Mental Health of People with Dementia During COVID-19 Pandemic: What Have We Learned from the First Wave?

Participants and data collection

We conducted an anonymous cross-sectional online survey in France from April 15 to June 15, 2020. The inclusion criteria for the study required participants to be a non-professional caregiver of a PWD living at home or in nursing-home and be at least 18 years old. In line with French regulations on health research, no ethics committee approval was required since collected data was anonymous. Participants were recruited from various sources: e-mails were sent to members of the France Alzheimer Association, information about the survey was given during medical phone appointments carried out during the lockdown or in the month following its end.

Questionnaires

Two different versions of the online survey were available, one for caregivers supporting a PWD living at home (home caregivers, HC), the other one for caregivers supporting a PWD living in a nursing home (nursing home caregivers, NHC). Additionally, printed questionnaires were provided to participants who did not have a web connection. The global questionnaire is presented in the Supplementary Material. Only items from this questionnaire exploring clinical issues were analyzed for this article.

The two versions of the survey shared common questions regarding sociodemographic data, dementia etiology, symptoms duration, caregiver and PWD personal and environmental conditions during lockdown, fear regarding COVID-19 for themselves and PWD, caregivers’ mental health (including anxiety, depression, stress, burden, and difficulties to feel well-rested or to practice physical activities).

Additionally, the questionnaire dedicated to HC addressed specific questions about the PWD living conditions, the support provided by the caregiver to the PWD and the intervention of professional caregivers at home.

The questionnaire completed by NHC explored the possibility offered by nursing homes to contact the PWD by phone or video or to visit him/her after the lockdown, when the measures were more flexible. Moreover, the quality of information provided to caregivers by the nursing home staff about the PWD health during the lockdown, was also investigated.

Caregivers’ mental health scales

Self-reported scales were used to assess caregivers’ mental health (including anxiety, depression, stress, and caregivers’ burden). Anxiety was assessed with the GAD-7 scale (The General Anxiety Disorder-7 [8]). This scale consists of 7 items measuring the presence and severity of anxiety symptoms, specifically linked to the DSM-IV criteria. Each of the 7 items is scored from 0 to 3, and the GAD-7 scale score ranges from 0 to 21. A score from 5 to 9 can be interpreted as mild anxiety, 10 to 14 as moderate anxiety, and 15 to 21 as severe anxiety. A score below 5 means no anxiety. A cut-off of 10 is recommended for identifying cases of generalized anxiety [8].

Depressive symptoms were assessed with the CES-D scale (Center for Epidemiologic Studies- Depression [9]), which contains 20 items assessing the subject’s mood by evaluating the experienced symptoms or behaviors associated with depression. The frequency of symptoms onset during the past week has been measured using a 4-point Likert scale. The highest score is 60 points. A cut-off of 20 is recommended to identify major depressive disorder [10].

The caregiver’s burden was assessed using the validated short version of the Zarit Burden Interview [11]. The continuous score ranged from 0 (no burden) to 7 (highest burden). This score is the sum of the answers to 7 questions to which the caregivers answered “never” (0), “sometimes” (0.5), or “nearly always” (1). A score from 0 to 1 corresponds to “no or slight burden”, 1.5 to 3 to “slight to moderate burden”, 3.5 to 5 to “moderate to severe burden” and 5.5 and over to “severe burden”.

The level of self-rated stress was assessed with a visual analog scale ranging from 0 (no stress at all) to 10 (highest stress you can imagine). This scale was constructed specifically for the purpose of this study.

To provide clinically relevant scores according to literature, each of caregivers’ mental health scores was transformed into binary variables: for the Mini-Zarit scale, a score from 0 to 5 corresponded to “no to moderate burden” and a score from 5.5 to 7 to “severe burden” [11]. Regarding the GAD-7, a score from 0 to 9 corresponded to “no or mild anxiety”, and 10 and over to “moderate or severe anxiety” [8]. For self-rated stress, a score from 0 to 5.9 corresponded to “no or mild stress” and 6 to 10 to “moderate to severe stress”. Finally, a score from 0 to 19 on CES-D scale corresponded to “no major depressive disorder”, and 20 and over “presence of major depressive disorder” [10].

Statistical analysis

Characteristics of caregivers and PWD were described and summarized using mean, SD, median, IQR or patient’s repartition (percentage), and compared between HC and NHC when appropriate, using Chi2 test for categorical variables and Student t-test or Mann-Whitney test for continuous variables, according to Kolmogorov-Smirnov test for normality.

