|Home | About | Journals | Submit | Contact Us | Français|
This study tested the hypothesis that women exposed to childhood abuse or neglect would have an increased likelihood of reporting hot flashes and night sweats during the menopausal transition.
This hypothesis was evaluated in 332 white and African American women participating in the Study of Women’s Health Across the Nation Mental Health Study, a prospective investigation of women transitioning through menopause. Childhood abuse and neglect were measured once with the Child Trauma Questionnaire. Vasomotor symptoms (any/none hot flashes, night sweats) were reported annually over 8 years. Associations between maltreatment and vasomotor symptoms were estimated with generalized estimating equations.
Childhood abuse or neglect was associated with increased reporting of hot flashes (odds ratio = 1.73, 95% CI: 1.23–2.43) and night sweats (odds ratio = 1.75, 95% CI: 1.26–2.43) in age-adjusted models. Results persisted in multivariable models and across several types of abuse and neglect.
The experience of childhood abuse and neglect is associated with increased vasomotor symptom reporting in adulthood. The sequelae of childhood abuse and neglect may persist well into adulthood to influence the occurrence of vasomotor symptoms at midlife.
Childhood abuse and neglect are highly prevalent in the United States, with estimates in 2004 from the Administration on Children, Youth, and Families indicating rates at 11.2 per 1,000 boys and 12.6 per 1,000 girls.1 Across multiple investigations experiences of abuse and neglect in childhood have been linked to significantly elevated risk of several adverse health outcomes, including a range of psychiatric disorders,2–5 substance abuse problems,2,6 eating disorders,7,8 and poor self-rated health.6,9 In the case of physical health conditions, significant associations are most consistently observed for symptomatic conditions that include headache, back pain, chronic pelvic pain, chest pain, functional gastrointestinal disorders, and premenstrual dysphoric disorder.10–14
Vasomotor symptoms, or hot flashes and night sweats, are experienced by the majority of US women during the menopausal transition. In the largest multiethnic study of women transitioning through menopause, the Study of Women’s Health Across the Nation (SWAN), 60% to 70%15 of women reported vasomotor symptoms at some point during the menopausal transition. Highest levels are observed among key ethnic minority groups, particularly African American women.15 Vasomotor symptoms are associated with reported sleep disturbance and fatigue,16–18 irritability and depressed mood,19–23 and impaired mental, physical, and social quality of life.24–26 Given the findings of risk associated with hormone therapy (HT),27 the leading treatment for vasomotor symptoms,28 vasomotor symptoms have emerged as an important public health issue. The risk factors for and cause of vasomotor symptoms are not fully understood. However, women who are smokers, are overweight or obese, or have elevated depressive symptoms have 60% to twofold odds of vasomotor symptoms,15,29 and those with elevated anxious symptoms have threefold odds of reporting vasomotor symptoms30 relative to women without these risk factors. Leading models characterize vasomotor symptoms as thermoregulatory events with involvement of the reproductive hormonal, central adrenergic, and seroto-nergic systems.30–35
The aim of this investigation was to examine the association between childhood abuse and vasomotor symptoms (hot flashes, night sweats) over the menopausal transition. Given the association between childhood abuse and other physical symptoms, reporting of vasomotor symptoms may be elevated among women who have experienced childhood abuse or neglect. Moreover, childhood abuse and neglect have been associated with behavioral, affective, and neuroendocrine factors also linked to vasomotor symptoms,15,30–35 including smoking and obesity,36–38 depression and anxiety,2 and alterations in key neuroendocrine39–43 and reproductive hormone44 systems. Thus, we hypothesized that women exposed to childhood abuse or neglect would be more likely to report vasomotor symptoms than women who had not been exposed to abuse or neglect. Secondary aims, examined in an exploratory fashion, were (1) to evaluate the association between different types of maltreatment (emotional abuse, physical abuse, physical neglect, emotional neglect, sexual abuse) and vasomotor symptoms and (2) to examine whether associations between abuse and neglect and vasomotor symptoms varied between African American and white women.
