Sei sulla pagina 1di 9


Complementary Therapies in Clinical Practice (2007) 13, 137–145

Evaluation of anxiety, salivary cortisol and

melatonin secretion following reflexology
treatment: A pilot study in healthy individuals
A.J. Mc Vicara, C.R. Greenwoodb, F. Fewella, V. D’Arcyb,
S. Chandrasekharanb, L.C. Alldridgeb,

Institute of Health and Social Care, Anglia Ruskin University, Bishop Hall Lane, Chelmsford,
Essex CM1 1SQ, UK
Helen Rollason Cancer Care Laboratory, Anglia Ruskin University, Bishop Hall Lane, Chelmsford,
Essex CM1 1SQ, UK

KEYWORDS Summary This pilot study sought to identify an appropriate methodology to

Reflexology; investigate the impact of reflexology in healthcare settings. The study involved
Stress; healthy volunteers to prevent unnecessary intervention to individuals who may
Biometric already be experiencing health related trauma. Thirty participants underwent either
evaluation; reflexology or no treatment (control), in a cross-over experimental design. Self-
Psychometric reported anxiety (Spielberger STAI), cardiovascular parameters (BP and pulse rate)
evaluation and salivary cortisol and melatonin concentrations were assessed before and after
reflexology. Control data were obtained at the same time points in identical
Reflexology had a powerful anxiety-reduction effect (‘state’; Po0.001) but no
significant effect on underlying anxiety (‘trait’). Cardiovascular parameters
decreased (Po0.001). Baseline salivary cortisol and melatonin were not significantly
correlated with STAI scores and did not change significantly following reflexology.
Reflexology reduced ‘state’ anxiety and cardiovascular activity within healthy
individuals, consistent with stress-reduction. Considering the connection between
stress/anxiety and well being, the effects of reflexology may have beneficial
outcomes for patients. These findings will be transferred to a study involving breast
cancer patients where effects may be more pronounced particularly since cancer
patients display disregulation of cortisol and melatonin secretion.
& 2006 Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +44 1245 493131x3098; fax: +44 1245 683095.
E-mail address: (L.C. Alldridge).

1744-3881/$ - see front matter & 2006 Elsevier Ltd. All rights reserved.
138 A.J. Mc Vicar et al.

