The impact of coronavirus disease 2019 on emotional and behavioral stress of informal family caregivers of individuals with stroke or traumatic brain injury at chronic phase living in a Mediterranean setting

Abstract Introduction Even in nonpandemic times, persons with disabilities experience emotional and behavioral disturbances which are distressing for them and for their close persons. We aimed at comparing the levels of stress in emotional and behavioral aspects, before and during coronavirus disease 2019 (COVID‐19), as reported by informal family caregivers of individuals with chronic traumatic brain injury (TBI) or stroke living in the community, considering two different stratifications of the recipients of care (cause and injury severity). Methods We conducted a STROBE‐compliant prospective observational study analyzing informal caregivers of individuals with stroke (IC‐STROKE) or traumatic brain injury (IC‐TBI). IC‐STROKE and IC‐TBI were assessed in‐person before and during COVID‐19 online, using the Head Injury Behavior Scale (HIBS). The HIBS comprises behavioral and emotional subtotals (10 items each) and a total‐HIBS. Comparisons were performed using the McNemar's test, Wilcoxon signed‐rank test or t‐test. Recipients of care were stratified according to their injury severity using the National Institutes of Health Stroke Scale (NIHSS) and the Glasgow Coma Scale (GCS). Results One hundred twenty‐two informal caregivers (62.3% IC‐STROKE and 37.7% IC‐TBI) were assessed online between June 2020 and April 2021 and compared to their own assessments performed in‐person 1.74 ± 0.88 years before the COVID‐19 lockdown. IC‐STROKE significantly increased their level of stress during COVID‐19 in five emotional items (impatience, frequent complaining, often disputes topics, mood change and overly sensitive) and in one behavioral item (overly dependent). IC‐TBI stress level only increased in one behavioral item (impulsivity). By injury severity, (i) mild (14.7%) showed no significant differences in emotional and behavioral either total‐HIBS (ii) moderate (28.7%) showed significant emotional differences in two items (frequent complaining and mood change) and (iii) severe (56.6%) showed significant differences in emotional (often disputes topics) and behavioral (impulsivity) items. Conclusions Our results suggest specific items in which informal caregivers could be supported considering cause or severity of the recipients of care.

By injury severity, (i) mild (14.7%) showed no significant differences in emotional and behavioral either total-HIBS (ii) moderate (28.7%) showed significant emotional differences in two items (frequent complaining and mood change) and (iii) severe (56.6%) showed significant differences in emotional (often disputes topics) and behavioral (impulsivity) items.

Conclusions:
Our results suggest specific items in which informal caregivers could be supported considering cause or severity of the recipients of care.

K E Y W O R D S
behavioral, COVID-19, emotional, informal caregivers, stress, stroke, traumatic brain injury INTRODUCTION Informal care is considered a cornerstone of all long-term care provision systems in Europe (EU, 2018). It is also gaining increasing recognition in international policy circles as a key issue for future welfare policy. Estimates suggest that as much as 80% of all long-term care in Europe is provided by informal carers (Frontiera et al., 2019). In the United States, approximately 43.5 million people are caregivers to a family member with a disability or illness (Gottschalk et al., 2020).
Caregiving can affect a caregiver's emotional well-being and their social activities as they often feel that they are not prepared for their role due to a wide variety of personal factors and lack of skills (Woodford et al., 2018).
A first step to developing therapeutic interventions provided by clinicians targeting caregiver preparedness is to better understand factors which influence such burden levels. This might enable the development of effective interventions to prepare and support caregivers in their role (Lieshout et al., 2020). A recent thematic analysis reporting on informal caregivers of community dwelling people with traumatic brain injury (TBI) highlighted the health systems' reliance on informal care and the importance of supporting them: Healthcare professionals need to consider and respond to the impact that changing circumstances have on the capacity of informal caregivers to manage their workload (McIntyre et al., 2018).