This study was conducted among participants in the SWAN Mental Health Study (MHS), an ancillary study to the Study of Women’s Health Across the Nation. SWAN is a multisite, community-based cohort investigation of menopause. The MHS, an ancillary study conducted at the Pittsburgh SWAN site only, was designed to collect detailed mental health measures during the first 8 years of follow-up of SWAN. Details of the SWAN study design and recruitment procedures have been previously reported.45 Briefly, each SWAN site recruited white women and a sample of a predetermined minority group. At the Pittsburgh SWAN site, African American and white women were recruited using established sampling techniques, random digit dialing, and a voter registration list. SWAN eligibility criteria included age 42 to 52 years, having an intact uterus and at least one ovary, not pregnant or breast-feeding, having menstruated in the past 3 months, and no use of reproductive hormones in the previous 3 months. SWAN and the MHS were approved by the University of Pittsburgh Institutional Review Board, and each participant provided written informed consent. SWAN and MHS baseline assessments were conducted in 1996 to 1997. Core SWAN data from baseline through study visit 7 and MHS data from baseline through study visit 8 were available for the present investigation.
Of women eligible for participation in SWAN, 51% (N = 3,302) enrolled. Of the 463 Pittsburgh SWAN participants eligible for the MHS, 96% (N = 443) enrolled. The MHS retention rate was more than 80% (N = 355) through visit 8. Measures of abuse and neglect were collected in MHS at visit 8, and only women with these data were included in the present analysis. Seventy-nine women dropped out of the MHS by visit 8, 13 active MHS participants did not complete visit 8, and an additional 19 women had missing or incomplete abuse/neglect data. Therefore, 111 MHS participants were excluded from the present analysis. These 111 women were more likely to be younger (P = 0.05), smokers (P = 0.01), later in the menopausal transition (P = 0.008), and to have lower education (P = 0.009) and more depressive symptoms (P = 0.04) at baseline than women included in the analysis. The final sample for evaluation of primary hypotheses included 332 (222 white, 110 African American) women.
SWAN participants completed an interview, physical examination, and questionnaires at the SWAN baseline and annually thereafter. Vasomotor symptoms (hot flashes and night sweats) were assessed annually during the core SWAN interview. As part of a self-report questionnaire, women were asked to indicate the number of days in the previous 2 weeks they had experienced hot flashes and night sweats (not at all, 1–5 days, 6–8 days, 9–13 days, every day) in the 2 weeks before the interview. Women were categorized as experiencing any or no hot flashes and night sweats. A three-level categorization was also considered (not at all, 1–5 days, ≥ 6 days), and although results were similar, there was some indication of a threshold effect at any versus no hot flashes and night sweats. Therefore, results using the any versus no categorization are presented here.
Childhood abuse and neglect were assessed using the 28-item short form of the Child Trauma Questionnaire (CTQ), a validated measure of childhood abuse and neglect experienced at or before age 18.46,47 The CTQ has strong test-retest reliability (0.79–0.86), internal consistency (Cronbach’s α = 0.74–0.95 across subscales in this investigation), and convergent validity with clinical interview.46–48 Examples of questions include “People in my family said hurtful or insulting things to me” and “I got hit or beaten so badly that it was noticed by someone like a teacher, neighbor, or doctor.” Respondents rate each item on a 5-point scale ranging from 1 (never) to 5 (very often true). Items are summed to yield scores on five subscales (emotional abuse, physical abuse, emotional neglect, physical neglect, and sexual abuse) with scores ranging from 5 to 25. CTQ short form clinical cut off points validated by Walker and colleagues48 have sensitivity and specificity at 0.85 or higher relative to clinical interview and are as follows, with scores at or above these thresholds classified as positive for abuse/neglect: emotional abuse = 10, physical abuse = 8, emotional neglect = 15, physical neglect = 8, and sexual abuse = 8. If scoring at or above the clinical threshold on any one subscale, individuals were classified as having been exposed to any abuse or neglect. If scoring below all clinical cut off points, an individual was classified as not exposed to abuse or neglect. The CTQ was also considered as a continuous measure.
