Medical Hypotheses (2004) 63, 268273

http://intl.elsevierhealth.com/journals/mehy

Do the emotional side-effects of

hormonal contraceptives come
from pharmacologic or psychological mechanisms?
Stephen A. Robinson*, Matt Dowell, Dominic Pedulla, Larry McCauley

St. Anthony Family Practice Residency Program, 608 NW 9th, Suite 1000, Oklahoma City, OK 73102, USA

Received 23 June 2003; accepted 23 November 2003

Summary Hormonal preparations have become one of the most popular methods used for controlling fertility. The
literature over the last 40 years continues to reveal how their numerous side effects negatively impact many users and
even society at large. Three large cohort trials were the first to demonstrate, on a grand scale, certain emotional and
behavioral associations with contraceptive use. Current contraceptive use was associated with an increase rate in
depression, divorce, tranquilizer use, sexual dysfunction, and suicide and other violent and accidental deaths. Despite
the advent of more user friendly contraceptives, the discontinuation rate secondary to side effects has changed
little through the years. While in rare cases hormonal preparations can be deadly to the user, there is substantial
evidence that their negative effect issues more from their emotional and behavioral properties. This paper reviews the
results of over seven studies which further characterize these prominent associations, particularly with hormonal
contraception, in an attempt to demonstrate their association with the intrinsic pharmacologic properties of hormonal
preparations. Hormonal contraceptive users, in contrast with non users, were found to have higher rates of depression,
anxiety, fatigue, neurotic symptoms, sexual disturbances, compulsion, anger, and negative menstrual effects. The
question of whether the association of these maladies is directly due to the effect of taking exogenous hormones versus
the psychological impact of the contraceptive behavior itself had yet to be studied. Seven small randomized-controlled
trials were found in a review of the literature which studied this hypothesis in a direct way. They do not support the
origination of these side effects being from the pharmacological properties of hormones. No association was found
between hormone levels and emotional functioning in females. Psychiatric evaluations among IUD and oral
contraceptive pill (OCP) users reveal no significant differences. Women who were given an OCP placebo experienced
a similar side effect profile of OCP users. Different hormonal concentrations and combinations made no significant
difference in the side effect profile. A study of women who were given either weak female hormones or a placebo
failed to duplicate the side effect profile found in all of the other studies where the hormones were labeled as
contraceptives. The evidence suggests that most of the side effects of hormonal contraception are a result of a
psychological response to the practice of contraception. More study is warranted to further understand this
psychological phenomenon, especially now that an effective non-contraceptive method of fertility regulation and more
reliable psychological instruments are available. Furthermore, it is reasonable to hypothesize, given the present data,
that contraceptive activity itself is inherently damaging to women.

c 2004 Elsevier Ltd. All rights reserved.

*
Corresponding author. Tel.: +1-405-272-7494; fax: +1-405-
272-6985.
E-mail address: stephenarobinson@email.com (S.A. Robin-
son).

0306-9877/$ - see front matter c 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.mehy.2004.02.013
Emotional side-effects of hormonal contraceptives
The desire to control procreation has lead to the
advent of widely available hormonal contracep-
tion. With this wide spread availability has come
evidence of both physical and psychological ad-
verse effects, some of which have greatly affected
our society. The purpose of this paper is to: (1)
recapitulate how the side effects of hormonal
contraception have impacted our society; (2) de-
termine the prevalence of emotional/behavioral
side effects; and finally (3) demonstrate how these
emotional/behavioral effects may be caused by
psychological versus pharmacologic mechanisms
intrinsic to the hormones properties and give
plausible explanations.
Hormonal contraceptive side effects
Although research on hormonal contraceptive side
effects has been vastly under reported in recent
times, an awareness of the adverse effects is
important for every patient and physician who
encounters hormonal contraceptive users. As with
any medication, hormonal contraceptives have
side effects, some of which can even be deadly.
