Evaluating the Effect of Depression,Anxiety,and Post-Traumatic Stress Disorder on Anti-Müllerian Hormone Levels Among Women Firefighters

Abstract

Objective:

To assess whether depression, anxiety, and post-traumatic stress disorder (PTSD) are associated with serum anti-Müllerian hormone (AMH) levels.

Study Design:

We used data from a sample of women firefighters from the Fire Fighter Cancer Cohort Study. Participant demographics, reproductive history, and self-reported clinical diagnosis of anxiety, depression, and PTSD were collected with serum for AMH analysis at enrollment.

Main Outcome Measure:

Linear regression models were used to estimate the association between anxiety, depression, and PTSD and log transformed AMH levels adjusted for age years (continuous and squared) and body mass index. Percent difference in AMH was calculated by [exp(β) – 1] × 100.

Results:

Among 372 participants, with mean ± standard deviation age 32.54 ± 6.32, clinical diagnoses were reported as follows: depression (15%), anxiety (18.2%), or PTSD (8.7%). No statistically significant association was observed between depression and AMH levels (−22%Δ, 95% confidence interval [CI]: −47.3, 14.5). Women firefighters with a history of anxiety (−33%Δ, 95% CI: −53.5, −4.2) and PTSD (−66%Δ, 95% CI: −79.1, −44.6) had lower serum AMH compared with participants without those conditions. When individuals with concurrent PTSD were excluded, the association between anxiety ceased to be statistically significant (26.7%Δ, 95% CI: −17.9, 92.6).

Conclusion:

A history of clinically diagnosed anxiety or PTSD was associated with statistically significantly lower AMH levels. This association offers insight into the potential biological mechanisms through which mental health conditions may influence reproductive health.

Introduction

Firefighters are exposed to multiple occupational psychosocial stressors increasing the risk of developing mental health conditions.¹ Firefighters have been shown to experience a higher prevalence of depressive symptoms and post-traumatic stress disorder (PTSD) compared with the general population,^2,3 and more women firefighters have PTSD symptoms than their male counterparts (20% versus 12%).⁴ Certain factors, such as workplace discrimination and harassment, may worsen symptoms of depression, anxiety, and PTSD among women firefighters.⁵ Depression, anxiety, and PTSD have been shown to be negatively correlated with ovarian reserve among women in the general U.S. population who are seeking fertility treatment.⁶ Additionally, depression has been observed to be associated with changes in the menstrual cycle including timing to menopause.⁷ Prior research among nonfirefighting women has demonstrated that those with a history of PTSD have an increased risk of experiencing gynecological complications including gynecological surgeries (hysterectomy, bilateral salpingo-oophorectomy), endometriosis, polycystic ovary syndrome (PCOS), and infertility.^8
–10

To date, the effect of mental health conditions on adverse reproductive outcomes in women firefighters remains unknown.^11

–15 Previous studies of women firefighters have found that they have a higher risk of miscarriage, premature delivery, and lower anti-Müllerian hormone (AMH) levels compared with the general population.^11,12,15 AMH, a member of the transforming growth factor β family, is secreted from granulosa cells in the ovaries and is considered a clinical marker of functional ovarian reserve.¹⁶ Notably, serum AMH levels may reflect a woman’s reproductive age more accurately than her chronological age and predict transition to menopause more effectively than other classical ovarian reserve markers.¹⁶ Moreover, decreased AMH has been correlated to decreased ovarian reserve and a shortened interval to menopause, whereas high levels of AMH may be associated with PCOS.^16,17 The purpose of this study was to determine whether mental health conditions, specifically self-reported clinical diagnoses of depression, anxiety, and PTSD, are associated with serum AMH levels among women firefighters. We hypothesized that these conditions would be associated with lower AMH levels.

Materials and Methods

Study population

Study participants were drawn from the Women Firefighter Study: Stress, Cancer, and Reproductive Toxicity, a component of the larger Fire Fighter Cancer Cohort Study. Participants included women firefighters in a fire training academy (recruits) and women firefighters who have been with the fire service for more than 1 year (incumbents), all of whom were 18 years of age and above. Study enrollment, completion of standardized surveys, and biological sample collection were performed at the participating fire departments. Study enrollment took place between September 2020 and September 2022. Standardized surveys collected information on participant demographics, cancer and behavioral risk factors, and firefighting history. Study data were collected and managed using REDCap electronic data capture tools at The University of Miami. An institutional review board at The University of Miami (#20170997) approved this research, and all participants provided informed consent.