Introduction the hypothalamus to directly influence circadian

rhythms related to light/dark cycles, especially the
Complementary therapy treatments are increas- sleep–wake cycle.23 Melatonin also acts as an
ingly being integrated into hospice and hospital immunoenhancer by increasing the mitogenic, and
settings to improve the client’s psychological and reducing apoptotic, responses.24 Immunoregulatory
physical well being.1 However, some Health Orga- roles of melatonin have been widely demonstrated
nisations do not partake, possibly due to concerns in cancer, aging, physical stress, and treatment
about the lack of appropriate evidence-based with corticoids.14,25–27 Interestingly, melatonin has
research to support therapeutic claims.2–4 Most been shown to be a hormonal regulator of
clients report a reduction in anxiety, improvement neoplastic cell growth28,29 and an association
in general well being3–5 and even reduced pain.6 between low levels of melatonin and human breast
However, it is difficult to determine whether these cancer have been reported.30 It has also recently
beneficial effects are the result of a specific been identified as a potential cancer-modulating
complementary therapy or of the therapeutic therapeutic.31
relationship and general relaxation. Better health outcomes for cancer patients who
The potential for reflexology to reduce anxiety is receive psychological support have been reported
of interest in healthcare due to profound connec- but the mechanisms behind such effects are
tions between anxiety following a stress event and unknown. A bio-psychosocial model has been
various disease processes.7 In addition, Schag and formulated in which coping and psychological
Heinrich8 reported that 44% of cancer patients adjustment is associated with alterations in cortisol
experience anxiety, which can influence pain levels as well as immune function and long-term
sensation, sleep disturbance and anticipatory nau- outcomes for breast cancer patients.32 Recently, it
sea,9 and our unpublished work with cancer was shown that massage therapy improves the
patients has anecdotally observed a reduction in medical and psychiatric condition of people,
anxiety following reflexology treatments. Stress is decreasing their stress perception, decreasing
speculated to be part of the cancer aetiology,10 cortisol levels, and improving sleep patterns and
which may be due to innumerable interactions, but immune function.33 Complementary therapies such
the immune system is postulated to play a as reflexology therefore could potentially have a
significant role.11 Whilst anxiety is normally viewed significant role in reducing patient stress.
as a cognitive experience and may be evaluated Despite claims that reflexology reduces anxiety,
quantitatively using psychometric measures,12 im- little is known of the magnitude of this effect, or of
mune changes are thought to involve metabolic and its impact on biometric, and psychometric mea-
endocrine responses that are also apparent. Two of sures. The aims of this study are to investigate the
the most important of these include variations in association between anxiety, salivary cortisol and
secretion of cortisol13 and melatonin.14 melatonin, blood pressure, pulse rate (PR) and
Cortisol exerts its action in a negative feedback reflexology in healthy volunteers as a prelude to a
loop on the hypothalamic pituitary adrenal axis, study involving cancer patients. This will establish
which is one of the primary biological pathways our methodology and so enable us to transfer it to a
linking psychological factors and the immune larger study that will evaluate the effects of
system.15 Cortisol suppresses the immune system reflexology on physiological, biochemical and psy-
and directly inhibits natural killer (NK) cell activ- chological measures of anxiety and stress in breast
ity.16 More recently cortisol has been shown to cancer patients.
down regulate the protein NKp30 which correlates
strongly with reduced NK cytolytic activity.17 A
high percentage of breast cancer patients Methods
have been shown to display abnormal secretion
patterns of the hormone cortisol such as flattened This study applied a cross-over design, with
diurnal cortisol, consistently high levels or erratic participants acting as their own controls on one
fluctuations18,19 which have been linked to poor occasion and receiving treatment on another, to
prognosis.20 The cause of this circadian deregula- strengthen the potential for detecting treatment
tion in cancer is unknown, but aberrations of effects in a small sample. Weaknesses are noted for
cortisol rhythms have been linked both with the this design but were negated:
physical stress of cancer21 and with psychological
stress.22  Carry-over effects. Possible impacts of receiving
Melatonin is secreted by the pineal gland and treatment on the second visit were reduced by
binds to receptors in the suprachiasmatic nuclei of ensuring at least 3 days between attendance.
A pilot study in healthy individuals 139

