Menopausal status, transition, and age at menopause with accelerated biological aging across multiple organ systems: findings from two cohort studies

Abstract

Background

Biological aging is a heterogeneous process that varies across organs and systems. The dynamic hormonal changes during the menopausal transition may have profound and organ-specific impacts on biological aging. However, the relationship between the menopausal transition and both comprehensive and organ-specific biological aging remains poorly understood. This study aimed to investigate the associations between menopausal factors and both comprehensive and organ-specific biological aging, as well as the modifying role of reproductive history.

Methods

This study included 37,244 women from the China Multi-Ethnic Cohort (CMEC) and 140,479 from the UK Biobank (UKB). Menopausal factors included menopausal status, menopausal transition, and age at menopause. Comprehensive and organ-specific biological ages (BAs) were calculated using the Klemera-Doubal method and clinical biomarkers and have been shown to predict age-related health outcomes. Multiple linear regression and change-to-change models were applied, with stratified analyses based on reproductive history.

Results

Compared with pre-menopausal women, those who were peri- or post-menopausal or had undergone hysterectomy or oophorectomy exhibited greater acceleration in comprehensive, liver, metabolic, and kidney BA. In longitudinal change-to-change models, women undergoing menopausal transition showed greater increases in comprehensive BA (CMEC: β = 1.33, 95% CI = 0.89, 1.76; UKB: β = 2.60, 95% CI = 1.91, 3.30), as well as liver, metabolic, and kidney BAs compared to those remaining pre-menopausal. Earlier age at menopause was associated with accelerated comprehensive BA in UKB (< 40 years: β = 0.69, 95% CI = 0.39, 0.98; 40–44 years: β = 0.24, 95% CI = 0.09, 0.40). Across organ-specific BAs, liver BA showed the strongest associations with menopausal factors. Reproductive history like age at live birth and number of live births emerged as potential modifiers of these associations.

Conclusions

Menopause, particularly the menopausal transition, was associated with accelerated comprehensive and organ-specific biological aging, with liver aging being most affected. These findings underscore the menopausal transition as a critical window for interventions to enhance women’s health and longevity.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12916-025-04223-7.

Background

Menopause is a natural biological transition that marks the end of reproductive capacity in women, characterized by the permanent cessation of the menstruation and the decline in ovarian function [1]. Beyond its reproductive implications, menopause has been associated with health declines across multiple organ systems, including increased risks for cardiovascular [2, 3], cognitive [4, 5], and musculoskeletal issues [6, 7]. Epidemiological studies have suggested that early menopause was associated with elevated risks of age-related morbidity and mortality [4, 8–10]. These findings have led to a hypothesis that menopause may accelerate the biological aging process [11], potentially reshaping the trajectory of women’s health throughout their extended post-menopause years.

Biological aging is a complex process that involves changes at multiple biological levels, from the molecular and cellular to the systemic [12]. While chronological age (CA) increases uniformly, biological aging can vary substantially between individuals and across life stages [13]. For women, the potential effects of menopause on biological aging suggested variations in biological aging rate at different reproductive stages. Several cross-sectional studies have explored the associations between age at menopause and biological aging markers such as epigenetic age [14], telomere length [15], frailty [16], and composite biomarkers [17, 18]. While these studies offered valuable insights, they capture only a limited aspect of the broader relationship between reproductive and biological aging. Menopause is not merely an endpoint but a dynamic, multi-year process, typically lasting 2–8 years [2], marked by dramatic hormonal changes that may trigger a cascade of biological effects [1]. Studying the menopausal transition period is crucial for understanding the mechanism through which menopause contributes to accelerated biological aging, as it captures dynamic alterations during this critical phase, providing insights that a static measure like age at menopause alone cannot reveal.

Emerging evidence indicated that biological aging exhibits distinct rates and patterns across various organ systems [19, 20]. Menopause and its associated hormonal changes influence not only reproductive functions but also a wide range of physiological systems, potentially accelerating aging in certain organs. Examining the effects of menopause on organ-specific biological aging may help identify systems that are particularly vulnerable to menopausal changes. However, studies linking menopause to organ-specific biological aging remain limited. Additionally, reproductive history factors beyond menopause, such as early menarche, have been associated with long-term health outcomes and the aging process [17, 21–23]. Despite these findings, whether the effects of menopause on biological aging vary by reproductive history remains unclear.