The coronavirus disease 2019 (COVID-19) began spreading in China
at the end of 2019 and, to date (October 2021) represents an international health emergency without precedents in terms of its health, economic and organizational effects on people's lives (WHO, 2021).
Spain has been one of the most affected countries in the world in terms of relative and absolute number of diagnosed cases. On March 13, 2020 (legally effective on March 15), the government declared a national state of alarm, with regulations targeted to facilitate diagnosis, ensure appropriate treatment of cases and reduce the spread of COVID-19, including measures of national lockdown, confinement of the population and restricted mobility (Laxe et al., 2020). While these experiences have been felt globally, the COVID-19 pandemic has introduced additional vulnerability and marginalization to those with some type of functional impairment-people with disabilities, chronic illness or frailty due to aging (Kang et al., 2020).
Evidence suggests, even in nonpandemic times, that persons with disabilities experience emotional and behavioral disturbances which are distressing for both individuals with disabilities and their close persons (Kim, 2007). There are limited data on the impact or added burden of a pandemic (or other calamities) on informal caregivers of persons with TBI or stroke.
TBI is a growing public health concern and represents the greatest contributor to disability globally among all trauma-related injuries (Dewan et al., 2018). There is consensus over the fact that the disability associated with TBI stems mainly from the cognitive, emotional, and behavioral alterations that complicate its course (Jorege et al., 2015). Furthermore, it has been recognized that these disturbances tend to be chronic, difficult to treat and, occasionally, progressive (Bavisetty et al., 2008).
Stroke is a major cause of long-term disability worldwide with increasing number of young people being affected by stroke in lowand middle-income countries (Katan et al., 2018). Behavioral disturbances have been reported as frequent complications in stroke survivors (Ferro et al., 2009). As reported in previous research, survivors of TBI very often experience symptoms ranging from irritability to aggressive outbursts (Dyer et al., 2006). Previous research remarked that increased severity of acquired injury is related to higher levels of caregiver stress. For example, Doser and Norub (2016) demonstrated over a long period (from 3 to 6 years) that the main factor influencing the burden of spouse-caregivers was the severity of the injury of the patients. Considering a shorter period after injury (2 years) Laratta et al. (2020) recently identified injury severity being associated with a higher level of spouse-caregiver burden. Similarly, the CONOCES study concluded that both the burden borne by informal caregivers and the likelihood of them being at a high risk of burnout (at 3 and 12 months post event) were associated with the severity of the stroke (Oliva Moreno et al., 2018). Meanwhile complaints on behavioral issues (such as low self-activation, lack of initiative or apathy) have been previously reported in survivors of stroke (Caeiro et al., 2013).
Consequently, we hypothesized that during COVID-19, specific emotional (e.g., anger, irritability and impatience) or behavioral (e.g., lack of initiative and overly dependency) aspects could be identified as sources of increased stress to caregivers when recipients of care are stratified by cause or by injury severity. Such aspects may be taken as an initial step for developing personalized therapeutic interventions provided by clinical professionals (e.g., as online remote services) to informal caregivers.
Therefore, this study aims at comparing the levels of stress in a set of 10 emotional and 10 behavioral standardized items, as reported by informal family caregivers of people with chronic TBI or stroke living in the community before and during the outbreak of COVID-19. We stratified the participants in (i) informal caregivers of people with TBI (IC-TBI) and informal caregivers of people with stroke (IC-STROKE) and (ii) informal caregivers of mild, moderate or severe recipients of care.

Study design
We conducted a prospective observational study analyzing informal caregivers of individuals with acquired brain injury (TBI or stroke) at chronic phase, with more than 3 years since injury, who were living in the community.

