BEHAVIORAL MEDICINE, 40: 1421, 2014

C Taylor & Francis Group, LLC

Copyright 
ISSN: 0896-4289 print/1940-4026 online
DOI: 10.1080/08964289.2013.829020

Dysmenorrhea, the Menstrual Cycle, and Sleep

Julie A. Woosley and Kenneth L. Lichstein
The University of Alabama

This study examined the relationship between dysmenorrhea and insomnia, as well as variability in sleep across the menstrual cycle. Participants were 89 women, ages 18 to 24
(M = 18.63, SD = 0.93), who completed daily surveys for five weeks. On the second day of
menses, they completed a questionnaire regarding dysmenorrhea. Participants having insomnia
rated their dysmenorrhea as being more severe and causing more interference with daily activities than did participants without insomnia. Insomnia severity was directly associated with
dysmenorrhea severity and interference. Sleep onset latency was longer and sleep efficiency
was lower in participants with severe dysmenorrhea than in those with mild dysmenorrhea.
Further, participants with mild dysmenorrhea reported significantly better sleep quality than did
those having moderate or severe dysmenorrhea. Additionally, wake time after sleep onset was
shortest and number of awakenings was lowest around the time of ovulation. Future research
should examine whether treating dysmenorrhea or insomnia alone results in improvements in
the other condition.
Keywords: dysmenorrhea, insomnia, menstrual cycle

Research has consistently shown that sleep disturbances are

more prevalent in women than in men.14 It seems plausible that this difference may be accounted for, in part, by
menstrual pain, as previous research has linked various other
pain conditions to disturbed sleep.5 Additionally, there is evidence that pain can be a risk factor in the development of
insomnia6. However, very little research has examined the
effects of dysmenorrhea (painful uterine cramps that begin
sometime between the hours preceding or the hours directly
following the onset of menses7,8) on sleep. One study9 found
that women with dysmenorrhea reported worse sleep quality ratings than asymptomatic women during menstruation.
Moreover, the dysmenorrheic participants had lower sleep
efficiency during menstruation relative to their other menstrual phases and to the asymptomatic participants during
menses. The authors attributed the reduced sleep efficiency
of the dysmenorrheics to increased wake time, movement
time, and Stage 1 sleep time.
In addition to sleep disturbances that may result from
dysmenorrhea, it seems that aspects of the menstrual cycle
(eg, recurring changes in temperature rhythms, variations
in the production of ovarian hormones, pituitary hormones,

Correspondence should be addressed to Julie A. Woosley, Department

of Psychology, Box 870348, University of Alabama, Tuscaloosa, AL 354870348. E-mail: jawoosley@gmail.com

melatonin, and cortisol10) could influence sleep. Indeed, the

literature suggests that womens core body temperatures are
most similar to those of men when women are in the follicular
phase, which includes menses and precedes ovulation.11 During the luteal phase, which follows ovulation and precedes
menses, womens body temperatures increase by approximately 4 C, with an attenuated decline in nocturnal temperature.1112 This increase in body temperature is likely due to
the rising level of progesterone during the luteal phase.11 It
seems plausible that increased nocturnal body temperature
during the luteal phase interferes with sleep.
In recognition of this possibility, researchers have examined the relationship between sleep and the menstrual cycle
and obtained significant results using subjective measures of
sleep. Studies have consistently found poorer sleep quality
(eg, worse sleep quality ratings, sleep onset latency, wake
time after sleep onset, and sleep efficiency) during the luteal
phase relative to other menstrual phases.1315 There is also evidence that sleep may be shorter14 and more disturbed around
the time of ovulation.15
Given that the majority of women of reproductive age
experience menstruation, and as many as 90% of women
aged 1845 report dysmenorrhea,16 the lack of research
in this area is surprising. This study is novel in that it is
the first to examine the relationship between insomnia and
dysmenorrhea. Additionally, the present study sought to
expand on the literature on the variability in sleep across the

DYSMENORRHEA AND SLEEP

menstrual cycle with a larger sample, as most of the aforementioned findings have been based on small sample sizes
(eg, n = 189, n = 812, n = 614). Further, many of the previous
studies examining variability in sleep across the menstrual
cycle have examined only a small sampling of nights from
each menstrual phase (eg, one night from the mid-follicular
phase, one night from the mid-luteal phase, and one night at
the onset of menses9; three nights from the follicular phase
and three nights from the luteal phase12). The present study
took a broader approach by collecting sleep data continuously throughout the menstrual cycle.
The current research sought to address the following research questions: (a) Do women with insomnia experience
more severe dysmenorrhea than women without insomnia?
(b) Is dysmenorrhea severity associated with commonly used
measures of insomnia? (c) Is there a relationship between
menstrual phase or dysmenorrhea and sleep diary parameters? and (d) Do dysmenorrhea and menstrual phase interact
to influence sleep diary parameters?

