Pain Catastrophizing Predicts Menstrual Pain Ratings in Adolescent Girls with Chronic Pain

Abstract

Objectives. The current study aimed to explore relationships among self-reported menstrual pain ratings, acute laboratory pain, pain catastrophizing, and anxiety sensitivity in a sample of girls without pain (No Pain group) and girls with a chronic pain condition (Chronic Pain group).

Setting. A laboratory at an off-campus Medical School office building.

Subjects. Eighty-four postmenarchal girls (43 No Pain, 41 Chronic Pain) ages 10–17 participated in the study.

Methods. All participants completed self-report questionnaires assessing menstrual pain, pain catastrophizing, and anxiety sensitivity and completed a cold pressor task. Pain intensity during the task was rated on a 0 (no pain) to 10 (worst pain possible) numeric rating scale.

Results. After controlling for age, average menstrual pain ratings (without medication) were significantly correlated with cold pressor pain intensity for the No Pain group only. In the Chronic Pain group, menstrual pain ratings were significantly correlated with pain catastrophizing and anxiety sensitivity. In a multiple linear regression analysis, after controlling for age, only pain catastrophizing emerged as a significant predictor of menstrual pain ratings in the Chronic Pain group.

Conclusion. Results demonstrate differences in relationships among menstrual pain, acute laboratory pain, and psychological variables in girls with no pain compared with girls with chronic pain. In addition, pain catastrophizing may be a particularly salient factor associated with menstrual pain in girls with chronic pain that warrants further investigation.

Introduction

Painful menstruation, known as dysmenorrhea, is a recurrent and disabling condition affecting almost 90% of adolescent girls [1]. In addition to missed school days, and impaired physical and social activities, adolescent girls with menstrual pain also have a significantly reduced quality of life [2,3]. Menstrual pain is associated with elevated acute pain responses in the laboratory [4–7] and is linked with chronic pain conditions such as migraine headache and irritable bowel syndrome [8–10]. A clearer understanding of modifiable constructs that influence the experience of dysmenorrhea may inform efforts to prevent and treat this disabling condition.

Psychological factors are likely to play a role in the development and maintenance of menstrual pain [6,11–13]. Dysmenorrhea shares a strong association with mood and anxiety disorders [14]. Self-report measures of depression and anxiety are related to increased menstrual symptoms (i.e., menstrual and premenstrual pelvic and back pain, premenstrual negative affect, and premenstrual water retention) in adolescents [11]. Another study of almost 12,000 US women showed that women with menstrual-related problems (e.g., heavy bleeding, bothersome cramping, or premenstrual syndrome) were significantly more likely to report feeling frequently depressed or anxious [13]. In a laboratory setting, adult women with dysmenorrhea demonstrated higher levels of state anxiety when tested during all phases of the menstrual cycle compared with women without dysmenorrhea [6]. Furthermore, in a nonlaboratory study of 424 adolescents, those with dysmenorrhea reported significantly higher levels of depression, general anxiety, and state and trait anxiety, as compared with girls without dysmenorrhea [12]. Despite these findings, data on the associations between menstrual pain and specific psychological processes are not well understood, particularly in younger populations.

Pain-specific anxiety may intensify attention to symptoms and thereby exacerbate pain responses [15] as demonstrated in both laboratory and clinical settings [16]. Pain catastrophizing is the tendency to magnify negative expectations about pain [17] and this attribute may elucidate the relationship between psychological symptoms and pain. Children with chronic pain problems report higher levels of pain catastrophizing than children without pain [18]. Pain catastrophizing has also been linked with acute and chronic pain prevalence, intensity, and disability in adults and children [19–22]. Laboratory studies have shown that women who report higher levels of pain catastrophizing report more severe menstrual pain, compared with those with low levels of pain catastrophizing [23]. Women with dysmenorrhea report higher levels of pain catastrophizing on the first day of menstruation compared with those without dysmenorrhea [24]. No research has yet examined pain catastrophizing and menstrual pain among adolescent girls. The majority of the data linking pain catastrophizing and pain is correlational, so a causal relationship cannot be inferred [16]. However, recent work has attempted to address this issue by examining the temporal relationship of pain catastrophizing to pain experience. Evidence suggests that changes in pain catastrophizing are associated with future clinical and experimental pain, and not the reverse [25,26]. Another recent study demonstrated the genetic heritability of pain catastrophizing, and linked pain catastrophizing with experimental pain, independent of the genetic influence [27]. Taken together, these data suggest that pain catastrophizing may be an independent construct that serves as a risk factor for the development of chronic pain. If relationship between pain catastrophizing and menstrual pain is found in younger populations, then modification of catastrophic thinking styles may be a component of an effective treatment strategy.