Participants
Eligible participants were informal caregivers of individuals with the diagnosis of stroke or TBI (at the moment of injury were aged ≥18), living in the community with electronical health records including complete data.
Individuals with stroke or TBI were excluded for the following reasons: diagnosis of concomitant comorbidity (e.g., spinal cord injury, brain tumors and anoxia) and a previous history of another disabling condition or not living in the community.
To satisfy the informal caregiver criteria, the participant had to be a first-degree relative (parent, sibling or spouse) who had spent at least 12 months looking after the person with TBI or stroke and who had the following characteristics: (i) is responsible for and makes decisions con-cerning the care of the person with TBI or stroke, regardless of whether he/she lives with the latter and/or (ii) spends most of the time, daily or weekly, accompanying/looking after the person with TBI or stroke. The informal caregiver might look after no more than one dependent person (De Arroyabe et al., 2013).
Every eligible participant was contacted as part of the routine clinical follow-up. Consequently, participants completed the online measures analyzed in this study as part of a virtual visit involving other assessments, not analyzed in this study. Such assessments include, for example the Patient Competency Rating Scale (PCRS) to assess selfawareness following brain injury.
Informal caregivers answered their follow-up online assessments within 10 days since contacted; therefore, all assessments were completed between June 2020 and April 2021.

Remote follow-up service and the Head Injury Behavior Scale
The online assessments were implemented during the COVID-19 lockdown in order to provide a remote follow-up service. Each participant received the online assessments by an SMS message sent to the participant's mobile phone. This SMS was sent by the professional from the hospital's Psychosocial Unit in charge of the online follow-up.
The online assessments include the same questionnaires and measurements as when participants were assessed (before the COVID-19 outbreak) in-person during their follow-up visits.
In this study, we focused on one of such measurements: the Head Injury Behavior Scale (HIBS) (Godfrey et al., 2003). The Psychosocial Unit of the hospital routinely performs HIBS follow-up every 3 years.
The HIBS was implemented to assess distress in caregivers of persons with acquired brain injury. It is a 20-item scale describing common psychological problems that occur following brain injury. For each of the items in the scale, the caregiver is asked, "Is the behavior a problem?" (yes/no) and "How much distress does this problem cause?." Distress ratings are recorded for the identified items using a fourpoint scale: 1 (no distress); 2 (mild distress); 3 (moderate distress) and 4 (severe distress). The HIBS includes two subscales: Emotional Regulation and Behavioral Regulation. The Emotional Regulation subscale assesses behaviors reflecting impaired emotional control and occurring during interactions with caregivers (e.g., sudden/rapid mood changes, depression, anger and aggression). The Behavioral Regulation subscale assesses problems that carry less emotional valence for caregivers (e.g., lack of control over behavior, inappropriate behavior for social situations, lack of motivation, or lack of interest in doing things) (Godfrey et al., 2003).
The HIBS has been previously used in assessing emotional and behavioral problems in individuals with chronic TBI (Villalobos et al., 2020) (Marsh et al., 2006) or stroke (Orfei et al., 2009). In this study, we used the version of the HIBS adapted to the Spanish language (De Arroyabe et al., 2013). A description of the 20 HIBS items is presented in Table S1.

Clinical and demographic variables
Demographics (age, sex, years of education and marital status), as well as time since onset of the injury, were collected. Stroke severity of the injury was assessed within 24 h before acute phase discharge by the National Institutes of Health Stroke Scale (NIHSS) and stratified as follows: mild (1-4), moderate  or severe (NIHSS ≥15) (Williams et al., 2000).
The Glasgow Coma Scale (GCS) (Teasdale et al., 1974) was used to assess TBI severity at admission and was stratified as mild (14-15), moderate (9-13) or severe (3-8) (Mena et al., 2011). All data were collected from the hospital' electronical health records of the individuals with TBI or stroke.

Ethical considerations
The study follows the Declaration of Helsinki and was approved by the Ethics Committee of Clinical Research of Institut Guttmann-Neurorehabilitation hospital. The participants were anonymized and nonidentifiable.

