Post-traumatic stress disorder symptoms and sexual function among midlife women

Two hundred and seventy-four women ages 45–67 recruited in 2017–2020 for the parent MsBrain study, a study of menopause and brain health [ 29 ]. Participants were recruited from the Pittsburgh, PA community via advertisements, registry mailings, and a menopause and cardiovascular health study [ 30 ]. Study exclusion criteria reflected the aims of the parent study on menopause and included pregnancy; hysterectomy and/or bilateral oophorectomy; history of stroke/cerebro-vascular accident; Parkinson's disease, chemotherapy; a history of dementia; seizure disorder; brain tumor; active substance abuse (established via urine toxicology screen); history of head trauma with loss of consciousness >60 min; contraindication to MRI; and use of systemic estrogen and/or progesterone, selective estrogen receptor modulators, aromatase inhibitors, gabapentin, selective serotonin reuptake inhibitors, or serotonin norepinephrine reuptake inhibitors. Of the primary sample of 274 women, 270 women completed the sexual function measure. Further, consistent with prior research [ 31 ], women who reported no sexual activity in the past month were excluded from analyses, to avoid conflating women who did not attempt sexual activity in the past month with sexual dysfunction. Participants underwent telephone and in-person screening, physical measurements, questionnaires, and a three-day sleep protocol via actigraphy. Study procedures were reviewed and approved by the University of Pittsburgh Human Research Protection Office. All participants provided written informed consent. The PTSD Checklist Civilian Version (PCL–C) for DSM-IV [ 32 ] is a validated, 17-item self-report inventory assessing PTSD symptoms over the past month [response options: 1 (not at all) to 5 (extremely)]. The PCL-C yields a total symptom severity score (range = 17–85), with higher scores reflecting greater PTSD symptom severity; total scores ≥30 reflect the clinical cutpoint for elevated PTSD symptoms among community samples. Additionally, per PCL-C scoring criteria [ 33 ], individual items can be scored as symptomatic (response options 3–5, Moderately or above) vs. non-symptomatic (response options 1–2, Not at all or A little bit), then categorized into scoring positive for each of the three symptom clusters based on DSM criteria: (1) Re -experiencing cluster: symptomatic response to at least one item on questions 1–5; (2) Avoidance/numbing cluster: symptomatic response to at least three items on questions 6–12; and (3) Hyperarousal/hypervigilance cluster: symptomatic response to at least two items on questions 13–17. For the present investigation, consistent with PCL-C scoring guidelines, primary analytic models used the continuous summed PTSD symptom severity score and exploratory models used scoring positive on each of the three PTSD symptom clusters (yes/no each symptom cluster; analyzed in separate models); in additional exploratory models, we also chose to examine continuous total scores on each PTSD symptom cluster. The Female Sexual Function Index (FSFI-6) [ 34 ] is a validated 6-item self-report measure of past-month sexual function. The FSFI-6 is a short form of the original 19-item FSFI scale [ 35 ]. Women are asked to consider partnered or unpartnered sexual activity, including caressing, foreplay, masturbation, and sexual intercourse (vaginal penetration). FSFI-6 items are rated on a 5-point Likert scale reflecting variations in frequency, intensity, or degree of satisfaction. The FSFI-6 yields a total score (range = 2–30, lower scores reflect lower overall sexual function) and six sub-domains of sexual function, each assessed by a single item, including: sexual desire or interest, sexual arousal, lubrication, orgasm, satisfaction, and discomfort/pain. FSFI-6 total scores ≤19 reflect clinical sexual dysfunction [ 34 ]. Consistent with prior work [ 34 ], we did not consider individual sub-domains in analyses, given validation of total scores only. Women underwent three days of objective sleep assessment via actigraphy. Women wore an Actiwatch-2 wrist actigraph on the non-dominant wrist (Respironics, Inc., Murrysville, PA) [ 36 ] and completed a sleep diary [e.g.,37] for three days. Actigraphy data were collected in 1-min epochs and analyzed with Philips Actiware v6.0.0 software, with a wake threshold of 40 and number of