A 3-UK-nation survey on dementia and the cost of living crisis: contributions of gender and ethnicity on struggling to pay for social care

Abstract Objectives The aim of this 3-UK-nation online survey was to explore the impact of the cost of living crisis on the lives of people with dementia and their carers, specifically on their ability to access social care and support services, and the contributions of gender and ethnic background. Methods A 3-UK-nation (England, Wales, Northern Ireland) 31-item online survey was conducted in October 2022 asking people with dementia, carers, and people knowing but not caring for someone with dementia about social care and support service access, cost of living crisis, and changes due to the cost of living crisis. Frequency analysis and Chi-square analysis were employed to assess whether forms of payment for services varied by gender. Pearson correlation analysis and binary logistic regression were used to assess whether gender and ethnicity were associated with struggling to pay for care since the crisis. Results A total of 1,095 people with dementia, unpaid carers, and people who knew but not cared for someone with dementia participated. Of those, 745 people with dementia were utilising community-based social care and support services. Twenty percent of those with complete data had reduced their spending on care services since the crisis. Men and those from non-white ethnic backgrounds were at significantly increased odds of struggling to pay for care services. Conclusions The cost of living crisis has led to exacerbated inequalities in accessing and using dementia care. Men and those from non-white ethnic backgrounds in particular need to receive greater support in accessing care.


Introduction
The number of people living with dementia continuously rises, predicted to reach 1 million in the UK in 2024 (Wittenberg et al., 2019), with over 55 million people estimated to live with the condition worldwide (ADI, 2022).With pre-pandemic dementia care costs estimated at £34.7 billion a year across the UK (Wittenberg et al., 2019), social care (publicly and privately funded) and unpaid care provided by family and friends represent the largest proportions of the costs, with 45% (£15.7 billion) and 40% (£13.9 billion), respectively.
Accessing and utilising social care and social support services prior to the pandemic and the cost of living crisis has been difficult for many people with dementia and their carers (Giebel et al., 2021a;Stephan et al., 2018;Watson et al., 2021), regardless of whether in high-income or lower-and middle-income countries.In the latter, people face additional barriers of high levels of stigma, poor recognition of dementia, as well as limited resources (Nguyen et al., 2023).Using day care, respite care, support groups, social activities, or paid home care however is important to support the mental and physical well-being of the person, and their unpaid carer (Finnanger-Garshol et al., 2022;Tretteteig et al., 2016).Utilising these community-based care services after a diagnosis can also delay care home entry, as lack of access to services and wider cognitive, physical, and social stimulation has been linked to faster deterioration and care home entry in a qualitative study into COVID-19 (Giebel et al., 2022).
Considering that many people with dementia, and their carers, have to pay for social care and support services either fully or partially themselves where services are known of and available (Giebel et al., 2021b), with commissioning of care services differing across areas (Davies et al., 2020), the cost of living crisis is likely to impact negatively on their utilisation.Those who have to fund care themselves either to top up care or by not qualifying for financial support despite coming from middle/lower socio-economic backgrounds are likely to struggle financing social care and may instead have to pay for other basic necessities.With a scarcity of existing research on the impact of the current cost of living crisis on the wider population, but evidence highlighting the link between socio-economic deprivation and poorer health and mental health outcomes, including in older adults (Fernandez-Nino et al., 2014;McCann et al., 2018), the cost of living crisis is likely causing severe implications on people living with dementia which are yet unknown.
A recent evaluation of primary care data reported increased incidence of dementia diagnosis in black women and men and reduced incidence in Asian women and men compared to people from a white ethnic background (Pham et al., 2018).Moreover, people from black ethnic backgrounds in the UK have a 22% higher likelihood to develop dementia compared to white people, with black and South Asian people with dementia dying sooner after the dementia diagnosis compared to white people (Mukadam et al., 2022).This shows how people from minority ethnic backgrounds can experience greater disadvantages in terms of dementia, which is amplified in the myriad of barriers which they experience in relation to accessing and using dementia care services.People from minority ethnic backgrounds are less likely to access dementia care services due to cultural stigma, whilst services are often not adapted to cultural (dietary, religious, cultural norms) and linguistic needs (Brijnath et al., 2022;Nielsen et al., 2020).There has been no evidence on the impact of the cost of living crisis on people with dementia from minority ethnic backgrounds and their ability to access social care and support services, to date.
The aim of this study across three UK nations (England, Wales, Northern Ireland) was to explore the impact of the cost of living crisis on the lives of people with dementia and carers, specifically on their ability to access social support and social care services, and the contributions of gender and ethnic background.We hypothesised that people with dementia would experience increased barriers in accessing and paying for social care and support since the cost of living crisis, and that people from minority ethnic backgrounds would face greater struggles than those from white ethnic backgrounds.No hypothesis was made about the contributions of gender based on lack of previous evidence.This is the first study to date to focus on this topic, and can provide important insights to inform Governmental funding support for some of the most vulnerable members of society in trying to access care.

