Content uploaded by Filipa M.B. Lã
Author content
All content in this area was uploaded by Filipa M.B. Lã
Content may be subject to copyright.
http://rsm.sagepub.com/
Education
Research Studies in Music
http://rsm.sagepub.com/content/24/1/75
The online version of this article can be found at:
DOI: 10.1177/1321103X050240010601
2005 24: 75Research Studies in Music Education
Filipa Lã and Jane W. Davidson
Investigating The Relationship Between Sexual Hormones And Female Western Classical Singing
Published by:
http://www.sagepublications.com
On behalf of:
Society for Education, Music and Psychology Research
can be found at:Research Studies in Music EducationAdditional services and information for
http://rsm.sagepub.com/cgi/alertsEmail Alerts:
http://rsm.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://rsm.sagepub.com/content/24/1/75.refs.htmlCitations:
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
75
© 2005 Callaway Centre
Investigating The Relationship Between Sexual
Hormones And Female Western Classical Singing
Filipa Lã & Jane W. Davidson
Abstract
There has not always been an awareness of the relationship between sexual
hormonal concentrations and quality of the singing voice in the western classical
singing world. Despite evidence for this connection existing since the 3rd century
BC, there are still controversies and a lack of information about this area, especially
regarding the effects on the female western classical singing voice. The aim of this
paper is to shed light on the importance of undertaking further research in this
field. Therefore, a revision of previous research, exploring the extent to which
sexual hormonal concentrations can contribute to the wellbeing of a singer’s voice
and career, from both psychological and biological points of view, will be
presented. Additionally, further investigation in the field of hormonal related
medication and its effects on singing voice quality will be undertaken.
t is estimated that malfunctions of the endocrine system are responsible for 15% of
voice disorders. For example, hypothyroidism is often responsible for a husky
‘broken-pot’ voice, and minor thyroid dysfunction can also cause mild dysphonia
and a sensation of “having a veil over the voice” (Brodnitz, 1971; Sataloff, 1987).
However, prominent changes in the quality of the voice occur mostly when there are
changes in the concentrations of sexual hormones. During the life span, the voice
undergoes changes related to a person’s sexual hormonal concentrations, and women
are notably affected by these hormonal fluctuations more than men. The male voice
changes in the early and later stages of sexual development, puberty and
andropause; however, the female voice is affected across her life span, and during
the reproductive years in particular this effect can be cyclical. Therefore, and bearing
in mind the complexity of the female sexual endocrinology system, this article
focuses on changes in female classical singing voices related to the hormonal
variations that are responsible for women’s menstrual cycles.
Historical contextualisation
Nowadays, it is known that the human larynx is a “hormonal target organ”
(Abitbol, Abitbol & Abitbol, 1999). However, the acceptance of this concept by the
western classical singing world has been a long and difficult process which still
raises some controversy.
The first written evidence of a strong connection between sexual hormones
and voice quality dates back to the 3rd century BC, when castration was reported as a
way of preserving a boy’s high vocal range for singing purposes beyond puberty.
Experiments with castration spread to all of Europe during the 8th and 9th centuries,
especially to Spain and to Italy, and by the 16th century the Church became the
greatest promoter of the castrato voice.
A castrato had a crystalline timbre and an exceptional range, and his voice
was extremely powerful (Weiss, 1950). Castrati developed a reputation as skilful
singers, able to sing the majority of operatic roles of their time, despite the fact that
they were originally vocally altered in order to sing sacred music in churches. Their
voice characteristics were a consequence of a paediatric laryngeal structure, vibrating
I
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
76
© 2005 Callaway Centre
in a female register, but having the breathing power and resonance of a man
(Abitbol, Abitbol & Abitbol, 1999).
Another source of evidence that demonstrates the interference of sexual
hormonal levels on voice quality has been obtained from pubertal vocal studies.
During puberty a set of bodily transformations occur. The increase in the activity of
the adrenal cortex and gonads is responsible not only for transformations in both
male and female reproductive systems, but for significant changes in the larynx,
especially for boys (Gelder, 1974). Vocal changes during puberty are a result of a fast
increase in the size of the larynx; there is a rapid and disproportional change in the
thyroid cartilage, and thus vocal fold length, which happens approximately between
the ages of 12 to 13 years of age (Titze, 1993). Physically, there is an increase in
breathing capacity as a consequence of an enlargement of the neck and chest. The
lengthening of the neck allows the descent of the larynx, which is more significant for
boys and especially in deeper voices (Cooksey, Beckett & Wiseman, 1984). Following
these transformations, voice becomes a sexual secondary characteristic, which means
that the gender of a person is likely to be detected by his/her vocal pitch (Abitbol,
Abitbol & Abitbol, 1999).
A third and important source of evidence for the hormonal-vocal
relationship has been the singers’ subjective reports of their experiences. Classically
trained female singers have complained about vocal alterations over several days
before the onset of menses (Frable, 1961). These reports have often been taken into
account by eastern European opera houses, where ‘respect days’ were written into
contracts (three days in which the singer can refrain from singing during vocal
premenstrual tension) in order to avoid negative effects on the quality of her voice
(Lacina, 1968). However, Eastern and Western Europe is divided in its opinion as to
whether singers should refrain from singing during the premenstrual and menstrual
phases of the cycle. In Western Europe, female operatic singers may have
professional engagements regardless of which phase of the menstrual cycle they are
in.