METHOD
Participants
The present study was conducted with undergraduate students because dysmenorrhea is more prevalent among
young women.1718 Research indicates that the prevalence of
dysmenorrhea among young women is between 64%19 and
approximately 72%.2021 For this study, 247 women were recruited and screened. In order to be included in the study, participants had to have regular menstrual cycles (with menses
occurring approximately once every four to five weeks). They
could not be pregnant or breastfeeding; use any type of hormonal birth control (eg, oral contraceptives, hormonallybased intrauterine devices, NuvaRing, implant, patch,
DepoProvera). Exclusion criteria were implemented in an
effort to minimize confounds. Women currently using
hormonal contraceptives were excluded because there is evidence that the exogenous hormones found in oral contraceptives influence core body temperature differently than
endogenous progesterone and estrogen.22 Additionally,
women taking oral contraceptives have a greater percentage of Stage 2 sleep and are less likely to experience dysmenorrhea than women with ovulatory cycles.20,22 Women
who had used hormonal birth control within the previous
three months were also excluded because recent research indicates that menstruation is altered for at least two months
following discontinuation of oral contraceptive pills.23 Additionally, women taking sedating or stimulating drugs (unless
these drugs were taken in the morning and not expected to be
sleep-active; women taking hypnotic medications for sleep
problems were included), those with unstable medical or psychiatric problems, and those with sleep disorders other than
insomnia were excluded. Women taking medications for in-

15

somnia were included so as to include as many participants

with insomnia as possible.
Of the 247 women recruited, 92 (37.2%) did not meet
criteria for the study due to symptoms of severe psychiatric
disturbance, use of sleep-active medications in the evenings,
presence of a sleep disorder (other than insomnia), current
hormonal contraceptive use, or recent discontinuation of hormonal contraceptive use; 13 participants dropped out of the
study; 14 were excluded due to irregular menstrual cycles;
and 39 were excluded due to missing or incomplete data. Participants were excluded from this study if they were missing
more than 50% of sleep diary entries during a particular
menstrual phase, if they missed more than seven sleep diary
entries, or if they missed entering data in the sleep diary on
more than three consecutive days. Due to the lack of prior
research monitoring sleep continuously across the menstrual
cycle, the aforementioned 50% cut-off was chosen in an effort to include data from as many participants as possible
while maintaining the integrity of the data. The participants
who were excluded due to missing data did not differ significantly from the final sample in terms of age (t = 2.11,
p = .054) or race ( 2 = 7.13, p = .129). The final sample
included 89 participants. All participants were given detailed
information about the study before agreeing to participate.
The study was approved by the University of Alabamas Institutional Review Board.
Measures

Screening Questionnaire. Participants were screened

to ensure they met inclusion criteria. The screening questionnaire included questions about demographics, typical
menstrual cycle length, use of hormonal contraceptives, and
pregnancy. The questionnaire also included questions from
the Structured Clinical Interview for DSM-IV Axis I Disorders, Clinical Version.24 Included were two questions inquiring about depressive symptoms, two about manic symptoms,
three about symptoms of substance abuse/dependence, one
about panic attacks, and one about dream symptoms of posttraumatic stress disorder (which could influence sleep). An
investigator-designed health survey was utilized to inquire
about physical and mental illness, symptoms of sleep disorders other than insomnia, and medication use. The final portion of the screening was the Pittsburgh Sleep Quality Index
(PSQI ),25 which assesses seven different components related
to sleep quality over the past month. A global score greater
than 5 is suggestive of significant sleep disturbance.25 The
PSQI has excellent psychometric qualities, with good internal consistency (.83.85) and testretest reliability (.85.87),
and a strong correlation with sleep diaries.2526 It also displays high sensitivity and specificity for people with insomnia compared to normal sleepers.2526
Insomnia Severity Index (ISI). The Insomnia Severity
Index (ISI) is a self-report questionnaire that includes three

16

WOOLSEY AND LICHSTEIN

questions related to sleep difficulties (ie, difficulty initiating or maintaining sleep, waking up too early) over the past
two weeks.27 Four other questions address perceptions about
sleep (eg, satisfaction with sleep, worry/distress about sleep
problems). Responses are summed, with a score of 07 indicating no clinically significant insomnia, 814 indicating
subthreshold insomnia, 1521 indicating clinical insomnia
of moderate severity, and 2228 indicating severe insomnia.
The ISI has strong psychometric properties, with good internal consistency (.74) and significant correlations with sleep
diary data.28