epochs of sleep/wake for sleep onset/offset of 10. Bedtime (time tried to go to sleep) and rise time (final wake time) were determined via self-reported sleep diary. Total sleep time (hours) was defined as: [(difference between actigraphy-measured sleep onset and actigraphy-measured final wake time) − (actigraphy-measured minutes of wakefulness between sleep onset time and final wake time)] [ 36 ]. Height was measured via fixed stadiometer and weight via balance beam scale, and body mass index (BMI) was calculated [weight (kg)/height 2 (m)]. Demographics and health behaviors including current alcohol use (yes/no) were self-reported via questionnaires. Medical history and medication use were documented via interview; medications were classified according to their indication (e.g., depression, vaginal estrogen) and coded for analysis as current vaginal estrogen use (yes/no) and current depression or anxiety medication use (yes/no). Menopause status (0 = pre-menopausal vs. 1 = postmenopausal) and number of years since final menstrual period (FMP) were obtained from reported menstrual bleeding patterns [ 38 ]. Race/ethnicity was self-reported and categorized for this analysis as 0 = minoritized race/ethnicity vs. 1 = white. Educational attainment was assessed as years of completed education and classified as 0 = less than a college degree vs. 1 = college degree or greater for analysis. Women reported partner status, categorized in this analysis as: 0 = widowed, separated, divorced/formerly lived with someone in a marital-like relationship, or never married/never lived with someone in a marital-like relationship vs. 1 = currently married and living together or living with someone in a marital-like relationship. Depressive symptoms were assessed via the Center for Epidemiologic Studies Depression (CESD) scale and were categorized according to the clinical cut point (CESD≥16; yes/no) [ 39 ]. Adult trauma was assessed via the validated Brief Trauma Questionnaire [ 40 ] which assesses lifetime exposure (yes/no) to nine traumatic events (e.g., car accidents, natural disasters, sexual assault); women who reported exposure to at least one event were classified as positive for adult trauma history. Childhood trauma was assessed in a subset of women ( N = 69) via the 28-item short form of the Child Trauma Questionnaire (CTQ), a validated multi-dimensional scale of childhood abuse and neglect experienced at or before age 18 [ 41 ] and categorized according to clinical cutpoints [ 42 ]. Variables were examined for distributions, outliers, and cell sizes. PTSD symptom severity scores and BMI were log transformed given skewness. Independent samples t -tests and chi-square tests were used to test differences in study variables for women who did vs. did not engage in sexual activity in the past month. For our primary analyses, linear regression models were used to test associations of PTSD symptom severity scores with sexual function (FSFI-6 total score) among the subsample of 121 women who reported past-month sexual activity, adjusting for age, race/ethnicity, education, partner status, vaginal estrogen use, alcohol use, and years since FMP (Model 1) and further adjusted for BMI, depressive symptoms, and actigraphy-assessed total sleep time (Model 2, fully-adjusted). In exploratory analyses, linear regression models were used to (1) examine associations between PTSD symptom clusters (i.e., women who did vs. did not score positive for re-experiencing, avoidance/numbing, and hypervigilance, respectively, as well as total scores on each of these clusters) and sexual function in separate fully-adjusted models, and (2) examine relationships between adult trauma exposure (any trauma and individual subtypes, respectively) and childhood trauma exposure (any trauma and individual subtypes, respectively) and sexual function in fully-adjusted models. Finally, in additional analyses, we considered use of medications for depression or anxiety as an additional covariate in fully-adjusted primary models (in addition to self-reported depressive symptoms) and examined associations between PTSD symptom severity scores and FSFI-6 scores consistent with clinical sexual dysfunction (FSFI-6 total ≤ 19, yes/no) via logistic regression models. All tests were two tailed with an alpha set to 0.05. Analyses were conducted using SPSS v28.0.1.0 (IBM SPSS Statistics) and SAS v9.4 (SAS Institute, Cary, NC).

