Archives of Gerontology and Geriatrics 66 (2016) 42–48

Contents lists available at ScienceDirect

Archives of Gerontology and Geriatrics

journal homepage: www.elsevier.com/locate/archger

Caregiver burden and fatigue in caregivers of people with dementia:

Measuring human herpesvirus (HHV)-6 and -7 DNA levels in saliva
Tohmi Osakia,b,* , Takako Morikawab , Hiroyuki Kajitab , Nobuyuki Kobayashic ,
Kazuhiro Kondoc, Kiyoshi Maedab
a
Medical Center for Dementia, Kobe University Hospital, 7-5-2, Kusunoki-cho, Chuo-ku, Kobe, Hyogo 650-0017, Japan
b
Kobe Gakuin University Faculty of Rehabilitation, 518 Arise, Ikawadani-cho, Nishi-ku, Kobe, Hyogo 651-2180, Japan
c
Department of Virology, The Jikei University School of Medicine, 3-25-8, Nishi-shimbashi, Minato-ku, Tokyo 105-8461, Japan

A R T I C L E I N F O A B S T R A C T

Article history: Purpose: We examined chronic fatigue, which has not been investigated in detail, in family caregivers for
Received 4 December 2015 people with dementia.
Received in revised form 25 April 2016 Methods and materials: Forty-four community-dwelling family caregivers (the caregiver group: CG) and
Accepted 26 April 2016
50 elderly control participants (the non-caregiver group: NCG) participated in this study. We measured
Available online 7 May 2016
salivary human herpesvirus (HHV)-6 and -7 DNA levels and the Chalder fatigue scale (CFS) to assess levels
of fatigue; we also measured the Center for Epidemiologic Studies-Depression Scale, Physical Activity
Keywords:
Scale for the Elderly, Zarit Caregiver Burden Interview, Mini-Mental State Examination, Assessment of
Dementia
Caregivers
Motor and Process Skills, and Dementia Behavior Disturbance Scale.
Fatigue Results: For CG, the salivary HHV-6 DNA levels and CFS scores were significantly higher than those in NCG.
Human herpesvirus 6 The salivary HHV-6 DNA levels in CG were significantly correlated with depressive symptoms, the
Chalder Fatigue Scale cognitive function of the patients, and the activities of daily living/instrumental activities of daily living
Psychological stress (ADL/IADL) abilities of the patients. The CFS scores in CG significantly correlated with caregiver burden,
depression symptoms, leisure physical activity, the number of other family caregivers, and the hours
spent for caregiving per week, as well as with behavior disturbances and ADL/IADL abilities.
Conclusions: The salivary HHV-6 DNA levels may be added as a new biomarker for caregiver exhaustion.
We concluded that fatigue assessments should be performed by not only a questionnaire, such as the CFS,
but also by a biomarker search, such as HHV-6, when estimating the caregiver burden for family
caregivers of people with dementia.
ã 2016 Elsevier Ireland Ltd. All rights reserved.

1. Introduction caregiving, it is reported that caregivers’ mortality (Schulz & Beach,

The caregiver burden in caregivers of people with dementia is
one of the most serious problems in medical and nursing care for
dementia (Brodaty & Donkin, 2009). According to chronic stress for
Abbreviations: CG, the caregivers group; NCG, the non-caregiver group; HHV,
human herpesvirus; DNA, deoxyribonucleic acid; CFS, Chalder fatigue scale; CFS-P,
CFS-physical; CFS-M, CFS-mental; CES-D, the Center for Epidemiologic Studies-
Depression Scale; PASE, Physical Activity Scale for the Elderly; PASE-L, PASE-leisure;
PASE-H, PASE-house; PASE-W, PASE-work; ZBI, Zarit Caregiver Burden Interview;
MMSE, Mini-Mental State Examination; AMPS, Assessment of Motor and Process
Skills; DBD, Dementia Behavior Disturbance Scale; ADL, activities of daily living;
IADL, instrumental activities of daily living; QOL, quality of life; PCR, polymerase
chain reaction; SD, standard deviation; IQR, interquartile range.
* Corresponding author at: Medical Center for Dementia, Kobe University
Hospital, 7-5-2, Kusunoki-cho, Chuo-ku, Kobe, Hyogo 650-0017, Japan.
E-mail addresses: tohmiiiii@gmail.com (T. Osaki),
maedak@reha.kobegakuin.ac.jp (K. Maeda).
1999) and risk of coronary heart disease is increased (Lee, Colditz,
Berkman, & Kawachi, 2003). The Zarit Caregiver Burden Interview
(ZBI; Zarit, Reever, & Bach-Peterson, 1980) has been used for the
assessment of family caregiver burden in people caring for patients
with dementia. ZBI estimate caregiver stress from caregiving,
economic burden, or restriction of social participation. ZBI,
however, is not enough to assess for caregivers’ health, because
it is a questionnaire that evaluates subjective caregiver burdens
instead of objective burdens.
In the current study, we focused on fatigue in family caregivers.
Fatigue is defined as an inimitable discomfort and a condition of
decreased physical activity with a desire for rest; it is caused by
excessive physical and mental activity or disease. Fatigue, pain, and
fever are the three alarms for an organism in crisis. We considered
that fatigue was a useful indicator that assessed the health risk
caused as a result of chronic stress from caregiving. In previous
studies of the burden on family caregivers of patients with