Dr. C.R. Kay and the Royal College of General
Practitioners were one of the first groups to study
the pill and its associated side effects. A cohort of
approximately 46,000 women were followed for
25 years. Six years into the study the first major
interim report was published which showed sev-
eral significant correlations among users. Among
those include: a 30% relative increase in depres-
sion, a fourfold decrease in libido, an increase in
divorce (RR 1.78) [1]. Twenty-five years after the
studys inception newer associations dealing with
physical side effects emerged; however the po-
tential psychological impact of this class of drugs
still remained exemplified by their association
with all violent and accidental causes of death. It
is interesting to note that in the first 10 years of
oral contraceptive pill (OCP) use, there was a
significant excess of mortality from all causes of
death. The authors of the study conclude from
review of the data that mortality fell with time to
become insignificant at the studys end. Some
commentators, however, have speculated that
the decline in mortality rate depicted at the end
of the study was rather due to study dynamics and
confounding influential factors that remain unac-
counted [5]. The 36% increase in neurotic de-
pression likewise was said to have declined with
discontinuation of OCP use. Given the large
number in the study and the length of follow up
time it is hard to specify the exact cause among
so many confounding variables. The study,
269
therefore, in a limited way concludes that the
effects of OCPs on mortality occur mainly in
current and recent users [2]. An interesting ob-
servation, however, is made in the interim report
concerning the fact that neurotic depression did
not vary with the hormonal dose given. The au-
thors speculate a possible non-pharmacological
effect as the cause [1].
Another cohort, the Oxford FPA study, also
begun in 1968 followed 17,000 young healthy up-
per-class women through 8 years. The purpose of
the study was to determine the beneficial and
harmful effects among OCPs, Diaphragms, and
IUDs. The harmful effects tended to be very sim-
ilar among all three groups which are thoroughly
categorized by medical systems. It is interesting to
note one of the few significant differences is that
OCP users had four times as many suicide attempts
compared with diaphragm users [3]. A third trial,
the Walnut Creek prospective cohort, of 10,135
women (less than half were taking OCPs when
enrolled) revealed similar side effects among ever
users of OCPs. Users were more likely to take
tranquilizers, more likely to die from accidents or
violence, and three times more likely to have ac-
cidents and suicides among ages 5064 [4]. An-
other study also of older women on hormone
replacement therapy (the British HRT Study
N 4544) likewise revealed a statistically in-
creased number of deaths from suicide [5]. Al-
though these side effects are considered by some
to be too small to discourage their use given the
benefit of such high contraceptive efficacy, many
users would disagree [6].
During the early years of hormonal contracep-
tive use, it was surprising to some how the side
effects could so lead to a very significant discon-
tinuation rate. Some of the aforementioned stud-
ies, as well as other recent studies, reveal
substantial rates of discontinuation secondary to
experiencing side effects. It appears that even with
the advent of newer more user friendly hormonal
preparations there has been little change in the
discontinuation rate and overall satisfaction. One
prospective trial of 76 women randomized to two
hormonal formulations showed a 33% discontinua-
tion rate because of side effects after six months of
use [7]. The various reasons for discontinuation
were as follows: 37% physical side effects; 33%
emotional side effects; 8% sexual side effects [7].
In 1993, Rosenfeld polled 257 women in rural
Tennessee in an attempt to uncover the level of
satisfaction with current contraceptive types
available. The level of dissatisfaction with OCPs
were as high as 43%. Interestingly, the degree of
satisfaction did not correlate with the percent
270
efficacy of the method used. Users cited their nu-
merous side effects as reason for dissatisfaction
[6]. Even the mainstream medical community gave
pause over the hormonal contraceptive impact on
mortality. In 1979, Valerie Beral proposed the in-
clusion of the number of deaths related to hor-
monal contraceptive use into the definition of
Reproductive Mortality because of its emerging
significance. While the mortality secondary to
pregnancy complications continued to decrease,
Beral [8] published a disturbing upward trend of
deaths related to complications from modern
contraceptive methods. Whether this trend has
continued is yet to be formally established. It is
clear from a review of the literature that the side
effects of hormonal contraceptives are significant
and have affected our society. They continue to
pose significant medical and psychosocial risks
which concern both users and physicians.
