Biomedical Research 29 (6) 317-320, 2008
The facial massage reduced anxiety and negative mood status, and increased
sympathetic nervous activity
Tomoko Hatayama, Shingo Kitamura, Chihiro Tamura, Mayumi Nagano and Koichiro Ohnuki
User Science Institute, Kyushu University, Fukuoka, Japan
(Received 20 October 2008; and accepted 24 October 2008)
The aim of this study was to clarify the effects of 45 min of facial massage on the activity of au-
tonomic nervous system, anxiety and mood in 32 healthy women. Autonomic nervous activity was
assessed by heart rate variability (HRV) with spectral analysis. In the spectral analysis of HRV, we
evaluated the high-frequency components (HF) and the low- to high-frequency ratio (LF/HF ra-
tio), reecting parasympathetic nervous activity and sympathetic nervous activity, respectively.
The State Trait Anxiety Inventory (STAI) and the Prole of Mood Status (POMS) were adminis-
tered to evaluate psychological status. The score of STAI and negative scale of POMS were sig-
nicantly reduced following the massage, and only the LF/HF ratio was signicantly enhanced
after the massage. It was concluded that the facial massage might refresh the subjects by reducing
their psychological distress and activating the sympathetic nervous system.
Body massage treating large muscles located in up-
per limb, shoulders and back brings many psycho-
logical and physiological effects such as decreasing
anxiety and depression, increasing attentiveness, re-
lieving stress and pain, and reducing blood pressure
(1–3, 10, 13).
In the case of facial massage, a popular cosmetic
technique preferred by women, much anecdotal evi-
dence has been reported that facial massage can
enhance relaxation just as well as body massage.
However, very few studies have provided scientic
evidence, such as physiological or psychological
data. To date, only two studies (7, 17) have reported
the change in psychological parameters and electro-
encephalogram (EEG) data following aesthetic facial
massage. They observed higher subjective scores in
both general deactivation and deactivation-sleep fac-
tors, signicant reduction of anxiety, and a marked
attenuation of the alpha-waves accompanied by a
small increase in theta-waves during massage, an ef-
fect which they attributed to sleep induction by fa-
cial massage. Therefore, facial massage might have
positive effects not only on beauty, but also on psy-
chological status or cerebral activity.
In the case of EEG, often measured in preceding
studies, it is difcult to avoid equipment noises and
the misalignment of electrodes in treating facial
muscles. Analysis of autonomic nervous activity us-
ing heart rate variability derived from electrocardio-
gram (ECG) through a chest lead is available as an
alternative physiological measurement. However,
there have been no studies evaluating the possible
effect of facial massage using such a method. Fur-
thermore, even though Inoue et al. (6) and Hughes
and Stoney (5) reported that changes in mood state
were associated with autonomic nervous system ac-
tivity, no study has investigated the effects of facial
massage from both physiological and psychological
The present study therefore tested the following 2
Address correspondence to: Tomoko Hatayama
User Science Institute, Kyushu University
4-9-1 Shiobaru, Minami-ku, Fukuoka Japan 815-8540
Tel & Fax: +81-92-553-9458
This research was undertaken as outsourced work from
the Ministry of Education, Culture, Sports, Science and
Technology (science and technology advancement ad-
justment expense; referred to as Kyushu University
User Science Institute).
T. Hatayama et al.318
low frequency (LF, 0.04–0.15 Hz) and high frequen-
cy (HF, 0.15–0.4 Hz). The ratio of LF power to HF
power (LF/HF ratio) was also assessed. The HF
component represents parasympathetic nervous ac-
tivity (11, 12) while the LF/HF component is a rep-
resentation of sympathetic nervous activity (8, 18).
POMS-short form. POMS-short form, which had
been shortened and translated into Japanese (19),
was administered. The POMS-short form comprised
of 30 questions about the current mood state. These
30 questions were classified in 6 sub-scales: T-A
(tension and anxiety), D (depression and dejection),
A-H (anger and hostility), V (vigor), F (fatigue),
and C (confusion). The subjects selected the raw
score from one of five values (0, 1, 2, 3 and 4,
where 0 = no such mood state and 4 = extreme
mood state). These raw scores in each sub-scale
were then added to generate each sub-scale score.
STAI-JYZ. The STAI (15) is a 20-item scale that
measures acute level of anxiety. The subjects select-
ed the raw score from one of four values (1, 2, 3
and 4, where 1 = not at all and 4 = very much). A
summary score is obtained by adding the weight of
each item. The STAI scores indicate an increase in
response to situational stress and a decline under re-
laxing conditions. In this study, STAI-JYZ, which
regards Japanese cultural factors better, was used.
The STAI-JYZ exhibits acceptable internal consis-
tency and test-retest reliability (4).
Statistical analysis. Each data was expressed as
means ± standard error (SE). Student’s paired t test
was used to evaluate, and statistics were calculated
with SPSS version 14 (SPSS Inc. Chicago, Illinois,
USA). Results were considered signicance at P<
LF/HF ratio remarkably increased from 0.81 ± 0.11
to 1.22 ± 0.19 after the massage (Fig. 1, P= 0.035),
even though the HF power and heart rate remained
unchanged (data not shown).