http://dx.doi.org/10.1016/j.archger.2016.04.015
0167-4943/ã 2016 Elsevier Ireland Ltd. All rights reserved.
 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48
dementia, fatigue has been assessed as a part of the quality of life
(QOL) assessment; but fatigue itself has not been investigated in
detail. In this study, we proposed to evaluate fatigue in these family
caregivers using two scales.
Firstly, we used the Chalder Fatigue Scale (CFS; Chalder et al.,
1993), which is a self-reported assessment and has often been used
to assess patients with chronic fatigue syndrome (Morriss,
Wearden, & Mullis, 1998). Secondly, we measured the saliva
DNA levels for human herpesvirus (HHV)-6 and -7. Most people get
infected with HHV-6 and -7 during childhood, and life-long latency
is established (Kondo & Yamanishi, 2007). The trigger that induces
viral reactivation has not been completely identified. However, it is
reported that when physical and mental stress result in fatigue,
human herpesvirus may be autonomously reactivated and shed in
the saliva (Kondo & Yamanishi, 2007; Whitley, Kimberlin, & Prober,
2007). Assessment of fatigue by using the salivary HHV-6 and -7
DNA levels has not been standardized. However, in a few previous
studies, it has been reported that HHV-6 and -7 DNA saliva levels
are useful for the assessment of chronic fatigue (Fukuda et al.,
2015; Ito et al., 2014; Morris, Berk, Walder, & Maes, 2015; Tanaka,
Shigihara, Funakura, Kanai, & Watanabe, 2012). Acute stress does
not influence these levels, and diurnal variation is little (Kondo,
2009). We surmised that the HHV-6 and -7 DNA levels in the saliva
were relevant in the assessment of caregivers who were exposed to
moderate- to long-term stress during caregiving.
In this study, we examined the levels of fatigue which has not
been investigated in detail in the family caregivers of patients with
dementia. In particular, we also wanted to investigate the
usefulness of the salivary HHV-6 and -7 DNA levels as a new
fatigue biomarker.
2. Methods
2.1. Subjects
We recruited 44 primary family caregivers that were caring for
relatives with dementia (the caregiver group: CG). Caregivers with
intellectual problems were excluded. The people with dementia
had been diagnosed by psychiatrists. Fifty non-caregiver controls
(the non-caregiver group: NCG) were recruited from the senior
school in the same area as the psychiatric hospital. Those who were
caring for someone already were excluded.
This study was approved by the Ethical Committee of Kobe
Gakuin University in December 2013 [approval number
HEB131218-1] and the Ethics Committee of the Jikei University
School of Medicine in February 2015 [approval number 23-316
(7822)]. All participants provided written informed consent to
participate in the study.
2.2. Assessments
For this cross-sectional study, we collected data between
January and April 2015. The evaluations of the people with
dementia had been obtained by an occupational therapist in the
psychiatric hospital. The questionnaires were self-reported from
home, and the saliva samples were either collected during the time
that questionnaires were distributed or collected at home or the
psychiatric hospital or school.
Saliva samples for the analysis of the HHV-6 and -7 DNA levels
were collected in a tube (Sallivette; Sarstedt, Tokyo, Japan) and
centrifuged at 3000g for 2 min at 4
C. These supernatants were
dispensed at 450 ml and stored at 8
C until analyzed. The DNA
samples were extracted from 400 ml of saliva with an EZ1 Virus
Mini Kit v2.0 (Qiagen, California, USA). DNA was eluted in 90 ml of
elution buffer. The HHV-6 and -7 DNA levels were quantified via
real-time polymerase chain reaction (PCR), which was completed
43
using an Applied Biosystems 7300 Apparatus (Life Technologies,
California, USA). The amplifications were performed in duplicate in
a total volume of 50 ml containing 25 ml of Premix Ex Taq (Perfect
Real Time; Takara Bio Inc., Shiga, Japan), 0.45 ml of PCR forward
primer (100 mM), 0.45 ml of PCR reverse primer (100 mM), 1.25 ml
of TaqMan probe (10 mM), 1 ml of Rox reference dye, 5 ml of the viral
DNA, and 16.85 ml of PCR-grade water. The primers used for real-
time PCR were as follows: HHV-6 forward primer, 50 -GACAATCA-
CATGCCTGGATAATG-30 ; HHV-6 reverse primer, 50 -
0 0
TGTAAGCGTGTGGTAATGGACTAA-3 ; HHV-6 probe, 5 -FAM-
AGCAGCTGGCGAAAAGTGCTGTGC-TAMRA-30 ; HHV-7 forward
primer, 50 -CGGAAGTCACTGGAGTAATGAC-30 ; HHV-7 reverse prim-
er, 50 -CCAATCCTTCCGAAACCGAT-30 ; and HHV-7 probe, 50 -FAM-
CCTCGCAGATTGCTTGTTGGCCATG-TAMRA-30 (Gautheret-Dejean
et al., 2002; Hara et al., 2002). The thermal profile was 95
C for
30 s, followed by 50 cycles of 95
C for 5 s and 60
C for 31 s. Data
analysis was performed using Sequence Detection Software
version 1.4 (Life Technologies, California, USA). The sample below
the detection limits (20 copies/mL) adopted the detection limit
value. The HHV-6 and -7 DNA levels (copies/mL) were log-
transformed (log 10).
The Chalder Fatigue Scale (CFS) for chronic fatigue assessment
contained 14 items, which had been divided into two subscales:
physical symptoms (CFS-P) and mental symptoms (CFS-M). Each
answer was scored from 0 to 3. The CFS-P contained eight items
(score range; 0–24), and the CFS-M contained six items (score
range; 0–18). A higher score indicated that the level of fatigue was
severe. In this study, we assessed the participants using the
subscales (CFS-P, CFS-M).
We also used other assessments during the study. Depressive
symptoms were assessed by the Center for Epidemiologic Studies-
Depression Scale (CES-D; Radloff, 1977), which contained 20 items
(score range; 0–60); a score of over 16 points was graded as being
depressed. Physical activity levels for the past week were assessed
by the Physical Activity Scale for the Elderly (PASE; Hagiwara, Ito,
Sawai, & Kazuma, 2008), which had 12 components pertaining to
leisure time activity (PASE-L; five components), household activity
(PASE-H; six components), and work-related activity (PASE-W; one
component) over the past seven days.
In addition, four assessments were carried out in CG. The
caregiver burden was assessed using the ZBI, and the cognitive
functioning of the patients with dementia was assessed with the
Mini-Mental State Examination (MMSE; Folstein, Folstein, &
McHugh, 1975). Daily living/instrumental activities of daily living
(ADL/IADL) score for the people with dementia was assessed using
the Assessment of Motor and Process Skills (AMPS; Fisher & Bray,
2012), which was an observational evaluation used by occupa-
tional therapists in the evaluation of the quality of ADL/IADL. The
motor ability (need for physical effort) and process ability
(efficiency) were computed from the raw scores by the AMPS
computer-scoring software. Eighty-six percent of persons with
motor ability measures above 1.5 logits and process ability
measures above 1.0 logits can be independent in the community.
Behavior disturbances for people with dementia were assessed
using the Dementia Behavior Disturbance Scale (DBD 13; Machida,
2012), which contained 13 items (score range; 0–52). Higher scores
indicated more severe behavioral disturbances.
2.3. Statistical analysis
Statistical analyses were carried out using SPSS version 17.0.
(IBM, New York, USA). A P value of <0.05 was considered
statistically significant. The Shapiro–Wilk test was used for the
test of normality. The t-test, Mann–Whitney test, or Chi-Square
test was used for the comparisons between CG and NCG. In all
subjects or in CG, we computed the Pearson’s partial correlation
44 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48