 Treatment sequence. An influence of time- Reflexology treatment

related events independent of treatment was
avoided by some participants attending on non- Reflexology is the systematic application of special
treatment (control) days followed by treatment touch or pressure to reflex points on the foot or hand.
days, whilst for others the sequence was The specific foot chart and technique employed here
reversed. is detailed by Fewell34 and uses gentle, precise
 Drop-out. The findings presented here are only pressure. The most commonly adopted movement is
those for participants who attended both occa- the ‘caterpillar’ technique and this is used by most
sions in the trial, and for whom full sets of data reflexologists as the precise, fluid movements allow
were obtained on each occasion. In reality, drop- for pressure to be exerted without inducing pain. At
out and/or non-completion of research tools was Anglia Ruskin University, we have developed a gentle
very low. approach to reflexology as opposed to the deeper
treatment sometimes applied by reflexologists. This
is based on our experiences from over nearly 30 years
Ethical considerations of working with patients presenting with complex
health conditions, pain and altered pain perception
All participants were informed of the details of the for example during chemotherapy. Our unpublished
study and gave their written consent. The study findings confirm that there are no negative effects
was approved by the Ethical Committee of Anglia from using gentle pressure.
Ruskin University. For both Trials, participants were divided into a
therapy group, who were treated with reflexology
for 60 min (12.00 noon–1.00 pm), and a non-
Participants treatment (control) group. Whilst the treatment
group received reflexology, the non-treatment
Thirty healthy volunteers were recruited through group sat quietly together in the adjoining room.
internal advertisements within the university. A The ambience (colour of the walls, temperature,
wide age range of 16–59 years was included as a background music) was the same in both rooms, in
pragmatic approach in consideration of the likely order to control for any changes in anxiety arising
participants within the follow-up study in a cancer from relaxation and social support. Three days
centre proposed to follow this pilot. Participants later the participants in the non-treatment group
with stimulated/deficient steroid secretion were underwent reflexology treatment while those in the
excluded: Cushing’s disease, Adrenal tumours, treatment group now formed the control group.
Addison’s disease, Adrenocorticotropic hormone This cross-over design allowed each participant to
deficiency disease, pregnancy, and women on oral provide their own control data.
contraceptive. Participants were requested not to
eat, drink or smoke 60 min before giving saliva
Psychometric assessment
samples. Health behaviours that potentially affect
the endocrine system were also recorded to detect
Spielbergers state trait anxiety inventory for adults
and eliminate unrelated sample variability, these
(STAI) was used to assess changes in transitory
included caffeinated drink and alcohol consump-
anxiety experienced by participants prior to and
tion (servings weekly), smoking (per day), exercise
following treatment.12 The STAI is a well-validated
(weekly), average sleep per night, self-rated
tool and differentiates between the temporary condi-
quality of sleep (poor, adequate good) and self-
tion of ‘‘state anxiety’’ and the more general and long-
rated quality of diet (poor/adequate/good).
standing quality of ‘‘trait anxiety.’’ Scores increase in
The recruited healthy volunteers were randomly
response to psychological stress, and decrease as a
divided into two trials. Trial A investigated the
result of relaxation training or therapies.
immediate effect of reflexology, whilst Trial B
repeated the main measures but expanded the
salivary cortisol and melatonin measurements. This Biometric measurements
was necessary to more precisely assess any impact
on the diurnal patterns of secretion through Blood pressure and pulse rate
measurement of night-time and waking levels. Systolic and diastolic blood pressure (BP) was
Immediate effects may not be seen due to time measured by auscultation over the brachial artery
required for production and secretion, whereas using a mercury sphygmomanometer, and radial PR
longer term measurements may be more informa- was measured manually, whilst sitting prior to and
tive. following reflexology treatment, and at identical
140 A.J. Mc Vicar et al.