Fig. 2 shows the differences in average scores be-
tween pre-massage and post-massage for the six
subscales of the POMS test. Most scores, which in-
dicate negative feelings, tended to decrease after the
massage. Especially, T-A (tension and anxiety) sig-
nicantly declined from 3.66 ± 0.62 to 1.16 ± 0.33,
and also F (fatigue) dropped 2.19 ± 0.58 to 0.81 ±
0.24 (P< 0.001). The anger and vigorous scores
1) There will be an enhancement of physiological
relaxed status in the activity of autonomic ner-
vous system assessed using a power spectral anal-
ysis of heart rate variability following a facial
2) There will be a reduction in psychological dis-
tress as evaluated using the Profile of Mood
States (POMS) and the State Trait Anxiety Inven-
tory (STAI) following a facial massage.
MATERIALS AND METHODS
Subjects. Data was collected from 32 healthy Japa-
nese women ranging in age from 20 to 40 years
(mean, 28.5 years). After each subject had been
briefed on the objective and method of the experi-
ment, written informed consent was obtained.
Protocol. Measurements were carried out before and
after a facial massage in a private room set at an
ambient temperature average 26.8 ± 1.2°C. The fa-
cial massage consisted of four steps: (1) cleansing
with cleansing cream by hand, (2) removal of kera-
tin with steaming, (3) ultrasonic wave facial treat-
ment with gel, and (4) application of lotion for
moisture retention. The total time for this treatment
was 45 min and it was carried out by trained, expe-
The subjects were first asked to complete the
POMS-short form and STAI-JYZ in sitting position,
keeping the electrodes on their chests. The ECG
was recorded for 10 min both before and after the
massage, with subjects lying on their backs. The
data obtained between third to sixth minute was
used for power spectral analysis. The details of
measurement are described below.
Electrocardiogram procedures. ECG was recorded
with MWM-01 monitor (GMS Inc, Tokyo, Japan)
and was analyzed by the MemCalc system (Mem-
Calc/Makin2; GMS Inc) on an on-line computer
(VAIO, VGN-T70B; Sony Co., Tokyo, Japan).
ECG was recorded from electrodes placed in the
CM5 position. In order to evaluate cardiac autonom-
ic nervous activity, a power spectral analysis of the
temporal intervals between heart beats (R-R inter-
vals) was used. Power spectral analyses using the
MemCalc system were performed for each 5 second
period. We obtained the R-R intervals (ms) as the
time domain index of heart rate variability (HRV).
The power of HRV was quantied by determining
the areas of the spectrum in 2 component widths:
The effects of facial massage 319
cal and psychological aspects, by evaluating auto-
nomic nervous activity and mental status.
Anxiety in the STAI and negative moods in
POMS were significantly decreased following the
massage. These results are consistent with previous
ndings (10, 13). In the present study, the score of
the STAI and the negative scales of POMS at base-
line were slightly lower as compared with the corre-
sponding normal values for healthy women cited in
the guidelines of Japanese version of the STAI and
POMS. Despite relatively low negative psychologi-
cal scores before the facial massage, the scores fol-
lowing facial massage significantly decreased, as
well as body massage. These results suggested that
the facial massage had strong effects on stress alle-
viation or psychological relaxation.
Contrary to our prediction, sympathetic nervous
activity increased following the massage. It is spec-
ulated that the discrepancy was simply due to differ-
ences in the timing of measurement. According to
preceding studies, heart rate and LF/HF ratio de-
clined during the massage, but these affects almost
subsided shortly after massage (13, 16). It is not
surprising that parasympathetic nervous activity was
enhanced during the massage, because the subjects
stayed in a relaxed position or their eyes closed dur-
ing the massage.
Selye (14) once announced that a stressor, which
was a stimulus inducing a stress reaction to biologi-
cal body, could not only be distress but also eu-
stress. Eustress contains positive stimuli such as
showed no signicant changes.
Similarly, the score for anxiety in the STAI sig-
nicantly declined from 36.25 ± 1.33 to 28.25 ± 1.04
following the massage (P< 0.001).
To date, many studies have reported positive psy-
chological and physiological effects of body mas-
sage when performed as a method of psychosomatic
conditioning. However, very little research exists on
aesthetic facial massage, which is usually performed
expecting not psychosomatic treatment but rather
beautication of facial skin. We therefore examined
the effects of facial massage from both physiologi-
Fig. 1 Change of LF/HF ratio before and after facial mas-
sage. Error bars indicate ± SE, n = 32, *:P< 0.05
Fig. 2 Changes in mean POMS test score before and after facial massage. Each value represents the mean score for six
sub-scales of the POMS test, and is expressed as the mean ± SE (n = 32, *:P< 0.05). T-A (tension and anxiety), D (depres-
sion and dejection), A-H (anger and hostility), V (vigor), F (fatigue), and C (confusion).