coefficient or Spearman’s partial rank correlation coefficient with Table 2

Characteristics of people with dementia.
control variables (age, sex, and the presence or absence of disease
that may cause chronic fatigue) to estimate the association Age (years) 80.0 (8.2)
between each assessment. Sex (male: female) 23:21
number of days that they used a day care service per week 3.0 (2.0–5.0)
3. Results number of days that they used a stay-care service per week 0.0 (0.0–1.8)
MMSE 13.0 (6.0–17.0)

3.1. Normality of the data AMPS

Motor ability (logits) 1.4 (0.8–1.8)
In CG, the data of age (both people with dementia and Process ability (logits) 0.2 (0.7)
caregiver), HHV-6, HHV-7, CFS-P, CFS-M, ZBI, CES-D, AMPS Process DBD 19.2 (9.2)

Ability, and DBD are normal distribution. In NCG, the data of
HHV-6, PASE total and PASE-L are normal distribution. In all
subjects, the data of caregivers’ (or controls) age and HHV-6 are
normal distribution.
3.2. Characteristics of CG and NCG
MMSE, Mini-Mental State Examination; AMPS, Assessment of Motor and Process
Skills; DBD, Dementia Behavior Disturbance scale.
The data of age, AMPS process ability, and DBD are expressed as the mean (standard
deviation).
The data of the number of days that they used for day care service per week, number
of days that they used for stay-care service per week, MMSE and AMPS motor ability
are expressed as the median (interquartile range).