Anxiety sensitivity has also been linked with pain conditions. Anxiety sensitivity is a fear of the physical sensations of anxiety, and there are strong links between anxiety sensitivity and acute and chronic pain in children. In healthy children, anxiety sensitivity is associated with heightened acute, experimental pain responses [28], and also predicts fear of pain in adolescents [29]. In children with chronic pain, higher anxiety sensitivity is significantly related to lower quality of life and poorer overall functioning [30]. Anxiety sensitivity is significantly associated with ratings of current bodily pain in children [18] and accounts for a significant part of the relationship between neuroticism and pain responses in children and adolescents [31]. However, these studies included both boys and girls. Most of the studies controlled for child sex in the analyses or found no statistically significant sex related differences in anxiety sensitivity, with the exception of Muris et al. [29]. However, no studies have yet evaluated the relationship between anxiety sensitivity and dysmenorrhea in postmenarchal female children and adolescents with and without a chronic pain condition.

Little is known about how menstrual pain differs in adolescents without a chronic pain condition compared with adolescents with other chronic pain conditions. Understanding the relationship between acute pain, pain catastrophizing, anxiety sensitivity, and menstrual pain may help further characterize dysmenorrhea in adolescent girls. Exploring these relationships in samples of adolescent girls with and without other chronic pain conditions may provide clues as to how these relationships change once other pain systems have been activated. The aim of the current study was to explore the relationships between menstrual pain and 1) acute laboratory pain and 2) pain-specific anxiety (pain catastrophizing and anxiety sensitivity) in a sample of adolescent girls both with and without another chronic pain condition. Based on previous research, we anticipated that menstrual pain severity would be associated with acute laboratory pain intensity and measures of pain-specific anxiety (pain catastrophizing and anxiety sensitivity) in both samples, although the relationships would be stronger in the sample of adolescent girls with chronic pain conditions as compared with the pain free sample. That is, we hypothesized that laboratory pain intensity, pain catastrophizing, and anxiety sensitivity would contribute to menstrual pain severity ratings in the chronic pain sample.

Methods

Participants

The data for the current study were drawn from a larger laboratory pain study of children aged 8–17 years [31–33]. Participants for the current study were 84 (43 No Pain, 41 Chronic Pain) female adolescents, with a mean age of 15.37 years (SD = 1.79, range = 10–17) who had begun menstruating. All participants completed the laboratory pain session on their own without the parent present. Demographics, including race/ethnicity, and for participants in the chronic pain group, their current most bothersome pain location are reported in Table 1. Written informed consent was obtained from parents and written assent from the pediatric participants. The study was approved by the UCLA Institutional Review Board. Each adolescent received $50 cash for participation.

Table 1.

Demographic and descriptive information for study participants

	Chronic Pain (n = 41)	No Pain* (n = 43)	P Value	Total Sample* (n = 84)
Age—M (SD)	15.92 (1.92)	14.84 (1.89)	0.01	15.37 (1.79)
Race—n (%)			0.17
Caucasian	28 (68%)	19 (44%)		47 (56%)
African American	4 (10%)	10 (24%)		14 (17%)
Asian	1 (2%)	1 (2%)		2 (2%)
Multiracial	8 (20%)	12 (28%)		20 (24%)
Ethnicity—n (%)			0.76
Non-Hispanic/Non-Latino	28 (68%)	28 (65%)		56 (67%)
Hispanic/Latino	13 (32%)	15 (35%)		28 (33%)
Most bothersome pain location—n (%)
Head	11 (27%)
Foot/ankles/leg/knee/hip	8 (20%)
Back/shoulders	7 (17%)
Stomach	5 (12%)
Neck	4 (10%)
Bones	1 (2%)
Chest	1 (2%)
Did not answer	4 (10%)

Note. * One participant in the No Pain group did not indicate race.

