Significant outcomes
Carers of people with dementia reported difficulties with their own mental health and observed worsening of behavioural and cognitive symptoms in patients during the first months of lockdown in the UK.
Worsening of behavioural and cognitive symptoms observed in patients was associated with higher frequency of carers’ mental health problems and their concerns about faster disease progression.
Limitations
The small number of participants recruited, especially with rare forms of dementia, limits any generalisation of findings to the whole clinical population of patients with dementia.
All carer-reported variables analysed in this investigation nested in the SOLITUDE study were collected cross-sectionally at baseline and, therefore, provide limited evidence of the long-term impact of lockdown.
Introduction
The social context in which a person with dementia (PWD) lives has an impact on both clinical and quality-of-life outcomes. Several dimensions of social engagement contribute to cognitive reserve by supporting cognitive performance, even in the presence of neurodegenerative pathology (Stern, Reference Stern2012). In fact, older adults with a large social network have been found to have reduced risk of dementia (Fratiglioni et al., Reference Fratiglioni, Wang, Ericsson, Maytan and Winblad2000) and they maintain cognitive performance within the normal range for a longer period of time before showing objective impairment (Wu et al., Reference Wu, Hasselgren, Zettergren, Zetterberg, Blennow, Skoog and Halleröd2020). A recent meta-analysis concluded that social isolation may significantly increase one’s likelihood of developing dementia (Penninkilampi et al., Reference Penninkilampi, Casey, Singh and Brodaty2018). In support of this clinical evidence, investigations into the neurobiological effects of the social environment have found that higher levels of serum brain-derived neurotrophic factor play a mediating role between emotional support gained through social engagement and risk of dementia (Salinas et al., Reference Salinas, Beiser, Himali, Satizabal, Aparicio, Weinstein, Mateen, Berkman, Rosand and Seshadri2017). Moreover, patients with Alzheimer’s disease (AD) with larger social networks appear to retain better cognitive performance despite high levels of brain amyloid plaques detected at post-mortem (Bennett et al., Reference Bennett, Schneider, Tang, Arnold and Wilson2006).
The social environment can also offer useful supporting resources to caregivers of PWD. A recent study found that carers of PWD believed that social support from family, friends and neighbours had positively contributed to enhancing their resilience (Donnellan et al., Reference Donnellan, Bennett and Soulsby2017). This effect seems mediated by positive social interactions that result in reduction of burden associated with caregiving (Han et al., Reference Han, Jeong, Park, Kim, Lee, Lee, Ryu, Kim, Yoon, Jhoo, Kim, Lee, Lee, Kwak, Kim, Park and Kim2014). Being connected to a supportive social network may be particularly beneficial for carers by limiting their feelings of loneliness (Beeson, Reference Beeson2003). Furthermore, better mental health in caregivers can also have a positive impact on patients’ care and has been shown to predict mortality rates among patients with neurodegenerative diseases leading to dementia (Lwi et al., Reference Lwi, Ford, Casey, Miller and Levenson2017).