Participants and recruitment
Eligible participants included anyone living with dementia, those knowing and caring for someone with dementia, and those knowing but nor directly caring for someone with dementia living in England, Wales, or Northern Ireland.People with dementia, carers, and those knowing but not caring for someone in Scotland were not included as the Alzheimer's Society does not cover Scotland, only Alzheimer's Scotland does.Inclusion criteria for knowing someone with dementia included present or former unpaid carer, healthcare professional who works or has previously worked with someone living with dementia, or an individual who currently holds a relationship to someone living with dementia but does not care for them.Quality control measures were also asked at the start of the survey to screen participants for accessibility eligibility.
Participants were identified through the Yonder Data Solutions Panel, a fieldwork and data collection company.The recruitment advertisement was put out to eligible participants on the panel (data are gathered on panel participants upon entry) based on the database, with a £2 incentive upon survey completion.Then the inclusion criteria from the survey questions would have filtered out respondents who are no longer associated with dementia (for example if the database is not up to date).Yonder Data Solutions also sought out at least 100 participants from each region in England, Wales, and Northern Ireland to ensure generalisability of the sample.
Yonder Data Solutions did not check for the mental capacity of people living with dementia specifically but included quality control measures within their survey where participants' answers would be excluded if they did not meet these.In addition, only people with dementia with mental capacity would be able to complete the survey remotely.

Ethical considerations
No ethical approval was sought as the Alzheimer's Society commissioned the survey company to collect the data within GDPR parameters, and was only provided with the anonymised data with anonymised IDs for each participant, with data not including any personal identifiable information.The research team only received anonymised secondary data, with full identifiable data stored with Yonder Data Solutions.Informed consent was taken at the beginning of the survey.Specifically, at the start of the survey, participants were shown a screen as required by GDPR to expressively ask permission from respondents to ask them about any sensitive issues.For mental health related questions for example, participants could opt to either answer or not answer those questions.

Data collection
Survey data were collected online via Yonder Data Solutions from the 21 st to the 30 th of October 2022 via an online link in an email to eligible participants.The survey was co-designed by the Alzheimer's Society, the University of liverpool, the NIHR Applied Research Collaboration North West Coast, and Dementia Voice (Alzheimer's Society volunteer network of people affected by dementia).The 31-item survey included basic demographic questions (age range, gender, ethnicity, dementia subtype, living situation, relationship to person with dementia (if carer), home ownership), as well as questions on social care and support service usage (specifically paid home care, support groups, day care, and respite care) and how these were funded, experiences and noted changes as a result of the cost of living crisis, and socialising and visitation).All questions focused on the characteristics and experiences of the person with dementia (except 'relationship to person with dementia').Thus, questions were either self-reported by people with dementia completing the survey, or proxy reports by carers or those knowing someone with dementia.Each survey question had different items, and a full list of the survey questions is attached in Appendix I.
Personal email addresses are held by the Yonder Data Solutions Panel through consent to take part in relevant surveys.

Data analysis
Data were analysed using SPSS Version 28.Demographic characteristics and all outcome variables were analysed using frequency analysis.A Chi-square analysis was conducted to assess whether forms of payment for services (i.e.self-funded, local authority, or mixed) varied by gender.Bivariate correlation analysis was employed to analysis the association between gender, ethnicity, and dementia subtype with the ability to struggle to pay for social care and social support (Q9).Variables found to be significantly associated with struggling to pay for care were entered into a binary logistic regression model, to explore significant associations and odds ratios, with Q9 (struggling to pay for social care and social support) as outcome variable.Responses on Q9 for 'struggling to pay for social care' and 'struggling to pay for social support' were merged, so that someone either struggled with one of these, or both.

Missing data
Some responses for Q2 were entered for different categories which are self-excluding (i.e.person with dementia and carer, or relative caring for someone with dementia and person knowing someone but not caring for them), so these had to be counted as missing data with cases excluded (n = 71).This resulted in a total sample size of 1,095.For Q4, where participants only ticked young-onset dementia, they were classed as 'not sure' , as they may have had Alzheimer's Disease dementia or different subtypes of fronto-temporal dementia for example (which were categorised) or other subtypes of dementia.Thus, these cases were categorised as 'not sure' .This is because young-onset dementia is not a subtype but an age-specific categorisation of dementia, contraire to late-onset dementia (diagnosis aged 65+).