Procedure

A detailed description of the procedures and study protocol were previously published [31]. Participants were recruited through advertisements, community events, and referrals from previous participants. Study advertisements were posted on online forums, libraries, pediatricians’ offices, festivals, and so forth. Previous participants were compensated $25 for each referred family that completed the study.

Eligibility was confirmed by telephone. Participants in the Chronic Pain group had a diagnosis of a chronic pain condition confirmed by one of the authors who is an expert in pediatric pain (L.Z.). Participants in the No Pain group reported that they did not have any chronic pain lasting 3 months or longer. Exclusionary criteria included: 1) acute illness or injury that may impact lab performance or that affects sensitivity of the extremities (e.g., Reynaud's disease, hand injuries); 2) daily use of opioids; or 3) developmental delay, autism, or significant impairment that could preclude understanding of study procedures. Only one child per family was enrolled in the study.

On arrival to the laboratory, participants were escorted to separate rooms; there was no contact between parent and child until after the session was completed. Participants provided written informed consent and then completed questionnaires using an online survey program and were interviewed by a research assistant about their menstrual history. Only those questionnaires relevant to the current study are discussed herein.

Participants were then escorted into the laboratory, where their height and weight were recorded. Participants were shown the 0 (none) to 10 (worst or most possible) Numerical Rating Scale and instructed on its use. These endpoints were identical for all Numeric Rating Scale-related questions about anticipatory anxiety, pain intensity, and pain bother. Participants were then administered a single trial of the cold pressor task (additional pain tasks were then administered but are presented in a separate report [31,33]). Participants placed the right hand in a cold pressor unit comprised of a Techne TE-10D Thermoregulator, B-8 Bath, and RU-200 Dip Cooler (Techne, Burlington, NJ). The unit maintained the water at a temperature of 5°C and kept the water circulating to prevent localized warming around the hand. Recent research has shown 5°C to be an appropriate temperature for use with children and adolescents in the age range of the current study [34]. Participants’ hands were submerged up to approximately 2″ above the wrist. Participants were instructed to keep their hand in the cold water for as long as they possibly could but that they could terminate the trial at any time. The task had an uninformed ceiling of 3 minutes. The administration of the cold pressor task was in line with the guidelines set forth in previous systematic reviews [35].

Measures

Self-Report Measures

Menstrual Pain Ratings. A menstrual history interview was developed for the purposes of the study. Participants were first asked if they had period cramps (yes/no). Those who answered “yes” were then asked, “When you have period cramps, what is your usual level of pain without medication?” using the 0 (none) to 10 (worst) Numeric Rating Scale. Participants were also asked 1) if they had any activity interference due to menstrual cramps/pain (yes/no); 2) if they took medication for menstrual cramps (yes/no); and 3) if they took hormones or birth control for menstrual cramps or for any other reason (yes/no).

The Pain Catastrophizing Scale—Child version [36], a revised version of the original scale [37], measures the extent to which participants worry, amplify, and feel helpless about the experience of pain. Thirteen items are rated from 0 (not at all) to 4 (extremely); total scores range from 0 to 52; higher scores indicate a greater tendency to engage in pain catastrophizing. The measure has demonstrated reliability and validity in children aged 8–16 years. Measures of internal consistency are high (α range from 0.84 to 0.89, depending on age) and the PCS-C also has good construct validity with a measure of trait anxiety (r = 0.53) [36]. For the current sample, internal consistency was very high (α = 0.93).