Race/ethnicity and educational attainment (years of completed education/highest degree, categorized into high school, some college or vocational school, college degree or higher) were derived from the baseline SWAN interview. Age, smoking status (current vs past/never), menopause status, and depressive and anxious symptoms were derived from the annual core SWAN interview. Menopause status was determined from self-reported bleeding patterns over the year preceding the visit categorized as follows: bleeding in the previous 3 months with no change in cycle predictability in the past year was considered premenopausal, bleeding the previous 3 months with a decrease in cycle predictability in the past year was considered early perimenopausal, fewer than 12 and more than 3 months of amenorrhea was considered late perimenopausal, and 12 months or more of amenorrhea was considered postmenopausal.49 Visits in which women reported taking hormones (HT or oral contraceptives) within the past month were excluded. However, women previously classified as pre- or perimenopausal who reported hormone use since the last study visit were classified as indeterminate status because of the potential impact of discontinuation of hormone use on bleeding patterns. Depressive symptoms were summed scores on the widely used and well-validated Center for Epidemiologic Studies Depression Scale,50 assessed annually. Anxious symptoms (irritability or grouchiness, feeling tense or nervous, heart pounding or racing, feeling fearful for no reason)15 were assessed annually. Body mass index (BMI) was derived from measures taken during the annual SWAN examination and was calculated as weight (kg)/height (m2).
Univariate associations between abuse/neglect, covariates, and hot flashes and night sweats were evaluated using t tests and χ2 analyses. Covariates were selected on the basis of a univariate association with hot flashes or night sweats at P less than or equal to 0.10. Associations between abuse/neglect, covariates, and hot flashes and night sweats were estimated using generalized estimating equations with a logit link, binomial distribution, and an unstructured correlation matrix. Hot flashes and night sweats were considered separately in all models. Models were estimated first controlling for age and additionally for race, education, and time-dependent covariates menopause status, BMI, smoking status, and depressive symptoms. Anxious symptoms were also considered as a covariate. However, given the correlation between anxious and depressive symptoms (r = 0.6) and the comparability of results when considering anxious rather than depressive symptoms as a covariate, only depressive symptoms were included in the model. Women having undergone a hysterectomy or oophorectomy (n = 35) were censored at the time of the surgery (ie, their data were no longer included in the analysis after the surgery). A large number of women used hormones (HT or oral contraceptives) intermittently during the study (n = 141). To retain the maximal number of women in the analysis, those study visits in which women reported taking hormones within the past month were excluded (14% of study visits). Excluding visits in which hormone use was reported since the last visit (in the past year) was also considered. Because results were largely unchanged, to retain the maximal number of visits the past month exclusion was used. Associations were first estimated for total abuse and subsequently for the five abuse subscales. Interactions between total abuse or neglect and race were estimated and models subsequently stratified by race. Statistical analyses were performed with SAS version 8.2 (SAS Institute, Cary, NC). All analyses were two sided at α = 0.05.
Approximately 38% (n = 126) of the sample reported experiencing some form of childhood abuse or neglect. Examination of subtypes indicated that 20.2% (n = 67) of women scored positive for emotional abuse, 17.2% (n = 57) for physical abuse, 15.1% (n = 50) for sexual abuse, 7.5% (n = 25) for emotional neglect, and 15.1% (n = 50) for physical neglect during childhood. Whereas women who reported any childhood abuse or neglect did not differ on baseline age, race, education, menopause status, and smoking status relative to those who reported no abuse or neglect, women reporting abuse or neglect had higher depressive symptoms and a slightly higher BMI (Table 1).
Abuse or neglect was associated with significantly increased odds of reporting both hot flashes and night sweats in age- and site-adjusted and in multivariable models (Table 2). Significant results were also observed when considering the CTQ as a continuous measure (hot flashes: odds ratio = 1.02, 95% CI: 1.01–1.02, P = 0.002; night sweats: OR = 1.02, 95% CI: 1.01–1.04, P = 0.0002, fully adjusted models).
