy and mindfulness techniques to improve sleep in older adults at risk for delirium. This study will also assess the intervention’s impact on reducing experiences of pain and anxiety in this population. Overall, this research seeks to improve the health and well-being of older adults in care settings by exploring the link between delirium and sleep disturbances and striving towards targeted interventions to enhance their experience and outcomes.

Multimorbidity is defined as the co-occurrence of two or more chronic conditions, and poses a significant challenge in healthcare, especially given the complexities introduced by patient heterogeneity. Previous research has successfully investigated the impact of factors such as age, sex, and socio-economic status on the prevalence of multimorbidity and disease clusters. However, understanding the longitudinal patterns of disease trajectories and accumulation still poses a challenge in current state-of-the-art research.

To address this challenge, this study will make use of longitudinal data from Electronic Health Records (EHRs), such as the UK Biobank or the Clinical Practice Research Datalink (CPRD), and advanced modelling techniques. Taking inspiration from treeLA, a methodology combining topic modelling and medical ontology, this work aims to integrate statistical and computational techniques from different disciplines to try and identify disease accumulation trajectories and relationships. Such methods include not only Natural Language Processing (NLP) in genetics, medical ontology and text, but also time-series forecasting methods (such as Long Short-Term Memory networks), Large Language Models, Bayesian Networks, and Bernoulli Mixture Models.

Throughout this project, the integration and application of theoretical knowledge drawn from different domains, such as medicine, computing, and social sciences through collaboration with experts ensures that the conclusions drawn are thorough and comprehensive. This demonstrates the inherent interdisciplinary nature of data science, as it combines insights across various disciplines to advance understanding of specific topics, such as multimorbidity.

Information infrastructures (II) are essential for the functioning of health and social care services, facilitating data exchange amongst the various actors and institutions comprising health/social care systems. II’s are constituted by a complex interweaving of information technologies, governing policies and healthcare practices, interacting to shape how actors – from healthcare staff to service users – access, share and deploy medical information.  

Various initiatives across the UK have attempted to integrate workflows both within and between health (primary, secondary & community) and social care organisations to improve service efficiency, reduce costs and establish personalised models of care. However, transforming existing IIs to accommodate integration efforts is a challenge – often requiring the decoupling of data from its current function. Actors navigating integration must contend with a messy landscape of sometimes incompatible IT systems and standards, embedded work behaviours, overbearing or vague governance directives and conflicting user expectations. Understanding the various tensions between these factors, and how they are interpretated/mediated by IT engineers, healthcare staff and service users, can therefore better inform future integration initiatives and II reform.   

This PhD project is concerned with the interaction between IT systems, healthcare practices, governing rules and user perspectives. It will draw on various disciplines, notably: Science & Technology Studies (STS), Health and Informatics. To understand this interplay, semi-structured interviews will be conducted with: 

1) Healthcare practitioners to gain insight into their day-to-day interactions with the II, and how it impacts their service delivery.   2) Actors involved in the development of the IT infrastructure and their interpretation of the II’s purpose. These may include IT architects, service-delivery managers and information governance specialists. 3) Service users to understand their expectations and experiences with the II.    These data will then be thematically analysed to identify the experiences of those involved in the development and deployment of IIs in integrated care settings. Obtaining rich accounts of these perspectives and detailing the development process itself can, therefore, help inform future integration projects. 

AI technologies are increasingly integrated into people’s homes, with the promise to support ageing and care. However, older people are often seen merely as recipients, rather than active users and makers of these technologies. The central tenet of participatory design is to ensure those affected by the introduction of new technologies have a direct involvement in their design. The complexity of data-driven technologies and the assumed “black box” nature of algorithmic systems make it especially challenging for meaningful involvement of older people in their design process.

This project aims to develop a range of methods to enable older adults to contribute directly to the design of AI-enabled products, making them useful and meaningful to the group.

The project, grounded in pluriverse and feminist theories, will employ a participatory action research methodology. First, it will analyse and catalogue examples of participatory methods for technologies and older people. Secondly, it will use interviews, participant observation, and culture probes to understand the current situation of older people and AI technologies, as well as identify local assets to collaborate in future phases. Thirdly, the project will recruit a diverse group of older people as co-researchers, and co-create a hackerspace equipped with tools, materials, tutorials, and support in the community. Through a series of workshops, the project aims to encourage the older group to learn about, play with, and create new technological tools for themselves. Lastly, the methodology developed through this co-working process will be shared and discussed with other researchers and practitioners.

