In the current investigation we applied principles of life history theory and specifically the psychosocial acceleration hypothesis to the examination of variability in reproductive aging. Specifically, we proposed a model in which environmental adversity is hypothesized to promote the allocation of resources toward greater reproductive readiness via increases in the volume of growing follicles (i.e. initial recruitment) at the cost of hastening the depletion of the ovarian reserve over time. In a large, community-based sample of 979 premenopausal women ages 25–45, we evaluated this model by examining the interaction of age and self-reported perceptions of stress in relation to AFC, a subset of growing follicles rescued during cyclic recruitment, that represent proportionately the finite pool of primordial follicles (i.e. ovarian reserve). Results showed that the rate of AFC loss across women was higher among women reporting more psychological stress. Associations were independent of statistical adjustment for potential confounding factors, including race/ethnicity, SES, menarcheal age, use of hormone-containing medication for BC, parity, cigarette smoking, BMI and WHR. When the sample distribution of psychological stress was divided into tertiles, women in the low-stress group showed an average follicle loss of −0.781/year, women in the mid-stress group showed an average follicle loss of −0.842/year and women in the high-stress group showed an average follicle loss of −0.994/year. Calculation of the x-intercepts, ages at which AFC is estimated to be zero (or projected ages of menopausal onset), were 54.3, 52.3 and 51.1 for the low-, mid- and high-stress groups, respectively.
In addition to the significant interaction showing psychological stress was related to a higher rate of AFC loss across women, a significant main effect was found in which greater psychological stress was related to higher AFC. Follow-up tests at pre-specified ages showed that the main effect of stress on AFC was present in the younger (ages 25, 30 and 35) but not in the older (ages 40 and 45) women. That psychological stress was related both to higher AFC among younger women and a higher rate of AFC loss across women suggests that fertility may initially be potentiated among women experiencing greater stress. Increased reproductive readiness (marked by higher AFC) among women experiencing greater stress appears to occur at the cost of having a higher rate of AFC loss. Although speculative, this may ultimately lead to earlier onset menopause and increased risk for diseases of aging that increase in prevalence post-menopausally. This apparent trade-off is consistent with life history theory and specifically the psychosocial acceleration hypothesis which proposes that environmental adversity biases women toward enhanced biological readiness for current versus longer term reproduction even at the cost of longer term consequences.
Although the current findings are consistent with the psychosocial acceleration hypothesis, a model derived from life history theory that has garnered a substantial amount of empirical support in its previous application to the investigation of pubertal timing (Moffitt et al., 1992
; Wierson et al., 1993
; Campbell and Udry, 1995
; Graber et al., 1995
; Ellis et al., 1999
, Ellis and Garber, 2000
; Ellis and Essex, 2007
; Belsky et al., 2007
; Saxbe and Repetti, 2009
), it is notable that alternate life history models would suggest that environmental adversity may actually suppress reproductive functioning based on evidence supporting the hypothesis that reproduction is forestalled when environmental conditions threaten the viability of the offspring (Wasser and Barash, 1983
; Ellison 1990
). This model is additionally supported by studies of Cynomolgus monkeys in which socially subordinate females (presumed to experience psychological stress) displayed suppressed ovarian function (Kaplan et al., 1996
, Kaplan and Manuck, 2004)
, as well as by studies showing stress-related activation of the hypothalamic–pituitary axis inhibits certain reproductive functions both in animal models and among women (Rivier and Rivest, 1991
; Kalantaridou et al., 2004
). Although speculative, it is possible that whether fertility, operationalized here as being reflected by the number of antral follicles, is enhanced or suppressed may depend on the severity of the adversity exposure. Because only normative variability in psychological stress was examined in the current investigation, it remains unclear whether the experience of more significant stressors or even traumatic events (that are more likely to threaten offspring viability) might have a different impact on folliculogenesis and reproductive aging. Future work is required to reconcile the current findings with previous evidence suggesting adversity exposures may have a suppressive effect on ovarian function and whether these conflicting outcomes may be explained by the severity of stress experienced.