Examination of individual CTQ subscales indicated a pattern of associations across most subscales (Table 3) with results most consistently observed for emotional and physical abuse. Interactions between race/ethnicity and abuse/neglect were not statistically significant (hot flashes: P = 0.54; night sweats: P = 0.24). However, fully adjusted results stratified by race/ethnicity suggested somewhat more pronounced associations between abuse or neglect and vaso-motor symptoms among African American women (Table 4).
This study is the first to examine the association between childhood abuse or neglect and the presence of vasomotor symptoms during the menopausal transition. Results indicate that childhood abuse or neglect is associated with significantly increased reporting of hot flashes and night sweats during midlife. Findings were consistent across hot flashes and night sweats and across multiple forms of abuse and neglect, and persisted after controlling for potential confounders, including psychological factors such as depressive symptoms.
The present findings are consistent with a burgeoning literature documenting increased adverse health conditions, particularly symptomatic health conditions, among adults with a history of childhood abuse or neglect. Childhood abuse is a known risk factor for many chronic pain syndromes in adulthood, such as headache, chronic pain back, chronic pelvic pain, chest pain, abdominal pain, and premenstrual complaints,10–14 as well as general symptom reporting.11,48 However, there has been little examination of menopausal symptoms. Findings from the Melbourne Women’s Health Project indicated that intimate partner violence was associated with reporting significantly more bothersome symptoms (including but not specifically vasomotor symptoms)51 and childhood sexual abuse with poor sexual functioning during menopause.52 However, there has been no examination of the association between childhood abuse or neglect and vasomotor symptoms, a prevalent symptom associated with impaired mental, physical, and social quality of life among midlife women.24–26 Moreover, there has been little examination of how relations between childhood abuse and any health outcome vary by race/ethnicity.
The mechanisms by which abuse or neglect may affect vasomotor symptoms are unclear, in part because of the incomplete understanding of the cause of vasomotor symptoms. Vasomotor symptoms may be linked to health behaviors, with vasomotor symptoms most prevalent among women who smoke or who are obese.15 Moreover, women exposed to childhood maltreatment may be more likely to smoke36 or be obese.37,38 However, in the present study, abuse or neglect was not significantly related to smoking and was only marginally associated with BMI, and associations persisted with adjustment for smoking and BMI.
A second link between abuse or neglect and vasomotor symptoms may be a psychological one. As with reporting of most physical symptoms,53 psychological factors may affect the reporting of vasomotor symptoms. For example, anxiety levels assessed premenopausally are among the strongest predictors of vasomotor symptom reporting at midlife, with findings from the SWAN and the Penn Ovarian Aging Study showing women with elevated levels of anxiety symptoms having an almost threefold higher odds of reporting vaso-motor symptoms relative to women with lower anxiety levels.15,30 Moreover, in research with physiological measures of hot flashes, depressive and anxious symptoms are associated with increased reporting of hot flashes not detected physiologically.54 Notably, childhood abuse has been demonstrated to be a significant risk factor for mood and anxiety disorders in adult women across multiple investigations.2–5 However, adjusting for these negative affective factors had little impact on the observed relations between abuse and vasomotor symptoms. Although other psychological characteristics not assessed here may be associated with symptom reporting,55 the present findings point to the importance of considering other mechanisms in explaining the association between abuse and vasomotor symptoms.
A third pathway linking maltreatment and vasomotor symptoms may be a direct physiological one. Leading models characterize vasomotor symptoms as hormonally mediated central nervous system thermoregulatory events,31 and empirical evidence suggests a role of reproductive hormones such as estradiol30,32,33 and central adrenergic34,35 and central seroto-nergic56,57 functioning in the occurrence of vasomotor symptoms. Notably, long-term changes in central nervous system functioning is widely postulated to occur with exposure to abuse in childhood.40–42 Although empirical evidence demonstrating persistent alterations among adults with a history of childhood abuse is limited, available evidence shows altered hypothalamic-pituitary-adrenal axis functioning42,43 among adults who have experienced abuse as children. Altered hypothalamic-pituitary-adrenal axis functioning may be in turn associated with impaired hypothalamic-pituitary-gonadal axis functioning,58,59 which may underlie the greater pregnancy complications,60 earlier age at menarche,61 and lower estradiol levels at midlife44 found among women exposed to childhood abuse. Furthermore, suggestive evidence indicates that childhood abuse may be linked to altered peripheral adrenergic functioning39 and trauma with altered serotonergic functioning.40 Thus, although speculative, long-term changes in central nervous system and hormonal functioning associated with abuse or neglect may render women more vulnerable to vasomotor symptoms.