The importance of the emotional side
effects of hormonal contraception
Ever since the interim results of the first major
cohort studies were released, more attention has
been focused on the emotional and behavioral side
effects of hormonal contraceptives. In the late
1960s, Udry and Morris [9] formally noted the
prevalence of two life disturbing behavioral side
effects of the hormonal preparations: depression
and decreased libido. Kane et al. in trying to fur-
ther characterize the emotional effects, studied
and published the results of a perplexing disposi-
tion of white contraceptive users toward increased
anxiety, depression, headaches, fatigue, and irri-
tability compared with black females who used
significantly less hormonal contraception at that
time. Among the many reasons cited for these
survey results is that [f]emininity and motherhood
seem to be for the black women, less conflicted
and thus more likely a positive source of self-
esteem [10]. Although this observational study
among dissimilar groups has little objective con-
clusive weight in itself, it reflects the scientific
intrigue with such emotional effects. Again Kane
[10] presented a review of the literature in 1976
capsulating the results of studies which reveal
disturbing emotional reactions to OCPs: increased
psychiatric morbidity including increased neurotic
symptoms; increased inclination for depression on
the Hamilton Depression scale; increased mental
and sexual disturbances in the first six months of
OCP use. Over all there were 46% with adverse
reactions to OCPs, mostly depressive in nature.
Robinson et al.
Again, many theories as to the causation were
presented by Kane, however speculation still re-
mained. The 1980s brought continued research in
this matter when Gonzalez-Olguin et al. sent
questionnaires to 3060 women in an effort to
characterize the psychological profiles of 58.3%
contraceptive and the 41.7% noncontraceptive us-
ers. Non-contraceptive users had no particular
psychological profile in common; however, in-
creased compulsion was noticed in injectable hor-
mone users and increased tendency for phobias was
associated with IUD users [11]. Rubino-Watkins
et al. [12] used several psychological inventories
administered to 29 OCP users and 47 non-users
to show an increase in the level of anger and
depression among the hormonal contraceptive us-
ers. Ross et al. [13] found OCP users had higher
levels of neuroticism and negative menstrual ef-
fects after a prospective 70 day administration of
the NEO Personality Inventory Revised neuroticism
scale. (Neuroticism is characterized by these six
traits: anxiety, angry hostility, depression, self-
consciousness, impulsiveness, vulnerability.) The
list of studies documenting such adverse emotional
effects continues to evolve, while the existence of
these side effects has been established through
clinical research.
While relatively few will suffer the conse-
quences of the contraceptives emotional side
effects to the extreme of committing suicide,
many have sympathized with those of depression,
anxiety, and even apparent subsequent sexual
dysfunction which is commonly cited in the lit-
erature. Rosemarie Lincoln [14] in writing on the
loss of libido points to the following effects as
leading many times to the loss of sexual interest
in a relationship: depression, grief, the need for
self-punishment or control, unconscious anger.
Likewise Bozeman and Beck [15] demonstrated
that anger reduces mens sexual desire and pe-
nile tumescence. Not only does emotional pa-
thology seem to pose a problem in itself to
hormonal contraceptive users, but it also seems
to lead to further psychosocial and relational
problems.
The demonstrable link of the emotional
side effects of hormonal contraceptives
to psychological and not pharmacological
causes
As early as the 1960s, Udry et al. made attempts
at explaining why these side effects occur. The
Emotional side-effects of hormonal contraceptives
theories ranged from psychosocial explanations to
pharmacological effects of progesterone and es-
trogen, as well as a certain female sensitivity to
her own endocrine function. The complicating
factor at the time was the lack of any unifying
conclusion in the scientific literature. Any prevail-
ing theory could be easily trumped by another
study. For example, the theory that progesterone
is responsible for decreased sexual interest, as is
also observed in primates, seems to be refuted by
observations that some women who discontinue
such contraceptives become less interested in sex
than others [9]. Rubino-Watkins et al. [12] also
noticed no relation between blood levels of estro-
gen and progesterone and cognitive and emotional
functioning. One study by Herzburg et al. followed
218 OCP users and 54 IUD users and found no dif-
ference in psychiatric evaluation [16]. Several
landmark studies were published in the 1970s
which shed some light on the subject. Aznar-Ramos
et al. in Mexico gave a placebo to 147 women ages
1641 who had just had a miscarriage. The women
were told that they were taking OCPs. The women
were surveyed in monthly intervals over 12 months
and the following side effects were ascertained:
decreased libido 29.5%, headache 15.6%, inter-
menstrual bleeding during the first 2 months. The
side effect rate was not measurably different from
women who take hormonal contraceptives [17].