Relationships between caregivers’ status and PWD characteristics and caregivers’ mental health scales, including Mini-Zarit, GAD7, CES-D and self-rated stress (considered as continuous variables) were assessed using Student t-test or Mann-Whitney test when appropriate.

To identify independent predictors of caregivers’ poor mental health for HC, the association between caregivers or PWD characteristics and caregivers’ poor mental health (using binary variables as defined in the ‘Caregivers’ mental health scales’ section) were assessed with multiple logistic regressions adjusted on caregiver and PWD age, sex, number of years since dementia symptoms onset and the fact that the caregiver lives with the PWD.

For each analysis, a p value < 0.05 was considered to be statistically significant. All statistical analyses were performed using SPSS version 21.

Environmental and relational aspects for caregivers and PWD

Caregivers’ mental health during the lockdown

Caregivers’ mental health

A major result of the present study is the strikingly high prevalence of PMH among caregivers of PWD during the lockdown period. Indeed, about half of them had a major depressive disorder or major anxiety, and one third a severe burden. These rates were higher than those reported in general population during COVID-19 pandemic, estimated to be between 20 and 30%for depression [12–14], and from 11 to 30%for anxiety [12, 13]. Outside a context of lockdown, supporting a PWD is already a stressful experience associated with detrimental effects on the caregiver’s psychological health [15] and responsible for a high level of burden [16]. A meta-analysis previously reported 22%of major depressive disorder in such a population [17]. Few studies have addressed caregivers’ mental health during COVID-19 lockdown. Previous findings [18] reported a similar level of anxiety (46%) but a lower depression rate (18%) than our study. However, these authors did not use standardized scales to evaluate mood disorders, the caregivers were only asked if they felt anxious, depressed, isolated or helpless [18]. Using self-administered questionnaires, another study asked 210 caregivers living with PWD if they felt an increase in their psychological or physical burden; possible answers were “not at all”, “some” and “a lot”. In this latter study, 35%of caregivers declared that their psychological burden increased “a lot” while 45%answered “some”.

Another finding needs to be underlined: similar PMH rates were observed in HC and NHC. These results suggest that increased material tasks may explain only a small part of caregivers’ distress. Generally, caregivers report difficulty coping with separation after PWD placement. In addition, visiting restrictions during lockdown may have increased such difficulties. A previous study during COVID 19 pandemic assessed the psychological impact of visiting restrictions on caregivers of PWD living in nursing-homes. A large proportion of respondents reported recent low well-being as well as feeling lonely and isolated: 44%reported well-being scores below 50%; furthermore, 17%of the caregivers reported a poor support from nursing-homes staff [6]. In our study, almost half of the NHC reported difficulties or impossibility to get information about PWD’s health status from the nursing-home staff, certainly contributing to worsen their psychological distress.

Factors associated with caregivers’ poor mental health

Regarding HC group, our results highlighted that providing more help to PWD, living with PWD and feeling isolated were linked to increased anxio-depressive symptomatology and subjective stress and burden. The reduction of the support provided by professional caregivers during lockdown may contribute to increase caregivers’ PMH. Additionally, the lockdown period probably negatively impacted caregivers’ daily life by decreasing intimacy and social life, while increasing a feeling of loss of control, all of this causing emotional burden.

Sleep disorders and difficulties to feel well-rested were associated with a higher level of stress, anxiety and depression. Lockdown, by modifying daily activities and life routines, can bring significant disturbances in biological rhythms and sleep, which can in turn exert detrimental effects on mental health. An Italian study [19] reported that 42.2%of respondents exhibited sleep disorders during lockdown. In addition, reduced physical activity seems to be linked to anxiety-depressive symptoms, stress and burden, as revealed from our results.

In line with previous reports [20], female gender can be identified as a risk factor for higher caregivers’ anxiety, depression, stress or burden. In contrast, caregivers’ age did not contribute to explain PMH risk.