RESULTS
The initial number of eligible participants, considering the criteria described in Section 2.2. was n = 191, four of them had difficulties in understanding or expressing in the Spanish language as detected by members of the research team, which prevented him/her from being able to collaborate. One hundred and eighty-seven participants received the SMS in their mobile phones, 24 (12.8%) were not a firstdegree relative (parent, sibling or spouse) who had spent at least 12 months looking after the person with TBI or stroke, 3 (1.6%) were looking after more than one dependent person, 11 (5.8%) did not have an available Functional Independence Measure (FIM) in person assessment and 27 (14.4%) did not complete the online HIBS assessment.

Stratification by IC-TBI or IC-STROKE
We then compared the emotional regulation subtotal, the behavioral regulation subtotal and the total HIBS score for IC-TBI (N = 46, 37.7%) and IC-STROKE (N = 76, 62.3%), as presented in Table 2.
When considering emotional regulation subtotals, no significant differences were found in IC-TBI (p = .109), but differences were significant for IC-STROKE (p = .001) and with the highest effect size Meanwhile, the opposed is observed when considering behavioral regulation subtotals, no significant differences were found in IC-STROKE (p = .072), but differences were significant for IC-TBI (p = .027). We then compared individuals with stroke with individuals with TBI considering demographics and clinical variables, as presented in Table 3.

TA B L E 2
Significant differences were found in age at the moment of online assessment. Individuals with TBI were significantly younger (p < .001) with 50% of them between 31 and 45 years old at the moment of the online assessment, whereas only 9.2% of the individuals with stroke were between 31 and 45 years old at the moment of online assessment (p < .001).
When comparing functional independence, as assessed using the FIM, cognitive FIM was significantly higher in individuals with stroke (p = .008).
We further compared each of the 20 HIBS items separately for IC-STROKE and IC-TBI, before and during COVID-19, in order to identify specific items causing (mild, moderate or severe) stress, as detailed in Table 4. For IC-TBI only in one of them (item 5: Impulsivity; does things without thinking), a significant increase of the level of stress was reported: 32.6% of the informal caregivers had stress due to this item before and during COVID-19, and the percentage of informal caregivers who had stress due to this item significantly increased to 52.2%, as detailed in Table 4.
Meanwhile, for IC-STROKE, significant differences were reported in six items: impatience, frequent complaining, often disputes topics, mood change, overly sensitive and overly dependent.
A graphical representation of Table 4 (considering only the emotional regulation items) is presented in Figure 1.

Stratification by injury severity
We then compared the emotional regulation subtotal, the behavioral regulation subtotal and the total HIBS for informal caregivers of the recipients of care, now stratified according to their injury severity (as defined in Section 2.4).
When considering individuals with mild injury (N = 18, 14.7%), no significant differences were found in the emotional regulation subtotal (p = .273), and the behavioral regulation subtotal (p = .09) either the total HIBS (p = .23).
When considering individuals with moderate injury (N = 35, 28.7%), we found significant differences in the emotional regulation subtotal (p = .04). No significant differences were found in the behavioral regulation subtotal (p = .62) either the total HIBS (p = .18). Finally, when considering individuals with severe injury (N = 69, 56.6%), we found significant differences in all three of them: the emotional regulation subtotal (p = .03), the behavioral regulation subtotal (p = .02) and the total HIBS (p = .01), as presented in As shown in Figure 2, individuals with mild severity get significantly higher total FIM scores when compared with individuals with severe injury, and this is represented with two asterisks in Figure 2. Differences in total FIM scores are not significant when comparing mild with moderate and moderate with severe. Nevertheless, Figure 2 clearly shows that as severity increases, the FIM score decreases.
We further compared each of the 20 HIBS items separately for moderate and severe recipients of care, before and during COVID-19, in order to identify specific items causing (mild, moderate or severe) stress, as detailed in Table 6.
For recipients of care with moderate injury, a significant increase of the level of stress was reported by the informal caregivers in two emotional items (frequent complaining and sudden/rapid mood change).
For severe recipients of care, a significant increase of the level of stress was reported in two items, one emotional (argumentative; often disputes topics) and the other behavioral (impulsivity; does things without thinking).
We did not include recipients of care with mild injury severity in Table 6 because (as reported in Table 5) no significant differences were found for them in the emotional regulation subtotal and the behavioral regulation subtotal either the total HIBS. Besides, only 14.7% of all recipients of care presented mild injury severity.