times during control sessions. Measures were taken (sem). Time matched data for all measures were
by a nurse assistant experienced in the methods. compared using an independent (Student) t-test for
between-group differences. Pre- vs. post-treatment
changes within groups were also analysed using a
Collection and analysis of saliva samples
paired t-test. Pre-treatment and time-matched con-
Cortisol is secreted diurnally with levels peaking in
trol data were also analysed for evidence of correla-
the early morning and decreasing over the course of
tion between the two psychometric scales of the
the day.35 Secretion of melatonin is stimulated by
STAI, and between these measures and hormone
the dark and inhibited by the light and so also
concentration, using bivariate analysis (Pearson’s).
exhibits a diurnal rhythm. These hormones are
excreted in saliva in their free form and salivary
concentrations have been shown to be an accurate Results
indicator of free or biologically active hormone in
human serum.36–38 Collecting saliva is a relatively ‘State’ and ‘trait’ anxiety scores following
non-invasive procedure and evaluating hormone reflexology
concentrations in saliva has become a widely used
method to evaluate serum changes, and hence The two trials were performed on separate occa-
changes in glandular secretion. sions, with 12.00 noon as the starting point on each
Participants collected saliva samples using cotton occasion. No significant differences were noted
swabs or ‘‘salivettes’’ (Sarstedt, Inc.) at time points between the baseline ‘state/trait’ anxiety data on
elucidated from a small preliminary trial where non-treatment or treatment days for each trial, or
saliva samples were collected from the research between trials. The ‘state’ and ‘trait’ anxiety
team over 2 days. The need to adhere to collection measures at this time point were significantly
times in early morning and in the evening, that is correlated (all participants: r ¼ 0:583, Po0.001,
when participants were away from the university, n ¼ 52) suggesting that perceptions of participants
were strongly emphasised. on the day of attendance (viz ‘state’) were
In Trial A, saliva samples were collected at midday, associated with their underlying anxiety trait.
1.15 pm (i.e. 15 min after reflexology treatment), and Trait anxiety is considered unlikely to alter in the
at 8 pm (i.e. 7 h post-treatment). Multiple samples short-term, and the ‘trait’ scores did not change
were taken for greater precision. This allowed significantly either on the non-treatment or treat-
comparison of short-term changes in hormonal levels. ment day (Table 1). No significant change in ‘state’
As no changes were apparent, samples in Trial B were score was observed with time on the non-treatment
taken at 1.15 pm, 11 pm and on waking the next day day for individual Trials but pooling the data for
to incorporate late evening and early morning values. both Trials and subjecting it to paired analysis
The data from both trials allowed comparison of indicated a small but significant reduction
hormone concentrations at the times allocated but (3.571.2, Po0.01) (Table 1).
also enabled evaluation of the early morning peaks for There was a profound decrease in the ‘state’
cortisol and the dark induced peak for melatonin to score following treatment with reflexology; the
assess ‘responders/reactors’.39 The non-treatment post-treatment value was significantly lower (all
data provided the mean base-line readings. participants, independent t-test Po0.001) than the
Saliva samples were centrifuged for 4 min at time-matched value observed on the non-treat-
2500 rpm to remove the saliva from the swab, stored ment day (Table 1) and within-group analysis also
at +4 1C and assayed within 5 days. Cortisol and identified a pre–post treatment change (10.47
melatonin concentrations in saliva were assayed in 1.7, Po0.001). Closer within-group analysis of the
duplicate by enzyme-linked immunosorbant assay 20 items that comprised the ‘state’ component of
(ELISA). (Cortisol: Salimetrics, State college PA; the STAI identified significantly reduced scoring in
Melatonin; Buhlmann laboratories AG, Switzerland.) 11 items (Q1,2,3,4,5,7,15,16,17,19 and 20) consis-
Standards (cortisol 1.8–0.007 mg/dl; melatonin tent with an increased feeling of ease and a
81–3 pg/ml) were assayed simultaneously. Samples reduction in anxiety.
were quantified by colorimetric analysis at 450 nm.
Systolic blood pressure, diastolic blood
Statistical analysis pressure and resting PR following reflexology
All data were collated using Statistical Package for
the Social Sciences (SPSS; Version 13). Data are No significant time-related changes in systolic
expressed as means7standard error of the mean blood pressure, diastolic pressure or PR were
A pilot study in healthy individuals 141

Table 1 Comparison of time-matched data (mean7sem) for ‘state’ and ‘trait’ anxiety scores from the non-
treatment and treatment days, for both trials.

‘State’ anxiety ‘Trait’ anxiety

12.00 h 13.15 h 12.00 h 13.15 h

Trial A
Non-treatment day 36.171.8 33.171.3 41.171.8 39.471.7
n ¼ 19
Treatment day 36.172.0 26.271.2 39.671.6 38.072.0
n ¼ 19
Trial B
Non-treatment day 37.473.8 32.873.0 40.973.5 41.073.9
Treatment day 40.373.8 28.773.2 43.873.2 42.875.0
Both trials
Non-treatment day 36.571.6 33.071.2 41.071.6 39.971.6
n ¼ 27
Treatment day 37.171.5 26.871.1 40.671.5 39.271.9
n ¼ 25
Between group differencea Not significant Po0.001 Not significant Not significant
Difference between groups shown. Difference within groups between 12.00 and 13.15 h: Po0.05, Po0.01, Pp0.001.