T. Hatayama et al.320
taking a bath, deep or sweet sleep, mental satisfac-
tion, and so on (9). It was thus speculated that the
increased sympathetic nervous activity following the
facial massage in the present study might be a posi-
tive stress reaction to a stressor which is classied
as eustress. Taking these results into consideration,
such a mixed status of psychologically relaxed and
physically-activated might well be regarded as re-
freshment rather than relaxation.
Heart rate after facial massage did not change
compared to pre-massage values in the present
study. It was guessed that the facial massage treat-
ing relatively-small facial muscles did not change
the circulation as much as heart rate was affected.
In conclusion, the present study could demon-
strate that facial massage reduces psychological dis-
tress and activates the subjects physically. Taken
together, it might be reasonable that facial massage
originally performed as a cosmetic treatment has an
effect of refreshment rather than relaxation. Thus, it
is expected that using psychological and physiologi-
cal evaluation methods simultaneously enables us to
extract more appropriate information and better as-
sess the effects of various stimuli.
The present study had several limitations, such as
the lack of measurement during the massage period
and a control group. Further study should be de-
signed as a randomized controlled trial, and the data
should be continuously collected through the experi-
This research was nancially supported by Tina-pri
Corporation. We thank the massage therapists and
the assistants who helped with this research.
1. Diego MA, Field T, Sanders C and Hernandez-Reif M (2004)
Massage therapy of moderate and light pressure and vibrator
effects on EEG and heart rate. Int J Neurosci 114, 31–44.
2. Field T, Morrow C, Valdeon C, Larson S, Kuhn C and
Schanberg S (1992) Massage reduces anxiety in child and
adolescent psychiatric patients. J Am Acad Child Adolesc
Psychiatry 31, 125–131.
3. Field T, Ironson G, Scafidi F, Nawrocki T, Goncalves A,
Burman I, Pickens J, Fox N, Schanberg S and Kuhn C (1996)
Massage therapy reduces anxiety and enhances EEG pattern
of alertness and math computations. Int J Neurosci 86, 197–
4. Hidano T, Fukuhara M, Iwawaki M, Soga S and Spielberger
CD (2000) New STAI Manual State-Trait Anxiety Inventory-
Form JYZ. Jitsumu Kyouiku Press, Tokyo. (in Japanese)
5. Hughes JW and Stoney CM (2000) Depressed mood is relat-
ed to high-frequency heart rate variability during stressors.
Psychosom Med 62, 796–803.
6. Inoue N, Kuroda K, Sugimoto A, Kakuda T and Fushiki T
(2003) Autonomic nervous responses according to preference
for the odor of Jasmine tea. Biosci Biotechnol Biochem 67,
7. Jodo E, Yamada Y, Hatayama T, Abe T and Maruyama K
(1988) Effects of facial massage on the spontaneous EEG.
Tohoku Psychologica Folia 47, 8–15.
8. Koizumi K, Terui N and Kollai M (1985) Effect of cardiac
vagal and sympathetic nerve activity on heart rate in rhyth-
mic regulation. J Auton Nerv Syst 12, 251–259.
9. Lazarus RS and Folkman S (1984) Stress, Appraisal, and
Coping. Springer Publishing Company, Inc, New York.
10. McNamara ME, Burnham DC, Smith C and Carroll DL (2003)
The effects of back massage before diagnostic cardiac cathe-
terization. Altern Ther Health Med 9, 50–57.
11. Pagani M, Lombardi F, Guzzetti S, Rimoldi O, Furlan R,
Pizzinelli P, Sandrone G, Malfatto G, Dell’Orto S, Piccaluga
E, Turiel M, Baselli G, Cerutti S and Malliani A (1986) Pow-
erspectral analysis of heart rate and arterial pressure variabil-
ities as marker of sympathovagal interaction in man and
conscious dog. Circ Res 59, 178–193.
12. Akselrod S, Gordon D, Ubel FA, Shannon DC, Berger AC
and Cohen RJ (1981) Power spectrum analysis of heart rate
uctuation: a quantitative probe of beat-to-beat cardiovascu-
lar control. Science 213, 220–222.
13. Sato T (2006) Effects of hand massage on the human auto-
nomic system and mood in healthy women. Yamanashi Nurs-
ing Journal 4, 25–32. (in Japanese with English abstract)
14. Selye H (1974) Stress Without Stress. Lippincott, Philadel-
15. Spielberger CD, Gorusch TC and Lushene RE (1970) The
State Trait Anxiety Inventory. Consulting Psychologists Press,
Palo Alto, CA.
16. Tochikubo O, Ri S and Kura N (2006) Effects of pulse-
synchronized massage with air cuffs on peripheral blood ow
and autonomic nervous system. Circ J 70, 1159–1163.
17. Yamada Y, Hatayama T, Maruyama K, Abe T and Suzuki Y
(1986) A psychological effect of facial estherapy. Tohoku
Psychologica Folia 45, 6–16.
18. Yeragani VK, Srinivasan K, Vempati S, Pohl R and Balon R
(1993) Fractal dimension of heart rate time series: an effec-
tive measure of autonomic function. J Appl Physiol 75,
19. Yokoyama K (2005) POMS-short form handbook and case
study. Kaneko Shobo, Tokyo. (in Japanese)