Survey data and serum for AMH levels measurement were collected at the time of enrollment for all participants (n = 372). Our analysis was restricted to participants under the age of 45 (n = 334). Individuals who had reported a history of chemotherapy (n < 10), a previous diagnosis of a medical condition known to affect AMH (n = 11), or a clinical diagnosis of PCOS (n = 22) were excluded. Because high levels of AMH are associated with PCOS, we excluded those with AMH ≥10 ng/mL (n = 17).¹⁶ Additional exclusions were current use of hormonal medications (n = 21) (i.e., hormonal contraceptives), currently pregnant (n < 10), postmenopausal (n = 11), or postoophorectomy (n < 10). Finally, participants with missing exposure variables (n < 10) were excluded. Our final sample consisted of 250 participants.

Primary exposure: Clinical diagnoses

Self-reported clinical diagnoses of depression, anxiety, and PTSD were ascertained from the enrollment questionnaire. Participants indicated whether a doctor or medical provider had ever given them a diagnosis (“Has a doctor or medical provider EVER told you that you had…”) of each of the conditions: depression: “a depressive disorder (including depression, major depression, dysthymia, or minor depression”; anxiety: “an anxiety disorder (including acute stress disorder, anxiety, generalized anxiety disorder, obsessive-compulsive disorder, panic disorder, phobia, social anxiety disorder, and PTSD)”; or PTSD. Participants who selected “yes” were categorized as having a clinical diagnosis of the indicated condition. Missing values were excluded.

Secondary exposures

Separate from the self-reported clinical diagnoses of depression, anxiety, and PTSD described above, we performed a secondary analysis using self-reported mental health scales assessed via survey for depression Patient Health Questionnaire (PHQ)-4, anxiety (PHQ-4), PTSD (PCL-C), and stress (Perceived Stress Scale; PSS-10) to evaluate their associations with AMH levels.^18,19 The PHQ-4 consists of four questions that allow for measurement of symptomatic depression (depression subscale ≥3) and anxiety (anxiety subscale ≥3) experienced in the past 2 weeks. The PCL-C consists of six items that measure response to stressful life experiences in the last month.²⁰ Participants with a total score of 14 or more were categorized as having symptomatic PTSD.²⁰ The PSS-10 is a 10-item scale ranging from 0 to 40 assessing stress experienced over the last month. The PSS-10 scores were categorized into tertiles (Low: 0–13, Moderate: 14–26, and High: 27–40).¹⁹ After excluding missing values of the self-reported mental health scales, there were 249 participants in the depression analysis, 247 participants in the anxiety analysis, 248 participants in the PTSD analysis, and 245 participants in the stress analysis.

Outcome: AMH sample collection and testing

Blood was collected by qualified phlebotomists in a 10.0 mL serum tube for AMH analysis, with additional blood collected for storage and future analyses. Blood was stored for 30 minutes to 1 hour before being chilled at 2–8°C and shipped priority overnight to the University of Arizona where it was processed. The serum tube was centrifuged at 1300 g for 15 minutes, and the serum was aliquoted in 0.5–1.0 mL aliquots. The serum was then stored frozen at −80°C until analyzed. The additional blood was processed and stored at −80°C for future analysis. Frozen serum samples were shipped on priority overnight on dry ice in accordance with International Air Transport Association guidelines to the Motive Biosciences laboratory (Webster, TX) for quantification of AMH. Serum AMH (nanograms per milliliter) levels were measured using the picoAMH log form method for all women that indicated their age at enrollment. Values below the lower limit of detection (LLD) were set to LLD/√2.

Statistical analyses

Primary analyses

All statistical analyses were conducted using R version 4.2.3 (2023-03-15). Descriptive statistics were determined for the total sample and stratified by self-reported clinical diagnosis of depression, anxiety, and PTSD. Continuous variables were described by the mean and standard deviation, whereas categorial variables were described using proportions. AMH levels were described using the geometric mean, standard deviation, and range. To determine the percent change in AMH levels by self-reported diagnosis of depression, anxiety, or PTSD, linear regression models were utilized. AMH was log-transformed to account for the non-normal distribution. Models were adjusted for centered age (continuous and squared) and body mass index (BMI) (continuous). The percent change in AMH was calculated using the following formula: [exp(β) – 1] × 100. Since the distribution of PTSD in our sample was highly correlated with age, we stratified the models by age group (<35, 35+) and tested for effect modification using the likelihood ratio test.

Sensitivity analyses

Three sensitivity analyses were performed on our inclusion/exclusion criteria. First, individuals more than 45 years of age were reintroduced in the sample (n = 272). A second sensitivity analysis was performed reincorporating those with suspected PCOS (AMH ≥10), those who had self-reported diagnosis of PCOS, and those who were using hormonal medications into the sample (n = 296). The third analysis performed restricted the sample to only those who had reported at least one pregnancy (n = 87). To evaluate the effect of PTSD on the anxiety model, we excluded those who reported a clinical diagnosis of both anxiety and PTSD (n = 236). To evaluate potential confounding on the PTSD model by frequency of heavy alcohol consumption (consuming ≥4 drinks within a 2-hour period in the past 30 days), we conducted an additional multiple linear regression model adjusted for alcohol consumption. A separate analysis was carried out to determine whether AMH differed by firefighting status (incumbent versus recruit). To evaluate the influence of infertility on our results, we ran sensitivity analyses adjusting for infertility history in our multivariable model. To evaluate for potential confounding by cigarette smoking, we excluded individuals who reported current use (n < 10).