observed within individuals on the non-treatment developed a methodology for measurement of

day (Table 2). Although no significant between- psychometric, biometric and hormonal levels be-
group differences were observed on the treatment fore and following reflexology treatment. These
day, a within-group comparison identified that data have provided the necessary information to
reflexology significantly reduced systolic blood enable us to proceed to a study of efficacy of
pressure (all participants, 6.371.6 mmHg, reflexology in breast cancer patients who have
Po0.001) and PR (all participants, 7.971.3 been shown to display different cortisol, melato-
beats/min, Po0.001), but not diastolic blood nin, biometric and anxiety baseline data.40 The
pressure (all participants, 2.471.3 mmHg, NS). methodology proved to be non-invasive, of minimal
disruption, accurate and suitable for transfer to a
larger study to acquire multidisciplinary scientific
Salivary cortisol and melatonin following data to help clarify the efficacy and impact of
reflexology treatment reflexology.
In addition to methodological advances, we have
No statistically significant differences in salivary also shown a highly significant decrease in state
cortisol or melatonin concentrations or the rele- anxiety following reflexology in healthy subjects.
vant peaks were evident between the treatment This reduction was considerably more pronounced
and non-treatment days in either Trial A or Trial B than that observed in the non-treatment group who
at any of the time points measured (Tables 3 and 4). simply relaxed in a similar environment: relaxation
and support were controlled for as these have been
shown to affect salivary cortisol.41 Closer analysis
Discussion of the types of questions in which there were
significant changes in participants’ mean responses
There is a dearth of scientific-based evidence to suggest that feelings of tension and worry were
consolidate therapeutic claims of complementary reduced following reflexology, whereas feelings of
treatments in healthcare settings. Such evidence well being (such as calmness, security, relaxation)
will be crucial in aiding informed decision making were increased.
for patients and future health strategies involving Baseline state and trait anxiety scales taken at
complementary therapies, including reflexology. noon were significantly correlated but there was no
This pilot study involving healthy subjects has evidence for an influence of ‘trait’ anxiety on the
142 A.J. Mc Vicar et al.

Table 2 Comparison of time-matched data (mean7sem) for cardiovascular parameters for both trials.

Non-treatment day Treatment day

12.00 h 13.15 h 12.00 h 13.15 h

Trial A
Systolic BP mmHg 126.975.5 122.774.4 130.473.4 124.573.7
n ¼ 13
Diastolic BP mmHg 69.972.8 68.572.3 74.572.1 71.972.5
n ¼ 13
Pulse beats/min 66.372.0 62.772.0 73.172.8 63.371.6
n ¼ 13
Trial B
Systolic BP mmHg 126.974.9 123.373.9 126.774.9 119.674.5
n ¼ 11
Diastolic BP mmHg 76.873.5 73.472.8 75.672.5 73.472.4
n ¼ 11
Pulse beats/min 68.472.4 65.472.2 68.471.7 62.671.6
n ¼ 11
Both trials
Systolic BP mmHg 128.173.6 123.072.9 128.772.9 122.572.9
n ¼ 24
Diastolic BP mmHg 73.172.3 70.771.8 75.071.6 72.671.7
n ¼ 24
Pulse beats/min 67.371.5 63.971.5 70.971.8 63.071.1
n ¼ 24

Time-matched, treatment vs. non-treatment differences between groups not significant. Within-group differences between
12.00 and 13.15 h: Po0.05, Po0.01, Pp0.001.

Table 3 Time-matched data for salivary cortisol (mg/ml; means7sem) on non-treatment and treatment days,
for each trial.

12.00 h 13.00 h 20.00 h 23.00 h Waking next day

Trial A
Non-treatment day 0.1570.03 0.1370.04 0.0670.01
n ¼ 10
Treatment day 0.1570.02 0.1570.03 0.0770.02
n ¼ 10
Trial B
Non-treatment day 0.2670.09 0.3370.11 0.3770.07
Treatment day 0.2470.10 0.3070.11 0.3970.05

No significant differences were seen at any time point either between or within groups.

‘state’ response to reflexology. The changes may be cally produce supportive effects and even reduce
a real effect of the therapy and so suggest that in anxiety.43 It may be that touch alone has a state
healthy participants the state-scale of the STAI is anxiety reducing effect.
sensitive to the effects of treatment. Whether the In addition, the control group were placed
effects resulted from the impact of reflexology or together in the same room and although they were
massage per se cannot be ascertained as the staff and students and thus ‘familiar faces’ this
additional benefit of thigmotropism (i.e. respond- situation may have caused anxiety to some parti-
ing to physical contact)42 was not specifically cipants. To maximise relaxation we created an
controlled for in this study, though it can intrinsi- ambience, played relaxing music and asked the
A pilot study in healthy individuals 143

Table 4 Time-matched data for salivary melatonin (pg/dl, means7sem) on non-treatment and treatment days,
for each trial.