In this study, we aimed to investigate how biological aging differs across reproductive stages, changes during menopausal transition, and is influenced by the timing of menopause. Thus, we focused on three key menopausal factors: menopausal status, menopausal transition, and the age at menopause. Given the potential heterogeneity in biological aging, we also investigated how menopause influences biological aging across various organ systems and whether reproductive history modifies these relationships. Using data from the China Multi-Ethnic Cohort (CMEC) Study [24] and the UK Biobank (UKB) [25], we employed the widely validated Klemera-Doubal method (KDM) [26] to calculate comprehensive and organ-specific biological ages (BAs). These BA measures serve as robust predictors of aging-related morbidity and mortality [27–29], thereby facilitating a systematic examination of menopause in relation to biological aging.

Methods

Study population

The CMEC is an ongoing prospective cohort study initiated in May 2018 across community populations in five provinces of Southwest China [24]. A total of 99,556 participants (n = 59,762 females) aged 30–79 years were recruited, given full consideration of ethnic characteristics, socioeconomic status, population size, and disease patterns. The first follow-up survey, conducted from August 2020 to July 2021, included approximately 10% of participants from the baseline survey. Data collection involved face-to-face interviews using electronic questionnaires, medical examinations, and clinical laboratory tests. All participants provided written informed consent, and the study received ethical approval from the Sichuan University Medical Ethical Review Board as well as local ethics committees at participating sites.

The UKB is a large-scale prospective study that recruited over 500,000 participants (n = 273,316 females) aged 37–73 from 22 assessment centers across the UK between 2006 and 2010 [25]. The first follow-up survey, conducted from 2012 to 2013, included a subset of 20,343 participants based at the Cheadle assessment center. At both surveys, comprehensive data were collected through touchscreen questionnaires, physical measurements, and biological samples. All participants provided electronic informed consent, and the study was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Multi-Centre Research Ethics Committee.

This study included female participants aged 40–65, with available data on menopausal status and biological aging at baseline. Participants were excluded if they were pregnant, reported an extreme age at menopause (< 30 or > 65 years), or had been diagnosed with breast, cervical, or ovarian cancer. The final analytic sample comprised 37,244 participants from the CMEC and 140,479 participants from the UKB (cross-sectional sample). For analyses examining menopausal transition, which required both baseline and follow-up data (longitudinal sample), we further excluded participants without follow-up information on menopausal status and biological aging, as well as those who had undergone oophorectomy or hysterectomy, resulting in 3441 participants from the CMEC and 1826 participants from the UKB (Fig. 1).

Fig. 1.

Flowchart of the study design

Assessment of menopausal factors and other reproductive history

Menopausal factors

In CMEC, participants were asked about their current menstrual status with the options: regular menstruation, irregular menstruation, peri-menopausal, and post-menopausal (cessation of menstruation for ≥ 12 months). Participants reporting post-menopausal status were further asked to provide their age at menopause. Information on surgical history, including unilateral or bilateral oophorectomy and hysterectomy, was also collected. In UKB, participants were asked, “Have you had your menopause (periods stopped)?” with options: yes, no, not sure—had a hysterectomy, not sure—other reason, and prefer not to answer. Data on age at menopause and any history of bilateral oophorectomy or hysterectomy were also obtained.

Using these data, menopausal status was classified into pre-menopause, peri-menopause (CMEC only), post-menopause, oophorectomy, and hysterectomy. In CMEC, pre-menopause was defined as regular or irregular menstruation; in UKB, it was defined as a “no” response to the menopause question. Participants with missing or uncertain menopausal status (e.g., “not sure—other reason” or “prefer not to answer”) were excluded from this study. For those who had experienced natural menopause, age at menopause was classified as < 40, 40–44, 45–49, 50–54, and ≥ 55 years [4, 16]. For analyses examining changes in menopausal status from baseline to follow-up, the CMEC menopausal categories were aligned with those in the UKB by reclassifying status as pre-menopause (including pre-menopause and peri-menopause) and post-menopause. This allowed us to define three transition groups: pre-menopause to pre-menopause (pre-pre), pre-menopause to post-menopause (menopausal transition), and post-menopause to post-menopause (post-post). Additionally, a more detailed classification was conducted for the CMEC, reconsidering peri-menopause as a distinct category, resulting in six transition groups: pre-pre, pre-peri, peri-peri, peri-post, pre-post, and post-post.