Secondary analysis

The geometric mean AMH values and standard deviation were described by mental health scale (PHQ-4, PCL-C, and PSS-10). Multivariable linear regression models were conducted to evaluate the association between log-transformed AMH levels and the self-reported screening mental health scales (PHQ-4, PCL-C, and PSS-10). Models were adjusted for age (continuous and squared) and BMI. The percent change in AMH was calculated using the formula above.

Results

Descriptive statistics

The average age of our firefighter participants was 32.5 (standard deviation 6.3) years, and the majority were White, non-Hispanic (60.5%) (Table 1). In our sample, 15.2% of participants reported a clinical diagnosis of depression (n = 38), 18.4% reported a clinical diagnosis of anxiety (n = 46), and 8.8% reported a clinical diagnosis of PTSD (n = 22). A small percentage of the sample reported both depression and PTSD (4.0%, n = 10), anxiety and PSTD (5.6%, n = 14), and all three conditions (4.0%, n = 10). A combination of depression and anxiety was the most common (12.0%, n = 30). Those with depression were more likely to be nulliparous (71.1%) compared with those with anxiety (65.2%) and PTSD (50.0%). Those with a diagnosis of PTSD reported the highest levels of infertility (18.2%) followed by anxiety (15.2%) and depression (13.2%). Similarly, those with a diagnosis of PTSD reported the highest levels of endometriosis (9.1%) followed by depression (5.3%) and anxiety (4.4%). However, comparing rates of infertility and endometriosis in participants with and without each mental health condition, the only significant finding was an association of anxiety with infertility (p-value 0.03, data not shown).

Table 1.

Description of Study Participants (n = 250)

	Total	Diagnosed depression	Diagnosed anxiety	Diagnosed PTSD
	Total	Yes (N = 38)	Yes (N = 46)	Yes (N = 22)
Age, mean (SD)	32.5 (6.3)	31.9 (6.5)	31.5 (7.2)	35.0 (7.5)
Race ethnicity, N (%)
White non-Hispanic	153 (60.5)	30 (79.0)	35 (76.1)	16 (72.7)
White Hispanic	46 (18.2)	<10	<10	<10
Other^a	54 (21.3)	<10	<10	<10
BMI (kg/m²) (continuous), mean (SD)	25.4 (4.0)	26.0 (3.6)	25.7 (3.8)	25.3 (3.7)
BMI (kg/m²), N (%)
<18.5	<10	<10	<10	<10
18.5–24.9	121 (47.8)	20 (52.6)	24 (52.2)	10 (45.5)
25.0–30.0	100 (39.5)	12 (31.6)	16 (34.8)	10 (45.5)
>30	27 (10.7)	<10	<10	<10
Years of career and volunteer firefighting, N (%)
<1 year (recruit)	89 (35.2)	13 (34.2)	15 (32.6)	<10
≥1 year (Incumbent)	163 (64.4)	25 (65.8)	31 (67.4)	18 (81.8)
Age menarche mean (SD)	12.8 (1.6)	12.5 (1.6)	12.7 (1.6)	12.8 (1.4)
History of infertility, N (%)	18 (7.1)	<10	<10	<10
Gravidity, N (%)
0	159 (62.9)	27 (71.1)	30 (65.2)	11 (50.0)
1	37 (14.6)	<10	10 (21.7)	<10
2	19 (7.5)	<10	<10	<10
3 or more	31 (12.3)	<10	<10	<10
Heavy alcohol consumption,^b mean (SD)	0.5 (1.6)	1.1 (3.4)	0.6 (1.0)	0.5 (0.8)

Values <10 suppressed to protect participant privacy.

Includes Black or African American Hispanic or non-Hispanic, Native Hawaiian or Pacific Islander Hispanic or non-Hispanic, American Indian or Alaskan Native Hispanic or non-Hispanic, Other unknown/unknown.

Frequency of ≥4 alcohol drinks consumed within a 2-hour period in the last 30 days.

BMI, body mass index; PTSD, post-traumatic stress disorder; SD, standard deviation.

Primary analyses

In our multivariable model adjusted for age and BMI (Table 2), women with a self-reported clinical diagnosis of depression had lower AMH levels compared with those without that condition (−22.0%Δ, 95% confidence interval [CI]: −47.3, 14.6); however, this association was not statistically significant. Women who had reported a clinical diagnosis of anxiety had a statistically significantly lower serum concentration of AMH compared with those without such a diagnosis (−33.0%Δ, 95% CI: −53.5, −3.4), and the greatest reduction in AMH was observed for participants with PTSD (−66.0%Δ, 95% CI: −79.1, −44.6) compared with those without.