Table 1 shows the characteristics of CG and NCG. Most

caregivers were female (77%) and spouses (64%) of the people
with dementia. Diseases that may cause chronic fatigue (such as
cirrhosis, chronic renal failure, cancer, and thyroid disease) were
found in approximately one-fourth of CG. The mean (standard
deviation; SD) score of the ZBI was 40.8 (17.4), which indicated that
many CG had moderate burdens during caregiving.
We also compared the characteristics of CG with those of NCG
(Table 1). There were no significant differences in the caregivers’
(or controls) age, sex, and the presence/absence of a disease that
may be the cause of chronic fatigue. The median (interquartile
range; IQR) score for the CES-D in CG was 15.0 (9.0–25.0), which
was significantly (P = 0.003) higher than that in NCG [9.0
(5.0–14.0)]. The median (IQR) score for the PASE-L in CG was
14.9 (5.0–25.8), which was significantly (P < 0.000) lower than that
in NCG [39.2 (21.5–62.6)]. The median (IQR) score for the PASE-H in
CG was 85.0 (85.0–121.0), which was significantly (P < 0.000)
higher than that in NCG [50.0 (50.0–86.0)]. There were no
significant differences between the median (IQR) of the PASE
total score and the PASE-W score.
3.3. Characteristics of people with dementia
Table 2 shows the characteristics of people with dementia. The
mean age (SD) of people with dementia was 80.0 (8.2) and the sex
ratio was almost 1:1. The median (IQR) score of the MMSE was 13.0
(6.0–17.0), which indicated that many of the people with dementia
had moderate cognitive disorders. The median (IQR) score for the
AMPS motor ability and the mean (SD) score for the AMPS process
ability were 1.4 (0.8–1.8) logits and 0.2 (0.7) logits, respectively,
which indicated that many people with dementia were unable to
live independently in their communities. In particular, the process
ability scores were low. The mean score (SD) of the DBD was 19.2
(9.2), which indicated that most people with dementia had some
behavioral disturbances.
3.4. HHV and CFS in CG and NCG
In all subjects, salivary HHV-6 or -7 DNA levels were below the
detection limits for one person or five persons. It was unclear that
the subjects get infected or not with the virus.
The mean level (SD) of HHV-6 DNA in CG was 3.04 (0.63), which
was significantly (P = 0.030) higher than that in NCG [2.78 (0.47)]
(Fig. 1(a)). There was no significant (P = 0.077) differences in the
median HHV-7 DNA level (IQR) between CG [4.19 (3.50–4.98)] and
NCG [4.78 (4.05–5.45)] (Fig. 1(b)).
The median score (IQR) of CFS-P in CG was 11.0 (9.0–16.0),
which was significantly (P < 0.000) higher than that in NCG [7.0
(4.0–9.0)] (Fig. 2(a)). The median score (IQR) of the CFS-M in CG

Table 1
Characteristics of CG and NCG.

CG (n = 44) NCG (n = 50) P-value

Age (years) 69.0 (60.0–75.0) 67.0 (66.0–72.0) 0.877
Sex (male: female) 10:34 18:32 0.160
Disease that may cause chronic fatigue (presence: absence) 12:31 8:42 0.254
Relation to people with dementia 28:15:1 – –
(spouse: child: others)
Number of other family caregivers 1.0 (0.0–1.0) – –
Caregiving hours per week 46.5 (23.1) – –
ZBI 40.8 (17.4) – –
CES-D 15.0 (9.0–25.0) 9.0 (5.0–14.0) 0.003
PASE total 129.6 (93.6–161.9) 112.6 (93.5–133.9) 0.229
PASE-L 14.9 (5.0–25.8) 39.2 (21.5–62.6) 0.000
PASE-H 85.0 (85.0–121.0) 50.0 (50.0–86.0) 0.000
PASE-W 0.0 (0.0–0.0) 0.0 (0.0–6.0) 0.501

CG, the caregiver group; NCG, the non-caregiver group; ZBI, Zarit Caregiver Burden Interview; CES-D, Center for Epidemiologic Studies-Depression scale; PASE, Physical
Activity Scale for the Elderly; PASE-L, PASE-leisure; PASE-H, PASE-house; PASE-W, PASE-work.
The data of caregiving hours per week and ZBI are expressed as the mean (standard deviation).
The data of age, number of other family caregivers, CES-D, PASE total, PASE-L, PASE-H, and PASE-W are expressed as the median (interquartile range).
In the question of presence or absence of a disease that may cause chronic fatigue, one subject gave no response.
Mann–Whitney test was used for comparison of the data concerning age, CES-D, PASE total, PASE-L, PASE-H, and PASE-W between groups.
Chi-Square test was used for comparison of the data of sex and the presence or absence of disease that may cause chronic fatigue.
 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48 45

Fig. 1. Comparison of salivary HHV-6 and -7 DNA levels between groups.

