Consequences of Anosognosia on the Cost of Caregivers ’ Care in Alzheimer ’ s Disease

11 Background: Anosognosia is common in patients with Alzheimer’s disease (AD) and it is frequently related to an increase in time of care demand. 12 13 Objective: The aim of the study was to examine the effect of anosognosia on the total costs of informal care in patients with AD. 14 15 Methods: This was a prospective longitudinal study with community-dwelling AD patients. Anosognosia, time of informal care, and the use of support services (e.g., day care centers) were recorded at baseline and after 24 months. The cost of informal caregiving was calculated as ‘market price’. 16


INTRODUCTION
The costs of dementia care are frequently divided into formal and informal costs.Informal costs refer Most patients with dementia are cared for at home, which causes an increase in informal costs, adding an economic burden on families instead of on healthcare systems [1].
Caring for a relative with dementia has a hard impact on the emotional, physical, and economical situation.The economic value of this impact has been estimated using the cost of professional caregivers as a basis (cost of formal care) or by providing a monetary value to the possible loss of opportunities due to the time spent as a caregiver [7,8].
Previous studies related informal costs to the progressive increase of the patient's dependency, to the severity of the dementia, and to the presence of behavioral and psychological symptoms of dementia (BPSD) [5,9].One of the BPSD that has been reported to increase the perceived caregiver's burden is anosognosia, the lack of disease awareness [10,11].
The prevalence of anosognosia in AD patients is over 25%, and it is cognitive decline-related, with around 80% of the patients with severe AD having anosognosia [12].Patients with anosognosia present more BPSD than patients with no anosognosia, such as delusions [13], disinhibition, and apathy [14].
BPSD cause an increase of the burden, the distress, and the cost of care [15].Besides, the quality of life perception and the efficiency of neurocognitive rehabilitation are negatively affected by the presence of anosognosia [16][17].All these factors associated with the presence of anosognosia, even when they may also appear in patients with mild cognitive impairment [18], have an effect on the impact of the informal care required by the patient.Furthermore, patients with anosognosia are more likely to engage in risky behaviors, which increases the burden of the caregivers due to the need of increased supervision and control, which may even lead to early institutionalization [19].Moreover, the presence of anosognosia has been linked to an increased sense of isolation, greater physical and emotional burden, and a patient-caregiver relationship characterized by greater dependency [20].Overall, therefore, the presence of anosognosia may be associated with greater care needs, either regarding the number of hours of care needed or the use of health and social support services.
To our knowledge, this is the first study to analyze the impact of anosognosia on informal costs.
Our hypothesis was that the presence of anosognosia may cause an increase in the cost of care due to the increased time of care, the increased use of resources, or to both.The cognitive assessment of the patients was based on the Mini-Mental State Exam (MMSE), a brief cognitive assessment tool with a score ranging from 0 to 30 (the lower the score the greater the cognitive deterioration) [25].
The functional assessment of the patient was based on the Disability Assessment for Dementia (DAD) [26].The DAD provides a measure of basic and instrumental ADL and was administered to the main caregiver.It comprises 40 items and its total score ranges from 40 to 80 (the higher the score the greater the patient's functional ability).
The severity of dementia was classified according to the criteria of the GDS, a tool designed to determine the stage of a patient's dementia.GDS 4 corresponds to mild dementia, GDS 5 to moderate dementia, and GDS 6 to moderately severe dementia.
The presence of BPSD was evaluated by means of the Neuropsychiatric Inventory (NPI) [27], which comprises 12 subscales that assess the frequency and severity of 12 neuropsychiatric symptoms (or BPSD), based on information provided by caregivers.The score ranges from 0 to 144, and the higher the score the greater the frequency and severity of neuropsychiatric symptoms.
Finally, anosognosia was assessed using the Anosognosia Questionnaire-Dementia (AQ-D) [28], a tool that is administered both to the patient and the caregiver.It comprises 30 items assessing cognitive/functional deficits and changes in the patient's behavior, with each item being rated according to the frequency of occurrence, from 0 (never) to 3 (always).The total score ranges from 0 to 90.The degree of anosognosia is estimated on the basis of the difference between the patient and caregiver's scores.The cut-off for the presence of anosognosia was established at difference ≥32 points.We identified asymptomatic cases when anosognosia was not present in any evaluation, incident cases when patients without anosognosia at baseline developed it during the follow up, and persistent cases when anosognosia was present both at baseline and during the follow up.