The direct outcome of this project — the hackerspace — benefits more than its immediate participants. It is a seed in the community that will, through the continuous participation of its members and the technological products created, eventually bear the fruit of a larger vision: the democratisation of technology, by developing a methodology generalisable to other communities to change the position of marginalised groups in technological development.

Machine learning has gained increasing popularity in disease diagnosis and healthcare outcome prediction. However, conventional machine learning models only consider correlation in data, leading to susceptibility to biases associated with variations in environmental deployment and data distribution. Causality, with its inherent stability across varying environments, offers a robust solution to this challenge by focusing on the immutable cause-effect relationships between treatments and outcomes. This project focuses on identifying causal relationships between treatments and outcomes and using such relationships to predict the best care decision. 

This project will start by integrating healthcare data for an exemplar caring decision and use a simple causal predictive model to develop a care application. As simple models cannot scale as the number of information sources increase, non-linear causal models will then be developed. This will require causal structure discovery: finding useful variables and their causal associations. To address this, we will combine representation learning and causality. Qualitative research methods, such as interviews with healthcare professionals, will also be conducted to assess the validity and applicability of the causal models in real-world clinical settings. 

By taking into account the true causes of each treatment outcomes, rather than just statistical correlations, healthcare professionals can make more accurate predictions and develop tailored care plans for individual patients. This ultimately leads to better health outcomes and improved quality of life. By synthesizing interdisciplinary insights from artificial intelligence, data science, healthcare, and sociology, this project underscores the essential role of causality in advancing care and highlights the potential for collaborative innovation across domains.

Healthier working lives (HWL) completed three years’ work in February of this year on retaining, recruiting, and improving the jobs of care workers aged 50 and over working in residential adult social care homes located in semi-rural, and rural locations in Scotland. The team carried out 44 interviews with care home workers, focused ethnographic observation in six homes, followed by creative, coactive and interactive work with a range of 310 workers, managers and relevant organisations. The HWL team are now moving into impact work in other parts of the UK.

Those participating became our intrapreneurs as we also worked closely with entrepreneurs in relevant sectors. As HWL moved into its final six months we formed multi sectoral groups to further develop ideas. One team developed a new Care Designer Role to create activities and initiatives to improve worker wellbeing. Another is exploring improvements to and the portability of training. The remaining innovation teams worked together to develop their initial ideas into clearly defined “problem statements”. Analysis of the ethnographic material has shown low-cost ways of improving retention through small actions to show workers they are valued and suggested the strategy of Retain to Recruit. Further exploration suggested the potential of information and technology to enhance Quality Time to Care. As we move into impact work, we will explore these findings through multi sectoral participation, including domiciliary and residential care workers, local government, businesses in the field and NGOs. We propose to undertake this this work in areas south of London serving ageing populations with a wide range of incomes. Find out more.

Findings from En-Route to Recovery: Diversity and Vulnerability in Care Work During And after The Covid-19 Pandemic - https://www.kcl.ac.uk/research/en-route-to-recovery

The ongoing physical, financial, and psychological impacts of the Covid pandemic exacerbated existing problems for recruitment and retention in social care. These problems are likely to worsen due to the ageing of the current care workforce, with over 1/4 of adult social carers aged over 55. The rising cost of living is also severely impacting care workers, who represent some of the lowest paid workers in the UK – 1 in 4 carers were living in or on the brink of poverty even before the Covid pandemic, causing widespread demoralisation.

While the Covid pandemic was unarguably an extremely difficult time for those in the sector, the conditions were not extraordinary for those who are doing this work every day. The additional challenges caused by the pandemic heightened and worsened existing systemic problems which have plagued the sector for decades and continue to do so. We used in-depth qualitative methodology (including in-depth interviews and sound-sourcing) to explore the experiences and concerns of a wide range of individuals ‘on the front lines’ of the current care crisis - including domiciliary care workers, managers, policy stakeholders, business owners and social care providers.

Here, we present a summary of this qualitative data, which explores first hand experiences and reflections around these emotional, financial, and societal challenges currently being faced in the sector. These narratives – particularly around the value (or lack thereof) accorded to care work and care workers – offer strong, emotive insight into the impact of these various and coalescing factors within individual lives.