Fig. 2. Comparison of CFS-P and CFS-M between groups.

Fig. 3. Correlations between HHV-6 and MMSE or CES-D.

46 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48
was 7.0 (5.0–11.0), which was significantly (P = 0.039) higher than
that in NCG [6.0 (4.0–8.0)] (Fig. 2(b)).
3.5. Correlation between HHV and CFS in all subjects
In all subjects, the Spearman’s partial rank correlation
coefficient with control variables [caregivers’ (or controls) age,
sex, and the presence or absence of a disease that may cause
chronic fatigue] was calculated to estimate the association
between HHV-6, HHV-7, CFS-P, and CFS-M. Except for the
correlations within assessments, there were no significant
correlations between HHV-6, HHV-7, CFS-P, and CFS-M. In
addition, there was no significant correlation between HHV-6
and HHV-7, and there was a significant correlation between CFS-P
and CFS-M (P < 0.000, r = 0.665).
3.6. Factors associated with HHV or CFS in CG
In CG, the Pearson’s partial correlation coefficient (shown as ‘r’)
or Spearman’s partial rank correlation coefficient (shown as ‘r’)
with control variables (caregivers’ age, sex, and the presence or
absence of a disease that may cause chronic fatigue) was calculated
to estimate the associations between HHV-6, HHV-7, CFS-P, or CFS-
M and other factors. The associations that had a significant
correlation were as follows: HHV-6 vs. MMSE (P = 0.014, r = 0.422)
(Fig. 3(a)), vs. AMPS motor (P = 0.047, r = 0.401), and vs. CES-D
(P = 0.008, r = 0.427) (Fig. 3(b)); CFS-P vs. ZBI (P < 0.000, r = 0.780),
vs. DBD (P = 0.011, r = 0.404), vs. CES-D (P < 0.000, r = 0.650), vs.
PASE-L (P = 0.011, r = 0.396) and vs. caregiving hours per week
(P = 0.044, r = 0.424); CFS-M vs. ZBI (P = 0.011, r = 0.399), vs. AMPS
Process Ability (P = 0.025, r = 0.439), vs. DBD (P = 0.008, r = 0.420),
vs. CES-D (P < 0.000, r = 0.693), vs. PASE-L (P = 0.006, r = 0.426)
and number of other family caregivers (P = 0.005, r = 0.436).
4. Discussion
In this study, we found that caregivers’ salivary HHV-6 DNA
levels were higher than non-caregivers’. Reactivation of HHV-6 has
been associated with over-production of some cytokines (Kondo,
2009). Therefore, HHV-6 might be reactivated by a strong
generalized immunological response (Kondo & Yamanishi,
2007). The reactivation of HHV-6 has been caused by overwork;
this has also been found in patients with chronic fatigue syndrome
(Kondo, 2007). It has been also reported that moderate- to long-
term fatigue after cognitive task trials was positively associated
with the salivary HHV-6 DNA levels (Tanaka et al., 2012). In this
study, we found that caregivers suffer from immune dysfunction
and chronic fatigue for the assessment of salivary HHV-6 DNA
levels. In family caregivers of dementia patients, high blood
pressure (Chattillion et al., 2013), high concentrations of cortisol
levels in hair (Stalder et al., 2014), and low nutritional status
(Rullier et al., 2014) have also been identified. HHV-6 reactivation
may be added as a new biomarker for exhaustion in the assessment
of caregiver burden.
In this study, caregivers had lower leisure physical activity,
higher household chore physical activity, and depression symp-
toms; these findings have been reported previously (Arai,
Kumamoto, Mizuno, & Washio, 2014; Hirano et al., 2011). Our
study confirmed the results of previous studies. Physical activity,
particularly leisure activities, was found to be inversely correlated
with care burden (Hirano et al., 2011). It is considered that the
caregivers had severe psychological stress. From a different
perspective, the high caregivers’ salivary HHV-6 DNA levels might
be induced by their psychological stress. Typically, if subjective
fatigue is severe, physical activity is performed at a low level
(Timmerman, Dekker-van Weering, Tönis, Hermens, &
Vollenbroek-Hutten, 2015). However, in this study, the caregivers
had a similar degree of physical activity, despite severe fatigue
(based on CFS score), compared with that in the non-caregivers.
The results suggest that the caregivers were engaged in the forced
activity of caregiving, although they had psychological problems.
In the patients with chronic fatigue syndrome, post-exertional
malaise which can occur even after minimal exercise is a common
symptom that includes increased fatigue, muscular pain, head-
ache, nausea, and physical weakness; it was found to be linked to
immunological dysregulations that persisted for up to 48 h after