The International Classification of Sleep Disorders,

Second Edition (ICSD-2) diagnostic questions. This
investigator-designed self-report questionnaire includes 14
questions pertaining to the International Classification of
Sleep Disorders, Second Edition (ICSD-2) diagnostic criteria
for insomnia.29 To meet these criteria one must have difficulty
initiating sleep, difficulty maintaining sleep, or early morning
awakenings; at least one impairment of daytime functioning
(eg, fatigue, mood disturbance or irritability, worries about
sleep); and a symptom duration of at least one month.
Brief Pain Inventory (BPI). This assessment included
the pain severity and pain interference scales of the Brief
Pain Inventory (BPI),30 with instructions to respond to the
questions only with regard to menstrual pain within the past
24 hours. Each participants average score on the inventory
(completed on the second day of menses) was used to determine the severity of the dysmenorrhea. The BPI displays
high internal consistency (.85 for severity items and .88 for
interference items) and is responsive in its detection of improvements in pain with treatment.31 Consistent with the suggestion of the creators of the BPI and with common practice,
the item pertaining to pain severity at its worst (ie, Please
rate your pain by choosing the one number that best describes
your pain at its WORST in the last 24 hours) served as the
primary response variable for pain severity in the present
study, and the mean of the ratings for the interference items
were used to measure pain interference.3233
Consensus Sleep Diary (CSD). The Consensus
Sleep Diary (CSD)34 consists of nine questions (eight of
which are open-ended) that are answered each day. Individuals are asked to indicate what time they went to bed, what
time they began trying to sleep, how long it took them to fall
asleep, how many times they awoke during the night, how
long (in total) their awakenings lasted, what time they awoke
for the last time, and what time they got out of bed for the
day. They are also asked to describe their sleep quality by selecting one the following: very poor, poor, fair, good, or very
good. Finally, participants are given a blank space in which
to comment about their sleep. A question was added to the
CSD to inquire about the presence of menstrual flow the pre-

vious day. This information was used to track participants

menstrual cycles.
Participants responses to the CSD were used to determine their bedtimes, out of bed times, sleep onset latency,
wake time after sleep onset, and number of awakenings. This
information allowed for calculations of total sleep time and
sleep efficiency. Participants time in bed was calculated from
their bedtimes and out of bed times. All wake time (sleep onset latency, wake time after sleep onset, and morning wake
time) was subtracted from time in bed to determine total
sleep time. To determine sleep efficiency, total sleep time
was divided by time in bed. These variables and their definitions conform to the recommendations of the Pittsburgh
Consensus Conference on evaluating insomnia.35 Further,
the researchers at this conference recommended including
sleep diaries in the standard set of research assessments for
insomnia.35
It is inherently difficult to establish the psychometric properties of the CSD for a number of reasons. As noted by the creators of the CSD, sleep can vary greatly from night to night,
making it difficult to establish testretest reliability. Further,
the CSD is intended to measure several different aspects of
sleep that are not expected to correlate with one another (eg,
number of awakenings and sleep onset latency).34 As such,
measurement of internal consistency would be inappropriate.
However, as asserted by a panel of sleep experts, sleep diary
data provide valuable information about perceptions of sleep
that may not necessarily agree with objective measures, as
objective measures and subjective measures evaluate different phenomena.34
Procedure
Participants were recruited from the psychology department
subject pool website and from a flyer emailed to an undergraduate class in the department of music. Participants
completed the entirety of the study online. Participants were
first asked to complete the screening questionnaire to determine their eligibility for the study. Those who met inclusion
criteria then completed the ISI and ICSD-2 diagnostic questions. After completing these questionnaires, they began the
35 days of sleep diaries. The online survey was designed
so that participants were prompted to complete the pain inventory when they responded affirmatively to the question
inquiring whether they had started menses the previous day.
Thus, they completed the pain inventory on the second day
of menses at the same time as they completed the CSD. Responses to all questionnaires were date- and time-stamped to
ensure timely completion.
Participants were instructed to complete the CSD as soon
as possible after awakening in the morning, and they received automated daily email reminders to complete their
sleep diary. On days when a participant failed to complete the
CSD prior to going to bed the following night, the data were
considered missing. As previously stated, participants were

DYSMENORRHEA AND SLEEP

excluded if they missed more than 50% of sleep diaries in a

particular menstrual phase, if they missed more than seven
sleep diaries, or if they missed sleep diaries on more than
three consecutive days. Each of the participants who were
included, on average, failed to complete the CSD on 2.73 of
35 days. At the end of the study, those participants who were
recruited from the subject pool received class credit for their
participation.

RESULTS
Participant Characteristics
Iincluded in these analyses were 89 participants ranging in
age from 18 to 24 years (M = 18.63, SD = 0.93). Of the sample, 83.1% identified as Caucasian/White, 10.1% as African
American/Black, 3.4% as Asian or Asian American, 1.1%
as Latina/Hispanic, 1.1% as Filipino American, and 1.1%
as Black/Hispanic. The mean body mass index for the sample was 23.56 (SD = 4.51). The mean score on the PSQI
(Cronbachs = .62) was 4.71 (SD = 2.42) and the mean
score on the ISI (Cronbachs = .87) was 5.21 (SD = 3.89).
These mean scores are below the thresholds that would indicate sleep problems, indicating that the overall sample was
comprised of relatively good sleepers.