Table 1 presents demographic and health characteristics for the full sample ( N = 270) and broken down by women who did vs. did not report sexual activity (partnered or unpartnered) in the past month. In the primary analytic sample of 121 women who reported past-month sexual activity, participants were on average 58.3 years old (range = 45–66 years). Women self-identified as Asian or Pacific Islander (2%), Black (17%), and Multiracial (1%), or White (81%). Over 80% of the women who engaged in sexual activity over the past month were currently married or partnered and 98.3% ( N = 119 women) were postmenopausal. Overall, average PTSD symptom severity score was 23.9 ( SD = 6.8, range = 17–49), with only 23 women (19.0%) scoring above the typical clinical cutoff for probable PTSD. Further, whereas the average sexual function scores fell within what are considered normal ranges ( M = 20.5, SD = 5.1, range = 6–29) on the FSFI-6, over a third of the scores (34.7%, N = 42) fell in the range of clinical sexual dysfunction. Women who reported past-month sexual activity were younger [ M = 58.3 vs. 59.7 years, t (268) = 2.66, p < .01], reported fewer number of years since FMP [ M = 8.2 vs. 10.1 years, t (268) = 3.20, p < .01], and more likely to be currently married/partnered relative to women who reported no sexual activity [84.3% vs. 51.0%; χ 2 (1, N = 270) = 32.94, p < .001]; see Table 1 . No other demographic or health differences emerged between women who did vs. did not report past-month sexual activity, including no differences in average PTSD symptom severity or proportion of women scoring positive for PTSD symptom clusters (see Table 1 ). In primary analyses, when considering associations between PTSD symptom severity and past-month sexual function, results indicated that greater PTSD symptom severity scores were associated with significantly poorer sexual function after adjustment for age, race/ethnicity, education, partner status, vaginal estrogen use, alcohol use, and years since FMP ( Table 2 , Fig. 1 ). This association remained significant with additional adjustment for BMI, depressive symptoms, and actigraphy-assessed total sleep time ( Table 2 ). In exploratory models, we examined associations between PTSD symptom clusters and past-month sexual function. Twenty-one (17.4%), 12 (9.9%), and 17 (14.0%) of women scored positive for re-experiencing, avoidance/numbing, and hypervigilance PTSD symptom clusters, respectively. Of these three clusters, women who scored positive (vs. those who did not) for avoidance/numbing reported poorer sexual function [B(95% CI) = −3.69 (−6.90, −0.48), p = .025] after adjustment for age, race/ethnicity, education, partner status, vaginal estrogen use, alcohol use, years since FMP, BMI, depressive symptoms, and actigraphy-assessed total sleep time. No associations emerged between re-experiencing or hypervigilance clusters and sexual function ( p s > 0.05). Relationships between continuously-scored PTSD symptom clusters and past-month sexual function are presented in Supplemental Table 1 . Additionally, we explored history of adult trauma and childhood trauma, respectively, in relation to past-month sexual function (see Supplemental Table 2 for prevalence of trauma exposure by past-month sexual function). Childhood emotional abuse and childhood emotional neglect, respectively, were related to poorer sexual function in fully-adjusted models ( Supplemental Table 3 ). No associations emerged between any adult trauma or adult trauma subtypes, respectively, and sexual function ( Supplemental Table 4 ). Finally, we conducted several additional models. First, for our primary analysis examining the relationship between PTSD symptom severity and sexual function, we included a covariate for depression or anxiety medication use in our fully-adjusted models and results were similar [B(95% CI) = −6.57 (−10.90, −2.24), p = .003] to those reported (see Table 2 , column 2). Second, when considering the relationship of PTSD symptom severity to an FSFI score consistent with sexual dysfunction (FSFI-6 total ≤ 19), associations were not statistically significant (OR [95%CI] = 3.66 [0.66, 20.19], p = .137) after adjustment for age, race/ethnicity, education, partner status, vaginal estrogen use, alcohol use, and years since FMP.

Among a well-characterized community sample of midlife women, higher PTSD symptom severity scores were related to poorer past-month sexual function. These associations persisted beyond adjustment for a range of potential confounders, including vaginal estrogen use, alcohol use, years since FMP, depressive symptoms, and sleep duration. Critically, although women in this sample had relatively low PTSD symptomatology, associations between PTSD symptoms and sexual function were observed even at these low, largely subclinical, symptom levels. Results highlight the importance of considering psychosocial sequelae of trauma history in midlife women's health and wellbeing. This study adds to the nascent literature on PTSD and sexual function in midlife women [ 12 , 20 ]. Select prior research has reported associations between PTSD symptoms and poorer sexual function [ 9 - 21 ], yet largely included Veterans (including men), younger reproductive-aged women, or clinical samples of women focused on reproductive conditions or emotion dysregulation. Our results are consistent with the few prior studies of women at midlife and older age, which found that greater PTSD symptoms were related to vaginal dryness, irritation, and pain with intercourse large multiethnic community cohort of women aged 40–80 years [ 20 ], and also to sexual dysfunction among women Veterans aged 45–64 years [ 21 ]. The present results extend and advance the literature by studying a community sample of midlife women, important as midlife can be a time of notable psychological and sexual challenge for women. We further advanced upon prior work by considering and adjusting for a range of relevant covariates with known associations with PTSD symptoms and sexual function, including vaginal estrogen use [ 26 ] and alcohol use [ 28 ], as well as actigraphy-measured sleep duration [ 27 ] in additional models, to avoid potential reporting biases with PTSD symptoms. In exploratory models, we considered specific subtypes of PTSD symptoms, finding that associations between PTSD symptomatology and lower sexual function may be pronounced among women experiencing avoidance/numbing symptoms. Given the small number of women in this sample who scored as elevated on the avoidance symptom cluster ( N = 12), this association should be interpreted with caution; however, it is consistent with prior findings in samples of women [ 9 , 15 ]. Avoidance/numbing