Two primary weaknesses of the current study were that the analysis was cross-sectional and that the evaluation of psychological stress was limited to the self-report of stress measured over the past month. First, the cross-sectional nature of the current analysis limits the conclusions that can be drawn regarding whether stress relates to rates of AFC loss over time. Rather, the current findings demonstrate an association between stress and AFC loss as examined across women. A longitudinal component of the OVA study is currently underway in which women are returning for a 3-year follow-up visit at which time the study measures, including AFC, are repeated. In the future when these data become available, it will be possible to examine stress in relation to the actual change in the AFC over this period. Secondly, the measurement of psychological stress was limited to the use of the PSS which assesses appraisals of stress over the past month only. Although the measurement of a psychological state or transient appraisal of stress would not be expected to relate to a biological process that unfolds over many years, PSS scores are likely to reflect stress experienced more chronically as is suggested by previous investigations relating PSS to longer term health outcomes and biological processes relevant to disease development (Cohen et al., 1999
; Ng and Jeffery, 2003
; Epel et al., 2004
; McDade et al., 2006
; Fredman et al., 2010
; Yu et al., 2010
) and/or to reflect a trait-like dimension (Federenko et al., 2006
; Conard and Matthews, 2008
; Bogdan and Pizzagalli, 2009
; Ebstrup et al., 2011
) which may increase one's proneness to experience stress.
Strengths of the current investigation were its novel focus on reproductive aging during the premenopausal period; use of a well-validated marker of total follicular reserve and recruitment of a sample that is large in size, healthy and community based. To date, while the premenopausal period is well recognized as an important area of study for improved understanding of variability in fertility as well as the emergence of pre-clinical risk factors for disease development post-menopausally (Kaplan and Manuck, 2004
; Kaplan and Manuck, 2008
), there has been a paucity of research examining the mechanisms underlying the loss of ovarian function during this period. In this respect, the contribution of the current study is unique and timely. The current study also possesses several methodological strengths through its use of AFC, a well-validated marker reflecting reproductive aging on a continuum versus traditional staging methods which provide only categorical determinations of reproductive age (pre-, peri-, post-menopause) on the basis of indicators (e.g. hormones) that can be unreliable or unchanged until the function of the ovary is severely compromised. Lastly, the current sample is large, well-characterized in terms of reproductive history and general health, and includes women screened to be regularly cycling and free from major medical illnesses, maximizing the generalizability of the current results.
To address the limitations of the current study, future investigations should be designed to assess AFC over time, to measure objective as well as subjective aspects of the environment that may contribute to the psychological well-being of women and to characterize features of early environments that might be particularly salient in setting the stage for reproductive development and senescence. Future studies should also attempt to elucidate the biological mechanisms, primarily the hormonal regulation of follicle growth and development, by which such apparent stress effects on the body occur. Lastly, the current analysis was limited in its focus on reproductive aging without specific consideration of how the current findings may relate to other relevant literatures, namely life history models of pubertal timing (Ellis, 2004
). Although findings generally suggest that pubertal and menopausal timing are unrelated (Treloar, 1974
; Snieder et al., 1998
), whether variation in pubertal timing may underlie aspects of ovarian function that are relevant to folliculogenesis and trajectories of reproductive aging is an important question for future research. Moreover, future research should be guided by the over-arching goal of developing better integrated models reflecting the inter-relation of major reproductive processes and events over a woman's life course.
In conclusion, results from the current investigation show that psychological stress was related to higher AFC among younger women and to a higher rate of AFC decline across women. This finding provides preliminary support for the proposed model of reproductive aging which suggests that environmental adversity may enhance the allocation of resources toward greater reproductive readiness via increases in initial recruitment of dormant primordial follicles into the growing pool of follicles, thereby enhancing fertility in the short term at the cost of hastening the depletion of the ovarian reserve in the long term.