The coronavirus disease 2019 (COVID-19) pandemic dictated abrupt, drastic and long-lasting limitations to social contacts, and these were enforced in many countries as an initial effort to contain viral spreading. In particular, older adults were confined to their homes for extended periods soon after COVID-19 was declared a pandemic as a preventative measure, given early evidence of them being at higher risk of mortality(Wynants et al., Reference Wynants, Van Calster, Collins, Riley, Heinze, Schuit, Bonten, Damen, Debray, De Vos, Dhiman, Haller, Harhay, Henckaerts, Kreuzberger, Lohmann, Luijken, Ma, Andaur Navarro, Reitsma, Sergeant, Shi, Skoetz, Smits, Snell, Sperrin, Spijker, Steyerberg, Takada, van Kuijk, van Royen, Wallisch, Hooft, Moons and van Smeden2020). Moreover, all non-emergency hospital and community services were suspended for several months and many have resumed with alternative consultation formats, limiting face-to-face contact only to special circumstances (Liu et al., Reference Liu, Howard, Banerjee, Comas-Herrera, Goddard, Knapp, Livington, Manthorpe, O’Brien, Paterson, Robinson, Rossor, Rowe, Sharp, Sommerlad, Suárez-González and Burns2021). Older adults with cognitive impairments have been particularly disadvantaged by this pandemic for several reasons: (1) they often have health conditions that confer vulnerability not only to dementia but also to COVID-19 (Hwang et al., Reference Hwang, Kim, Park, Chang and Park2020; Korczyn, Reference Korczyn2020) and its consequences on the nervous system (Manca et al., Reference Manca, De Marco, Ince and Venneri2021a); and (2) they have been unable to access any relational and healthcare support services, leading to worsening of their mental health early on after lockdown enforcement (Giebel et al., Reference Giebel, Lord, Cooper, Shenton, Cannon, Pulford, Shaw, Gaughan, Tetlow, Butchard, Limbert, Callaghan, Whittington, Rogers, Komuravelli, Rajagopal, Eley, Watkins, Downs, Reilly, Ward, Corcoran, Bennett and Gabbay2020; Manca et al., Reference Manca, De Marco and Venneri2020). Similarly, carers of PWD have been unable to access support services, resulting in limited social contact with professionals and peers.
To respond to the above-mentioned concerning changes affecting PWD, the SOcial LImitations Turn Up DEmentia (SOLITUDE) study (Manca et al., Reference Manca, De Marco, Blackburn, Russell, Evans, Kirkland, Amin, Davies, Firth, Raymont, Colston, Collett, Kumar, Balckman, McCoy, Davies, Robertson, van Diepen and Venneri2021b) was designed with the aim of addressing the central knowledge gap (Liu et al., Reference Liu, Howard, Banerjee, Comas-Herrera, Goddard, Knapp, Livington, Manthorpe, O’Brien, Paterson, Robinson, Rossor, Rowe, Sharp, Sommerlad, Suárez-González and Burns2021) of the long-term impact of social isolation due to the COVID-19 pandemic on cognitive and mental health of PWD and on their carers’ burden in the United Kingdom (UK).
Methods
Clinical sample
Eighty-three participants (38 PWD/carer dyads and 7 carers alone) were recruited for this study between September 2020 and March 2021 from six centres across the UK.
Clinical records were used to identify potential participants meeting the following criteria: (1) a clinical diagnosis of dementia due to a predominantly neurodegenerative aetiology; (2) a Mini-Mental State Examination (MMSE) score ≥18. Exclusion criteria were as follows: major medical diagnoses (other than dementia) of clinical concern with impact on patients’ or carers’ physical and mental wellbeing; a non-neurodegenerative primary cause of dementia; history of life-long psychiatric conditions; history of significant acute neurological events (e.g. stroke, traumatic brain injury); absence of a reliable carer cohabiting with the patient; major sensory or speech impairment that would prevent a telephone conversation; no telephone service in place; insufficient use/comprehension of English.
Study protocol
All potential participants were contacted by telephone by a clinician to explore their interest in receiving information about the SOLITUDE study. Within one week, all people willing to take part in the study provided their informed consent, which was audio-recorded over the telephone. All study procedures were carried out in compliance with the Declaration of Helsinki. The study received ethical approval from the NHS Health Research Authority, North West – Preston Regional Ethics Committee, reference n. 20/NW/0305 (protocol version 1).