The present study suggested a stronger association between abuse or neglect and vasomotor symptoms among African American versus white women. As indicated by the lack of a significant interaction term, these racial differences were not statistically significant and therefore should be interpreted with caution. However, consistent with previous findings,1,15 higher rates of vasomotor symptoms and slightly higher rates of abuse and neglect were observed among African American women, which may affect the ability to detect statistically significant differences in this group. Moreover, some previous evidence has suggested that associations between parental physical abuse and internalizing symptoms, including somatic symptoms, may be particularly strong among African American as compared with white adolescents.62 Furthermore, suggestive evidence indicates differences in symptom perception and endogenous pain regulatory mechanisms between African American and white racial/ethnic groups,63,64 which may influence how stressors influence the occurrence and reporting of vasomotor symptoms.
These study results should be interpreted in the context of several limitations. Childhood abuse and neglect were assessed using retrospective report, which has known limitations, particularly underreporting.65,66 Thus, there may have been misclassification of the exposure, although this misclassification would have decreased the likelihood of detecting significant associations. Vasomotor symptoms were assessed using brief self-report measures about number of days experiencing vasomotor symptoms, which yield some information about frequency but not symptom intensity or bothersomeness. Retrospective self-report measures may also be vulnerable to the impact of psychological factors,54,67,68 although adjusting for affective factors did not attenuate results. As with the majority of investigations of abuse and physical symptoms, use of self-report measures for abuse and symptoms may be associated with common method variance. Moreover, this investigation included white and African American women; the degree to which findings generalize to other racial/ethnic groups is unknown. Power to detect racial/ethnic differences of this modest magnitude was also limited. Although abuse pertaining to childhood was examined in relation to vasomotor symptoms occurring during midlife, the causal nature of these associations cannot be established. Finally, the ability to evaluate potential pathophysiologic mechanisms was limited in this study. Future prospective research should be conducted to more completely identify potential pathways that may link childhood abuse or neglect with increased vasomotor symptoms.
This study has several notable strengths. It included a well-characterized sample of African American and white midlife women. Childhood abuse and neglect were assessed with a widely used and well-validated scale that allowed for examination of associations across subtypes of abuse and neglect. Vasomotor symptoms, as well as multiple covariates, were assessed longitudinally over the course of the menopausal transition.
This study is the first to examine the association between childhood abuse and the presence of vasomotor symptoms in adulthood. It indicated that women with a history of childhood abuse are more likely to report vasomotor symptoms during midlife. These findings may suggest that the sequelae of childhood abuse may extend well into midlife to affect health during the menopausal transition.
Clinical Center: University of Pittsburgh, Pittsburgh, PA: Karen Matthews, principal investigator (PI). National Institutes of Health Program Office: National Institute on Aging, Bethesda, MD: program officers Marcia Ory (1994–2001) and Sherry Sherman (1994-present), National Institute of Nursing Research, Bethesda, MD. Coordinating Center: New England Research Institutes, Watertown, MA: Sonja McKinlay, PI (1995–2001); University of Pittsburgh, Pittsburgh, PA: Kim Sutton-Tyrrell, PI (2001-present). Steering Committee: Chris Gallagher, chair, and Susan Johnson, chair. The authors thank the study staff at each site and all the women who participated in SWAN.
Funding/support: The Study of Women’s Health Across the Nation has grant support from the National Institutes of Health, Department of Health and Human Services, through the National Institute on Aging, the National Institute of Nursing Research, and the Office of Research on Women’s Health (AG012546, AG012553). Supplemental funding from the National Institute of Mental Health (MH59689) and the Pittsburgh Mind Body Center (HL076852/076858) are also gratefully acknowledged.
Financial disclosure: None reported.