Sample characteristics
A total of 1,095 people with dementia (n = 17), carers (n = 256), and people who know someone with dementia in a personal or professional capacity but are not providing care (n = 822) participated in the survey.The findings refer solely to the person with dementia (either via self-report or proxy report).The majority of people with dementia was female (n = 646, 59.0%), white/ white British (n = 991, 90.5%), and aged 65 or over (n = 951, 8.2%).The most common subtype of dementia was Alzheimer's Disease dementia (n = 451; 41.2%), followed by mixed (15.5%),vascular (13.9%), lewy body dementia (2.3%), Fronto-temporal dementia (1.2%), and other (1.6%).One quarter of participants was unsure of the precise subtype diagnosis (24.3%).A small proportion of people with dementia (5.8%) was living with young-onset dementia.The majority of people with dementia resided in their own home (52.4%), with over a quarter of people with dementia residing in a care home.People with dementia lived across different regions in England, as well as in Wales and Northern Ireland, with the majority of participants from across England (n = 888, 81.1%).Table 1 outlines the sample characteristics.
Details about changes in service use costs were also elicited from respondents and are shown in Table 2. Of those who answered this question (n = 640), 46% reported some increase in costs, 38% were unsure and 16% reported no changes.
Most participants (73%) fully self-funded their service usage, 10% relied on local Authority funding, while 9% reported partially funding their service usage from a mixture of self and other sources of funding.Pearson Chi-square test revealed no significant variations in how service were paid for between women and men [X 2 (5,640) =0.299].
Two thirds (68%) of respondents experienced increased cost of living in many aspects of their lives, including rent/mortgage, utility bills, fuel, food, taxes, health and social care, clothing, entertainment, and transportation.Of those accessing social care and support services and who responded (n = 518) to the question on whether they struggled to pay for various items, 16% (n = 82) reported difficulties in paying for social care services in the last month.

Associations between cost of living crisis, social background and social care and support service usage
Participants were also asked specifically about whether they had reduced their usage of social care and support services as a result of increased costs.While only 209 participants of the 745 who utilised services responded to the question, one fifth of them reported that they had reduced service usage to save money.
Bivariate correlation analysis showed that gender [r(518) = −.110,p < .05]and ethnicity [r(518)= −.101, p < .05]were significantly negatively associated with struggling to pay for social care and support.Specifically, women and people with dementia from white ethnic backgrounds were found to struggle less than men and those from other ethnic backgrounds.
Binary logistic regression analysis showed how both gender (p < .05)and ethnicity (p < .05)were significantly associated with struggling to pay for social care and support.Men and those from non-white ethnic backgrounds were at greater odds of struggling to pay for social care and support with odds ratios below 1 for both gender and ethnicity, and 'male' and 'non-white ethnic background' coded as '0′, compared to 'female' and 'white ethnic background' coded as '1′ (see Table 3).

Discussion
This is the first study to report on the impact of the cost of living crisis on people with dementia, showing the negative effect on their ability to finance and utilise community-based social care and social support services, thereby confirming our hypothesis.Specifically, men and those from non-white ethnic backgrounds were at increased odds of struggling to pay for social care and support services as a result of the cost of living crisis.Findings of this 3-UK-nation survey extend previously experienced inequalities for people from minority ethnic backgrounds in accessing dementia care and receiving a diagnosis in the first place (Nielsen et al., 2020;Parveen et al., 2017).In addition to existing barriers, the cost of living crisis has exacerbated access issues to necessary care and is thus creating further inequalities.External support services are already under-utilised by people with dementia from diverse ethnic backgrounds, with many cultures tending to care for their relatives solely or primarily within their families (Brijnath et al., 2022).This is also based on cultural stigma surrounding the diagnosis and wider mental health (Parveen et al., 2017), so that removing access to the already limited used services will create further caring duties for family carers, most of whom receive no support themselves for their caring role.The lack of carer support is not unique to different ethnic backgrounds though, and commonplace across the unpaid carer population, leading to increased unmet needs (Zwingmann et al., 2019).To ensure that people with dementia from any ethnic background continue to and increase their access to social care and support within their community, based on the evidence from this survey, further financial subsidies need to be in place as otherwise many will have to continue choosing basic necessities such as food and heating over care.
Men were also found to be more disadvantaged than women with dementia in terms of struggling to pay for care services since the cost of living crisis.Comparing the sources of funding for social care and support services, there were no significant variations between men and women, so that men were not more likely to pay for care themselves for example.Similarly, we did not collect any income or wealth data on participants, so are unable to compare the economic background of people with dementia by gender.Differences in financial backgrounds may explain why men were struggling to a greater extent in funding social care and support services.However, this male sample of people with dementia may have similar average financial background compared to female people with dementia, whilst men may have more diverse outgoings and thus had to cut back on paying for care services.Qualitative follow-up research to contextualise these findings and understand these nuanced variations by gender, and ethnicity, is required as a next step.
Besides ethnicity and gender as key barriers to accessing care, this 3-UK-nation survey clearly indicates the generic difficulties for some people with dementia in accessing care since the cost of living crisis.This is in addition to existing difficulties in accessing care regardless of the current financial and economic situation, and highlights the difficulties for many who are faced with financial barriers to receiving care.This key barrier could be addressed by the initially announced, and then delayed and cancelled increase of support for social care funding for those from more disadvantaged backgrounds (HM Treasury, 2022).With a lack of comparable national and international evidence, future research ought to explore the impact of the continued cost of living crisis and ongoing system barriers within social care for people with dementia by building on this survey and advancing it with more detailed background demographics and qualitative contextual research.This will be of particular interest for lower-and middle-income countries also, where dementia care is more limited than in high-income countries and already fewer people access support (ADI, 2022;Ferri & Jacob, 2017;Jacob, 2017).