The SOLITUDE study (osf.io/9chet) is a multi-centre observational longitudinal study in which participants undergo a total of three telephone-based assessments: at baseline (within a week from informed consent), at 3 months and at 6 months. A series of tests and questionnaires were completed by the PWD and their carers to assess patients’ cognitive performance, mental health and quality of life, and carers’ burden (for the full protocol, see Manca et al., Reference Manca, De Marco, Blackburn, Russell, Evans, Kirkland, Amin, Davies, Firth, Raymont, Colston, Collett, Kumar, Balckman, McCoy, Davies, Robertson, van Diepen and Venneri2021b). At baseline, each carer completed also a customised semi-structured interview with questions on daily-life activities carried out during lockdown and health conditions suffered by them and by the PWD (all questions can be viewed in Table 1). This interview was adapted from one used in previous studies (Cagnin et al., Reference Cagnin, Di Lorenzo, Marra, Bonanni, Cupidi, Laganà, Rubino, Vacca, Provero, Isella, Vanacore, Agosta, Apollonio, Caffarra, Pettenuzzo, Sambati, Quaranta, Guglielmi, Logroscino, Filippi, Tedeschi, Ferrarese, Rainero and Bruni2020; Rainero et al., Reference Rainero, Bruni, Marra, Cagnin, Cupidi, Laganà, Rubino, Vacca, Di Lorenzo, Provero, Isella, Vanacore, Agosta, Apollonio, Caffarra, Bussè, Sambati, Quaranta, Guglielmi, Logroscino, Filippi, Tedeschi and Ferrarese2021).
Table 1. Results of the carers’ interview on life conditions and changes in patients’ symptoms during lockdown
* Mean ± SD.
The aim of this study was to report the findings from the two sections of this interview: (1) carers’ responses to questions on their own demographic and social characteristics, life conditions during lockdown, support received and mental health; and (2) carers’ responses to questions on the PWD and, specifically, about their functional independence before lockdown, changes in existing and/or emergence of new neuropsychiatric symptoms, changes in motor and cognitive abilities and concerns about progression of dementia.
Statistical analyses
Demographic and clinical characteristics, and interview results were summarised to provide an overview of the sample and of the impact of lockdown on participants of the SOLITUDE study.
Subsequently, analyses were carried out to assess the relationship between subjective and objective descriptors that emerged from the interview and symptoms experienced by PWD and carers. First, chi-square and Fisher’s exact tests were run to explore the impact of the support received and that of changes in PWD symptoms on carers’ mental health, carers’ impression of faster disease progression during lockdown, carers’ worries about the consequences of lockdown on PWD and changes in their relationship with PWD. Second, we investigated whether a series of objective variables linked to social isolation (i.e. time spent in lockdown, calculated in days from 23rd March 2020 to the date of baseline assessment; living alone/with carer; frequency of socialising before lockdown) significantly predicted changes in PWD’s symptoms using logistic regression models and including the following covariates: last MMSE score available before recruitment, time between last MMSE assessment and baseline (in days), age and years of educational attainment. We also investigated the impact of these predictors on four primary outcomes (coded as binary variables: changes vs no changes), that is, any changes in behavioural symptoms, emergence of any new behavioural symptoms, any changes in motor abilities and any changes in cognitive symptoms. Exploratory analyses were subsequently carried out for each individual symptom (either existing or new) associated with the four primary outcomes. Bonferroni’s correction was applied to the significance p-value = 0.05 to account for multiple comparisons in each set of analysis.
Results
The majority of patients had a clinical working diagnosis of AD (34/45), including one case of early-onset AD, one case of logopenic primary progressive aphasia and two cases of frontal variant AD. Additionally, the sample included five cases with a working diagnosis of dementia due to mixed aetiology, three with dementia with Lewy bodies, two with posterior cortical atrophy and one case of corticobasal degeneration. All participants were of White British ethnic background and native English speakers. Patients were on average 70.04 ± 9.33 year old and had 12.96 ± 3.01 years of education. Twenty-five out of 45 were women, and the majority were right-handed (42/45). The average pre-recruitment MMSE score of the sample was 23.94 ± 3.37 that had been obtained 202.62 ± 166.69 days before baseline assessment. Patients had spent an average period of 259.62 ± 71.28 days in lockdown, and most of them were married and lived with their spouse (37/45). Most carers were women (27/45) and spouses or partners (38/45) of patients. The carers’ group had an average of 69.24 ± 10.23 years of age and 13.67 ± 2.99 years of education.