Unfortunately, the study did not employ a com-
parison or control arm of either hormonal contra-
ception users or women given nothing. Their
conclusion did, however, demonstrate not only the
difficulty in evaluating side effects of a drug due to
the effects of a placebo, but also how likely it is
that the emotional/cognitive side effect profile of
hormonal contraceptives could easily be induced
with a mere placebo. Goldzieher et al. in one study
found no difference in symptomatology among IUD
and OCP users after administering questionnaires.
This result would as well support the view that it is
not the pharmacodynamics that primarily impact
the individuals psyche and subsequent emotions
and behavior, but rather the belief that one is
contracepting is causing such a phenomenon.
Goldzieher et al. in 1971 decided to further study
this phenomenon. Prior studies had already sug-
gested such a non-pharmacologically induced
trend. For example, Richter showed that by
changing the color of the pill he was able to induce
a decrease in libido which then would return over
the ensuing months [18]. Goldziehers randomized
double-blinded placebo controlled trial placed 398
women (with no prior OCP use) into five groups:
placebo; high estrogen pill; combo pill; combo pill;
no estrogen pill. The women agreed to use a
271
standard vaginal contraceptive cream or foam in
addition to the OCP if requested. (The study also
employed a crossover design in the latter cycle.
After following the women over five cycles (1523
cycles total) the surveys revealed no significant
difference in the side effect profile between the
placebo and hormonal preparations, which again
strengthened the theory that the emotional side
effects were not pharmacologically induced.) Al-
though it does not prove that any significant dif-
ference in the side effect profile exists, it does say
that the possible pharmacologic significance is ac-
tually quite lower than expected given the study
sample size. They comment that two definite as-
sociations are unique to OCP users: they smoke
more heavily than the non-OCP user; they have a
higher prevalence of cervical dysplasia [19]. Al-
though trivial in statistical importance secondary
to size, Silbergeld et al. [20], in a small prospective
placebo controlled trial of eight women, again was
unable to find major changes in emotional effects
such as hostility, fear, anger, anxiety, or regret
among placebo and OCP users. Any correlations are
weak at best.
After noticing such an unusual phenomenon, a
few plausible explanations emerged. Johan Cull-
berg [21] reviewed these as well as other studies
and hypothesized that the seemingly psychologi-
cally induced side effects were due its symbolic
property; that it is something that prevents sexual
intercourse from being followed by pregnancy and
an unwanted child. In order to reduce these
symbolic effects, he designed a prospective study
geared toward studying the mental and menstrual
side effects of hormonal preparations with a pla-
cebo control. This design allowed for the dismissal
of contraceptive terminology and the ability to
only tell the women that they were taking weak
female hormones or a sugar pill [21]. The results
revealed a small but significant increase in de-
pressivedysphoric type reactions in the estrogen
predominant group compared to placebo, but the
mental effects shown to result directly from hor-
monal action were overall low in magnitude. Sur-
prisingly, there was no direct evidence of the
hormonal preparations causing adverse sexual re-
actions such as loss of libido; however, there was a
direct link between adverse mental reactions and
decreased sexual interest [21]. Cullbergs hypoth-
esis is further expounded by psychiatrist Bourgeois
[22] in his article Le psychiatre et les controles de
la procreation. He lists four levels at which psy-
chiatrists should be concerned with the imple-
mentation of hormonal contraception: (1) the
intrinsically linked side effects; (2) the ramifica-
tions of sterility; (3) the underlying motivations and
272
plans for OCP use in high risk psychiatric individu-
als; (4) the subconscious effects on societys view
of sexuality and procreation. Bourgeois also sees
the effects such as depression and loss of libido
stemming from primarily intrinsic emotional origin
and not pharmacologic. Like Cullberg, Bourgeois
sees the taking of a drug labeled as a contracep-
ceptive as having a significant effect on the psy-
che of the user. Contraception leads to a complete
dissociation between the pleasurable/relational
and procreational components which in turn places
a different value on sexuality itself [22,23]. If his
reasoning is correct, it may not be too difficult to
see why any intent to use contraception would it-
self have profound psychological influences and
deep emotional effects [2426].