Procedure
Neurologists from the Dementia Unit selected eligible patients according to the inclusion criteria.The sample was recruited between January and December 2011 and the study finished on March 2014.The aims of the study were explained to all the participants in an introductory interview, and an informed consent  Longitudinal data were analyzed by means of 217 generalized linear models (GLM), because some 218 dependent variables were not normally distributed 219 [33][34].This approach enabled us to examine the  Thus, the time requested by GDS4 patients was 2.6 h/day (SD = 1.7), while for GDS6 patients it was 7.0 h/day (SD = 1.8) (F = 274.8;df = 2; p < 0.001).
The time of care showed a moderate correlation with BPDS (rho = 0.405; p < 0.001) but when including the NPI score as an independent variable in the GLM, the relation with cost over 24 months was not significant (F = 1.449; df = 487; p = 0.229).
The resources used by the patients and their caregivers are described in Table 3.At the 24-month follow-up, 74.5% of cases used, at least once, the available health and social services.The proportion of patients using these resources did not change during the two-year study period: 68.8% at baseline, 74.7% at 12 months, and 66.1% at 24 months (χ 2 = 4.6; df = 2; p = 0.101).Across the 24 months of follow-up 4.7% of patients were institutionalized.

Impact of anosognosia on the care received by patients
Anosognosia was observed in 54.3% of the patients at baseline.Over the follow-up period, 27.2% of the cases (n = 140) remained asymptomatic, 22.2% (n = 114) presented with anosognosia at either 12 or 24 months (patients without anosognosia at baseline who developed it during the follow up), and anosognosia persisted in 50.6% (n = 260) of the cases (anosognosia present at baseline and during the follow up).A request for help from the relatives and/or the use of services was registered in 82.9% of the cases with anosognosia versus 61.8% of those without anosognosia (χ 2 = 28.9;df = 2; p < 0.001).
From baseline onwards, patients with anosognosia required more hours of care than did their asymptomatic counterparts (Wald = 37.1; df = 2; p < 0.001).

319
The presence of anosognosia was associated with 320 an increase in the time dedicated to the informal care 321 of the patients, regardless of the dementia severity 322 (Wald = 9.0; df = 2; p = 0.011).Figure 2 shows the port services was significantly higher at 24 months 342 (F = 3.6; df = 2; p = 0.026).

343
In The main purpose of this work was to describe 362 the effect of anosognosia in the cost of care, using the daily hours of care and the use of resources as cost-indicators.
The informal care provided by relatives is fundamental to AD patients' attention.Principal characteristics of AD, such as functional dependence and neuropsychiatric symptoms are the main factors related with an increase of the time of care.However, both the perceptions of caregiving stress and the positive aspects of caregiving are appraised through a cultural/ethnic lens [35], and whereas Caucasians generally place earlier the loved ones in care facilities [36], African Americans and Latinos tend to delay institutionalization [37].Informal care is usually provided by close relatives, especially if one of these is a woman.In agreement with previous research, our results show that external support services still account for a relatively small proportion of the overall care that is provided (around 15% of the total cost).other kind of formal support, such as tele-care [38].
Several studies have reported specific data about the cost of formal and informal care to patients with dementia [39][40][41].The factors that influence the cost of care can be classified in two groups: dementia-directly related factors and factors related to caregiver's burden.The first group encloses functional disability, anosognosia, and neuropsychiatric symptoms, among others [42].In our analyses neuropsychiatric symptoms were not significant, which may be due to the fact that dementia severity is influenced-by the presence of neuropsychiatric symptoms.In another study, the NPI scores were barely related with an increment of the caregiving costs, and it was the dependence scale that largely explained the variance of the caregiving costs [43].
The second group is related with the caregiver burden perception.We previously reported an increase of the burden perception associated with anosognosia [20], and there is evidence that greater burden is related to a greater use of resources [44].We did not include the burden in our study due to its strong relationship with anosognosia.[40,48], similar to the figures in Ireland [49], and slightly less than the cost documented in Germany, the UK, and France [50].The cost of care increases with the severity of dementia, and most studies coincide in terms of the support services requested by caregivers and the factors that lead to greater care needs (functional disability and BPSD) [51].The present study adds to this picture by showing that the presence of anosognosia is another factor that increases the cost of care over 300D /month on average.
Interestingly, we did not find an inverse relationship between the number of hours of informal care provided and the use of support services, which suggests that the availability of such services improves the quality of care provided, and the quality of life of patients and caregivers, rather than reducing the number of hours of informal care per se.
An increase in the amount of time spent by informal caregivers directly correlates with an increase in their perceived burden [52][53][54].In a previous study we observed that the cost of informal care increased in line with the patient's physical disability and cognitive impairment, as well as if the relative was the sole caregiver, with the cost of informal care, explaining 6.7% of the total variance in the perceived burden of caregivers [40].We also reported that the presence of anosognosia was a determining factor in terms of greater caregivers' perceived burden [20,55].The results of the present study highlight the need for more specific help to be offered to caregivers in relation to the impact that anosognosia in the patient can have, and the implications it has for home-based care [15,53].Future research may include a perceived economic burden measure, as well as some data regarding the "opportunity costs" related with a possible loss of opportunities due to the time spent as a caregiver.In the context of resource management, maintaining or improving the quality of life of patients and caregivers is one of the main ways in which the costs of health and social care can be reduced [56].
Previously, many studies examined the indirect cost based on caregivers' self-reported number of hours with no external validation.Also, our study did not include the "opportunity costs" to care-