Table 2.

Multiple Linear Regression Models to Evaluate the Effect of Depression, Anxiety, and PTSD on AMH Levels Among Women Firefighters (n = 250)

	AMH (ng/mL) geometric mean (±SD)	Percent difference of AMH (95% CI)
	AMH (ng/mL) geometric mean (±SD)	Model 1	Model 2
Self-reported clinical diagnosis of depression
No	2.24 (3.50)	referent	referent
Yes	1.86 (5.80)	−22.8 (−47.8, 14.3)	−22.0 (−47.3, 15.6)
Self-reported clinical diagnosis of anxiety
No	2.33 (3.13)	referent	referent
Yes	1.61 (7.27)	−33.5 (−53.8, −4.2)	−33.0 (−53.5, −3.4)
Self-reported clinical diagnosis of PTSD
No	2.47 (3.02)	referent	referent
Yes	0.58 (11.57)	−65.9 (−79.0, −44.5)	−66.0 (−79.1, −44.6)

Model 1: Adjusted for centered age and centered age squared.

Model 2: Adjusted for centered age, centered age squared, and BMI (continuous).

Percent change calculated via [exp(β) – 1] × 100.

AMH, anti-Müllerian hormone; BMI, body mass index; CI, confidence interval; PTSD, post-traumatic stress disorder; SD, standard deviation.

There was statistically significant effect modification by age (<35, ≥35) for the association of anxiety (p = 0.001) and PTSD (p = 0.01) with AMH (Table 3). Among those under 35, the association between previous diagnosis of depression, anxiety, or PTSD and AMH attenuated (depression: −9.9%Δ, 95% CI: −30.8, 17.3; anxiety: −1.4%Δ, 95% CI: −23.5, 3.1; and PTSD: −31.4%Δ, 95% CI: −56.8, 8.9). However, among ages 35–45, the association between previous diagnosis of depression, anxiety, or PTSD and AMH was more pronounced (depression: −51.4%Δ, 95% CI: −85.3, 60.3; anxiety: −72.2%Δ, 95% CI: −89.8, −24.3; and PTSD: −82.5%Δ, 95% CI: −93.3, −54.1).

Table 3.

Multiple Linear Regression Models Stratified by Age (<35, 35–45)

	Age <35 (N = 168)		Age ≥35 (N = 82)
	AMH (ng/mL) geometric mean (±SD)	Percent difference of AMH (95% CI)	AMH (ng/mL) geometric mean (±SD), [range]	Percent difference of AMH (95% CI)	p-Value for interaction model
	AMH (ng/mL) geometric mean (±SD)	Model	AMH (ng/mL) geometric mean (±SD), [range]	Model	p-Value for interaction model
Self-reported clinical diagnosis of depression					0.16
No	3.4 (1.8)	referent	1.0 (5.7) [0.2, 5.7]	referent
Yes	3.1 (2.4)	−9.9 (−30.8, 17.3)	0.5 (15.2) [0.03, 7.0]	−51.4 (−85.3, 60.3)
Self-reported clinical diagnosis of anxiety					0.001
No	3.3 (1.89)	referent	1.2 (4.7) [0.3, 5.5]	referent
Yes	3.5 (2.1)	−1.4 (−23.5, 3.1)	0.3 (16.0) [0.02, 4.5]	−72.2 (−89.8, −24.3)
Self-reported clinical diagnosis of PTSD					0.01
No	3.4 (1.9)	referent	1.2 (4.8) [0.3, 5.8]	referent
Yes	2.6 (2.2)	−31.4 (−56.8, 8.9)	0.3 (14.7) [0.02, 3.7]	−82.5 (−93.3, −54.1)

Model adjusted for centered age and centered age squared and BMI (continuous).

Percent change calculated via [exp(β) – 1] × 100.

p-Value for interaction calculated using likelihood ratio test.

AMH, anti-Müllerian hormone; BMI, body mass index; CI, confidence interval; SD, standard deviation.