One aim of this paper, therefore, is to increase awareness of the relationship
between sexual hormones and quality of classical singing voice and the potential
vocal problems related to it. To achieve this, an overview of previous research and
the investigative approaches employed will be discussed, followed by an assessment
of the psychological and physiological effects of sexual hormonal variations on
western classical singers’ performances. Finally, there is a discussion on how further
research in this area could help to improve the care of the professional voice at all
levels.
Summary of key research
Several previous studies have been concerned with the effects of cyclical
hormonal variations during the menstrual cycle on the quality of the female voice;
however, findings amongst these studies are inconsistent. Whilst there is little
consensus in one group of studies in demonstrating a clear link between the
menstrual cycle and vocal aberrations, another group of studies provides evidence
that supports such a link. For example, Frable (1961) reported the complaints of three
female voice users, all non-trained singers, who mentioned symptoms of hoarseness,
vocal breaks, loss of control, decrease in pitch and huskiness of the voice during the
premenstrual period of the menstrual cycle. However, the results of laryngoscopic
examinations undertaken at several times during the menstrual cycle showed normal
larynxes (Frable, 1961).
Silverman and Zimmer (1978) analysed the spectrograms of three sustained
vowels from twenty undergraduate students at ovulation and premenstruation.
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
77
© 2005 Callaway Centre
From their investigation, they concluded that hoarseness is not a symptom of
premenstruation, since there were no significant changes between ovulation and
premenstruation (Silverman & Zimmer, 1978).
Flach, Schwickardi and Simon (1968) studied the voices of 187 singing
students during their menstrual cycle, in order to assess whether vocal changes were
observed; however, no direct correlation between voice changes and menstruation
was found (Flach, Schwickardi & Simon, 1968). Wilson and Purvis (1980) compared
the range and fundamental frequency of the voice of ten singing students. They
analysed the production of a sustained vowel, /a/, and the highest and lowest notes
of the singers’ ranges, at three specific points of the menstrual cycle: two days before
the onset of menses, and on days 14 and 21 of the menstrual cycle. The results of this
study did not show significant acoustical changes between the three recorded times;
however, seven singers perceived negative changes in the quality of their voices
during menstruation and premenstruation. Their reports included comments such
as: “cloudiness of the voice”, loss of brilliance, loss of flexibility, changes in timbre
towards a darker and hoarse colour (Wilson & Purvis, 1980). Brown and Hollien
(1981) reported the results of a study undertaken by Coleman and Hyler (1981 as
personal communication) who recorded three young women with untrained voices
reading a standard passage everyday of one menstrual cycle. The research questions
were concerned with changes on the fundamental frequency of untrained female
voices during the menstrual cycle and its comparison with the fundamental
frequency of the trained female voice. The results did not indicate any significant
differences (Brown & Hollien, 1981).
On the other hand, studies have shown a direct relationship between certain
phases of the menstrual cycle and voice changes. Lacina (1968) highlighted the
existence of premenstrual and menstrual vocal symptoms by studying 42 opera
singers in the National Opera House in Prague. During the phases of the menstrual
cycle, some singers commonly reported loss of high notes and uncertainty of pitch. A
laryngeal examination showed vocal haemorrhage and a small oedema. With vocal
rest and small dosages of prednisolone the voice was restored, but these singers were
advised not to sing during the premenstrual phase of their menstrual cycle. In the
face of the gravity of these symptoms, Lacina (1968) advised avoiding severe
strenuous use of the vocal folds during these times of the menstrual cycle.
Brodnitz (1971) presented the case of a young opera singer who lost her voice
every month during premenstruation. Laryngoscopies were performed and revealed
small mucous membrane haemorrhages premenstrually. Also, hormonal analysis
showed low levels of luteinising hormone (LH). If the singer did not sing at all
during this phase of her ovarian cycle, vocal problems were reported as being
avoided (Brodnitz, 1971). Whitehead, Kohler and Schlueter (1974) reported that the
vocal quality of some singers could be affected some days before and during
menstruation. They studied the effects of the menstrual cycle on the harmonic and
non-harmonic portions of sung vowels in twelve adult females during one menstrual
cycle. Spectrographic analysis was performed and the results indicated that there
was an increase in non-periodic components in the vowel spectra during
premenstruation (Brown & Hollien, 1982).
Bearing in mind the severity of the vocal symptoms reported in some case
studies, Gelder (1974) considered vocal aberrations during premenstruation and
menstruation as a discrete diagnosis, laryngopathia menstrualis. According to the
author, this cyclical dysfunction of the voice could be seen more frequently in
singers, since they demand more from their vocal mechanism. The vocal symptoms
could include dull, colourless, raucous voice and hoarseness as consequences of
oedema or even haemorrhages in the vocal folds. The tension of the vocal folds was
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
78
© 2005 Callaway Centre
reduced, resulting in a husky voice. Gelder strongly advised singers not to perform
during these phases of the menstrual cycle, since vocal haemorrhages could cause an
irreversible damage of the tissues involved in phonation (Gelder, ibid.). Isenberg,
Brown and Rothman (1983) included in their study additional physiological
measures, particularly body weight and body temperature. They chose a population
of singers with irregular menstrual cycle or amenorrhea, since, according to them,
this population is the most affected by vocal problems, such as a ‘crack’ in the voice;
breathiness and weakness in a given area that persists even after several years of
vocal training; inability to phonate on given pitches; tonal ‘stops’ and lack of
flexibility. During their experiment, these authors found that singers who had an
increase in body temperature had an increase in their speaking fundamental
frequency. On this basis, they suggested that a fluctuation in hormone levels and the
fundamental frequency of the speaking voice might have a direct relationship, since
release of progesterone following ovulation induces a rise in basal body temperature
in the second half of the menstrual cycle. Additionally, measurements of serum
testosterone were made, since this hormone can have masculinising effects on the
female voice. The authors concluded that women with menstrual problems had
higher levels of testosterone when compared to the group of singers who had regular
menstrual cycles (Isenberg, Brown & Rothman, 1983).