The Children's Anxiety Sensitivity Index [38] is an 18-item scale assessing the specific tendency to interpret anxiety sensations as dangerous. Items are scored on a 3-point scale (none, some, a lot), with total scores calculated by summing all items (total score range 18–54). Higher scores reflect greater endorsement of anxiety sensitivity. This measure has demonstrated good internal consistency (α = 0.87) and adequate test–retest reliability over 2 weeks (range = 0.62–0.78) [38]. Construct validity is supported by good correlations with measures of trait anxiety (rs = 0.55–0.69); however, it also accounts for variance in fear that is not attributable to trait anxiety measures [39]. Internal consistency was good in the current sample (α = 0.84).

Laboratory Pain Measures

Numeric Rating Scale. The Numeric Rating Scale was used to assess pain intensity following the cold pressor task. Participants were asked “at its worst, how much pain did you feel?” The Numeric Rating Scale has been shown to be valid and reliable for children as young as 8 years [40] and results from the current study indicated that participants understood the scale [33].

Pain tolerance. Pain tolerance was assessed by calculating the amount of time that each participant was able to keep her hand in the cold water. There was no significant difference in pain tolerance between the No Pain and Chronic Pain groups so these data will not be reported further.

Menstrual Cycle Assessment

Menstrual cycle phase at the time of the laboratory session was assessed using a combination of self-reported last menstrual period and salivary progesterone. Unstimulated saliva samples were collected from each participant on arrival by asking participants to spit into a small cryovial, which was then stored in a medical grade freezer. Samples were tested for progesterone using commercially available immunoassay and kinetic reaction kits, without modification to the manufacturer's protocol (Salimetrics, LLC). In line with the protocol, progesterone values ≤100 pg/mL identified participants in the follicular phase; progesterone values of ≥100 pg/mL were considered representative of the luteal phase. Report of last menstrual period was then used to confirm the menstrual cycle phase determined by progesterone level.

Statistical Analysis

Preliminary analyses were conducted using independent samples t-tests to evaluate age differences between the groups. Chi-square tests were used to test whether the groups differed in menstrual cycle phase at the time of the laboratory session. Independent samples t-tests were conducted on measures of menstrual pain ratings, cold pressor pain intensity, and anxiety sensitivity scores to evaluate any group differences. Independent samples t-tests were done to evaluate for differences in menstrual cycle phase (follicular vs luteal) in cold pressor pain intensity, pain catastrophizing, and anxiety sensitivity scores for each the Chronic Pain and No Pain groups. Because pain catastrophizing scores were not normally distributed, Mann–Whitney U-tests were used to evaluate differences between the No Pain and Chronic Pain Groups and, subsequently, between follicular and luteal phases within each group.

A Chi-square test was used to evaluate group differences in rates of activity interference due to menstrual pain. Partial correlation analyses controlling for child age were conducted separately for each group to examine the relationships among menstrual pain ratings, cold pressor pain intensity, and anxiety measures. Sequential multiple linear regression analysis was conducted separately for each group with menstrual pain ratings as the dependent variable, and any variables found to be significantly correlated in the bivariate analyses as the independent variables (controlling for child age, which was entered on the first step).

Results

Preliminary Analyses

Participants in the Chronic Pain group were significantly older than those in the No Pain group (P < 0.01), so age was controlled for in all subsequent analyses. There were no significant racial or ethnic differences between the groups (P = 0.16, P = 0.76, respectively). We then tested for group differences in menstrual cycle phase on the day of the laboratory session. Three participants (2 No Pain, 1 Chronic Pain) did not have data on menstrual cycle phase available and were excluded from these analyses. In the No Pain group, 22 girls (51.2%) were in the follicular phase and 13 girls (30.2%) were in the luteal phase. In the Chronic Pain group, 18 girls (43.9%) were in the follicular phase and 20 girls (48.8%) were in the luteal phase. For 6 girls (4.7%) in the No Pain group and 2 girls (4.9%) in the Chronic Pain group, menstrual cycle phase could not be determined (i.e., last menstrual period indicating follicular phase, while progesterone level indicating luteal phase). There were no significant differences in menstrual cycle phase between the groups, irrespective of whether the 8 girls with undetermined menstrual cycle phase data were included in the analysis (χ² [2, N = 81] = 3.87, P = 0.14), or were included and classified as luteal phase based on the salivary progesterone levels only (χ² [1, N = 81] = 0.61, P = 0.44). We also ran an additional analysis to classify participants as being in the menstrual phase if their self-reported last menstrual period was within 5 days of the laboratory session date, irrespective of progesterone levels. Based on this, seven participants in the No Pain group and seven participants in the Chronic Pain group would have been classified as in the menstrual phase. However, with this additional classification, there were no significant differences in menstrual cycle phase between the No Pain and Chronic Pain groups (χ² (3, N = 81)= 4.57, P = 0.21).