12.00 h 13.15 h 20.00 h 23.00 h Waking next day

Trial A
Non-treatment day 2.6470.88 3.2071.64 6.672.38
n ¼ 10
Treatment day 0.7470.34 2.3170.93 2.8170.02
n ¼ 10
Trial B
Non-treatment day 0.8170.19 7.9171.88 5.6572.66
Treatment day 1.4970.65 5.0971.30 2.9071.41

No significant differences were seen at any time point between or within groups.

participants to sit quietly. In this environment we were already low and hence any responses con-
saw no evidence of collective increased anxiety in sistent with a reduction in stress might not be
fact there was a small reduction in state score on discernible. A major consideration for the follow-
the non-treatment day. The main study that will up study involving cancer patients will be that
follow this pilot will incorporate a design to avoid changes in secretion may be more pronounced as
strangers being placed in the same room as a chronic stress levels appear to alter the secretion
control group. patterns of stress-related hormones. If reflexology
Trait anxiety, on the other hand was not can help to maintain and/or restore the normal
significantly affected by the reflexology (nor were cortisol and melatonin secretion patterns, and to
changes observed in the non-treatment data). This reduce stress and the physiological and biochemical
is an expected result, since trait anxiety, unlike effects of stress, then patient outcomes may be
state, is not a transitory state and any changes in improved. Through evaluation of the responses in
trait anxiety would not be expected in the healthy individuals this study has provided valuable
time-course of this study. From a methodological parameters and a suitable methodology for asses-
viewpoint, the divergence of the state anxiety and sing the responses of cancer patients in a subse-
trait anxiety data following reflexology in this quent study.
particular setting supports a sensitivity of the STAI Potential limitations of this pilot are encom-
tool and also reduces concerns over the reapplica- passed in the reliable measurement of salivary
tion of the tool so soon after the initial scores were hormones. The range of time points, information on
taken. food intake and health behaviours helped to
We have also shown that systolic BP and PR are minimise the propensity for variation in cortisol
reduced following reflexology. These responses are and melatonin between individuals. In addition it
consistent with a reduction in sympathetic activity, cannot be ascertained whether the effects resulted
which would support a reduction in state anxiety. from the impact of reflexology or massage per se.
However, reflexology appeared to have no immedi- However, the study does exclude the significance of
ate or delayed effect on the secretion of cortisol the therapeutic relationship that builds over time
and melatonin. Thus, in healthy individuals changes with one practitioner, as each participant only
in cognitive stress perceptions may not be simulta- received one reflexology treatment.
neously translated into endocrine responses. The
lack of correlation between hormone concentra-
tions and state or trait STAI scores could also Conclusions
support dissociation with cognitive appraisal, but
‘spot’ hormone concentrations at specific time This study has defined safe non-invasive, accurate
points may not be an accurate evaluation of and multidisciplinary methodological standards to
stress-related secretion. However, the lack of an begin evidenced-based research into the efficacy of
observable impact on diurnal patterns of hormone complementary therapies for cancer patients and
secretion also suggests the dissociation. It is also of other healthcare settings.
note that baseline measures and the application of By evaluating psychometric and biometric
reflexology were at a time when secretion rates changes following reflexology, this pilot study has
144 A.J. Mc Vicar et al.