92
This was a longitudinal study involving a 24-month 93 follow-up of a consecutive sample of outpatients seen 94 at the Dementia Unit (Department of Neurology) of 95 Bellvitge University Hospital (Hospitalet de Llobre-96 gat, Barcelona).They were all diagnosed as either AD 97 according to the criteria of the Diagnostic and Statis-98 tical Manual of Mental Disorders [21] or probable 99 AD according to the criteria of the National Institute 100 of Neurological and Communicative Disorders and 101 Stroke / Alzheimer's disease and Related Disorders 102 Associations (NINCDS-ADRDA) [22].The main 103 caregiver was defined as the person with ongoing 104 responsibility for helping the patient with activities 105 of daily living (ADL).All the caregivers were rela-106 tives of the patient, mainly daughters/sons or spouses.107 Informed consent was obtained for all participants.108 Patients were excluded if they presented with vascular 109 or traumatic events, alcohol or substance dependency 110 or abuse, and if they had severe communication prob-111 lems or had a severity of Global Deterioration Scale 112 (GDS) stage 7 [23] that prevented them from respond-113 ing adequately to the assessment questions.The study 114 was approved by the hospital's Clinical Research tasks such as toilet visits, eating, dressing, grooming, 131 walking and bathing?" as BADL, and "On a typi-132 cal care day during the last 30 days, how much time 133 per day did you assist the patient with tasks such 134 as shopping, food preparation, housekeeping, laun-135 dry, transportation, taking medication and managing 136 financial matters?" as a question of IADL.
189 then interviewed separately by two clinical psychol-190 ogists trained in the administration of the study 191 protocol instruments.192 The economical cost of time of care was evaluated 193 as market price, thus attributing the cost per hour of 194 private professionals [29].The attribution of costs and 195 the cost of using different health and social resources 196 are shown in Table 1.The cost of day care centers 197 and residential care was derived from the costs estab-198 lished by the Catalan government [30], the cost per 199 hour for home care services was based on the figures 200 set by the Barcelona City Council [31], and the cost 201 of attending memory-training programs and day care 202 centers was taken as the mean cost of 10 hours/month 203 of such services in the area where the patient lived.204 The hourly cost of informal care was taken to be 205 equivalent to the cost per hour of a geriatric nursing 206 assistant providing help with BADL.The standard for 207 costs per hour was obtained from the Catalan Office 208 of National Statistics [32].209 Statistical analysis 210 Differences between missing cases and those that 211 completed the follow-up were analyzed using either 212 parametric or non-parametric tests according to the 213 criteria of normality, for continuous variables, and 214 with the Pearson chi-square test for categorical vari-215 ables.

Fig. 1 .
Fig. 1.Flowchart showing changes in the sample's characteristics over the 24-month period.

323
number of hours dedicated per day to provide help 324 with BADL and IADL, according to the severity of 325 dementia and stratified by the patient's anosognosia 326 status.
However, although traditional roles and relationships persist within the provision of informal care in Spain, research suggests that caregivers are beginning to ask for different kinds of help from government agencies [1].Whereas twenty years ago caregivers were most likely to request financial assistance in the form of a monthly caregiver allowance, the main demand nowadays is for home care services and greater training.These home care services would include the use of day care centers and the access to

Fig. 2 .
Fig. 2. Number of hours dedicated per day to helping with basic and instrumental ADL, according to the severity of dementia and stratified by the patient's anosognosia status.

488
givers and only included the "market price" for each 489 hour reported.These limitations have influenced both 490 the number of hours reported and the associated 491 attributed cost.However, both figures are similar to 492 those documented by other studies in Spain and inter-493 nationally [24, 48-49].Even though the resources and 494 their access are not equal everywhere, we decided 495 to describe their use in a specific work-class neigh-

Table 1
Monthly cost of the health and social support services considered in the study approximate cost for 10 h/month; * * Data from the Catalan government for such services in 2010; † Price/hour established by Barcelona City Council for home care services in 2013; Occasionally = 1 day/week; Weekly = 3-4 days/week; Daily = 5-7 days/week.
*was obtained from both patients and caregivers before 188 proceeding.The patients and their caregivers were

Table 3
Frequency of use of different support services according to the presence or absence of anosognosia, stratified by assessment point the generalized linear model the total cost of 347ther was there an interaction effect (Time × Severity 348 of Dementia × Anosognosia) on this total cost 359 GDS 6 (2,424.8D/month).360 DISCUSSION 361

Table 4
Total combined monthly cost of informal care by family caregivers and external support services, stratified by severity of dementia and anosognosia status 419 the number of hours of informal care required and 420 the use of health and social support services.How-421 ever, the inclusion of the presence of anosognosia as 422 a variable in our analysis revealed important differ-423 ences both in care hours and the use of resources.The 424 total cost of care increased in an almost linear manner