Abramson et al. (1984) distributed a questionnaire to 120 trained singers
concerning their own perception of the quality of their voices across their menstrual
cycles. Their results showed that changes in a singer’s voice during the menstrual
cycle might be a result of elevated oestrogen levels rather than drops in the levels of
oestrogens, since some of their responses highlighted vocal changes during the
ovulatory phase of the cycle, i.e. when levels of oestrogens are high (Abramson et al.,
1984). Brown and Rothman (1985) added a new technique for studying the quality of
the voice through the menstrual cycle. They used an index of hoarseness based on a
relation between the levels of harmonics versus the levels of noise (H/N ratio), and
they also analysed a jitter factor. These two measurements provide a more accurate
view of the acoustical characteristics of the voice, where ‘noise’ can be understood as
the non-periodic components of the vowel spectra. They compared a group of non-
singers, elementary school teachers, with a group of college students studying vocal
music, during the day prior to menses, one day in early menses and one day out of
premenstrual and menstrual influence. Their results suggested significant differences
in these parameters for the three recorded times. When the H/N ratio decreased,
indicating greater ‘noise’ in the vocal signal, the jitter values increased. This
phenomenon occurred during clear and premenstrual moments, when compared
with menses, results that the authors did not expect. There was more noise when
sustaining the three vowels /i/, /a/ and /u/ softly, which was the most difficult to
control. The authors did not find significant differences during menstruation for
values of fundamental frequency (F0). F0 was higher when sustaining the three
vowels loudly when compared with soft and conversational levels. The authors
reached the conclusion that although teachers and singers showed similar results, the
latter demonstrated more voice control and less overall aperiodicity in the voice
signal than the former, suggesting that singers may be more capable of overcoming
premenstrual vocal problems than non-singers.
Wicklund et al. (1988) compared the longitudinal effects of the menstrual
cycle on perturbation, phonatory range and mean fundamental frequency of singers
and non-singers with regular menstrual cycles. They analysed a spoken text, the
‘Rainbow Passage’ (Fairbanks, 1940), and three sustained vowels /a/, /i/ and /u/,
during pre, peri and postmenstrual phases of the menstrual cycle. They also analysed
diaries of the menstrual cycle (menstruation was day one), which embraced
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
79
© 2005 Callaway Centre
questions such as: perception of energy levels, emotions, irritability and hoarseness.
Generally, the results showed higher F0 for singers than for non-singers, which was
expected. In the same way, singers were more aware of vocal aberrations and were
more sensitive to them, looking to care for their voices. Differences in both singers
and non-singers F0 were found for the different recorded periods. Singers showed
lower F0 premenstrually, higher F0 during the menstrual period and even higher F0
postmenstrually. These results correspond with earlier theories of fluid retention in
the vocal folds during the premenstrual and menstrual phases of the cycle (and fluid
retention relates to increased mass in the vocal fold tissue and lower vocal pitch). On
the other hand, non-singers showed higher F0 during premenstruation. This could be
evidence of their increased effort to compensate for physiological changes. Non-
singers might increase the adductory forces to maintain a normal sound during
premenstruation. This hypothesis of an increase in adductory forces was supported
by an increase in the levels of loudness (Wicklund et al., 1988).
Abitbol et al. (1989) carried out an investigation aimed at assessing the
prevalence of premenstrual hoarseness and whether changes in voice quality
accompanied biological and/or hormonal changes during the premenstrual phase of
the cycle. The subjects sang Frère Jacques on three successive starting notes. The
recordings were made at ovulation (scheduled 14-15 days following onset of
menses), and for one to three days before the onset of menses, during two successive
cycles. Each subject’s experimental behaviour was videotaped and dynamic vocal
exploration (DVE) was performed. DVE offers three kinds of data, all synchronised:
acoustic, visual and glottographic. Vocal range, vocal quality and vocal intensity
were recorded and analysed by later audition. Visually, they investigated the
mobility of arytenoids, vocal mass and the appearance of the epithelium during
singing, all evaluated with videostrobofiberoscopy. A rigid laryngeal telescope was
used to magnify the capillaries after the exercises to assess oedema in detail and look
for possible signs of muscle fatigue. Electrolaryngography enabled the action and the
impedance of the vocal folds to be reviewed during phonation. This was a major
piece of research which confirmed that hormonal levels influence the woman’s voice.
The researchers also established that there were hormonal receptors in the larynx:
oestrogen target cells were found in the larynx, as well as androgen and steroid
receptors in human pharyngolaryngeal mucosa and epithelium. Furthermore, these
authors found a similarity between vocal fold and cervical epitheliums as an
outcome of performing a vocal fold smear and a cervical smear test. These
observations could help to explain why large numbers of singers had premenstrual
symptoms such as hoarseness and vocal fatigue.