There were no significant differences between follicular and luteal phases in cold pressor pain intensity for either the Chronic Pain (P = 0.53) or No Pain group (P = 0.24). Similarly, there were no significant differences in anxiety sensitivity scores between follicular and luteal phases for either the Chronic Pain (P = 0.90) or No Pain group (P = 0.54). Results of the Mann–Whitney U-tests suggest retaining the null hypothesis that there are no significant differences in pain catastrophizing scores between the follicular and luteal phases for the Chronic Pain group (P = 0.74) or No Pain group (P = 0.51).

Primary Analyses

As shown in Table 2, girls in the Chronic Pain group demonstrated significantly higher menstrual pain ratings than No Pain girls, [t(82) = −2.13]. Results of the Mann–Whitney U-test suggest that chronic pain participants reported significantly higher pain catastrophizing scores compared with the No Pain group. There were no significant differences between the groups on anxiety sensitivity or cold pressor pain intensity scores. Girls with chronic pain endorsed activity interference due to menstrual pain at a higher rate than No Pain girls, χ² (1, N = 64) = 7.58, P < 0.01. No Pain girls were as likely to take pain medication for menstrual cramps as girls with chronic pain. Girls with chronic pain were taking hormones or birth control for menstrual cramps or other reasons (e.g., acne, irregular periods, etc.) at a significantly higher rate than girls without pain [χ² (1, N = 64) = 15.72, P < 0.01; χ² (1, N = 64) = 14.68, P < 0.01], respectively.

Table 2.

Comparison of Chronic Pain and No Pain groups on measures of pain and anxiety, rates of interference from pain and medication use

Measures	Chronic Pain M (SD)	No Pain M (SD)	95% C.I.	P Value
Menstrual pain ratings	5.05 (3.54)	3.53 (2.96)	[−2.93, −.10]	0.04
Pain catastrophizing (PCS-C)	23.66 (11.90)	16.23 (8.25)	[−11.9,−2.95]	<0.01
Anxiety sensitivity (CASI)	30.37 (7.03)	29.23 (5.09)	[−3.81, 1.54]	0.40
Cold pressor pain intensity	5.78 (2.82)	5.86 (2.64)	[−1.11, 1.26]	0.89
	% of Chronic Pain group answering “yes”	% of No Pain group answering “yes”	χ 2 value	P value
Activity interference due to menstrual pain	69.2	30.8	7.58	<0.01
Takes medication for menstrual cramps	61.3	48.5	1.06	0.30
Takes hormones or birth control for menstrual cramps	38.7	0	15.72	<0.01
Takes hormones or birth control for other reasons	29.3	0	14.68	<0.01

Note. PCS-C = Pain catastrophizing scale for children; CASI = Children's anxiety sensitivity index.

Partial correlations controlling for age are presented in Tables 3 and 4. As shown in Table 3, menstrual pain ratings of No Pain girls were significantly correlated with cold pressor pain intensity ratings (P = 0.03), but were not significantly correlated with measures of pain catastrophizing or anxiety sensitivity. However, for girls with chronic pain, menstrual pain ratings were significantly correlated with pain catastrophizing and anxiety sensitivity scores, all Ps < 0.01 (see Table 4), but not with cold pressor pain intensity.

Table 3.