symptoms may have adverse impacts on relationships and social support [ 43 ], which may contribute to sexual dysfunction. Indeed, it may be the case that avoidance/numbing symptoms contribute to a disconnect between mind and body (e.g., psychological arousal and physical arousal) that leads to lower sexual satisfaction and/or sexual function. Ultimately, more work is needed, particularly among midlife women, to better identify what aspects of PTSD are relevant for sexual function outcomes. Results from exploratory analyses also suggested that childhood emotional abuse or emotional neglect, respectively, were related to poorer past-month sexual function, consistent with a growing literature describing the potency of emotional/psychological trauma for health outcomes [e.g., 44 ]. No other childhood or adult trauma subtypes were associated with sexual function, including sexual trauma (see also [ 8 ]). Indeed, these results further underscore the value of considering psychological sequelae of trauma on women's sexual function, and not solely the exposure to trauma. Critically, our findings concerning trauma subtypes must be interpreted with caution, given the exploratory nature of these analyses and the fact that only a subset of women ( N = 68) completed the measure of childhood trauma. Several mechanisms may link PTSD symptoms to sexual dysfunction, including prolonged sympathetic nervous system arousal [ 8 ]. Future work is warranted to study alterations in hormonal (e.g., oxytocin, testosterone) and neural networks (e.g., amygdala activity) relevant to attachment, arousal, and/or fear, as well as neural pathways involved in the development and consolidation of traumatic memories [ 8 ]. Alcohol use, poor sleep, and depressive symptoms have been associated with PTSD and worse sexual function in prior work [ 25 , 27 , 28 ]; while our associations were not explained by these factors, they are relevant to consider in future work, in addition to other health behaviors such as physical activity; see [ 8 ] for a comprehensive review of potential physiological, psychological, behavioral, and neural mechanisms. Results highlight the need for healthcare providers to adopt a trauma-informed approach and query all midlife women about trauma history as well as sexual healthcare needs in routine clinical care [ 45 ]. Many women are interested in behavioral rather than pharmacological treatments for low libido [ 46 ]. Limited evidence suggests that gold-standard psychological treatments for PTSD (e.g., prolonged exposure therapy, cognitive processing therapy), may be helpful in improving sexual function, although improvement may depend on the magnitude of reduction in PTSD symptoms [ 11 , 17 ] and type of trauma exposure. Emerging data suggests mindfulness-based therapies may be helpful for reducing sexual distress, including among midlife women and those with histories of interpersonal violence including emotional abuse [ 46 , 47 ]. Overall, a range of psychological and behavioral treatments may be beneficial for sexual dysfunction, but more research is needed (e. g., [ 48 ]). This investigation has several limitations. We did not conduct diagnostic clinical interviews, nor did we assess PTSD treatment history. Per FSFI scoring, analyses excluded women who did not report past-month sexual activity (partnered or unpartnered). Results from this convenience sample may not be generalizable to all midlife women or to women who are less sexually active. We used a validated 6-item short version of the FSFI scale, which assesses sexual function domains via single items, precluding the investigation of PTSD symptoms with sub-domains of sexual function. We did not conduct clinical interviews to diagnose PTSD, and participants had relatively low levels of PTSD symptoms, however, associations between PTSD symptoms and decreased sexual function were observed even at these low symptom levels. PTSD symptoms were measured via the PCL–C, which is based upon DSM-IV criteria and captures only three symptom clusters, rather than measures based upon current expanded DSM-5 criteria (e.g., PCL-5), which captures additional symptoms and a broader range of symptomatology, including a fourth symptom cluster (negative alterations in cognition and mood). While prior evidence indicates high agreement between PCL-C and PCL-5 [ 49 ], future work should use the PCL-5 or other instruments that align with DSM-5 criteria. We did not assess relationship quality, intimate partner violence, nor the sexual function of women's partners, which could have impacted the participant's sexual function. While midlife is often associated with natural declines in sexual function for many women in part due to the menopause transition, it is possible that results could be confounded by effects of the menopausal transition; however, our sample was almost exclusively composed of postmenopausal women and associations between PTSD and sexual function persisted after adjusting for years since FMP. All women identified as cisgender and the vast majority of the sample was non-Hispanic White or Black. Future work is needed in larger, more diverse samples. Findings from exploratory analyses should be considered with caution and replicated in larger samples, given the multiple tests conducted and small numbers of women who scored positive for PTSD symptom clusters and who endorsed individual childhood and adult trauma subtypes, that PCL-C scoring guidelines do not include creating continuous total scores of PTSD symptom clusters, and the fact that childhood trauma was only assessed in a subset of women. We cannot make claims regarding directionality or causality, given the observational and cross-sectional nature of this study. This study also has several notable strengths. It included a well-characterized community cohort of midlife women. Women taking common antidepressants or who had active substance abuse that can contribute to sexual dysfunction were excluded. We leveraged validated and widely-used measures of both PTSD symptoms and sexual function. We adjusted for a range of covariates in our analytic models, including socio-demographics, medication use, health behaviors, menopause effects, and depressive symptoms, allowing us to account for factors that have known associations with both PTSD symptomatology and sexual function. In conclusion, PTSD is a significant health issue for women. Among midlife women, higher PTSD symptom severity scores, particularly avoidance/numbing symptoms, were related to worse past-month sexual function. Findings underscore the need to assess and address PTSD symptoms in routine clinical care in order to improve midlife and aging women's health and wellbeing.