Other reproductive history

Other reproductive history factors included age at menarche, age at first live birth, age at last live birth, number of live births, history of miscarriage (including stillbirths, spontaneous miscarriages, and terminations), ever taken oral contraceptive pill, and ever used hormone-replacement therapy (HRT, UKB only).

Assessment of biological aging

In this study, we constructed comprehensive and organ-specific BAs using clinical biomarkers and anthropometric data, based on the Klemera and Doubal method (KDM). KDM is a widely used method for constructing composite biomarker BA and has been demonstrated good performance in predicting age-related health outcomes in both Chinese and UK populations [27–29].

Biomarker selection

First, candidate biomarkers were selected based on their role in the aging process, previous usage in related literature, and their missingness in the datasets. Next, we retained only biomarkers with a correlation coefficient of |r|> 0.1 with CA. To minimize redundancy, we excluded biomarkers that potentially reflected similar aspects of aging based on current knowledge and observed inter-correlations among biomarkers.

This process yielded 15 biomarkers for comprehensive BA in the CMEC, which included systolic blood pressure (SBP), waist-to-hip ratio (WHR), peak expiratory flow (PEF), low-density lipoprotein cholesterol (LDL-c), high-density lipoprotein cholesterol (HDL-c), glycated hemoglobin (HBA1C), triglyceride (TG), aspartate aminotransferase (AST), gamma-glutamyl transpeptidase (GGT), albumin (ALB), alkaline phosphatase (ALP), creatinine, urea, mean corpuscular volume (MCV), and platelet count (PLT). In the UKB, 18 biomarkers were selected, including SBP, WHR, body fat percentage (BFP), forced expiratory volume in 1 s (FEV1), MCV, ALP, AST, C-reactive protein, cystatin C, GGT, heel quantitative ultrasound index, HBA1C, insulin-like growth factor 1 (IGF-1), TG, urate, urea, ALB, and vitamin D.

These indicators were then categorized into four systems based on the organ or system function they represent: cardiopulmonary, metabolic, liver, and kidney systems. Cardiopulmonary BA included SBP and PEF/FEV1; metabolic BA included LDL-c, HDL-c, HBA1C, TG, and WHR for CMEC (and WHR, BFP, HBA1C, and TG for UKB); liver BA included AST, GGT, ALP, ALB, and IGF-1 (UKB only); the kidney BA included creatinine, urea, and cystatin C (UKB only).

Construction of comprehensive and organ-specific BA acceleration

BA was constructed by extracting information from the relationships between each biomarker and the aging process. Briefly, CA was regressed on m selected biomarkers in a reference population, allowing for the calculation of estimated BA as follows:

$${\mathit{BA}}_{\mathit{EC}}=\frac{{\sum }_{j=1}^m(x_j-q_j)\frac{k_j}{s_j^2}+\frac{\mathit{CA}}{s_{\mathit{BA}}^2}}{{\sum }_{j=1}^m{\left(\frac{k_j}{s_j}\right)}^2+\frac{1}{s_{\mathit{BA}}^2}}$$

Here, ${\mathit{BA}}_{\mathit{EC}}$ represents the estimated biological age. $x_j$ is the measured value of the jth biomarker, and $k_j$, $q_j$, and $s_j$ correspond to the slope, intercept, and root mean squared error from the jth biomarker’s regression on CA, respectively. $s_{\mathit{BA}}^2$ represents the estimated variance in CA explained by the selected biomarker set. Since BA alone cannot fully capture individual differences in biological aging, BA acceleration was derived by subtracting CA from BA. A positive value of BA acceleration indicates advanced biological aging, and vice versa. All analyses were performed using BA acceleration. More details on the construction and validation of BAs can be found in our previous study [29, 30]. In our study population, the median comprehensive BA at baseline was 51.83 (44.81, 58.69) years in CMEC and 55.77 (47.95, 62.10) years in UKB. Detailed distributions of organ-specific BAs and their component biomarkers are presented in Additional file 1: Table S1.