Comparison of Insomnia and Non-Insomnia

Groups
Using participants responses to the ICSD-2 questionnaire
(Cronbachs = .83; three linearly dependent items that prevented matrix inversion were omitted. Cronbachs was calculated from the remaining 11 items) and quantitative sleep
diary criteria36 (ie, average sleep onset latency or wake time
after sleep onset 20 minutes over the five-week period),
14 of 89 participants (15.7%) met criteria for a diagnosis of
insomnia. Participants with insomnia (PWI) were compared
to participants not having insomnia (PNI) on six sleep diary
parameters. Using a Bonferroni-adjusted alpha level of .008,
PWI had significantly lower sleep efficiency and longer sleep
onset latency and wake time after sleep onset than PNI (see
Table 1).
To examine the relationship between insomnia diagnosis and dysmenorrhea, two one-way ANOVAs were conducted with dysmenorrhea severity at its worst and dysmenorrhea interference (Cronbachs = .93) as outcomes.
Using a Bonferroni-adjusted alpha level of .025, insomnia diagnosis significantly predicted dysmenorrhea severity at its
worst. Participants with insomnia rated their worst menstrual
pain as being significantly more severe than PNI (Table 1).
Additionally, PWI experienced significantly more interference with daily activities due to dysmenorrhea than PNI
(Table 1).

17

TABLE 1
Sleep and Pain Measures by Insomnia Status

Measure

Insomnia Present Insomnia Absent

Mean (SD)
Mean (SD)

Sleep onset
33.57 (12.55)
latency (min)
11.34 (8.04)
Wake time after
sleep onset
(min)
Number of
0.94 (0.55)
awakenings
Total sleep time 461.58 (40.61)
(min)
Sleep efficiency
89.03 (3.75)
(%)
Sleep quality
3.67 (0.36)
rating
Dysmenorrhea
6.32 (3.17)
severityworst
Dysmenorrhea
3.32 (2.57)
interference

16.67 (10.12) 30.42 <.001 .259

6.50 (4.75)

9.59

.003 .099

0.88 (0.73)

0.10

.759 .001

453.32 (43.44)

0.44

.511 .005

93.06 (3.27)

17.09 <.001 .164

3.91 (0.46)

3.32

.072 .037

4.42 (2.30)

7.11

.009 .075

1.85 (1.94)

6.03

.016 .065

Dysmenorrhea and Sleep

To further examine the relationship between insomnia and
dysmenorrhea, the relationship between dysmenorrhea severity/interference and two additional sleep measures was considered: the PSQI and the ISI. To determine whether dysmenorrhea severity is related to sleep disturbance as measured by
the PSQI and the ISI, bivariate correlations were conducted
between participants dysmenorrhea ratings and these two
sleep measures. Severity ratings of dysmenorrhea at its worst
correlated significantly with ISI scores, with greater pain
severity associated with more disturbed sleep (r = .33, p =
.001), but not with PSQI scores (r = .18, p = ns). Dysmenorrhea interference (ie, interference with daily functioning as
a result of dysmenorrhea) was significantly correlated with
scores on the ISI (r = .37, p < .001) and the PSQI (r = .26,
p = .02), such that greater pain interference was associated
with increased sleep disturbance.
Relation of Menstrual Cycle and Dysmenorrhea
with Sleep Variables
Sleep diary data for each participant were divided into follicular phase, ovulation phase, and luteal phase, and the mean
of each variable for each phase was calculated. The twelve
nights prior to the onset of menses were included in the luteal
phase. The ovulation phase included the nights that occurred
13 to 15 days prior to the onset of menses. The remaining
nights, which included the menstrual period, were included
in the follicular phase. A mean for each variable within each
phase was calculated.
Participants were grouped according to dysmenorrhea
severity based on their worst level of pain at the onset
of menses. Typically on the BPI, a worst pain score of

18

WOOLSEY AND LICHSTEIN

TABLE 2
Sleep Measures by Menstrual Phase

Measure
Sleep onset latency (min)
Wake time after sleep onset (min)
Number of awakenings
Total sleep time (min)
Sleep efficiency (%)
Sleep quality rating

Follicular
Mean (SD)

Ovulation
Mean (SD)

Luteal
Mean (SD)

18.60 (12.47)
7.39 (6.70)
0.91 (0.70)a
455.37 (48.99)
92.68 (3.59)
3.84 (0.47)

20.17 (19.91)
5.54 (6.76)a
0.69 (0.71)b
453.52 (83.42)
92.45 (6.74)
3.95 (0.70)

19.92 (14.40)
7.62 (6.74)b
0.92 (0.84)a
454.48 (44.12)
92.07 (4.38)
3.88 (0.50)

0.48
4.99
7.68

.619
.010
.001

.006
.055
.082

Note. Follow-up testing was not conducted for total sleep time, sleep efficiency, and sleep quality rating due to a non-significant MANOVA. Means with
different subscripts differ at p < .05.