exertion (Lengert & Drossel, 2015). Although the subjects in our
study were not the patients with chronic fatigue syndrome, it is
considered that caregivers were forced physical activity despite
their severe psychological stress. Therefore, caregivers’ salivary
HHV-6 DNA levels may be higher than non-caregivers’ due to the
similar way of post-exertional malaise.
We also found that the HHV-6 DNA saliva levels were positively
correlated with the MMSE scores and the AMPS motor ability
scores and negatively correlated with CES-D. The trend for the high
caregivers’ salivary HHV-6 DNA levels may be induced by a high
MMSE score and a high AMPS motor ability score can be explained
as follows. In this study, the median (IQR) of MMSE was 13.0 (6.0–
17.0) and the AMPS motor ability was at a higher level compared
with the process ability. Therefore, the almost dementia subjects
were in the middle- to severe-stages of progression. At these
stages, caregivers of people with dementia who are in the middle
stage of progression may have more stressful experience than
those of them who are in the severe stage of progression. Kamiya
et al. (2014) demonstrated that the caregiver burden of caregivers
of people with Alzheimer’s disease that have an MMSE score of 12–
17 is associated with more factors (such as behavioral and
psychological symptoms of dementia, IADL/ADL abilities of
dementia, and presence or absence of comorbid conditions of
geriatric syndrome) than those with an MMSE score of 0–11. From
this, the caregivers’ salivary HHV-6 DNA levels may be associated
with the level of caregiver burden caused by the level of dementia.
The trend for the low HHV-6 DNA levels in caregivers who had a
high CES-D score can be simply explained. Generally, if the
depressive symptoms got worse, physical activity was suppressed.
The caregivers’ HHV-6 DNA saliva levels may be associated with
the changes in the physical activity levels, which may be caused by
a mental condition.
We also found that both caregivers’ CFS-P and CFS-M scores,
which is commonly used in the assessment of chronic fatigue, were
higher than non-caregivers’. Both CFS-P and CFS-M were positively
correlated with ZBI, CES-D, and DBD and negatively correlated with
PASE-L. The CFS-P score increased with a rise in caregiving hours
and the CFS-M score increased with a decrease in the number of
other family caregivers. The results that CFS correlated with multi
dimensions suggested that CFS was a useful assessment technique
for the estimation of the caregivers’ burden and fatigue. Both CFS-P
and CFS-M, however, were not associated with salivary HHV-6 and
-7 DNA levels. This may have occurred because CFS is a
questionnaire that estimates “subjective” fatigue. Because subjec-
tive fatigue may be influenced by several factors (e.g., reward and
sense of accomplishment), the discrepancy between subjective
and objective fatigue may be very common (Kondo, 2009). From
the results that indicated the factors associated with salivary HHV-
6 DNA levels, CFS-P, or CFS-M were different, each assessment may
estimate the fatigue from different viewpoints.
Although HHV-7 is a close relative to the HHV-6 virus, there was
no significant difference in the HHV-7 DNA saliva levels between
the two groups, and there were no significant associations between
the HHV-7 DNA saliva levels and other variables. In addition, non-
caregivers’ salivary HHV-7 DNA levels tended to be higher than
caregivers’. The reasons for these findings are unclear. HHV-6 is
 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48
believed to have a life-long latency in the macrophage. If the virus-
infected cell is stimulated, reactivation may occur in the salivary
gland and virus will be shed in the saliva. In contrast, HHV-7 virus
is always maintained in the body at low level, and is shed in saliva
through viral reactivation and multiplication in salivary gland
(Kondo, 2007). These results may be caused by the difference in the
system of reactivation or the difference in the factors, which induce
reactivation.
There were several limitations in this study. Firstly, the cross-
sectional design of this study was a limitation. Therefore, even if
we had found significant correlations between some variables, it
would be particularly difficult to determine if those were causal
relationships or not. Secondly, it may be necessary to determine
HHV-6 and -7 DNA saliva levels at two points. These levels were
significantly different inter-individually. We might have to