Assessment of covariates

Covariate information was mainly obtained through questionnaires, including sociodemographic characteristics, lifestyle and behavior factors, self-reported diseases, and reproductive history. Referring to previous studies [17, 23], in the final models we adjusted for age, ethnicity, education level, household income (CMEC only), Townsend deprivation index (UKB only), marital status (CMEC only), employment status, dietary score, total energy intake (CMEC only), smoking status, alcohol consumption, tea consumption, physical activity, body mass index (BMI), insomnia, depressive symptoms, anxiety symptoms, self-reported chronic diseases (cardiovascular disease, diabetes, cancer, and chronic obstructive pulmonary disease), age at menarche, number of live births, ever taken oral contraceptive pill, and ever used HRT (UKB only). Detailed description of covariates can be found in Additional file 1: Supplementary methods [31, 32].

Statistical analysis

Baseline characteristics

We described the baseline characteristics of study participants by menopausal status for the cross-sectional sample and by changes in menopausal status between baseline and follow-up for the longitudinal sample. Continuous variables were described as median (25th, 75th percentile), while categorical variables were presented as count (percentage). To assess the representativeness of the longitudinal sample, we compared baseline characteristics of women who completed the follow-up survey with those who did not in each cohort.

Analysis of three menopausal factors with comprehensive and organ-specific BA acceleration

To explore biological aging across distinct reproductive stages, we first examined the associations between menopausal status and BA acceleration using baseline data, analyzing both comprehensive and organ-specific BAs. We used multiple linear regression models, with pre-menopause as the reference group, and adjusted for baseline covariates mentioned above.

Using follow-up data, we further examined the relationship between changes in menopausal status (pre-pre, menopausal transition, and post-post) and changes in both comprehensive and organ-specific BA acceleration. We applied a change-to-change analysis with linear regression models, using the pre-pre group as the reference. Change-to-change analysis could reduce unmeasured time-invariant confounding by using only within-individual information. Models were adjusted for baseline time-invariant covariates, including demographics, reproductive history, and self-reported chronic diseases, as well as baseline and concurrent changes of time-variant covariates such as age, lifestyle factors, and mental status. Details of change-to-change analysis and covariate adjustment can be found in Additional file 1: Supplementary methods [33–36].

For women who had experienced natural menopause, we assessed the associations between age at menopause and BA acceleration. Age at menopause was analyzed as both a continuous and a categorical variable (< 40, 40–44, 45–49, 50–54, and ≥ 55 years, with the 45–49 age group as the reference [4, 37]). Multiple linear regression models were used, adjusting for the same baseline covariates.

Several sensitivity analyses were also performed to ensure the robustness of our findings. First, we reanalyzed the data after excluding females with self-reported chronic diseases, which included cardiovascular disease, diabetes, cancer, and chronic obstructive pulmonary disease. Second, missing covariate data were handled using multiple imputation via chained equations, and the analyses were repeated. Third, analyses were repeated with further adjustment for medication use, including anti-hypertensive and anti-diabetic medications (available in both cohorts), and lipid-lowering medications (available in UKB only). Lastly, we calculated E-values to assess the potential influence of unmeasured confounding [38].

Subgroup analysis by reproductive history and socioeconomic factors

To assess potential modifiers of the association between menopausal factors and comprehensive BA acceleration, we performed subgroup analyses primarily focused on reproductive history variables. These included age at menarche (< 15 or ≥ 15 years), age at first live birth, age at last live birth, number of live births (0–1 or ≥ 2), ever taken oral contraceptive pill (yes or no), and history of miscarriage (yes or no). Given the differences in childbearing age between the two cohorts, age at first live birth was grouped as < 23 or ≥ 23 years in CMEC and < 26 or ≥ 26 years in UKB, while age at last live birth was classified as < 26 or ≥ 26 years in CMEC and < 31 and ≥ 31 years in UKB. Additionally, subgroup analyses were conducted based on socioeconomic factors, including education and income level. Heterogeneity across groups was assessed using Cochran’s Q test.