0 indicates no pain, a score in the range of 1 to 4 indicates mild pain, a score in the range of 5 to 6 indicates moderate pain, and a score of 710 indicates severe
pain.3739 Because only one participant rated her worst menstrual pain as 0, she was included in the mild pain group,
resulting in three pain severity groups (mild, moderate, and
severe).
A mixed design 3 3 (menstrual phase as repeated
measure dysmenorrhea severity) MANOVA evaluated
measures of wakefulness (sleep onset latency, number of
awakenings, and wake time after sleep onset). There were significant main effects for menstrual phase (F 6,81 = 2.38, p =
.036; Wilks  = 0.85, 2 = .150) and dysmenorrhea severity (F 6,168 = 2.41, p = .029; Wilks  = 0.85, 2 = .079),
but no significant interaction. Univariate follow-up testing for
menstrual phase revealed that wake time after sleep onset and
number of awakenings both reached statistical significance
(see Table 2). Subsequent Bonferroni-adjusted repeated measures t-tests were used for follow-up testing. These revealed
a significant difference in wake time after sleep onset between the luteal and ovulation phases (t = 2.65, p = .010),
with wake time after sleep onset being significantly longer
during the luteal phase (see Table 2 for phase means) than
during the ovulation phase. There was also a significant difference in number of awakenings between the follicular and
ovulation phases (t = 3.23, p = .002) and between the ovulation and luteal phases (t = 3.51, p = .001), with number of

awakenings being significantly less during the ovulation

phase than during the follicular or luteal phases.
The results for the dependent variables for each dysmenorrhea severity group were also considered separately,
with sleep onset latency reaching statistical significance (see
Table 3). Bonferroni and Newman-Keuls pairwise comparisons did not reveal any significant differences. Consequently, Duncans new multiple range test was used. This
revealed that participants having mild dysmenorrhea had
shorter sleep onset latency than those having severe dysmenorrhea (p < .05; see Table 3 for means).
A mixed design 3 3 (menstrual phase as repeated measure dysmenorrhea status) MANOVA examined global
measures of sleep (sleep efficiency, total sleep time, and
sleep quality ratings). This analysis did not yield a significant main effect for menstrual phase or significant interaction
effects. However, there were significant main effects for dysmenorrhea severity groups (F 6,168 = 3.32, p = .004; Wilks
 = 0.80, 2 = .106). Sleep efficiency and sleep quality
ratings both reached statistical significance (see Table 3).
Bonferroni pairwise comparisons revealed that participants
having mild dysmenorrhea had significantly greater sleep efficiency than those having severe dysmenorrhea (p = .013;
see Table 3 for group means). Additionally, participants having mild dysmenorrhea reported significantly better sleep
quality ratings than participants having moderate or severe
dysmenorrhea (p = .031 and p = .023, respectively).

TABLE 3
Sleep Measures by Dysmenorrhea Severity

Measure

Mild
Mean (SEM)

Moderate
Mean (SEM)

Severe
Mean (SEM)

Sleep onset latency (min)

Wake time after sleep onset (min)
Number of awakenings
Total sleep time (min)
Sleep efficiency (%)
Sleep quality rating

16.08 (1.79)a
5.85 (0.76)
0.76 (0.10)
451.90 (7.28)
93.53 (0.56)a
4.05 (0.07)a

22.35 (2.83)
6.44 (1.21)
0.66 (0.15)
464.43 (11.50)
91.80 (0.88)
3.72 (0.11)b

23.67 (2.35)b
8.88 (1.01)
1.10 (0.13)
451.99 (9.57)
90.85 (0.73)b
3.74 (0.09)b

3.91
2.96
3.08
0.47
4.56
5.42

.024
.057
.051
.626
.013
.006

.083
.064
.067
.011
.096
.112

Note. Means with different superscripts differ at p < .05.

DYSMENORRHEA AND SLEEP

Examination of Hypnotic Use as a

Potential Confound
To determine whether participants taking hypnotics for sleep
problems (a possible confound in the above analyses) and
participants not taking hypnotics differ on total sleep time,
sleep efficiency, sleep onset latency, number of awakenings,
and wake time after sleep onset, participants were divided
into three groups based on their responses to the PSQI item
inquiring about sleep medication use (ie, no medication use
in the past month, less than once a week, or once or twice
a week; none of the participants indicated that they used
medication three or more times a week). For each of the
aforementioned outcome variables, a one-way ANOVA was
conducted, and all were not significant at p < .05.