DISCUSSION
We studied the impact of COVID-19 from a psychological perspective (using the HIBS 10 emotional and 10 behavioral items), on informal caregivers of individuals living in the community with chronic TBI or stroke, stratified by cause of injury and by injury severity of the recipients of care.
A recent survey conducted in Nashville, United States, by Morrow et al. (2021) on 47 individuals in the chronic phase of moderate-severe TBI reported that for 51.06% of the participants, the pandemic has affected their sense of mental and physical well-being. In our case, we analyzed 20 psychological items and only in one of them a significant increase of the level of stress was reported (Impulsivity; does things without thinking). Meanwhile as detailed in Table 4, in caregivers of people with stroke, significant differences were reported in six items: impatience, frequent complaining, often disputes topics, mood change, overly sensitive and overly dependent. Prepandemic studies among family caregivers of community dwelling stroke survivors report that informal caregivers' total burden score was largely driven by their feelings of overly dependence (Morimoto et al., 2001), being dependency and immature behavior major causes of stress (Dankner et al., 2016). Cohen et al. (2021) analyzed self-reported changes in burden (CB) and intensity due to COVID-19 (n = 835) of informal caregivers living in the United States. They found that men with higher initial levels of confidence interval were more likely to have an increase in CB due to the pandemic, but the association was not significant for women. There is no clear explanation for this finding although it may be due to gender differences in resilience among informal caregivers. Female caregivers may be more resilient than male caregivers (Gaugler et al., 2007), and this resilience may become magnified under periods of extreme stress and uncertainty, such as the COVID-19 pandemic. In our case, as presented in Table 1, 71.3% of care recipients were males, suggesting that most of their informal caregivers could be females, though we did not collect demographic data on the informal caregivers. Nevertheless, it is known that women are more likely than men to be informal caregivers, among the whole population of informal caregivers themselves (Bauer et al., 2015). Cohen et al., concluded that a notable finding of their study was that higher initial CB was associated with a higher likelihood of increased CB during the COVID-19 pandemic. This finding may be counterintuitive, as one could expect that there is a potential for a ceiling effect of CB-in other words, CB could not increase substantially due to the pandemic simply because it was already high (Hagell et al., 2017). One possible explanation for this finding is that the extra burdens of the pandemic, whether due to increased anxiety or other stressors, magnified the effects of caregiving on those with an already high level of CB. That was not the case in our IC-TBI sample, a ceiling effect may be occurring for IC-TBI, causing that an already high level of stress (as shown in Figure 1) masked the effect of the pandemic situation.   2,3,4,6,14,15,16,17 ,19,7,8,9,10,11,12,13,18,20. limitation of the study was that authors were not able to determine whether the caregivers were living with the care recipients and if they were the sole caregivers of the care recipients.  Table 4 may be due to a similar masking effect, as shown in Figure 1 their stress levels were already high before the COVID-19 outbreak. In relation to informal caregivers of individuals with stroke, we found significant differences in five out of the 10 HIBS emotional items, indicating a clear impact of the pandemic situation on them. Though, we did not stratify them considering informal caregivers of individuals with aphasia.
Injury severity has been reported to play a large role in the type of sequelae that result following TBI and thus influence the caregiver/family's function over time (Thompson et al., 2009). In our case, for mild injuries we identified no differences before and during COVID-19. For moderate injuries, we identified differences in the total emotional regulation score and when injuries were severe, we identified differences in the three of the assessed totals. Therefore, our results suggest that for recipients of care with moderate injuries, support to informal caregivers should be focused on emotional aspects (mainly in frequent complaining and sudden/rapid mood change as shown in Table 6) rather than on behavioral. Meanwhile, for severe injuries, support should be focused on both emotional (specifically in argumentative; often disputes topics) and behavioral aspects (specifically in impulsivity; does things without thinking). Acquired brain injury is considered a stressor that changes caregiver's psychosocial