All statistical analyses were conducted using R version 4.3.2. A two-sided P < 0.05 was considered statistically significant. The study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Additional file 2).

Results

Baseline characteristics of study population

Table 1 summarizes the baseline characteristics of study participants from CMEC and UKB based on change in menopausal status. Among 3441 participants aged 40–65 from CMEC, the median age was 50.61 (45.95, 55.77) years. During a median follow-up of 2.01 (1.81, 2.23) years, 684 (19.9%) women transitioned from pre-menopause to post-menopause. Among 1826 participants aged 40–65 from UKB, the median age was 57.17 (50.77, 61.42) years. During a median follow-up of 4.50 (3.92, 5.00) years, 223 (12.2%) women transitioned from pre-menopause to post-menopause. In both cohorts, compared with pre-menopausal women, those undergoing the menopausal transition exhibited higher increases in comprehensive BA acceleration as well as in the BA acceleration of most organ systems between the baseline and follow-up (Additional file 1: Fig. S1). Descriptive results for cross-sectional sample are presented in Additional file 1: Tables S2 and S3. Additionally, women who completed the follow-up survey tended to have higher economic and educational levels (Additional file 1: Table S4).

Table 1.

Baseline characteristics of study populations according to change in menopausal status

Characteristica	CMEC				UKB
	Overall	Pre-pre	Menopausal transition	Post-post	Overall	Pre-pre	Menopausal transition	Post-post
No. of participants	3441	1107	684	1650	1826	292	223	1311
BA acceleration (years)
Comprehensive BA	0.24 (−3.48, 3.71)	−1.34 (−5.02, 1.92)	0.05 (−3.83, 3.37)	1.47 (−2.04, 4.77)	−1.96 (−5.39, 2.05)	−3.57 (−7.22, 1.07)	−3.61 (−7.35, 0.97)	−1.48 (−4.71, 2.37)
Cardiopulmonary BA	−0.09 (−0.96, 0.82)	−0.06 (−0.92, 0.83)	−0.16 (−1.01, 0.81)	−0.08 (−0.96, 0.84)	−0.17 (−0.64, 0.34)	−0.19 (−0.59, 0.31)	−0.09 (−0.59, 0.48)	−0.17 (−0.68, 0.33)
Liver BA	−0.15 (−7.43, 6.95)	−4.31 (−11.30, 3.21)	−0.81 (−7.92, 6.61)	2.55 (−4.13, 9.18)	−0.46 (−5.26, 4.19)	−2.75 (−9.27, 2.26)	−3.53 (−8.15, 1.20)	0.35 (−4.06, 5.22)
Metabolic BA	0.55 (−3.57, 4.49)	−0.28 (−4.50, 3.45)	−0.16 (−3.98, 3.97)	1.52 (−2.56, 5.48)	−2.58 (−8.04, 3.69)	−3.46 (−8.61, 3.06)	−4.37 (−10.54, 2.61)	−2.04 (−7.33, 4.06)
Kidney BA	0.30 (−2.63, 3.00)	−0.17 (−3.24, 2.39)	0.35 (−2.47, 2.95)	0.59 (−2.34, 3.37)	−0.73 (−4.20, 2.61)	−0.96 (−4.49, 1.96)	−2.21 (−6.20, 1.25)	−0.39 (−3.90, 2.93)
Age (years)	50.61 (45.95, 55.77)	44.48 (42.29, 46.73)	49.40 (46.99, 51.23)	55.86 (53.06, 60.89)	57.17 (50.77, 61.42)	44.50 (42.50, 46.83)	50.17 (47.92, 52.50)	59.75 (56.33, 62.33)