Surprisingly, while old people are known to be at risk of severe forms of the disease, a moderate proportion of caregivers exhibited a high perceived risk of severe form of COVID-19: one fourth for themselves, and half of them for PWDs. These results are in contrast with a previous study reporting that 75%of the caregivers felt that the COVID-19 pandemic was a threat for their own health, whereas 82.1%reported that it was a health threat for PWD [7]. The difference between results may be due to methodological diversity in study design and/or scale to evaluate caregivers’ feeling. Notably, fear of severe infection was not related to caregivers’ PMH in our results (with the exception of anxiety and fear for the PWD). Finally, no association was observed between time since lockdown onset and caregivers’ PMH, both in HC and NHC. These results are in contrast with previous findings, highlighting a positive link between time of isolation and anxiety or depression risk in caregivers [21]. However, our participants’ enrollment began after the first month of lockdown, although the consequences of lockdown on mental health were shown to rapidly worsen after a few days only in previous reports [21, 22]. This may explain such discrepancies.

Strengths

The present study has several strengths. Our study was the first to explore the psychological impact of lockdown on a large sample of caregivers supporting PWD living at home or in nursing-home, using standardized scales. Secondly, while most of the previous studies assessing the impact of lockdown on caregivers’ mental health solely focused on the relationship between BPSD and caregivers’ burden or stress, our study explored the consequences of lockdown on specific caregiver’s mental health symptoms such as depression or anxiety. In this regard, the use of validated scales and clinical cutoffs gives clinical significance to our results. Finally, the inclusion of a large number of potential confounding factors in the multivariate analyses (caregivers’ sex, living or not with PWD and dementia duration) may be highlighted as another strength.

Weaknesses

Our findings also have limitations, some of which are inherent to survey methodology. Indeed, since participants completed the survey on a voluntary basis, only caregivers interested and motivated by the subject of the study responded, which reduces the representativeness of our sample. In addition, no direct measure of PWD cognitive status was available in our study. However, the symptoms duration was used as an indirect measure of dementia evolution. While educational level has been shown to be a protective factor for psychosocial outcome during lockdown in PWD’s caregivers [5], this variable was not collected in our study. Another limitation is the missing information on the severity of dementia, which could influence the mental health of caregivers. The time of onset of symptoms was collected, but it is only partially related to the severity of the disease.

In addition, due to the cross-sectional design, no evidence of a temporal relationship can be established between the patient’s medical information and the current lockdown. Lastly, it would have been interesting to explore caregivers’ mental health before lockdown and its evolution during and after this period. However, the survey methodology (anonymous online questionnaire, cross-sectional design) did not allow us to collect this data.

Clinical implications and recommendations for the future

Our findings have clinical implications. In the present study, half of our population presented with current anxiety/depressive disorders. While this is a matter of concern, the persistence of the virus and the current new lockdown is likely to worsen the health status of the caregiver population already weakened by the first lockdown. According to our findings, we should pay attention to both caregivers of patients at home and in nursing homes. In case of new lockdowns, it seems therefore essential to 1) identify caregivers with symptoms of anxiety, depression or severe burden. In this regard, a systematic evaluation should be carried out in memory centers and in other structures receiving PWD and their caregivers, e.g., with rapid screening scales. This identification should pay particular attention to women and caregivers living with their relatives because they are more at risk to develop anxiety and depression, 2) promote access to a comprehensive assessment carried out by psychologists or psychiatrists to specify the disorders in the identified caregivers, 3) refer caregivers to structures dedicated to them or to liberal professionals (e.g., psychologists) for therapeutic follow-up. Interventions and support to caregivers should be reinforced or at least maintained during the lockdown, face to face if possible or by video if necessary.

Furthermore, specific measures should be adopted to prevent occurrence or worsening of mood symptoms among caregivers during future lockdowns. For HC, it is important to maintain professional helps already in place while respecting the barrier measures. This would avoid disrupting the PWD’s routine. In addition, caregivers would feel less isolated and would not have to increase the time they spend helping PWD. Access to respite solutions should be facilitated in cases of heavy burden, for example with the intervention of paid home caregivers, as group care should be avoided. Access to parks or forests should be allowed, in order to prevent patients from feeling cramped in their homes. For NHC, it would be important to authorize visits by developing appropriate health protocols. In addition, contact with family through video or telephone should be encouraged. In this regard, the development of technologies adapted to the elderly may be relevant and their use thereafter should be encouraged.

The new measures taken by the French government in the second lockdown period are consistent with some of these points, in particular regarding visits in nursing-homes. In the future, it will be necessary to pay attention to these people after the crisis in order to estimate the longer-term impact on their mental health.