DISCUSSION
Participants with insomnia had higher scores on dysmenorrhea severity and interference than PNI, dysmenorrhea
severity was directly related to ISI scores, and dysmenorrhea interference was directly related to scores on the ISI and
the PSQI. Additionally, sleep fragmentation (wake time after
sleep onset and number of awakenings) varied between menstrual phases. Sleep onset latency, number of awakenings,
sleep efficiency, and sleep quality ratings varied across dysmenorrhea severity groups. Scores on sleep variables were
comparable for those participants using hypnotics and those
not using hypnotics, suggesting that this is not a confounding
factor.
The findings that dysmenorrhea severity and interference
is higher in women with insomnia than in women not having
insomnia, and that measures of dysmenorrhea are directly
related to measures of sleep disturbance are consistent with
past research on the relationship between sleep and pain.
However, in light of research suggesting that pain can influence sleep and vice versa,eg,40 the direction of causality in
the present study cannot be ascertained. Longitudinal studies,
particularly those studying girls around the age of menarche,
would be most beneficial in determining the primary directionality of this relationship.
The finding that wake time after sleep onset was shortest
and that number of awakenings was lowest during the ovulation phase is a relatively novel finding. However, it seems
that only two studies on sleep across the menstrual cycle have
examined sleep during the time of ovulation. One of these
studies14 found that estimates of total sleep time provided by
six participants were shortest around the time of ovulation,
and the other found (using a five-factor rating scale) that difficulty initiating and maintaining sleep were worse around the
time of ovulation and prior to menses.15 These findings were
not replicated in the present study. Given the differences in
measures used, it is not surprising that the results from previous research and the present study are not fully congruent.

19

Perhaps the longer wake time after sleep onset and increased
number of awakenings observed in the present study during
the follicular and luteal phases are due to such factors as dysmenorrhea and other premenstrual symptoms (eg, dysphoria,
bloating, intestinal upset, etc.) in the late luteal and early follicular phases, the increase in estrogen in the late follicular
phase, and the increase in progesterone (and thus, core body
temperature) in the luteal phase.
The lack of significant differences between other sleep
variables across the menstrual cycle is somewhat inconsistent
with previous research. This could be due in part to methodological differences. A number of previous studies have examined a small number of nights from each phase, whereas
the present study included data from nights across each phase.
For example, one study41 compared nights from the midfollicular phase with nights from the mid-luteal phase, and
another13 compared nights from the mid-follicular phase to
those from the late luteal phase. Further, there are differences in how phases are defined between studies. The algorithm used to divide nights into phases in the current study
is described in the procedures section. In contrast, Manber
and Bootzin13 categorized the 6 days preceding menstruation
as late luteal and days 7 through 12 of the cycle as midfollicular. The algorithm used in the present study seems
preferable, as it adjusts for variation in the length of the
menstrual cycle across individuals.
The current findings pertaining to dysmenorrhea are relatively consistent with previous research, which has found
(using polysomnography) that women with dysmenorrhea
have lower sleep efficiency during the follicular and luteal
phases when compared with their asymptomatic counterparts.9 However, in contrast to previous findings, the present
study found that women having mild dysmenorrhea reported
better sleep quality ratings than those having moderate or
severe dysmenorrhea throughout the menstrual cycle and not
only during menstruation. Additionally, the present study
failed to find the significant interaction effects found in previous research.9 This could possibly be attributed to sleep
measurement differences (ie, sleep diary vs. polysomnography) and pain measurement differences (ie, Brief Pain Inventory vs. visual analog scale and McGill Pain Questionnaire).
Additionally, Baker and colleagues9 asked participants to
rate pain severity first thing in the morning prior to taking
any pain medication. The time of completion of the pain
measure relative to the time of ingestion of pain medication
was not considered in the present study.
The absence of phase dysmenorrhea interactions in the
present study suggests that the relationship between menstrual pain and sleep is not just limited to the time of menses,
but persists throughout the cycle. Indeed, although Baker
and colleagues9 did find the aforementioned interaction,
they also found physiological differences between women
with and without dysmenorrhea that persisted throughout the
menstrual cycle. Specifically, they found that symptomatic
women experienced elevated estrogen levels and body