More recently, Abitbol, Abitbol and Abitbol (1999) undertook a systematic
study on the effects of the menstrual cycle on voice quality in 97 vocal professionals,
at ovulation and in the premenstrual phase, during three consecutive menstrual
cycles. They performed video-recordings of vocal fold anatomy and a laryngeal
chrono-kinetic study. Spectrographic vocal analyses were also performed. The
participants sang Frère Jacques to test the agility of their voices; they also sang staccato
notes on the /i/ vowel to allow the assessment of mobility and suppleness of the left
and right cricoarytenoid joints. The vibrations of the vocal fold mucosa during
phonation were analysed in slow motion using stroboscopy. The results of this study
indicated: (i) swelling of the vocal folds, with thickened mucous membrane and loss
of capacity of distension during the premenstrual phase of the cycle for the 97
patients analysed; (ii) 71 patients showed dilatation of microvarices in the vocal
folds, submucosal vocal fold haematoma and vocal fatigue; (iii) 59 subjects presented
a decrease in muscular tone, diminished power of contraction of the vocal muscles,
decreased range and vocal fold nodules. Results suggested that during the
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
80
© 2005 Callaway Centre
premenstrual and menstrual periods the laryngeal mucous is thicker, leading to a
frequent throat clearing and a decrease in the levels of hydration of the free edges of
the vocal folds, so vocal lubrication is reduced (Abitbol, Abitbol, & Abitbol 1999).
The studies described above are significant and we believe that they are
sufficiently compelling to support the concept of ‘respect days’ in female singers’
contracts. Nowadays, performers, voice educators, voice scientists, medical doctors
and behavioural therapists work in a close relationship, so they need to be aware of
these recent systematic studies.
Physiological and psychological effects of sexual hormonal variations
Having considered the research presented above, it seems that there is
evidence to support the proposition that negative vocal symptoms are related to
premenstrual and menstrual phases of the menstrual cycle. Therefore, the
physiological and psychological evidence for this assumption will now be explored.
From a physiological and endocrinological point of view, there are several
possible explanations for the effects of variations across the menstrual cycle on the
quality of the voice.
According to the pioneer of the study of vocal behaviour across the
menstrual cycle, Frable (1961), there is a strong influence of levels of oestrogen on the
state of the ground substance. Drawing on the work of Schiff and Burn (1961), Frable
(1961), Owen (1975) and Dalton (1977) suggested that the decreasing levels of
oestrogens just before and during the menstrual phase of the menstrual cycle cause
water retention in the vocal tissue, and that this physiological change would account
for vocal aberrations. Pressman and Kelerman (1970) explained that vocal changes
during the premenstrual and menstrual phases of the menstrual cycle were as a
result of an increase of blood supply to the vocal folds, due to an increase of thyroid
gland activity. Vocal hoarseness would be therefore the most common vocal
symptom at these phases of the menstrual cycle. Isenberg, Brown and Rothman
(1983) claimed that the cause for vocal changes during the menstrual cycle was not
the actual dropping in the levels of oestrogens before menstruation, but the
inconstancy of hormonal levels during the whole menstrual cycle. They also
hypothesised that changes in testosterone levels might also account for vocal changes
across the menstrual cycle, thus these changes should be closely monitored.
Abramson et al. (1984) also argued that for a female singer it is much more
difficult to overcome constant physiological changes in the larynx in response to
constant changing levels of oestrogen, than with a more prolonged change during a
particular moment of her menstrual cycle. Higgins and Saxman (1989) maintained
that the inconstancy of the sexual hormonal levels across the menstrual cycle would
interfere with the laryngeal neuromotor movement and sensory thresholds, so voice
production would be affected. More recently, Abitbol, Abitbol and Abitbol (1999)
studied the effects of oestrogens and progesterone on the vocal folds. They claimed
that oestrogens cause an increase on the secretion of the glandular cells above and
below the vocal folds’ edges. This increases mucous production both before
ovulation and when the levels of oestrogen fall before menses. These modifications in
the mucosa of the vocal folds may lead to minor changes in the voice, and these are
enhanced by progesterone production. When progesterone is secreted, major changes
in vocal quality occur due to the fact that this hormone increases the viscosity and
acidity of the secretions of the glandular cells, but decreases their volume, causing a
relative dryness. The dryness of the vocal folds, the increase in the acidity level, the
reduced tonicity of the laryngeal muscles, oedema of the vocal folds and venous
dilatation of the microvarices all combine and result in a premenstrual voice
syndrome. They also found (besides the fluctuations in hormonal levels) that cyclical
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
81
© 2005 Callaway Centre
changes in the mucosa of the vocal folds are similar to those occurring in the cervix
of the uterus, since both tissues are physiologically and structurally similar. The
squamous ectocervical mucosa has two layers: the lamina propria and the chorion,
which can itself be divided into the basal and parabasal membranes. During the first
part of the cycle, the junction between these two layers of cells is larger than during
the second half of the cycle, leading to the conclusion that the volume of intracellular
space in the cervical mucosa is hormone dependent. The endocervix is also hormone
dependent. It is ciliated and has serous and mucous glands. Oestrogens produce a
thin mucous and progesterone leads to a thick mucous. According to these authors,
both tissues have a hormonal sensitive intracellular space, so that oedema caused by
a thicker mucous during the luteal phase of the menstrual cycle (i.e. the last phase of
the menstrual cycle) may cause changes in vocal quality, such as decreased
flexibility, loss of high notes and breathy voice.
Alongside the understanding of physiological vocal changes that cause
premenstrual and menstrual vocal distress, it is important to recognise the
psychological impact of this in singers’ performances. A singer is always aware of the
next performance, and being obliged to sing when there is vocal distress can be very
soul-destroying. The reports of singers constitute the clearest currently available
evidence for an association between certain phases of the menstrual cycle and vocal
difficulties. To understand better the degree of vocal distress that singers feel during
the menstrual cycle, we report data gathered from face-to-face, semi-structured
interviews we undertook with five professional female singers (age range between
34-48 years, all in normal menstruating cycles) living in the UK. For reasons of
confidentiality, the five singers will be identified by the pseudonyms of Anna,
Beatrice, Catherine, Danielle and Eva. These qualitative inquiries encouraged the
exploration of singers’ perceptions, feelings and understandings of their voices
regarding their hormonal background, and the first reports concern the singers’
perceptions of their voices during the premenstrual and menstrual phases of the
menstrual cycle.