Most women in the U.S. will experience at least one major traumatic event in their life, which can include combat exposure, natural disasters, or interpersonal violence [ 1 ]. Roughly 10% of women will develop post-traumatic stress disorder (PTSD) following trauma exposure, a doubling of risk relative to men [ 2 ]. Moreover, PTSD is more persistent and severe among women [ 3 ]. Key clusters of PTSD symptoms include re-experiencing the event, avoidance of reminders of the event, hyperarousal, and negative alterations in mood and cognitions [ 3 ]. Many women will also experience elevated PTSD symptoms that do not meet the clinical threshold; elevated PTSD symptoms have been associated with adverse mental and physical outcomes [ 4 ], emotion dysregulation [ 5 ], and difficulties with social functioning [ 6 ], including increased conflict in intimate relationships [ 7 ]. Greater PTSD symptomatology is also related to poorer sexual function, an important dimension of health and intimate relationships. Yehuda et al. [ 8 ] provide a comprehensive conceptual model of sexual dysfunction in the context of PTSD symptomatology, summarizing it as “an inability to regulate and redirect the physiological arousal needed for healthy sexual function away from aversive hyperarousal and intrusive memories.” PTSD symptoms can have broad impacts on sexual function, including desire, arousal, orgasm, and satisfaction [ 8 ]. Evidence derives mainly from observational studies of male and female Veteran samples and indicates that PTSD symptoms are related to worse overall sexual function, including lower sexual desire and satisfaction ([for a review, see 9 ]), erectile difficulties, and vaginal pain [ 10 ]. Additionally, more significant improvements in PTSD symptoms following PTSD treatment predict better sexual function in samples of female Veterans and civilians [ 11 , 12 ]. Beyond Veteran samples, studies of PTSD symptoms and sexual function are focused on select populations, including younger, reproductive-aged women [ 13 - 15 ], crime victims [ 16 ], and clinical samples of women receiving treatment for emotion regulation difficulties [ 17 , 18 ] or endometriosis [ 19 ]. Only two studies to date have focused specifically on midlife and aging women, providing initial evidence for associations between PTSD symptoms and poorer sexual health [ 20 , 21 ]. Despite accumulating knowledge that PTSD is related to worse sexual function, there has been little focus on community-based samples of women at midlife and older ages. Midlife is a particularly important time to consider women's psychological as well as sexual function. For women, midlife includes the menopause transition, which is characterized for many women by changes in sexual function [ 22 ] and mood [ 23 ] that can negatively impact sexual function. Data from large cohort studies suggest that sexual activity declines for some women during midlife [ 22 ], yet many midlife women also emphasize the importance of a satisfying sex life for health and well-being [ 24 ]. PTSD symptoms reflect a psychosocial risk factor that may be relevant to sexual function yet is poorly understood among midlife women and is not regularly queried in routine clinical practice. Moreover, prior work does not always consider the range of potential covariates relevant to both midlife sexual function and PTSD symptoms, including mood symptoms 30 [ 25 ], vaginal estrogen use [ 26 ], and health behaviors like sleep problems [ 27 ] and alcohol use [ 28 ], highlighting the need to investigate these complex relationships, especially among midlife and aging women. In the present study, we tested the association between PTSD symptoms and sexual function in a well-characterized community sample of midlife women. We hypothesized that women who reported greater PTSD symptom severity would report lower sexual function, adjusting for socio-demographics, health factors, and depressive symptoms. To determine which aspects of PTSD symptoms may be particularly relevant to women's sexual function, in exploratory models we also examined relationships between PTSD symptom clusters (re-experiencing, avoidance/numbing, and hypervigilance, respectively) as well as trauma type with sexual function.