determine the HHV DNA saliva levels on two different occasions.
A randomized controlled trial, especially one that assesses HHV-6
and -7 DNA levels and CFS after an intervention to relieve the
caregiver’s burden, is warranted. Thirdly, we reported several
positive correlations that are not statistically significant. Howev-
er, there might be a relatively high likelihood of a Type II
statistical error. Our study had a relatively small number of
subjects, and therefore, it might have insufficient power. It might
be necessary to conduct research with a larger sample size.
Finally, from the median (IQR) score of MMSE of people with
dementia, the dementia subjects in this study were heteroge-
neous. It has been reported that the MMSE of people with
Alzheimer’s disease or related disorders was negatively correlat-
ed with caregiver burden (Dauphinot et al., 2015). In this study,
we could not divide the dementia subjects into several homoge-
neous groups, because of the small sample size. In future studies,
the group of people with dementia should be more homogeneous
for the MMSE parameter.
5. Conclusions
In this study, we found that salivary HHV-6 DNA levels in CG
were higher than those in NCG. This result suggested that salivary
HHV-6 DNA levels may be added as a new biomarker for
exhaustion in caregivers. We also found that salivary HHV-6
DNA levels in CG may be associated with the restriction of leisure
physical activity, the level of dementia, and the depressive
symptoms. We concluded that the assessment of fatigue, which
was performed not only by using a questionnaire such as the CFS
but also by using a biomarker such as the HHV-6, should be
conducted when estimating the caregiver burden for family
caregivers of dementia patients.
Conflicts of interest
Kazuhiro Kondo submitted a patent application titled “Methods
For Assessing Fatigue Level and Applications Thereof”
(US20110020789 A1). The other authors report no competing
interests.
Acknowledgments
We thank all participants for their cooperation. This study was
supported by the Ministry of Education, Culture, Sports, Science
and Technology (MEXT) KAKENHI Grant Number 26461758 and
MEXT-Supported Program for the Strategic Research Foundation at
Private Universities [grant number S1201032].
The authors would like to thank Enago (www.enago.jp) for the
English language review.
47
References
Arai, Y., Kumamoto, K., Mizuno, Y., & Washio, M. (2014). Depression among family
caregivers of community-dwelling older people who used services under the
Long Term Care Insurance program: a large-scale population-based study in
Japan. Aging & Mental Health, 18(1), 81–91.
Brodaty, H., & Donkin, M. (2009). Family caregivers of people with dementia.
Dialogues in Clinical Neuroscience, 11(2), 217–228.
Chalder, T., Berelowitz, G., Pawlikowska, T., Watts, L., Wessely, S., Wright, D., et al.
(1993). Development of a fatigue scale. Journal of Psychosomatic Research, 37(2),
147–153.
Chattillion, E. A., Ceglowski, J., Roepke, S. K., von Känel, R., Losada, A., Mills, P. J., et al.
(2013). Pleasant events, activity restriction, and blood pressure in dementia
caregivers. Health Psychology, 32(7), 793–801.
Dauphinot, V., Delphin-Combe, F., Mouchoux, C., Dorey, A., Bathsavanis, A.,
Makaroff, Z., et al. (2015). Risk factors of caregiver burden among patients with
Alzheimer’s disease or related disorders: a cross-sectional study. Journal of
Alzheimer’s Disease, 44(3), 907–916.
Fisher, A. G., & Bray, K. (2012). Assessment of motor and process skills, 7th ed.
Colorado: Three Star Press, Inc..
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-Mental State: a practical
method for grading the cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12, 189–198.
Fukuda, S., Koyama, H., Kondo, K., Fujii, H., Hirayama, Y., Tabata, T., et al. (2015).
Effects of nutritional supplementation on fatigue, and autonomic and immune
dysfunction in patients with end-stage renal disease: a randomized, double-
blind, placebo-controlled, multicenter trial. PLoS One, 10(3) .
Gautheret-Dejean, A., Manichanh, C., Thien-Ah-Koon, F., Fillet, A. M., Mangeney, N.,
Vidaud, M., et al. (2002). Development of a real-time polymerase chain reaction
assay for the diagnosis of human herpesvirus-6 infection and application to
bone marrow transplant patients. Journal of Virological Methods, 100(1-2), 27–
35.
Hagiwara, A., Ito, N., Sawai, K., & Kazuma, K. (2008). Validity and reliability of the