When I do my daily practice I feel tired, either the day before or the first day of the period. But the
biggest thing I notice, which always happens and which always has happened to me, is that my voice is
much heavier. The day before and the first day is like lifting a sack of potatoes. It’s probably the only way
to describe it. Every exercise is like lifting a sack of potatoes. The voice is very heavy and I find that it is
like the sack is on my back. The pianissimos are also affected. For instance, in the cadenza of Lucia di
Lammermoor, there is a very high section that is all in high B flat, C and D pianissimo; this is very high
in the voice, and it is right near the passaggio which is not good to sing—the passaggio and pianissimo.
During my period, at times it just doesn’t work, and I have just to accept that maybe on these days it’s just
not going to work. [Anna]
I feel dragged down, so that the problems are more in the lower parts of the body. It is like there is
dissociation between the several parts of my body. With menstrual pain, there is so much the feeling of
being dragged down that there is no more connection. [Beatrice]
It seems evident from the above reports that singers are aware of a strong
relationship between vocal tiredness, sensation of heavier voice, loss of pianissimo
effects on high notes and loss of vocal support during premenstrual and menstrual
phases of the menstrual cycle.
One of the reasons to explain why singers agree to sing in such demanding
and potentially professionally dangerous conditions might be because they do not
recognise these symptoms as being associated with their menstrual cycles. Because of
the personal nature of these issues, they might not have been discussed at a
professional level. Beatrice and Catherine had the opinion that consciousness about
this phenomenon among the ‘world of classical female singers’ might be the only
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
82
© 2005 Callaway Centre
way to overcome possible long-term vocal injuries associated with hormonal
imbalances.
I never had discussed these matters with other colleagues. [Beatrice]
Singers are more or less aware of this relationship between the hormonal cycle and the quality of the
voice, but I think they are not conscious of it. It’s like all kinds of things; it needs to be brought into
attention…I think if I had recognised the biological things much sooner, I might have done something
about it. [Catherine]
At this point of the interviews, a discussion emerged about whether there
should be ‘respect days’. The interviewees were aware that vocal damage could
occur at premenstrual and menstrual phases of the menstrual cycle, and that ‘respect
days’ exist in certain European countries.
As a singer, I don’t know whether ‘respect days’ should exist in England. I don’t know whether the
menstrual cycle affects a large number of singers in England. As far as I can see, I think colleagues of
mine tend to have the opinion that they just have to get on with it and put it down to a bad day. If it is a
performance day, then it is tough. The performance might not be so good, but it is just part of life, there is
no other way around it. I suppose that the fact that other countries have ‘respect days’ is a question of
temperament of people, and culture. Other European countries might have more respect and there is still
more respect for this kind of profession. The average working singer cannot refuse to sing, and the
average company would not be prepared for the expense and the bother of putting in another singer, just
because she is having maybe a slightly not-so-good performance. I think they would not be that bothered,
but if you have the clout, if you have a big reputation, people would listen to what you would say. Yes, I
think that is perhaps why it is more prevalent in other countries in Europe. People in positions that have
much more to say and are more outspoken about what they want, achieve conditions that they are
prepared to work in. [Danielle]
My singing teacher actually experienced a haemorrhage during her menstruation, in one menstrual cycle,
because her vocal folds were very swollen. One of my friends is a Russian soprano, and when she sang
back home, she didn’t have to sing during the premenstrual phase of her menstrual cycle. It was in her
contract that during the days before her period she was not booked to do an opera, and every concert and
every opera for her was arranged that way. In England they are not aware of all the problems that female
singers have during their cycles, and it can be very soul-destroying. They do need to know because it’s
not that you want to cancel anything, it’s just that the voice is not there and you can’t change your body.
[Anna]
In these interviews, the singers showed how concerned they feel about their
voices, and that they are always thinking about how their voices will be in their next
performance. It is urgent to bring out these issues for a broader discussion, so a
consensus could be reached for all (European) countries about the importance and
practice of ‘respect days’.
Further research
This article is focused on the physiological and psychological effects of the
menstrual cycle on the female operatic singers and singing voices. However, in the
field of endocrinology and hormonal-related medication applied to vocal quality,
there is still great scope for research. Hormones have been used as therapy for many
years; however, they have frequently been used without knowledge of their side
effects on the voice. Earlier research concerning female synthetic sexual hormones
and voice quality suggested a predominance of adverse effects of oral contraception
on singers’ voices. Gelder (1974) enhanced the study by Zilstorff (1965) in which the
author reported the case of two Denmark sopranos who showed deeper voices after
taking an oral contraceptive pill (OCP) Enovid®. The natural pitch of their voices
recovered after stopping the use of the OCP (Gelder, 1974).