Sensitivity analyses

When expanding our analysis to include participants >45 years old, the association with depression and AMH remained not statistically significant (10.4%Δ, 95% CI: −25.7, 64.0), and the results attenuated for anxiety (−20.7%Δ, 95% CI: −45.9, 16.2) and PTSD (−50.1%Δ, 95% CI: −69.6, −18.0) (Supplementary Table S1). In the sensitivity analysis, expanding the sample to include those with suspected and diagnosed PCOS and those using hormonal birth control, the magnitude of association was similar to the primary model for depression (−19.7%Δ, 95%CI −43.9, 15.0) and also did not appreciably change for anxiety (−32.0%Δ, 95% CI −51.2, −5.3) or PTSD (−68.2%Δ, 95% CI −79.8, −50.5). When restricting to only those with one or more pregnancies (n = 87), the magnitude of association increased for all three diagnoses (depression: −63.7%Δ, 95% CI: −86.3, −3.7; anxiety: −69.9%Δ, 95% CI: −87.1, −29.7; and PTSD: −81.2%, 95% CI: −92.9, −50.5). In the analysis of anxiety where those with a history of PTSD were excluded, there was no longer a statistically significant association between history of anxiety and AMH levels (26.7%Δ, 95% CI: −17.9, 92.6) (Supplementary Table S2). There was no change in the model for PTSD after adjusting for heavy alcohol consumption (PTSD −65.9%Δ, 95%CI: −79.0, −44.4) (Supplementary Table S3). Adjusted AMH levels did not differ between recruit and incumbent firefighters (Supplementary Table S4). Upon removal of those who reported smoking some days, the association between a previous diagnosis of PTSD and AMH levels attenuated (−56.6 95%Δ CI −75.5, −23.1) (data not shown).

Secondary analyses

We did not find a statistically significant association between the self-reported screening measures for depression (33.3%Δ, 95% CI: −28.0, 146.8), anxiety (11.6%Δ, 95% CI: −18.6, 53.1), and PTSD (−26.7%Δ, 95% CI: −49.7, 7.0) (Supplementary Table S5) with AMH. We found a nonstatistically significant reduction in AMH in among those with moderate and high PSS-10 scores compared with those with low PSS-10 scores (Moderate PSS-10 scores: −14.1%Δ, 95% CI: −36.1, 15.5; High PSS-10 scores: −11.1%Δ, 95% CI: −60.6, 100.6) (Supplementary Table S6).

Discussion

In our study of women firefighters, self-reported clinical diagnoses of anxiety and PTSD were associated with substantial reductions in serum AMH levels. However, once individuals with concurrent PTSD and anxiety diagnosis were removed from the analysis, the association between anxiety and AMH levels no longer remained statistically significant. Our finding suggests that adverse influence on the ovaries with accompanying reduction in AMH could be a potential mechanism explaining the previously observed association between PTSD and infertility in the general population.⁸ The observed magnitude of association between PTSD and AMH (66% lower AMH levels among those with a history of PTSD in our study) is larger than many other environmental exposures known to reduce AMH levels, including indoor fuel burning (wood: −36.0% or artificial fire logs −45.8%) and cigarette smoke exposure (≥20 cigarettes/day −56.2%, 10+ years of adult environmental smoke −31.3%).²¹

In our analysis of the self-reported mental health scales, we did not find a statistically significant association between those who had scored symptomatic for anxiety, depression, or PTSD and AMH levels. Those with self-reported clinical diagnosis of the conditions may represent more severe and persistent conditions than those measured via the scales. Furthermore, the self-reported clinical diagnosis captures the lifetime prevalence of the conditions compared with the scales over a shorter time period. For example, the PHQ-4 asks specifically about 2 weeks before the questionnaire and the PCL-C captures the last month. Our data suggest that lifetime prevalence of these conditions may have a greater impact on AMH levels than a cross-sectional scale of mental health symptoms.

Previous literature of AMH suggests that a long duration of exposure is needed to lead to a reduction in AMH levels. Indeed, while certain exogenous chemical exposures such as chemotherapy may lead to an instant drop in AMH levels, other exposures such as stress appear to have more of an effect when they are more chronic.²² We chose to investigate both the self-reported prevalent clinical diagnosis of the conditions as a proxy for chronic exposure in addition to the validated scales, which capture symptomatic disease in a short time frame.

Limited studies have evaluated how well duration/severity of disease is quantified in the self-reported diagnoses versus validated scales. Previous research in active-duty service members and veterans has shown that those with more severe symptoms of PTSD are more likely to receive a clinical diagnosis.²¹ Active-duty service members with a clinical diagnosis of PTSD present with a longer duration of symptomatic PTSD compared with those who did not receive a clinical diagnosis.²³ Additionally, those who screened positive for PTSD and received a clinical diagnosis may present with more severe and complex cases of PTSD with poorer symptom recovery.^23,24 Finally, the Brain Health Registry has shown that self-reported diagnosis accurately captures population prevalence.²⁵

A few studies have assessed the association between anxiety and depression on AMH levels. A study of 95 health care professionals found that severity of anxiety symptoms was associated with lower AMH levels.²⁶ Previous studies evaluating the association of depression with AMH levels did not observe statistically significant associations^27,28 similar to our findings. However, we did observe a statistically significant association between self-reported clinical diagnosis of depression and AMH levels among study participants who had experienced at least one pregnancy.