Carers’ life conditions and the perceived impact of lockdown are reported in part A of Table 1. Most PWD lived only with their carer, and, overall, they had been safe during lockdown, with just one couple reporting having accessed emergency services. None of the participants had tested positive for COVID-19 before recruitment. However, the majority of carers were worried about the consequences of the pandemic on PWD and would have welcomed more information and deployment of dedicated remote support. Over 60% of carers received help during lockdown, mostly from other family members.
Most carers (67%) observed no changes in their relationship with the PWD and only 15% reported a worsening. Over 40% also reported experiencing problems with their own mental health as a consequence of social isolation, especially anxiety and irritability.
The interview revealed that the vast majority of the PWD had an active social life before lockdown was enforced in the UK, and all but one were aware of the COVID-19 pandemic (part B of Table 1). Half of the carers reported worsening of previously existing behavioural/neuropsychiatric symptoms in the PWD, especially apathy and mood problems, that were reported for about 25% of PWD.
Onset of novel behavioural symptoms was reported by 25% of the carers. Incidence of individual symptoms was low, but apathy and irritability were the most common symptoms. Only one patient required pharmacological therapy to treat new neuropsychiatric symptoms. A third of the PWD also showed some decline in motor abilities, especially slowing of walking pace. Changes in cognition were very frequent (73% of participants), mainly worsening of general confusion and memory decline in over 50% of PWD.
Moreover, worsening of PWD’s behavioural and cognitive symptoms was associated with carers’ perception of faster disease progression during lockdown (Table 2). Other associations, such as that between carers’ anxiety and help received from family members, did not survive correction for multiple comparisons.
Table 2. Associations between carer-reported symptoms of PWD during lockdown and carers’ mental health and concerns (Chi-square or Fisher’s exact test and associated p-values)
* Fisher’s exact test significance.
† p < 0.05, not significant after correction for multiple comparisons.
‡ Significant association after correction for multiple comparisons.
Exploratory analyses showed several significant associations between changes in specific symptoms of PWD and carers’ mental health: (1) worse agitation with carers’ distress (Fisher’s exact test p = 0.003); (2) worse depression (Fisher’s exact test p = 0.004) and disorientation in time (Fisher’s exact test p = 0.006) with carers’ depression; (3) increased confusion with carers’ impression that the disease had progressed more quickly (χ 2 = 13.542, p = 2.33 × 10−4); (4) worse sleep with changes in relationship with PWD (χ2 = 16.724, p = 2.34 × 10−4), although both worsening and improvements in the relationship with PWD were reported by carers who observed sleep changes.
Logistic regression analyses found no significant association between the three objective predictors (i.e. time spent in lockdown, living alone and frequency of socialising) and changes in patients’ symptoms. Associations between age and changes in motor and cognitive abilities did not survive correction for multiple comparisons (Supplementary materials). Similarly, no associations were found with individual symptoms.
Discussion
Our study has identified that carers of PWD, while putting governmental guidelines into practise to maximise their safety, experienced symptoms of anxiety and irritability, possibly due to social isolation during lockdown. Changes in PWD’s behavioural and cognitive symptoms were also observed, especially worsening of apathy, mood, confusion and memory problems. Such changes were associated with worsening of carers’ mental health and their perception of a faster disease progression in PWD since lockdown enforcement. However, it is possible that these carer-reported changes were of mild severity, especially the appearance of new behavioural symptoms, given that only one patient out of 45 required additional pharmacological treatment. Demographic characteristics, time spent in lockdown and pre-lockdown lifestyle did not predict symptom changes in PWD.