Additionally, Brodnitz (1971) described symptoms of hoarseness and loss of
high notes in a 19-year-old singing student who took an OCP containing
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
83
© 2005 Callaway Centre
norethyodrel mestranol (synthetic progestogen) for two months, and for three opera
singers who were prescribed with this OCP for longer periods of time. For both
cases, the abovementioned vocal symptoms stopped after cessation of this
contraceptive method (Brodnitz, 1971). Dordain (1972) recorded the extemporaneous
speaking voice, the projected speaking voice and the highest and lowest notes in the
vocal ranges of singers who were not taking any kind of oral contraception, singers
who were taking an OCP and singers who were using mechanical contraception. The
results identified a decrease in the vocal range in professional singers who took an
OCP, even over a brief period of time (Dordain, 1972). Other studies have also found
OCPs containing 19-norsteroid have a negative effect on the voice (Damsté, 1967;
Pattie et al., 1998; Davies & Jahn, 1999; Baker, 1999). The authors hypothesised a
swelling of the vocal folds under the influence of synthetic progestogens; however,
an increased mass of the vocal folds could not be demonstrated. Another OCP,
Ovosiston®, containing 2 mg Chlormadione acetate and 0.1 mg mestranol, was also
associated with virilisation of the voice (Wendler et al., 1995).
Regarding the observations obtained from the abovementioned studies,
during the later 1960s and 1970s professional voice users who were taking an OCP
were advised to avoid it, or undertake regular phoniatric examinations by ear, nose
and throat (ENT) surgeons. In the 1980s, as a consequence of the pharmacological
evolution of the OCP, in which hormonal dosages were decreased and other active
ingredients were used, there was a reduction of complaints of vocal changes.
On the other hand, recent studies have been unable to find any effects of
OCP use in voice quality. In a study by Wendler et al. (1995), the effects of two
different OCPs on voice quality were assessed: Diane-35® (0.035 mg ethinylestradiol
and 2 mg cyproterone) and Microgynon® (0.03 mg ethinylestradiol and 0.15 mg
levonorgestrel). The study assessed the mean speaking frequency, pitch range, voice
range profile and sound quality coefficient for women who were non-singers, aged
between 18 and 35 years old, during thirteen cycles. Phoniatric observations were
done between the 10th and 14th days of the 3rd, 6th and 12th cycles. Perceptual,
laryngological and acoustical observations of the quality of the voice and the state of
the vocal folds were performed. Perceptual observations were based on singers’
reports of vocal changes, such as complaints about vocal fatigue, sensation of lump
in the throat, need to clear throat, unpleasant sensations in the throat, hoarseness and
changes in the speaking range. Laryngological examinations were made by an ENT
surgeon. And finally, acoustical observations assessing the speaking frequency and
the voice quality before and 20 minutes after voice production, were performed by a
panel of five phoniatricians and speech pathologists, who assessed three different
voice categories: (1) hoarseness (total impression of the noise); (2) roughness (noise
component based on irregularities in the vocal cord vibration); (3) breathiness (noise
component based on turbulence resulting from insufficient closure of the glottis).
These three categories were quantified in a scale between 0 = not present, 1 = slight,
2 = moderate and 3 = severe; recordings of pitch range (in Hz) and dynamic range (in
dB) were made from the softest to the loudest voice intensities over the entire pitch
range, from the lowest to the highest tones using vowels /a/, /e/, /i/ and /u/.
The subjective reports of the singers did not show any vocal symptoms
related to this medication, such as vocal fatigue, feeling of lump in the throat, feeling
a need to clear the throat, unpleasant sensation in the throat, hoarseness and changes
in speaking range. Also, from the observations done by the ENT surgeon, no
significant alterations were noticed; however, a slight increase in loosening of the
vocal mucosa and greater irregularities and incomplete closure of the glottis could be
observed in some of the participants taking Microgynon®. Regarding the several
acoustic parameters that were studied, there were no alterations noticed in the mean
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
84
© 2005 Callaway Centre
speaking frequency or reported changes in the quality of the sound. Additionally, the
vocal ranges did not show significant alterations, although an increase of one octave
above could be observed in the 12th cycle. Therefore, this research did not find a
connection between negative vocal changes and the intake of an OCP. There were no
alterations in the acoustic characteristics of the voice or in the quality of the voice.
These findings lead us to believe that, with the pharmacological evolution of the new
OCPs, side effects on the voice have been reduced. However, the authors still advise
monitoring of professional voice users who take OCPs (Wendler et al., 1995).
Other authors have proposed that an OCP might help singers to overcome
vocal aberrations connected with the menstrual cycle. In a study by Isenberg, Brown
and Rothman (1983) the authors tested the effects of an oral contraceptive pill, Ortho-
Novum®, on the singing voice. They chose five singers who had irregular menstrual
cycles and previous problems of vocal symptoms connected with the menstrual
cycle, and they recorded their speaking voice reading the ‘Rainbow Passage’
(Fairbanks, 1940) and singing a high and a low note during two cycles. At the end of
the experiment, which was completed by three singers, their results suggested that
the OCP stabilised the body temperature and the speaking fundamental frequency
was raised to a higher pitch. From their findings, they proposed that an OCP may
help singers who had serious hormonal imbalances to stabilise their voices (Isenberg,
Brown & Rothman, 1983).
A similar idea was proposed by Amir, Kishon-Rabin and Muchnik (2002).
They compared a group of non-singers, five who were taking an OCP and five who
were not, over a period of approximately forty days, sustaining two vowels /i/ and
/a/ for five seconds. Acoustic parameters such as fundamental frequency (F0), jitter,
amplitude, shimmer and harmonic-to-noise ratio (HNR) were assessed. The results
suggested that there were significant differences in shimmer and jitter values
between the two groups of women. Women taking an OCP showed lower
perturbation values and smaller variance—parameters associated with healthier
voices—during the menstrual cycle. Furthermore, fundamental frequencies were
higher for those who were not OCP users. This study supported the hypothesis that
the OCP might stabilise the vocal quality across the menstrual cycle (Amir, Kishon-
Rabin & Muchnik, 2002).