No Pain group partial correlations (controlling for age) between menstrual pain ratings, cold pressor pain intensity, and anxiety measures

Measure	Menstrual Pain Rating	Cold Pressor Pain Intensity	Pain Catastrophizing	Anxiety Sensitivity
	(1)	(2)	(3)	(4)
(1)		0.33*	0.06	−0.08
(2)			0.38*	0.40**
(3)				0.62**

Note. * P < 0.05; ** P < 0.01.

Table 4.

Chronic Pain group partial correlations (controlling for age) between menstrual pain ratings, cold pressor pain intensity, and anxiety measures

Measure	Menstrual Pain Rating	Cold Pressor Pain Intensity	Pain Catastrophizing	Anxiety Sensitivity
	(1)	(2)	(3)	(4)
(1)		0.21	0.61**	0.41**
(2)			0.27	0.28
(3)				0.68**

Note. ** P < 0.01.

A multivariate analysis was not conducted in the No Pain group because only one variable other than age (i.e., cold pressor pain intensity ratings) was found to be correlated with menstrual pain ratings in the bivariate analysis. In the Chronic Pain group, a multiple linear regression was conducted predicting menstrual pain ratings from pain catastrophizing and anxiety sensitivity scores. After entering child age on the first step, only pain catastrophizing scores remained a significant predictor of menstrual pain ratings (see Table 5).

Table 5.

Hierarchical linear regression of age and pain-related anxiety measures on menstrual pain ratings in girls in the Chronic Pain group (n = 41)

Step	Variables Entered	B	SE B	β	R 2 Change
Menstrual pain ratings (Dependent variable)
1					0.01
	Child age	0.44	0.31	0.11
2					0.36**
	Pain catastrophizing	0.18	0.05	0.61**
	Anxiety sensitivity	−0.001	0.09	−0.003

Note. ** P < 0.01.

Confirmatory Power Analysis

Estimated from our data, the standard deviations of menstrual pain ratings and pain catastrophizing in the Chronic Pain group are 3.54 and 11.90, respectively. The regression coefficient in a linear regression model using pain catastrophizing to predict pain ratings is estimated to be 0.18. Based on these estimates, the sample size of 41 patients provides >95% power for a two-sided test at the significance level 0.05. Actually in this regression analysis, the minimum detectable effect size (as measured by B, the regression coefficient) is 0.122 to achieve 80% power with an N of 41.

Discussion

The current study showed significantly higher menstrual pain ratings and higher rates of activity interference due to painful menses in adolescents with a chronic pain condition compared with those without a chronic pain condition. Furthermore, the relationships among menstrual pain, acute laboratory pain, and pain-related anxiety were different for each group. For girls without pain, menstrual pain was significantly correlated with acute cold pressor pain intensity; whereas for girls with chronic pain, menstrual pain was significantly correlated with pain catastrophizing and anxiety sensitivity. In the multivariate analysis, only pain catastrophizing accounted for significant variance in menstrual pain ratings among girls with chronic pain.

The finding that girls with chronic pain reported higher menstrual pain than those without chronic pain is in line with previous research suggesting comorbidity between dysmenorrhea and other chronic pain conditions [8–10,41]. Higher levels of menstrual pain may suggest an overall increased sensitivity to pain, possibly reflecting alterations in central pain processing in girls with chronic pain and dysmenorrhea. More recently, brain imaging research has highlighted the role of central pain mechanisms in women with dysmenorrhea [7], but adolescents have not been studied. [4,5,42]. However, our laboratory data did not show differences in pain responses between the Chronic Pain and No Pain groups, so it is possible that central pain mechanisms in response to an acute pain experience are not relevant or have not yet emerged as a salient factor in adolescents with dysmenorrhea. Further research in this area is warranted. In this study, girls with chronic pain also reported higher rates of activity interference due to menstrual pain and higher rate of medication use for menstrual cramps as compared with girls without chronic pain. These findings suggest greater impairment due to menstrual pain in girls who already struggle with chronic pain.