Physical Activity Scale for the Elderly (PASE) in Japanese elderly people.
Geriatrics and Gerontology International, 8(3), 143–151.
Hara, S., Kimura, H., Hoshino, Y., Tanaka, N., Nishikawa, K., Ihira, M., et al. (2002).
Detection of herpesvirus DNA in the serum of immunocompetent children.
Microbiology and Immunology, 46(3), 177–180.
Hirano, A., Suzuki, Y., Kuzuya, M., Onishi, J., Hasegawa, J., Ban, N., et al. (2011).
Association between the caregiver’s burden and physical activity in
community-dwelling caregivers of dementia patients. Archives of Gerontology
and Geriatrics, 52(3), 295–298.
Ito, T., Urushima, H., Sakaue, M., Yukawa, S., Honda, H., Hirai, K., et al. (2014).
Reduction of adverse effects by a mushroom product, active hexose correlated
compound (AHCC) in patients with advanced cancer during chemotherapy—the
significance of the levels of HHV-6 DNA in saliva as a surrogate biomarker
during chemotherapy. Nutrition and Cancer, 66(3), 377–382.
Kamiya, M., Sakurai, T., Ogama, N., Maki, Y., & Toba, K. (2014). Factors associated with
increased caregivers’ burden in several cognitive stages of Alzheimer’s disease.
Geriatrics and Gerontology International, 14(Suppl. 2), 45–55.
Kondo, K., Yamanishi, K., et al. (2007). HHV-6A, 6B, and 7: molecular basis of latency
and reactivation. In A. Arvin, G. Campadelli-Fiume, E. Mocarski, P. S. Moore, B.
Roizman, & R. Whitley (Eds.), Human herpesviruses: biology, therapy, and
immunoprophylaxis (pp. 843–849).Cambridge: Cambridge University Press.
Kondo, K. (2007). Chronic fatigue syndrome and herpesvirus reactivation. Nihon
Rinsyo, 65(6), 1043–1048 (in Japanese).
Kondo, K. (2009). Biomarker for fatigue: human herpesvirus 6 (HHV-6) reactivation
in saliva. Igakuno Ayumi, 228(6), 664–668 (in Japanese).
Lee, S., Colditz, G. A., Berkman, L. F., & Kawachi, I. (2003). Caregiving and risk of
coronary heart disease in U.S. women: a prospective study. American Journal of
Preventive Medicine, 24(2), 113–119.
Lengert, N., & Drossel, B. (2015). In silico analysis of exercise intolerance in myalgic
encephalomyelitis/chronic fatigue syndrome. Biophysical Chemistry, 202, 21–31.
Machida, A. (2012). Estimation of the reliability and validity of the short version of
the 28-item Dementia Behavior Disturbance scale. Nihon Ronen Igakkai Zasshi,
49(4), 463–467 (in Japanese).
Morris, G., Berk, M., Walder, K., & Maes, M. (2015). The putative role of viruses,
bacteria, and chronic fungal biotoxin exposure in the genesis of intractable
fatigue accompanied by cognitive and physical disability. Molecular
Neurobiology (Epub ahead of print).
Morriss, R. K., Wearden, A. J., & Mullis, R. (1998). Exploring the validity of the Chalder
Fatigue scale in chronic fatigue syndrome. Journal of Psychosomatic Research, 45
(5), 411–417.
Radloff, L. S. (1977). The CES-D scale: a self-report depression scale for research in
the general population. Applied Psychological Measurement, 1, 385–401.
Rullier, L., Lagarde, A., Bouisson, J., Bergua, V., Torres, M., & Barberger-Gateau, P.
(2014). Psychosocial correlates of nutritional status of family caregivers of
persons with dementia. International Psychogeriatrics, 26(1), 105–113.
Schulz, R., & Beach, S. R. (1999). Caregiving as a risk factor for mortality: the
Caregiver Health Effects Study. Journal of the American Medical Association, 282
(23), 2215–2219.
Stalder, T., Tietze, A., Steudte, S., Alexander, N., Dettenborn, L., & Kirschbaum, C.
(2014). Elevated hair cortisol levels in chronically stressed dementia caregivers.
Psychoneuroendocrinology, 47, 26–30.
48 T. Osaki et al. / Archives of Gerontology and Geriatrics 66 (2016) 42–48

Tanaka, M., Shigihara, Y., Funakura, M., Kanai, E., & Watanabe, Y. (2012). Fatigue-
associated alterations of cognitive function and electroencephalographic power
densities. PLoS One, 7(4), e34774.
Timmerman, J. G., Dekker-van Weering, M. G., Tönis, T. M., Hermens, H. J., &
Vollenbroek-Hutten, M. M. (2015). Relationship between patterns of daily
physical activity and fatigue in cancer survivors. European Journal of Oncology
Nursing, 19(2), 162–168.
Whitley, R., Kimberlin, D. W., Prober, C. G., et al. (2007). Pathogenesis and disease. In
A. Arvin, G. Campadelli-Fiume, E. Mocarski, P. S. Moore, B. Roizman, & R. Whitley
(Eds.), Human herpesviruses: biology, therapy, and
immunoprophylaxisCambridge: Cambridge University Press.
Zarit, S. H., Reever, K. E., & Bach-Peterson, J. (1980). Relatives of the impaired elderly:
correlates of feelings of burden. Gerontologist, 20, 649–655.