20

WOOLSEY AND LICHSTEIN

temperatures as well as reduced REM sleep throughout the

menstrual cycle when compared with asymptomatic women.
There were several limitations in the present study that
should be noted. One major limitation is that hormone levels
were not measured. Therefore, ovulation could not be confirmed. Additionally, although the sample size in the present
study was greater than the sample sizes used in previous
research in this area, it would have been helpful to have additional participants to increase power and confidence in these
findings.
One final limitation is that participants were not given any
instructions regarding the timing of pain medication administration, which may have led to confounds in the present data.
For example, if participant A rated her pain as being severe
prior to taking pain medication, but then took pain medication
continuously throughout the rest of menses and participant B
rated her pain as being mild after taking pain medication, but
then stopped taking pain medication for the rest of menses,
participant B may have experienced more sleep disturbance
related to pain than participant A. Such a phenomenon would
be consistent with previous research that found that women
with dysmenorrhea who were administered diclofenac potassium (a non-steroidal anti-inflammatory drug) showed improvement in sleep efficiency (as measured by polysomnography) and sleep quality ratings relative to when they were
administered a placebo.42
Future research should be conducted in an effort to replicate these findings and those from previous studies. It would
be helpful for future studies to include both objective and subjective measures of sleep. It would also be useful to conduct
additional research to determine whether treating dysmenorrhea (eg, using pain medications, mindfulness-based cognitive therapy for pain) results in improved prevention and/or
treatment of insomnia. Similarly, research could investigate
whether treating insomnia (eg, using cognitive-behavioral
therapy for insomnia or hypnotic medications) influences
dysmenorrhea severity and interference with daily activities.
Such treatment studies could allow for casual inferences.
Additional research should also examine other aspects of
dysmenorrhea that could influence sleep in addition to severity and interference. For example, it may be helpful to know
how long each participant experiences dysmenorrhea and
when the pain is at its worst (eg, in the morning, afternoon,
or evening). It could also be useful to assess participants
cognitions or emotions about their pain and determine how
those factors relate to sleep outcomes.
This study provides evidence of a relationship between
insomnia and dysmenorrhea severity and interference. The
results also indicate that sleep fragmentation is decreased
around the time of ovulation relative to the follicular and
luteal phases. The finding that insomnia and dysmenorrhea
are related has important implications for young women,
who are more likely to experience dysmenorrhea.1718 It is
the authors hope that these findings will be an impetus for
future research into interventions that will ultimately improve

the physical and emotional health and functioning of young

women.

REFERENCES
[1] Roberts RE, Shema SJ, Kaplan, GA. Prospective data on sleep complaints and associated risk factors in an older cohort. Psychosom Med.
1999;61:188196.
[2] Lichstein KL, Durrence HH, Riedel BW, Taylor DJ, Bush AJ.
Epidemiology of sleep: age, gender, and ethnicity. Mahwah, NJ:
Lawrence Erlbaum Associates; 2004.
[3] Leger D, Guilleminault C, Dreyfus JP, Delahaye C, Paillard M. Prevalence of insomnia in a survey of 12,778 adults in France. J Sleep Res.
2000;9:3542.
[4] Paine SJ, Gander PH, Harris R, Reid P. Who reports insomnia? Relationships with age, sex, ethnicity, and socioeconomic deprivation.
Sleep. 2004;27:11631169.
[5] Moldofsky H. Sleep and pain. Sleep Med Rev. 2001;5:387398.
[6] Leblanc M, Merette C, Savard J, Ivers H, Baillargeon L, Morin CM.
Incidence and risk factors of insomnia in a population-based sample.
Sleep. 2009;32:10271037.
[7] Durain D. Primary dysmenorrhea: assessment and management update. J Midwifery Wom Heal. 2004;49:520528.
[8] Lefebvre G, Pinsonneault O, Antao V, et al. Primary dysmenorrhea
consensus guideline. J Obstet Gynaecol Can. 2005;27:11171146.
[9] Baker FC, Driver HS, Rogers GG, Paiker J, Mitchell D. High nocturnal
body temperatures and disturbed sleep in women with primary dysmenorrhea. Am J Physiol Endocrinol Metab. 1999;277:E10131021.
[10] Moline ML, Broch L, Zak R, Gross V. Sleep in women across
the life cycle from adulthood through menopause. Sleep Med Rev.
2003;7:155177.
[11] Baker FC, Driver HS. Circadian rhythms, sleep, and the menstrual
cycle. Sleep Med. 2007;8:613622.
[12] Shibui K, Uchiyama M, Okawa M, et al. Diurnal fluctuation of sleep
propensity and hormonal secretion across the menstrual cycle. Biol
Psychiat. 2000;48:10621068.
[13] Manber R, Bootzin RR. Sleep and the menstrual cycle. Health Psychol.
1997;16:209214.
[14] Patkai P, Johannson G, Post B. Mood, alertness and sympatheticadrenal medullary activity during the menstrual cycle. Psychosom
Med. 1974;34:503512.
[15] Ishizuka Y, Usui A, Shiraishi K, et al. A subjective evaluation of sleep
during the menstrual cycle. Sleep Res. 1989;18:421.
[16] Jamieson DJ, Steege JF. The prevalence of dysmenorrhea, dyspareunia, pelvic pain, and irritable bowel syndrome in primary care practices. Obstet Gynecol. 1996;87:5558.
[17] Latthe P, Mignini L, Gray R, Hills R, Khan K. Factors predisposing women to chronic pelvic pain: a systematic review. BMJ.
2006;332:749.
[18] Patel V, Tanksale V, Sahasrabhojanee M, Gupte S, Nevrekar P. The
burden and determinants of dysmenorrhoea: a population-based survey of 2262 women in Goa, India. BJOG. 2006;113:453463.
[19] Ortiz MI. Primary dysmenorrhea among Mexican university students:
prevalence, impact and treatment. Eur J Obstet Gynecol Reprod Biol.
2010;152:7377.
[20] Andersch B, Milsom I. An epidemiologic study of young women with
dysmenorrhea. Am J Obstet Gynecol. 1982;144:65560.
[21] Unsal A, Ayranci U, Tozun M, Arslan G, Calik E. Prevalence of
dysmenorrhea and its effect on quality of life among a group of female
university students. Ups J Med Sci. 2010;115:138145.
[22] Baker FC, Mitchell D, Driver HS. Oral contraceptives alter sleep
and raise body temperature in young women. Pflugers Archiv: Eur J
Physiol. 2001;424:729737.