These findings highlight that COVID-19-related social limitations have had an impact on PWD’s and carers’ mental health in line with observations made in Italy, a country affected by the pandemic in its earlier stage to a similar level as the UK (Rainero et al., Reference Rainero, Bruni, Marra, Cagnin, Cupidi, Laganà, Rubino, Vacca, Di Lorenzo, Provero, Isella, Vanacore, Agosta, Apollonio, Caffarra, Bussè, Sambati, Quaranta, Guglielmi, Logroscino, Filippi, Tedeschi and Ferrarese2021). In fact, a qualitative study showed that carers of PWD reported higher stress, loneliness and concerns about care for PWD (Tam et al., Reference Tam, Dosso and Robillard2021). Closures of general practices and hospital services appeared to have added to the sense of responsibility of carers (Tam et al., Reference Tam, Dosso and Robillard2021). In fact, the relationship between PWD and their carers, who made crucial daily-life adjustments to limit COVD-19-related risks, emerged as a pivotal factor contributing to patients’ wellbeing during lockdown (Tuijt et al., Reference Tuijt, Frost, Wilcock, Robinson, Manthorpe, Rait and Walters2021). Moreover, the stress stemming from unprecedented life changes and from worsening of patients’ symptoms exacerbated burden and impacted on mental health of some carers (Cagnin et al., Reference Cagnin, Di Lorenzo, Marra, Bonanni, Cupidi, Laganà, Rubino, Vacca, Provero, Isella, Vanacore, Agosta, Apollonio, Caffarra, Pettenuzzo, Sambati, Quaranta, Guglielmi, Logroscino, Filippi, Tedeschi, Ferrarese, Rainero and Bruni2020; Pongan et al., Reference Pongan, Dorey, Borg, Getenet, Bachelet, Lourioux, Laurent, Rey and Rouch2021).
The SOLITUDE study provides additional evidence on the impact that the UK lockdown has had on PWD and on how worsening of patients’ symptoms has added to the burden of their carers. Indeed, it is likely that worsening of patients’ behavioural and cognitive symptoms may have had a detrimental impact on their carers’ mental health, although the directionality of this relationship was not specifically tested in our analyses and we cannot exclude that carers with a tendency to experience mental health problems were more likely to report changes in patients’ symptoms.
Despite the small sample size, these findings are consistent with those observed in other countries and highlight the importance of addressing the consequences of prolonged social isolation on PWD and their carers. The small number of patients with rare forms of dementia prevented any sub-group analyses; however, Zucca et al. (Reference Zucca, Isella, Di Lorenzo, Marra, Cagnin, Cupidi, Bonanni, Laganà, Rubino, Vanacore, Agosta, Caffarra, Sambati, Quaranta, Guglielmi, Apollonio, Logroscino, Filippi, Tedeschi, Ferrarese, Rainero and Bruni2021) found that carer’s burden and distress did not differ depending on the aetiology, but were rather affected by severity of cognitive impairment, a theme that also emerged from our data.
In conclusion, the first lockdown due to the COVID-19 pandemic appears to have had a detrimental impact especially on mental health of PWD and their carers. Although carer-reported data might not be as reliable as those collected by trained clinical researchers due to possible lack of objectivity, this study and previous investigations consistently showed that some subjectively reported symptoms, especially mood alterations, were common during lockdown among both PWD and carers. Future investigations will be needed to assess objectively the potential long-lasting consequences of social isolation on cognitive decline and to provide insights on how to intervene early to mitigate any negative impact. Indeed, evidence from our study, and from similar investigations across the world, should provide insight to policymakers on how to support PWD during this and future pandemics requiring social isolation.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/neu.2022.12.
Acknowledgements
The authors thank Dr Kirsty Harkness, Dr Aijaz Khan, Ms Zoe Collett, Ms Claire Firth, Ms Lynn Davies, Ms Selina Roberts, Dr Erik van Diepen, Ms Shani McCoy, Ms Kimberley Evans and Mr Jason Cook for supporting the identification of potential participants and the recruitment process.
Authors contributions
RM contributed to the study inception and participant recruitment, collected (part of), analysed and interpreted the data, drafted, revised and approved the final version of the manuscript for submission. MDM conceived the study design, contributed to data interpretation, revised and approved the manuscript for submission. AC, VR, JA, RD, PK, GR and DJB led site-specific recruitment and data collection, revised and approved the manuscript for submission. AV conceived the study, contributed to recruitment and data interpretation, revised and approved the manuscript for submission.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of interest
No conflict of interest to declare.
Ethical standards
The authors declare that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Open access