also showed that reflexology significantly reduced human natural killer cell line. Cell Immunol 1997;178:
‘state’ anxiety and related cardiovascular para- 108–16.
meters, but had no significant effect on stress- 17. Mavoungou E, Bouyou-Akotet MK, Kremsner PG. Effects of
prolactin and cortisol on natural killer (NK) cell surface
related secretion of cortisol or melatonin in healthy expression and function of human natural cytotoxicity
individuals, at the time points chosen. receptors (NKp46, NKp44 and NKp30). Clin Exp Immunol
18. van der Pompe G, Antoni MH, Heijnen CJ. Elevated basal
cortisol levels and attenuated ACTH and cortisol responses
Acknowledgements to a behavioural challenge in women with metastatic breast
cancer. Psychoneuroendocrinology 1996;21:361–74.
This work was supported by Anglia Ruskin University 19. Toutiou Y, Bogdan A, Levi F, Benavides M, Auzeby A.
Disruption of the circadian patterns of serum cortisol in
and Helen Rollason Heal Cancer Care Charity.
breast and ovarian cancer patients: relationships with
tumour markers antigens. Br J Cancer 1996;74:1248–52.
20. Sephton S, Sapolsky R, Kraemer H, Spiegel D. Early mortality
in metastatic breast cancer patients with absent or
References abnormal diurnal cortisol rhythms. J Natl Cancer Inst
1. Bell L, Sikora K. Complementary therapies and cancer care. 21. Mormont MC, Levi F. Circadian-system alterations during
Complementary Ther Nurs Midwifery 1996;2:57–8. cancer processes: a review. Int J Cancer 1997;70:241–7.
2. Mackereth P, Tiran D, editors. Clinical reflexology a guide 22. Deuschle M, Schweiger U, Weber B, Gotthardt U, Korner A,
for health professionals. London: Churchill Livingstone; Schmider J, et al. Diurnal activity and pulsatility of the HPA
2002. system in male depressed patients and healthy individuals.
3. Hillier D, Fewell F, Caan W, Shephard V. Wellness at work: J Clin Endocrinol Metab 1997;82:234–8.
enhancing the quality of working lives. Int Rev Psychiatry 23. Weaver DR, Stehle JH, Stopa EG, Reppert SM. Melatonin
receptors in human hypothalamus and pituitary: implica-
4. Fewell F, Mackdrodt K. An investigation into the potential
tions for circadian and reproductive responses to melatonin.
awareness and practice of complementary therapies in
J Clin Endocrinol Metab 1993;76(2):295–301.
hospital and community settings within the Mid Essex
24. Guerrero JM, Reiter RJ. Melatonin-immune system relation-
Region. Complementary Ther Nurs Midwifery 2004;11(2):
ships. Curr Top Med Chem 2002;2:167–79.
25. Maestroni GJ, Conti A, Pierpaoli W. Role of the pineal gland
5. Downer S, Cody M, McClus P, Wilson P, Arnott S, Lister T, et
in immunity. Circadian synthesis and release of melatonin
al. Pursuit and practice of complementary therapies by
modulates the antibody response and antagonizes the
cancer patients receiving conventional treatment. Br Med J
immunosuppressive effect of corticosterone. J Neuroimmu-
nol 1986;13(1):19–30.
6. Stephenson NL, Weinrich SP, Tavakoli AS. The effects of foot
26. Maestroni GJ, Conti A, Pierpaoli W. Pineal melatonin, its
reflexology on anxiety and pain in patients with breast and
lung cancer. Oncol Nurs Forum 2001;3:445–6. fundamental immunoregulatory role in aging and cancer.
7. McEwen BS. Protective and damaging effects of stress Ann N Y Acad Sci 1988;521:140–8.
mediators. New Engl J Med 1998;338:171–9. 27. Caroleo MC, Nistico G, Doria G. Effect of melatonin on the
8. Schag CA, Heinrich RL. Anxiety in medical situations: adult immune system. Pharmacol Res 1992;26(Suppl 2):34–7.
cancer patients. J Clin Psychol 1989;45:20–7. 28. Cos S, Sanchez-Barcello EJ. Melatonin and mammary patholo-
9. Ferrell-Torry AT, Glick OJ. The use of therapeutic massage as gical growth. Front Neuroendocrinol 2000;21:133–70.
a nursing intervention to modify anxiety and the perception 29. Cos S, Sanchez-Barcello EJ. Melatonin, experimental basis
of cancer pain. Cancer Nurse 1993;16:93–101. for a possible application in breast cancer prevention
10. McCarty R, Gold PE. Catecholamines, stress, and disease: a treatment. Histol Histopathol 2000;15:637–47.
psychobiological perspective. Psychosom Med 1996;58: 30. Davis S, Mirick DK, Stevens RG. Night shift work, light at
590–7. night, and the risk of breast cancer. J Natl Cancer Inst
11. Manjili MH, Wang XY, Park J, Facciponte JG, Repasky EA, 2001;93:1557–62.
Subjeck JR. Immunotherapy of cancer using heat shock 31. Mills E, Wu P, Seeley D, Guyatt G. Melatonin in the treatment
proteins. Frontiers Biosci 2002;7:43–52. of cancer: a systematic review of randomised controlled
12. Spielberger C. State-trait anxiety inventory for adults. trails and meta-analysis. J Pineal Res 2005;39:360–6.
California USA: Mind garden Inc.; 1983 Available at 32. Luecken LJ, Compass BE. Stress, coping, and immune
/www.mindgarden.comS. function in breast cancer. Ann Behav Med 2002;24:336–44.
13. Chrousos G, Gold PW. A healthy body in a healthy mind—and 33. Field T. Massage therapy. Med Clin North Am 2002;86:
vice versa—the damaging power of ‘uncontrollable’ stress. 163–71.
J Clin Endocrinol Metab 1998;83:1853–9. 34. Fewell F, Hodges M. Complementary medicine for oncology
14. Pierpaoli W, Maestroni GJM. Melatonin a principle neuroim- clients; making a difference using aromatherapy and
munomodulatory and anti-stress hormone. Immunology Lett reflexology education to inform practice. Int J Clin
1987;16:355–62. Aromatherapy 2005;2:5–9.
15. Kirschbaum C, Hellhammer DH. Salivary cortisol in psycho- 35. Stone AA, Schwartz JE, Kirschbaum C, Cohen S, Hellhammer D,
neuroendocrine research: recent developments and applica- Grossman S. Individual differences in the diurnal cycle of
tions. Psychoneuroendocrinology 1994;19:313–33. salivary free cortisol: a replication of flattened cycles for some
16. Zhou J, Olsen S, Moldovan J, Fu X, Sarkar FH, Moudgil VK, individuals. Psychoneuroendocrinology 2001;26:295–306.
et al. Glucocorticoid regulation of natural cytotoxicity: 36. Laudat MH, Cerdas S, Fournier C, Guiban D, Guilhaume B,
effects of cortisol on the phenotype and function of a cloned Luton JP. Salivary cortisol measurement: a practical
A pilot study in healthy individuals 145