DYSMENORRHEA AND SLEEP

[23] Nassaralla CL, Stanford JB, Daly KD, Schneider M, Schliep KC,
Fehring RJ. (2011). Characteristics of the menstrual cycle after discontinuation of oral contraceptives. J Womens Health. 2011;20:169
177.
[24] First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical
Interview for DSM-IV Axis I Disorders, Clinical Version. Washington,
DC: American Psychiatric Press; 1996.
[25] Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The
Pittsburgh Sleep Quality Index (PSQI): a new instrument for psychiatric research and practice. Psychiat Res. 1989;28:193213.
[26] Backhaus J, Jughanns K, Broocks A, Riemann D, Hohagen F.
Testretest reliability and validity of the Pittsburgh Sleep Quality Index in primary insomnia. J Psychosom Res. 2002;53:737
740.
[27] Morin CM. Insomnia: Psychological Assessment and Management.
New York: The Guilford Press; 1993.
[28] Bastien CH, Vallieres A, Morin CM. Validation of the Insomnia Severity Index as an outcome measure for insomnia research. J Sleep Med.
2000;2:297307.
[29] American Academy of Sleep Medicine. The International Classification of Sleep Disorders, Second Edition (ICSD-2): Diagnostic
and Coding Manual. Westchester, IL: American Academy of Sleep
Medicine; 2005.
[30] Cleeland CS. Measurement of pain by subjective report. In: Chapman
CR, Loeser JD, eds. Advances in Pain Research and Therapy: Issues
in Pain Measurement. Vol. 12. New York: Raven Press; 1989. 391
403.
[31] Tan G, Jensen MP, Thornby JI, Shanti BF. Validation of the Brief
Pain Inventory for chronic nonmalignant pain. J Pain. 2004;5:133
137.
[32] Cleeland CS, Ryan KM. Pain assessment: global use of the Brief Pain
Inventory. Ann Acad Med Singapore. 1994; 23:129138.

21

[33] Atkinson TM, Mendoza TR, Sit L, Passik S, Scher HI, Cleeland C,
Basch E. The Brief Pain Inventory and its pain at its worst in the
last 24 hours item: clinical trial endpoint considerations. Pain Med.
2010;11:337346.
[34] Carney CE, Buysse DJ, Ancoli-Israel S, et al. The Consensus
Sleep Diary: standardizing prospective sleep self-monitoring. Sleep.
2012;35:287302.
[35] Buysse DJ, Ancoli-Israel S, Edinger JD, Lichstein KL, Morin CM.
Recommendations for a standard research assessment of insomnia.
Sleep. 2006;29:11551173.
[36] Lineberger MD, Carney CE, Edinger JD, Means MK. Defining insomnia: quantitative criteria for insomnia severity and frequency. Sleep.
2006;29:47985.
[37] Pain Research Group. (n.d.). The University of Texas MD
Anderson Cancer Center web site. http://www3.mdanderson.org/
depts/prg/bpi.htm.
[38] Li KK, Harris K, Hadi S, Chow E. What should be the optimal cut points for mild, moderate, and severe pain? J Palliat Med.
2007;10:13381346.
[39] Serlin RC, Mendoza TR, Nakamura Y, Edwards KR, Cleeland CS.
When is cancer pain mild, moderate or severe? Grading pain severity
by its interference with function. Pain. 1995;61:277284.
[40] OBrien EM, Waxenberg LB, Atchison JW, Gremillion HA, Staud
RM, McCrae CS, Robinson ME. Intraindividual variability in daily
sleep and pain ratings among chronic pain patients: bidirectional association and the role of negative mood. Clin J Pain. 2011;27:425433.
[41] Shechter A, Varin F, Boivin DB. Circadian variation of sleep during the follicular and luteal phases of the menstrual cycle. Sleep.
2009;33:64756.
[42] Iacovides S, Avidon I, Bentley A, Baker FC. Dicoflenac potassium
restores objective and subjective measures of sleep quality in women
with primary dysmenorrhea. Sleep. 2009;10191026.

Copyright of Behavioral Medicine is the property of Taylor & Francis Ltd and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.