Conclusion
This review of the literature highlights the trends
in contraceptive research regarding its observed
emotional and behavioral effects. Although it is
not exhaustive, it demonstrates the plausibility for
a primary psychological and not pharmacological
induction of adverse emotional effects in hor-
monal contraceptive users. Each study has its own
strengths and weaknesses, and it is not the intent
to twist the research into proving that such an
hypothesis is in fact true. The intent is to show
that such an hypothesis exists and continues to
bear validity over 40 years of research. It is
therefore still worth considering and studying. As
research continues, new tools are continually be-
ing developed to further aid in identifying cause
and effect relationships from correlatives and as-
sociations. The primary study designs in the above
studied literature have been limited to large co-
horts, small RCTs using placebo and non-users,
cross sectional analyses, and the use of standard-
ized psychological instruments. Given the signifi-
cant emotional disturbances associated with the
intentional administration of what a woman thinks
is a contraceptive is enough to warrant further
study. For if this association is accurate, then one
would have to consider the possibility that con-
traceptive behavior itself has inherent detrimental
side effects. Now is the time for a definitive
breakthrough in characterizing this relationship.
There is now solid research identifying a new ef-
fective way to postpone pregnancy called natural
family planning (NFP) [27]. By definition it would
technically not fall into the category of contra-
ception and would seem to not have such associ-
ated emotional side effects if the hypothesis
Robinson et al.
proposed is true. There are also very well vali-
dated psychological instruments available which
would allow research to uncover accurate associ-
ations of emotional disturbances in comparison to
a normal population control group. It is therefore
proposed that a study be designed to compare the
emotional reactions of contraceptive users with
NFP users using the most accurate psychological
instruments available, such as the CookMedley
[28] scale from the MMPI. After all, the sooner we
are able to discover the true cause(s) of such
emotional disturbances, the sooner we can find a
remedy.
References
[1] Kay CR, Crombie DL, Kuenssberg EV, et al. Oral contracep-
tives and health: an interim report from the oral contra-
ception study of the Royal College of General Practitioners.
London: The Whitefriars Press Ltd; 1974.
[2] Beral V, Hermon C, Kay C, Hannaford P, Darby S, Reeves R.
Mortality associated with oral contraceptive use: 25 year
follow up of cohort of 46,000 women from Royal College of
General Practitioners oral contraceptive study. BMJ 1999;
318:96100.
[3] Vessey MP, Doll R, Peto R, Johnson B, Wiggins P. A long-
term follow-up study of women using different methods of
contraception: an interim report. Journal of Biosocial
Science 1976;8:373424 [Oxford/FPA].
[4] Ramcharan S, Pellegrin FA, Ray R, et al. The walnut creek
contraceptive drug study: a prospective study of the side
effects of oral contraceptives. Vol III. Publication no. 81-
564 Center for Population Research Monograph, NIH,
Bethesda; 1981.
[5] Grant E. Sexual chemistry. Reading, Birkshire: Cox &
Wyman Ltd; 1995.
[6] Rosenfeld JA, Zahorik PM, Saint W, Murphy G. Womens
satisfaction with birth control. J Fam Pract 1993;36(2):
16973.