Given the cross-sectional nature of our study, it is difficult to assess causality between mental health conditions and serum AMH levels. However, chronic inflammation provides a potential mechanism explaining this association. In a meta-analysis, anxiety and PTSD have been associated with chronic inflammation,²⁹ although these findings have been restricted to PTSD in some analyses. Chronic inflammation has also been associated with lower levels of AMH.³⁰ In a separate meta-analysis, individuals with PTSD have been shown to exhibit elevated levels of serum proinflammatory cytokines interleukin-1 beta (IL-1β), IL-6, tumor necrosis factor alpha, interferon gamma, and C-reactive protein (CRP) in comparison to individuals without PTSD.³¹ Moreover, poorer PTSD outcomes have been associated with elevated white blood cell count, CRP, fibrinogen, and erythrocyte sedimentation rate.³⁰ Although studies of inflammatory cytokines and serum AMH are limited, studies involving women with chronic inflammatory diseases including rheumatoid arthritis, Behçet’s disease, and spondylarthritis have indicated lower serum AMH levels compared with women without these conditions.³² Additionally, IL-1α and IL-6 levels have been observed to be elevated in women with primary ovarian insufficiency³³ and premature ovarian failure, respectively.³⁴

The strengths of this study include its focus on a population known to have a high prevalence of mental health conditions, notably PTSD.⁴ Additionally, we were able to perform sensitivity analyses to rule out potential confounding by alcohol consumption and other gynecological factors including PCOS diagnosis and gravidity. The study limitations included use of self-reported survey data to ascertain both the clinical mental health diagnoses and model covariates. We did not validate the self-reported clinical diagnosis of the mental health condition; however, these questions are used in the Behavioral Risk Factor Surveillance System conducted by the Centers for Disease Control and Prevention and have been shown to be valid in prior research compared to clinical diagnoses.^35,36 Furthermore, we were unable to account for the effect of pharmacological management of these conditions and their potential effect on AMH levels. Finally, due to the cross-sectional nature of the study, we were unable to capture the temporality of the observed associations. It is also important to recognize that the elevated prevalence of depression, anxiety, and PTSD inherent in firefighting might limit the generalizability of these findings to nonfirefighter populations. Firefighters may have a different distribution of lifestyle characteristics than the general population such as low levels of cigarette smoking and higher physical activity further limiting the generalizability to nonfirefighter populations. However, in sensitivity analyses where individuals who reported smoking were excluded (n = 10), we observed similar associations. Due to the high correlation between age and years in the fire service, as women get older, they will have higher levels of environmental/occupational exposures that may impact ovarian reserve. The study was conducted between September 2020 and September 2022; therefore, the incidence of mental health conditions may have been impacted by the COVID-19 pandemic among this group of first responders.

Further research is needed to validate these findings and evaluate the longitudinal effect that mental health conditions may have on ovarian reserve. Given the higher prevalence of anxiety, depression, and PTSD among women firefighters,^2
–4 it is imperative to advance this research in this population to identify potential mechanisms that may increase the risk of adverse health conditions and to identify areas for potential interventions. Additionally, further research should evaluate optimum treatments for PTSD in this population and whether treatment of these conditions could mitigate their adverse health effects.

Conclusions

In our study of women firefighters, those with a self-reported clinical diagnosis of anxiety or PTSD had significantly lower serum AMH levels than study participants who had not received such a diagnosis. This association was more pronounced among women between the ages of 35 and 45. Due to the high correlation between age and years in the fire service, as women get older, they will have higher levels of environmental/occupational exposures that may impact ovarian reserve. Upon exclusion of those with concurrent PTSD diagnosis, the association between anxiety and AMH no longer remained statistically significant. Among individuals with at least one pregnancy, a clinical diagnosis of depression was also associated with significantly lower AMH levels.

Footnotes

Acknowledgments

The authors would like to thank their fire service partners for their invaluable commitment and dedication to this project as well as all study participants—without them, this work would not be possible. Additionally, they would like to thank their laboratory staff for their time and effort provided in data collection and analysis.

Authors’ Contributions

M.A.V.: Conceptualization; methodology; project administration; and writing—original draft preparation. L.V.F.: Conceptualization; methodology; supervision; and writing—original draft preparation. K.H. and Y.L.: Methodology; formal analysis; and writing—original draft preparation. S.C.B.: Project administration and writing—original draft preparation. S.A.J.: Writing—review and editing. B.H.: Writing—review and editing. C.C.S.C.: Writing—review and editing. J.J.G.: Writing—review and editing. J.J.K.: Writing—review and editing. D.J.U.: Writing—review and editing. P.L.-F.: Writing—review and editing. J.B.B.: Writing—review and editing. K.L.J.: Writing—review and editing. A.J.C.-M.: Writing—review and editing. J.M.G.: Writing—review and editing. J.L.B.: Conceptualization; methodology; project administration; funding acquisition; and supervision. All authors reviewed and approved the final version, and no other person made a substantial contribution to the article.