approach to assess pituitary-adrenal function. J Clin 40. Abercrombie HC, Giese-Davis J, Sephton S, Epel ES, Turner-
Endocrinol Metab 1988;66:343–8. Cobb JM, Spiegel D. Flattened cortisol rhythms in metastatic
37. Vining RF, McGinley RA. The measurement of hormones in breast cancer patients. Psychoneuroendocrinology 2004;29:
saliva: possibilities and pitfalls. J Steroid Biochem 1987;27: 1082–92.
81–94. 41. Turner-Cobb JM, Sephton SE, Koopman C, Blake-Mortimer J,
38. Papanicolaou DA, Mullen N, Kyrou I, Nieman LK. Night time Spiegel D. Social support and salivary cortisol in women with
salivary cortisol: a useful test for the diagnosis of Cushing’s metastatic breast cancer. Psychosom Med 2000;63:337–45.
syndrome. J Clin Endocrinol Metab 2002;87(10):4515–21. 42. Montague A. Touching the human significance of the skin.
39. Vedhara K, Tuinstra J, Miles JNV, Sandeman R, Ranchor AV. New York: Columbia University Press; 1971.
Psychosocial factors associated with indices of cortisol 43. McNamara P. Massage for people with cancer: a working
production in women with breast cancer and controls. paper. London: Wandsworth Cancer Support Publication;
Psychoneuroimmunology 2006;31(3):299–311. 1993, 1999.