[7] Sanders SA, Graham CA, Bass JL, Bancroft J. A prospective
study of the effects of oral contraceptives on sexuality and
well-being and their relationship to discontinuation. Con-
traception 2001;64:518.
[8] Beral V. Reproductive mortality. BMJ 1979;2:6324.
[9] Udry JR, Morris NM. The biological effects of family
planning C. Behavioral effects of contraception. J Med
Educ 1969;44(Suppl 11):2837.
[10] Kane FJ, Lachenbruck PA, Lokey L, et al. Motivational
factors affecting contraceptive use. Am J Obstet Gynecol
1971;110(8):10505.
[11] Gonzalez-Olguin A, Romero-Gutierrez G, Malacara JM, Diaz
de Leon-Medina J. Personality profiles and socioeconomic
factors in contraceptive use. Contraception 1988;38(5):
5519.
[12] Rubino-Watkins MF, Doster JA, Franks S, et al. Oral
contraceptive use: implications for cognitive and emotional
functioning. J Nerv Ment Dis 1999;187:27580.
[13] Ross C, Coleman G, Stojanovska C, et al. Relationship
between the NEO personality inventory revised neuroticism
scale and prospectively reported negative affect across the
menstrual cycle. J Psychosom Obstet Gynecol 2001;22:
16576.
Emotional side-effects of hormonal contraceptives
[14] Lincoln R. Loss of libido. Practitioner 1989;233(1476):
13158.
[15] Bozeman AW, Beck JG. Covariation of sexual desire and
sexual arousal: the effects of anger and anxiety. Arch Sex
Behav 1991;20(1):4760.
[16] Kane FJ. Evaluation of emotional reactions to oral contra-
ceptive use. Am J Obstet Gynecol 1976;126(8):96872.
[17] Aznar-Ramos R, Giner-Velazquez J, Lara-Ricalde R, Marti-
nez-Manautou J. Incidence of side effects with contracep-
tive placebo. Amer J Obstet Gynec 1969;105:1144.
[18] Goldzieher JW, Moses LE, Averkin E, Scheel C, Taber BZ.
Nervousness and depression attributed to oral contracep-
tives: a double-blind, placebo-controlled study. Am J
Obstet Gynecol 1971;111(8):101320.
[19] Goldzieher JW, Moses LE, Averkin E, Scheel C, Taber BZ. A
placebo-controlled double-blind crossover investigation of
the side effects attributed to oral contraceptives. Fertil
Steril 1971;22:60923.
[20] Silbergeld S, Brast N, Noble EP, et al. The menstrual cycle:
a double-blind study of symptoms, mood and behavior, and
biochemical variables using enovid and placebo. Psychosom
Med 1971;33(5):41128.
273
[21] Cullberg Johan. Mood changes and menstrual symptoms
with different gestagen/estrogen combinations. Acta Psy-
chiat Scand 1972;236:186.
[22] Bourgeois M. Le psychiatre et les controles de la procre-
ation. LEncephale 1975;1:25963.
[23] Pedulla M. Abstract: The psychodynamics of pregnancy-
aversive sexuality (PAS). Postersession 3 Gender differ-
ences and psychiatric disorders II. U.K.2001:1-10: Available
from: www.akm.ch/wmh2001/programme/wed.htm.
[24] Bardwick JM. Psychology of women. NY: Harper & Row,
Pub.; 1971.
[25] Rohrbaug JB. Women, psychologys puzzle. NY: Basic Books
Inc, Pub.; 1971.
[26] van den Aardweg GJM. Contraceptive emotional and per-
sonality damage in women [A letter] Borgerweg 1,2111 CG
Aerdenhout, Netherlands, Nov 21; 2001.
[27] Secretariat for pro-life activities of the United States
Conference of Catholic Bishops under the microscope: NFP
& Marital Dynamics Help or Hinder?. CMR 2002;13(1,2):
1521.
[28] Cook WW, Medley DM. Proposed hostility and pharisaic-
virtue scales for the MMPI. J Appl Psychol 1954;6:4148.