Ethical Statement

This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of the University of Miami (#20170997). All participants provided informed consent.

Data Availability

Data used in this study are available on request from the corresponding author.

Author Disclosure Statement

The authors declare no conflicts of interest.

Funding Information

This work was supported by the Federal Emergency Management Agency (grants EMW-2015-FP-00213 and EMW-2019-FP-00526).

Supplementary Material

Supplementary Table S1

Supplementary Table S2

Supplementary Table S3

Supplementary Table S4

Supplementary Table S5

Supplementary Table S6

References

MacDermid

, Lomotan

, Hu

. Canadian career firefighters’ mental health impacts and priorities. Int J Environ Res Public Health, 2021; 18(23); doi: 10.3390/ijerph182312666

Martin

, Vujanovic

, Paulus

, et al. Alcohol use and suicidality in firefighters: Associations with depressive symptoms and posttraumatic stress. Compr Psychiatry, 2017; 74:44–52; doi: 10.1016/j.comppsych.2017.01.002

Heinrichs

, Wagner

, Schoch

, et al. Predicting posttraumatic stress symptoms from pretraumatic risk factors: A 2-year prospective follow-up study in firefighters. Am J Psychiatry, 2005; 162(12):2276–2286; doi: 10.1176/appi.ajp.162.12.2276

Noor

, Pao

, Dragomir-Davis

, et al. PTSD symptoms and suicidal ideation in US female firefighters. Occup Med (Lond), 2019; 69(8–9):577–585; doi: 10.1093/occmed/kqz057

Jahnke

, Haddock

, Jitnarin

, et al. The prevalence and health impacts of frequent work discrimination and harassment among women firefighters in the US Fire Service. Biomed Res Int, 2019; 2019:6740207; doi: 10.1155/2019/6740207

Cogendez

, Kumru

, Soysal

, et al. Evaluation of psychological distress in infertile women who underwent ART cycle during the COVID-19 pandemic. Gynecol Obstet Reprod Med, 2022; 29(1):1–9; doi: 10.21613/GORM.2022.1293

Bleil

, Adler

, Pasch

, et al. Depressive symptomatology, psychological stress, and ovarian reserve: A role for psychological factors in ovarian aging? Menopause, 2012; 19(11):1176–1185; doi: 10.1097/gme.0b013e31825540d8

Wamser-Nanney

. Trauma exposure, PTSD and indices of fertility. J Psychosom Obstet Gynaecol, 2020; 41(2):116–121; doi: 10.1080/0167482x.2019.1619691

Dobie

, Kivlahan

, Maynard

, et al. Posttraumatic stress disorder in female veterans: Association with self-reported health problems and functional impairment. Arch Intern Med, 2004; 164(4):394–400; doi: 10.1001/archinte.164.4.394

10.

Nishimi

, Thurston

, Chibnik

, et al. Posttraumatic stress disorder symptoms and timing of menopause and gynecological surgery in the Nurses’ Health Study II. J Psychosom Res, 2022; 159:110947; doi: 10.1016/j.jpsychores.2022.110947

11.

Jung

, Jahnke

, Dennis

, et al. Firefighter occupational factors and the risk of preterm birth: Results from a survey of women firefighters in the USA. Occup Environ Med, 2023; 80(2):77–85; doi: 10.1136/oemed-2022-108332

12.

Jung

, Jahnke

, Dennis

, et al. Occupational factors and miscarriages in the US fire service: A cross-sectional analysis of women firefighters. Environ Health, 2021; 20(1):116; doi: 10.1186/s12940-021-00800-4

13.

Jahnke

, Poston

WSC

, Jitnarin

, et al. Maternal and child health among female firefighters in the U.S. Matern Child Health J, 2018; 22(6):922–931; doi: 10.1007/s10995-018-2468-3

14.

Engelsman

, Banks

APW

, He

, et al. An exploratory analysis of firefighter reproduction through survey data and biomonitoring. Int J Environ Res Public Health, 2023; 20(8); doi: 10.3390/ijerph20085472

15.

Davidson

, Jahnke

, Jung

, et al. Anti-Müllerian hormone levels among female firefighters. Int J Environ Res Public Health, 2022; 19(10); doi: 10.3390/ijerph19105981

16.

di Clemente

, Racine

, Pierre

, et al. Anti-Müllerian hormone in female reproduction. Endocr Rev, 2021; 42(6):753–782; doi: 10.1210/endrev/bnab012

17.

Visser

, Schipper

, Laven

JSE

, et al. Anti-Müllerian hormone: An ovarian reserve marker in primary ovarian insufficiency. Nat Rev Endocrinol, 2012; 8(6):331–341; doi: 10.1038/nrendo.2011.224

18.