The literature on the use of OCP is too scant for definitive conclusions
concerning the effects of the OCP on the quality of the classical singing voice to be
drawn. Some studies support the idea that an OCP might decrease the fundamental
frequency and range of the voice, mainly using older ‘first generation’ OCPs, initially
used in the early 1960s. Lower amounts of synthetic hormones with the same
contraceptive effectiveness are currently used in new OCPs, so fewer side effects are
expected. Other studies support the hypothesis that ovulatory inhibitors like the
OCP might reduce certain singers’ vocal difficulties connected with hormonal
fluctuation during the menstrual cycle, although the majority of studies concern only
the speaking voice. Those studies which looked at the singing voice were not
performed under a situation analogous to performance of repertoire. Given that a
high percentage of young singers use this contraceptive method during their training
and early career, it is important to investigate the effects of OCPs on their
performance.
To contribute to further research in this area, we are currently undertaking a
systematic investigation on the effects of a third generation OCP on the female
operatic singing voice, in collaboration with Professor William Ledger, Professor
David Howard and Dr Georgina Jones. This ongoing study is a double blind,
randomised, placebo-controlled trial involving ten healthy operatic singers who are
taking a placebo and a third generation OCP (Yasmin®) each during three
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
85
© 2005 Callaway Centre
consecutive menstrual cycles. Acoustic measurements of vocal fold vibrations during
the performance of a Lied, a French melody and one operatic aria from each singer’s
repertoire are measured using an electrolaryngograph. Blood samples are collected at
the end of each recording session to measure levels of oestrogen, progesterone,
testosterone, LH and FSH. The recording sessions and blood collections are
performed for each part of the study (at the third month of use of placebo and third
month of use of Yasmin®) at three different phases of the menstrual cycle: at the
second day of menstruation, and during follicular (around the 10th day of the
menstrual cycle) and luteal (around the 25th day of the menstrual cycle) phases of the
menstrual cycle. Through this fieldwork, we hope to advance research in this area, to
extend our understanding and to provide information that is likely to be of great
concern to the female singer.
Acknowledgments
This paper refers to research work being undertaken at the Music
Department of the University of Sheffield, with the collaboration of Professor
William Ledger, Academic Unit of Reproductive and Developmental Medicine,
University of Sheffield; Professor David Howard, Electronics Department, University
of York; Dr Georgina Jones, Institute of General Practice & Primary Care, University
of Sheffield and with the support of the Department of Clinical Chemistry, Royal
Hallamshire Hospital; Pharmacy Services Directorate, Royal Hallamshire Hospital;
Schering Health Care, Ltd UK and Fundação para a Ciência e a Tecnologia, Portugal.
References
Abitbol, J., Abitbol, P., & Abitbol, B. (1999). Sex hormones and the female voice. Journal of
Voice, 13, 424-446.
Abitbol, J., Brux, J., Millot, G., Masson, M. F., Mimoun, O. L., Pau, H., & Abitbol, B. (1989).
Does a hormonal vocal cord cycle exist in women? Study of vocal premenstrual
syndrome in voice performers by videoscopy-glottography and citology on 38 women.
Journal of Voice, 3, 157-162.
Abramson, A. L., Steinberg, B. M., Gould, W. J., Bianco, E., Kennedy, R., & Stock, R.
(1984). Estrogen receptors in the human larynx: clinical study of the singing voice. In
V. Laurence (Ed.), Transcripts of the thirteenth symposium: care of the professional
voice, Part II (pp. 409-413). New York: The Voice Foundation.
Amir, O., Kishon-Rabin, L., & Muchnik, C. (2002). The effect of oral contraceptives on
voice: preliminary observations. Journal of Voice, 16, 267-273.
Baker, J. (1999). A report on alterations to the speaking and singing voice of four women
following hormonal therapy with virilizing agents. Journal of Voice, 13, 496-507.
Brodnitz, F. S. (1971). Hormones and the human voice. Bulletin of the New York Academy of
Medicine, 47, 183–191.
Brown, W. S., & Hollien, H. (1981). Effect of menstruation on fundamental frequency of
female voices. In V. Lawrence (Ed.), Transcripts of the tenth symposium for the care
of the professional voice, Part I (pp. 94-101). New York: The Voice Foundation.
Brown, W. S., & Hollien, H. (1982). Effects of menstruation on the singing voice. In V.
Laurence (Ed.), Transcriptions of the eleventh symposium for the care of the
professional voice, Part I (pp. 140-147). New York: The Voice Foundation.
Brown, W. S., & Rothman, H. (1985). The menstrual cycle and its relationship to the highly
female voice. In V. Lawrence (Ed.), Transcripts of the fourteenth symposium for the
care of the professional voice (pp. 114-123). New York: The Voice Foundation.
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Number 24, 2005 Research Studies in Music Education
86
© 2005 Callaway Centre
Cooksey, J. M., Beckett, R. L., & Wiseman, R. (1984). A longitudinal investigation of
selected vocal, physiological, and acoustics factors associated with voice maturation in
the junior high school male adolescent. In E. M. Runfola (Ed.), Proceedings: research
symposium on the male adolescent voice, Part I (pp. 4-60). Buffalo, NY: State
University of New York at Buffalo Press.
Dalton, K. (1977). Premenstrual syndrome and progesterone therapy. Chicago: Year Book
Medical Publishers.
Damsté, P. H. (1967). Voice change in adult women caused by virilizing agents. Journal of
Speech and Hearing Disorders, 32, 126-132.