Ethnicity, majorityb	1857 (54.0)	633 (57.2)	358 (52.3)	866 (52.5)	1781 (97.5)	282 (96.6)	214 (96.0)	1285 (98.0)
Educationc
Less than high school	2887 (83.9)	862 (77.9)	594 (86.8)	1431 (86.7)	120 (6.6)	4 (1.4)	4 (1.8)	112 (8.5)
High school or equivalent	331 (9.6)	121 (10.9)	47 (6.9)	163 (9.9)	901 (49.3)	135 (46.2)	102 (45.7)	664 (50.6)
College or above	222 (6.5)	124 (11.2)	43 (6.3)	55 (3.3)	798 (43.7)	153 (52.4)	117 (52.5)	528 (40.3)
Townsend deprivation indexd	–	–	–	–	−2.69 (−3.99, −0.74)	−2.36 (−3.71, −0.06)	−2.24 (−3.66, 0.09)	−2.87 (−4.04, −0.95)
Household income
<￥12,000	560 (16.3)	120 (10.8)	124 (18.1)	316 (19.2)	–	–	–	–
￥12,000–19,999	681 (19.8)	182 (16.4)	151 (22.1)	348 (21.1)	–	–	–	–
￥20,000–59,999	1344 (39.1)	471 (42.5)	243 (35.5)	630 (38.2)	–	–	–	–
￥60,000–99,999	465 (13.5)	175 (15.8)	96 (14.0)	194 (11.8)	–	–	–	–
￥100,000–199,999	319 (9.3)	124 (11.2)	59 (8.6)	136 (8.2)	–	–	–	–
≥￥200,000	70 (2.0)	34 (3.1)	11 (1.6)	25 (1.5)	–	–	–	–
Employment status
Employed	3017 (87.7)	1049 (94.8)	623 (91.1)	1345 (81.5)	1197 (65.6)	289 (99.0)	211 (94.6)	697 (53.2)
Unemployed	123 (3.6)	46 (4.2)	27 (3.9)	50 (3.0)	4 (0.2)	0 (0.0)	0 (0.0)	4 (0.3)
Retired	299 (8.7)	10 (0.9)	34 (5.0)	255 (15.5)	607 (33.2)	1 (0.3)	7 (3.1)	599 (45.7)
Married	3072 (89.3)	1017 (91.9)	624 (91.2)	1431 (86.7)	–	–	–	–
Current smoking	28 (0.8)	17 (1.5)	5 (0.7)	6 (0.4)	86 (4.7)	19 (6.5)	16 (7.2)	51 (3.9)
Current drinking	901 (26.2)	326 (29.4)	198 (28.9)	377 (22.8)	1718 (94.1)	278 (95.2)	216 (96.9)	1224 (93.4)
Current tea drinking	605 (17.6)	205 (18.5)	130 (19.0)	270 (16.4)	1540 (84.3)	234 (80.1)	185 (83.0)	1121 (85.5)
Healthy diete	1832 (53.2)	640 (57.8)	368 (53.8)	824 (49.9)	742 (40.6)	103 (35.3)	83 (37.2)	556 (42.4)
Total energy intake (kcal/d)	1680.63 (1330.43, 2092.78)	1697.29 (1344.46, 2093.28)	1663.14 (1299.51, 2144.17)	1676.38 (1329.24, 2064.92)	–	–	–	–
Physical activity (MET-h/wk)	170.33 (92.98, 274.08)	189.65 (112.02, 286.81)	191.78 (102.58, 286.07)	148.90 (78.40, 257.80)	–	–	–	–
Physical activity (d/wk)f
0	–	–	–	–	172 (9.4)	35 (12.0)	22 (9.9)	115 (8.8)
1–2	–	–	–	–	286 (15.7)	38 (13.0)	29 (13.0)	219 (16.7)
3–4	–	–	–	–	360 (19.7)	55 (18.8)	38 (17.0)	267 (20.4)
≥5	–	–	–	–	966 (52.9)	157 (53.8)	128 (57.4)	681 (51.9)
Insomnia symptom	1579 (45.9)	404 (36.5)	290 (42.4)	888 (53.6)	527 (28.9)	52 (17.8)	50 (22.4)	425 (32.4)
Depressive symptom	188 (5.5)	48 (4.3)	39 (5.7)	101 (6.1)	61 (3.3)	12 (4.1)	7 (3.1)	42 (3.2)
Anxiety symptom	254 (7.4)	65 (5.9)	55 (8.0)	134 (8.1)	64 (3.5)	7 (2.4)	6 (2.7)	51 (3.9)