functioning (Perlesz et al., 2000). Higher distress and lower life satisfaction of caregivers are associated with low longterm functional outcomes of persons with acquired brain injury, even after accounting for injury severity and cognitive functioning of care recipients (Vangel et al., 2011). As the person with acquired brain injury and his or her caregiver belongs to the same family system, both are influenced and influential agents (Vangel et al., 2011). Our findings are in line with such results; furthermore, in Table S3 we analyzed a stratification in two groups according to their FIM assessments (independent vs. dependent). The emotional, behavioral subtotals and total HIBS comparisons yielded that the informal caregivers to the dependent group presented higher effect sizes in all three of them.
Several limitations to this study are worth mentioning. First, the data were collected from informal caregivers of individuals living in the community but who had previously undertaken rehabilitation in one single tertiary center in Catalonia. As presented in Table 1, 72.8% of them were living in the province of Barcelona. Therefore, the generalization of these results should be considered carefully. Nevertheless, assessments by standardized tools (HIBS, FIM, NIHSS and GCS) allow for similar comparative studies, and the restricted physical locations allow for controlled variability in regional pandemic circumstances.
Second, the evolving nature of the pandemic may produce psychological effects, for example the COVID-19 vaccination campaign started on December 27, 2020 in Spain (Vaccination Program, 2021).
The initial phase of vaccination plan covered residents of senior homes and their carers and continued with healthcare workers and adults with major dependencies. Therefore, it did not involve any of the participants in this study, but it could somehow impact in their psychological status. Other contextual circumstances may have had an emotional impact (e.g., summer in Spain from June 2020 to September 2020).
Third, male gender accounts for about 70% of the individuals with stroke and individuals with TBI, suggesting a sex bias. Nevertheless, the proportion is similar to recent studies in similar settings (as presented in Supporting Information).
Fourth, demographic data (e.g., age, gender, level of education) were not collected on informal caregivers. The online assessments were implemented during the COVID-19 lockdown in order to provide a remote follow-up service, as performed in-person before the COVID-19 outbreak. Therefore, remote service included the same psychological assessments performed as in-person visits, which did not collect demographic data on informal caregivers.
Fifth, online assessments have been criticized in previous studies. Future work includes (i) analyzing informal caregivers' demographical variables, psychological status (e.g., HADS assessment) and burden (e.g., ZARIT assessment), (ii) conducting a specific analysis on caregivers of individuals with aphasia (approximately 40% of the hospital admissions) and (iii) in Table S2, we included a stratification by age in two groups (≤54 years (N = 55) and > 55 years (N = 67)) and found that differences were nonsignificant in the HIBS behavioral subtotal for the ≥55 years group, whereas in the younger group, differences were significant in both emotional and behavioral subtotals, leaving room for future analysis.

CONCLUSIONS
For the first time, we reported results of the COVID-19 impact from a psychological perspective, considering 10 emotional and 10 behavioral items on informal family caregivers of individuals with stroke or TBI living in the community. We identified six specific HIBS items that have been significantly impacted by the COVID-19 in informal caregivers of individuals with stroke. We found one item in caregivers of individuals with TBI, suggesting in their case a possible masking effect due to a preexisting situation of emotional instability.
When considering the severity of the recipients of care, we identified two emotional items for informal caregivers of moderate recipients of care and one emotional as well as one behavioral item for severe recipients of care. We identified a higher impact on emotional and behavioral subtotals as well as in total HIBS as severity of injury increased, further confirming them to be closely related to their longterm functional independence outcomes Our results can be used to suggest possible therapeutic interventions or support provided by clinicians to informal caregivers, specifically targeting the identified items or subtotals.

FINANCIAL SUPPORT
This research was partially funded by H2020 Personalised Medicine by Predictive Modeling in Stroke for better Quality of Life (PRECISE4Q) Grant Nr 777107.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1002/brb3.2440