Davies, D. G., & Jahn, A. F. (1999). Care of the professional voice: a management guide for
singers, actors and professional voice users. Oxford: Butterworth Heinemann.
Dordain, M. (1972). Etude statistique de l’influence des contraceptives hormonaux sur la
voix. Folia Phoniatrica, 24, 86-96.
Fairbanks, G. (1940). Voice and articulation drillbook. New York: Harper and Brothers.
Flach, M., Schwickardi, H., & Simon, R. (1968). Welchen Einfluss haben Menstruation und
Schwangerschaft auf die ausgebildete Gesangsstimme? Folia Phoniatrica, 21, 199-
210.
Frable, M. A. N. (1961). Hoarseness, a symptom of premenstrual tension. Archives of
Otolaryngology, 75, 66-68.
Gelder, L. V. (1974). Psychosomatic aspects of endocrine disorders of the voice. Journal of
Communication Disorders, 7, 257-262.
Higgins, M. B., & Saxman, J. H. (1989). Variations in vocal frequency perturbation across the
menstrual cycle. Journal of Voice, 3, 233-243.
Isenberg, H., Brown, W. S., & Rothman, H. B. (1983). Effects of menstruation on the singing
voice; Part II: Further developments in research. In V. Laurence (Ed.), Transcriptions
of the twelfth symposium for the care of the professional voice, Part I (pp. 117-123).
New York: The Voice Foundation.
Lacina, V. (1968). Der Einfluss der Menstruation auf die Stimme der Sängerinnen. Folia
Phoniatrica, 20, 13-24.
Owen, J. A. (1975). Physiology of the menstrual cycle. The American Journal of Clinical
Nutrition, 28, 333-338.
Pattie, M. A., Murdoch, B. E., Theodoros, D., & Forbes, K. (1998). Voice changes in women
treated for endometriosis and related conditions: the need for comprehensive vocal
assessment. Journal of Voice, 12, 366-371.
Pressman, J. J., & Kelerman, G. (1970). Physiology of the larynx. Revised by John Kirchner.
Rochester, Minnesota: American Academy of Ophthalmology and Otolaryngology.
Sataloff, R. T. (1987). The professional voice: Part I. Anatomy, function, and general health.
Journal of Voice, 1, 283-292.
Schiff, M., & Burn, H. (1961). The effect of intravenous estrogens on ground substance.
Archives of Otolaryngology, 73, 43-51.
Silverman, E. M., & Zimmer, C. H. (1978). Effect of the menstrual cycle on voice quality.
Archives of Otolaryngology, 104, 7-10.
Titze, I. (1993). Critical periods of vocal change: puberty. The NATS Journal, 49, 24.
Weiss, D. A. (1950). The pubertal change of the human voice. Pholia Phoniatrica, 2, 126-
159.
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
Research Studies in Music Education Number 24, 2005
87
© 2005 Callaway Centre
Wendler, J., Siegert, C., Schelhorn, P., Kingler, G., Gurr, S., Kaufmann, J., Aydinlik, S., &
Braunschweig, C. (1995). The influence of Microgynon® and Diane-35®, two sub-fifty
ovulation inhibitors, on voice function in women. Contraception, 52, 343-348.
Whitehead, R. L., Kohler, R., & Schlueter, S. (1974). The effect of the menstrual cycle on
female vowel spectra. Presented at the annual meeting of the American Speech and
Hearing Association, Houston, Texas.
Wicklund, K., Seikel, A., Dilman, R., Flynn, S., Bond, V., Murphy, S., Neal, H., Newton, T.,
Hipp, E., Archambault, N., Jones, C., & Zarko, K. (1988). Fundamental frequency
changes in singers and non-singers related to menstruation. Medical Problems of
Performing Arts, 13, 100-103.
Wilson, F., & Purvis, J. (1980). A study of selected vocal behaviors during the menstrual
cycle of trained singers. Journal of Research in Singing, 10, 16-23.
About the Authors
Filipa Lã undertook her PhD studies at the Music Department of the
University of Sheffield, and recently has presented her study on the Effects of the
Menstrual Cycle and the Oral Contraceptive Pill on the Female Operatic Singing
Voice at the Second International Physiology & Acoustics of Singing Conference,
Denver, USA (2005), and at the 5th Pan European Voice Conference, Graz, Austria
(2003). She is taking up a post-doctoral research fellowship in Portugal in autumn
2005. Filipa has a background in biology and vocal performance, and is a
professional singer.
Jane W. Davidson is currently Chair of Music at both University of Sheffield
and The University of Western Australia. She has a background in music psychology,
musicology, vocal performance and contemporary dance. A former editor of
Psychology of Music, she is currently Vice-president of the European Society for the
Cognitive Sciences of Music. She has taught at undergraduate and postgraduate levels
for many years, contributing to courses on psychological approaches to performance,
development of musical ability, psychology for musicians, music therapy, music in
the community, gender studies in music, opera and music theatre studies, vocal
pedagogy and movement classes. Jane has written more than ninety scholarly
contributions on performance, expression, therapy and the determinants of artistic
abilities. Her edited volume, The music practitioner, explores the uses of research for
the practising musician. She has held visiting posts at Hong Kong Institute of
Education, University of New South Wales, University of Western Sydney, Guildhall
School of Music and Drama and the Luzern Konservatorium. Jane also works as a
professional stage director in opera and music theatre, having collaborated with
Andrew Lawrence-King, Opera North and Drama per Musica.
at Universidade De Aveiro on February 20, 2011rsm.sagepub.comDownloaded from