All girls in the Chronic Pain group reported higher levels of pain catastrophizing compared with No Pain girls, although cold pressor pain intensity did not differ between groups. These data reflect previous research demonstrating higher levels of pain catastrophizing in chronic pain populations [16], and is in line with some previous studies showing no differences in cold pressor pain intensity between Chronic Pain and No Pain groups [32]. This may be because our population is younger and differences in laboratory-based pain responses have not yet emerged. However, these findings also support a temporal relationship between pain catastrophizing and dysmenorrhea. The Chronic Pain group reported higher levels of menstrual pain, yet showed similar levels of pain reactivity to an objective pain stimulus in the laboratory, indicating that the Chronic Pain group experiences higher levels of menstrual pain, specifically, rather than an increased sensitivity to painful stimuli, in general. As we did not assess pain catastrophizing prior to the development of menstrual pain, we cannot exclude the possibility that the experience of physical pain might lead to the development of pain catastrophizing. However, it is also reasonable to conclude, in line with recent research, that pain catastrophizing is a trait-like construct that contributed to the development or experienced intensity of menstrual pain and its associated pain-related impairment of function [25–27].

In girls without chronic pain, cold pressor pain intensity, but not anxiety or catastrophizing, was significantly correlated with menstrual pain ratings. This finding suggests more consistency between menstrual pain and laboratory pain responses in girls without chronic pain, where sensitivity to acute laboratory pain and the pain experienced during their menstrual period appear to be independent of psychological influences. However, in the girls with chronic pain, menstrual pain severity was related primarily to pain-specific anxiety (pain catastrophizing) rather than anxiety sensitivity. These data suggest that while psychological constructs reflecting fear of physical sensations, such as anxiety sensitivity, may be important in dysmenorrhea, as the initial analyses reflected, pain-specific cognitive constructs such as pain catastrophizing may play a critical role in those who already have another chronic pain condition, as demonstrated in the regression analyses. Furthermore, anxiety sensitivity did not differ between the No Pain and Chronic Pain groups, although pain catastrophizing was significantly higher in the Chronic Pain group. This finding suggests that pain catastrophizing is a more salient psychological construct than anxiety sensitivity for girls with chronic pain. Elevated pain catastrophizing as a result of another pain condition or contributing to another pain condition may create vulnerability for menstrual pain, or vice versa. Addressing the cognitive aspects of pain, such as catastrophizing, may prove beneficial for adolescent girls with chronic pain.

Limitations to this study include less precise assessment of menstrual cycle phase as ovulation testing was not included as part of the protocol. There was also a disproportionate number of participants taking some form of exogenous hormones in the Chronic Pain group as compared with the No Pain group, which may have affected pain responses [43]. Research has suggested elevated levels of pain catastrophizing during menstruation in healthy women [24], so menstruation of some participants at the time of the laboratory assessment may have influenced the results. Our estimate of menstruation at the time of the lab session based on a last menstrual period of 5 days or less from the day of the lab session did not yield any significant differences between the groups so it is unlikely that variation in timing of menstruation accounted for the results. Furthermore, due to our sample size, we were unable to compare differences in menstrual pain ratings and relationships to lab pain and anxiety for each of the chronic pain conditions individually.

Our findings highlight the importance of assessing psychological factors, particularly pain castastrophizing, in adolescents with menstrual pain and coexisting chronic pain conditions. Mitigating this cognitive response to pain in girls with dysmenorrhea might decrease the likelihood of these girls developing other chronic pain, as well as minimizing pain-related disability during menstruation. Future studies may consider looking at the role of central pain processes in relation to dysmenorrhea in girls with chronic pain, especially from a developmental standpoint studying girls premenarche to postmenarche.

Funding sources: This research was supported by a grant from the National Institute of Child Health and Human Development (K23HD077042; PI: Laura A. Payne), National Center for Advancing Translational Sciences UCLA Clinical and Translational Science Institute Grant Number KL2TR000122 (PI: Laura A. Payne), and National Institute for Dental and Craniofacial Research Grant Number R01DE012754 (PI: Lonnie K. Zeltzer).

Conflict of interest: None of the authors have conflicts of interest to disclose. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.