Limitations
Whilst this survey benefited from having recruited participants from across three UK nations, involving people from diverse backgrounds, there were some limitations to this externally organised survey.There was no confirmation on dementia diagnosis and subtype, and this was based on personal information from the participants instead.However, this is a common approach in the field, and enables larger sample sizes to be  collected, without necessitating the need to approach memory clinics and the National Health Services for subtype diagnosis.Moreover, the analysis of this study was not focusing on specific dementia subtypes, but focused on a general impact of the cost of living crisis.An additional limitation of the study was the low proportion of people with dementia who participated themselves, and the large proportion of people who knew someone with dementia well, yet were not providing unpaid care to them.This study aimed to provide a timely overview and insights into the situation, so that equal numbers of people with dementia, carers, and non-carers could not be effectively recruited.
However, future research needs to specifically investigate the experiences of people with dementia directly, also by contextualising the survey data via more nuanced qualitative explorations.This may also be the result of only recruiting through and conducting the survey via Yonder Data Solutions, which has members of the general public on their registered panel.
No dementia targeted recruitment approach, such as via Join Dementia Research, was utilised, due to the time-sensitive nature of the survey as well as the outsourcing of data collection to a company.In terms of the survey variables and missing data, whilst nearly 1,100 people participated in the survey, 745 of whom who had experiences of social care and support service usage, of those, over 200 participants had missing data on their experiences of struggling to pay for items, including care services.This is to be expected in a large survey, and did not impact on generating significant results in the correlation and logistic regression modelling.

Conclusions
The cost of living crisis has exacerbated existing difficulties in utilising vital social care and support services for dementia, by hindering more people from accessing the care they need, specifically men and those from non-white ethnic backgrounds.If people with dementia fail to access these services, they are more likely to deteriorate faster and enter residential long-term care, which is less affordable than community-based care and the heaviest cost factor of the dementia trajectory (Wittenberg et al., 2019).In addition, without the social care support to keep well and independent at home, people are more likely to utilise health care services and reach crisis point with their health.After the COVID-19 pandemic, the cost of living crisis generates new barriers and underscores the need for systemic changes and support for people with dementia to access care within their community.Future research needs to expand on these survey findings and explore the impact of the crisis in other countries and settings, especially lower-and middle-income countries.
Appendix I. a. They/I often forget to turn off lights b.They/I often forget to turn off taps c.They/I often forget to close the fridge door d.They/I have special dietary requirements e. They/I find it difficult to control their thermostat or other heating controls f.They/I use the washing machine / dryer regularly due to incontinence g.They/I feel the cold more than they used to h.They/I frequently have the lights on at night, for example, due to insomnia or fall risk i.None of the above j.Not sure 18 Which of the following benefits are you/the person living with dementia currently in receipt of?

Table 1 .
Demographic characteristics of people living with dementia.

Table 2 .
Social care service usage payments.

Table 3 .
Binary logistic regression model on social background and struggling to pay for social care and support.
They/I do not receive any benefits o.The benefits they/I receive are not listed p.Not sure 19 Have you/the person living with dementia ever used a price comparison site?Price comparison websites let you filter and compare products a. Yes, by themselves/myself b.Yes, with some help, for example from a friend or family member c.No, but they/I do use the Internet d.No, they/I do not use the Internet at all e.Not sure Cost of living' refers to everyday costs such as rent / mortgage, utility bills, fuel, food, taxes, health and social care, clothing, entertainment, and transportation.What negative impact, if any, do you think the cost of living has had on the number of times you/the person living with dementia has visited, or been visited by, friends / family in the past three months?