Kroenke

, Spitzer

, Williams

, et al. An ultra-brief screening scale for anxiety and depression: The PHQ-4. Psychosomatics, 2009; 50(6):613–621; doi: 10.1176/appi.psy.50.6.613

19.

Cohen

, Kamarck

, Mermelstein

. A global measure of perceived stress. Journal of Health and Social Behavior, 1983; 24(4):385–396; doi: 10.2307/2136404

20.

Lang

, Stein

. An abbreviated PTSD checklist for use as a screening instrument in primary care. Behav Res Ther, 2005; 43(5):585–594; doi: 10.1016/j.brat.2004.04.005

21.

White

, Sandler

, D’Aloisio

, et al. Antimüllerian hormone in relation to tobacco and marijuana use and sources of indoor heating/cooking. Fertil Steril, 2016; 106(3):723–730; doi: 10.1016/j.fertnstert.2016.05.015

22.

Romito

, Bove

, Romito

, et al. Ovarian reserve after chemotherapy in breast cancer: A systematic review and meta-analysis. J Pers Med, 2021; 11(8); doi: 10.3390/jpm11080704

23.

Holowka

, Marx

, Gates

, et al. PTSD diagnostic validity in Veterans Affairs electronic records of Iraq and Afghanistan veterans. J Consult Clin Psychol, 2014; 82(4):569–579; doi: 10.1037/a0036347

24.

Morgan

, Kelber

, O’Gallagher

, et al. Discrepancies in diagnostic records of military service members with self-reported PTSD: Healthcare use and longitudinal symptom outcomes. Gen Hosp Psychiatry, 2019; 58:33–38; doi: 10.1016/j.genhosppsych.2019.02.006

25.

Sordo Vieira

, Nguyen

, Nutley

, et al. Self-reporting of psychiatric illness in an online patient registry is a good indicator of the existence of psychiatric illness. J Psychiatr Res, 2022; 151:34–41; doi: 10.1016/j.jpsychires.2022.03.022

26.

Yeğin

, Desdicioğlu

, Seçen

, et al. Low Anti-Mullerian hormone levels are associated with the severity of anxiety experienced by healthcare professionals during the COVID-19 pandemic. Reprod Sci, 2022; 29(2):627–632; doi: 10.1007/s43032-021-00643-x

27.

Mínguez-Alarcón

, Hammer

, Williams

, et al. Earth Study Team. Self-reported history of comorbidities and markers of ovarian reserve among subfertile women. J Assist Reprod Genet, 2022; 39(12):2719–2728; doi: 10.1007/s10815-022-02643-5

28.

Golenbock

, Wise

, Lambert-Messerlian

, et al. Association between a history of depression and anti-müllerian hormone among late-reproductive aged women: The Harvard study of moods and cycles. Womens Midlife Health, 2020; 6(1):9; doi: 10.1186/s40695-020-00056-x

29.

Renna

, O’Toole

, Spaeth

, et al. The association between anxiety, traumatic stress, and obsessive-compulsive disorders and chronic inflammation: A systematic review and meta-analysis. Depress Anxiety, 2018; 35(11):1081–1094; doi: 10.1002/da.22790

30.

Eswarappa

, Neylan

, Whooley

, et al. Inflammation as a predictor of disease course in posttraumatic stress disorder and depression: A prospective analysis from the Mind Your Heart Study. Brain Behav Immun, 2019; 75:220–227; doi: 10.1016/j.bbi.2018.10.012

31.

Kim

, Lee

, Yoon

. Inflammation in Post-Traumatic Stress Disorder (PTSD): A review of potential correlates of PTSD with a neurological perspective. Antioxidants (Basel), 2020; 9(2); doi: 10.3390/antiox9020107

32.

Henes

, Froeschlin

, Taran

, et al. Ovarian reserve alterations in premenopausal women with chronic inflammatory rheumatic diseases: Impact of rheumatoid arthritis, Behçet’s disease and spondyloarthritis on anti-Müllerian hormone levels. Rheumatology (Oxford), 2015; 54(9):1709–1712; doi: 10.1093/rheumatology/kev124

33.

Yang

, Pang

, Miao

. Ovarian IL-1α and IL-1β levels are associated with primary ovarian insufficiency. Int J Clin Exp Pathol, 2018; 11(9):4711–4717.

34.

Sun

, Chen

, Zheng

, et al. Analysis on the level of IL-6, IL-21, AMH in patients with auto-immunity premature ovarian failure and study of correlation. Exp Ther Med, 2018; 16(4):3395–3398; doi: 10.3892/etm.2018.6592

35.

The Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. 2024. Available from: https://www.cdc.gov/brfss/ [Last accessed: September 18, 2024 ].

36.

Stuart

, Pasco

, Jacka

, et al. Comparison of self-report and structured clinical interview in the identification of depression. Compr Psychiatry, 2014; 55(4):866–869; doi: 10.1016/j.comppsych.2013.12.019

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.01 MB