BMI (kg/m2)	23.97 (21.84, 26.62)	23.83 (21.92, 26.18)	23.94 (21.76, 26.31)	24.06 (21.83, 27.06)	25.00 (22.78, 28.35)	24.00 (22.12, 26.99)	24.27 (22.27, 27.44)	25.27 (23.03, 28.65)
Self-reported diseases
Cancer	7 (0.2)	1 (0.1)	1 (0.1)	5 (0.3)	68 (3.7)	4 (1.4)	2 (0.9)	62 (4.7)
CVD	596 (17.3)	95 (8.6)	92 (13.5)	409 (24.8)	332 (18.2)	24 (8.2)	24 (10.8)	284 (21.7)
Diabetes	114 (3.3)	19 (1.7)	14 (2.0)	81 (4.9)	36 (2.0)	6 (2.1)	4 (1.8)	26 (2.0)
COPD	179 (5.2)	37 (3.3)	30 (4.4)	112 (6.8)	9 (0.5)	0 (0.0)	0 (0.0)	9 (0.7)
Age at menarche (years)	15.00 (13.00, 16.00)	14.00 (13.00, 15.00)	14.00 (13.00, 16.00)	15.00 (14.00, 17.00)	13.00 (12.00, 14.00)	13.00 (12.00, 14.00)	13.00 (12.00, 14.00)	13.00 (12.00, 14.00)
Age at first live birth (years)	23.00 (21.00, 25.00)	23.00 (21.00, 25.00)	23.00 (21.00, 25.00)	23.00 (21.00, 25.00)	26.00 (23.00, 29.00)	27.00 (24.00, 30.00)	27.00 (24.00, 30.00)	26.00 (23.00, 29.00)
Age at last live birth (years)	26.00 (24.00, 29.00)	27.00 (24.00, 30.00)	26.00 (24.00, 29.00)	26.00 (24.00, 29.00)	31.00 (28.00, 34.00)	32.00 (29.00, 35.00)	32.00 (28.00, 35.00)	31.00 (27.00, 34.00)
Number of live births	2.00 (1.00, 2.00)	2.00 (1.00, 2.00)	2.00 (1.00, 2.00)	2.00 (1.00, 2.00)	2.00 (1.00, 2.00)	2.00 (0.00, 2.00)	2.00 (0.00, 2.00)	2.00 (1.00, 2.00)
History of miscarriageg	2052 (59.6)	686 (62.0)	431 (63.0)	935 (56.7)	533 (29.2)	74 (25.3)	72 (32.3)	387 (29.5)
Ever taken oral contraceptive pill	516 (15.0)	205 (18.5)	116 (17.0)	195 (11.8)	1542 (84.4)	258 (88.4)	197 (88.3)	1087 (82.9)
Ever used hormone-replacement therapy	–	–	–	–	542 (29.7)	3 (1.0)	13 (5.8)	526 (40.1)

Abbreviations CMEC China Multi-Ethnic Cohort, UKB UK Biobank, BA biological age, METs-h/wk hours of metabolic equivalent tasks per week, BMI body mass index, CVD cardiovascular disease, COPD chronic obstructive pulmonary disease

^aData are presented as median (25th, 75th percentile) for continuous variables and count (percentage) for categorical variables. The numbers of missing covariates in CMEC/UKB were as follows: ethnicity (0/5), education (1/7), Townsend deprivation index (2 in UKB), household income (2 in CMEC), employment status (2/18), marital status (1 in CMEC), alcohol consumption (1/2), tea consumption (2/2), healthy diet (10/0), total energy intake (10 in CMEC), physical activity (16/42), insomnia symptom (11/0), depressive symptom (11/98), anxiety symptom (11/46), BMI (8/0), cancer (0/4), CVD (0/2), age at menarche (1/37), age at first live birth (57/625), age at last live birth (57/625), number of live births (15/2), history of miscarriage (36/28), ever taken oral contraceptive pill (0/3), and ever used hormone-replacement therapy (3 in UKB)

^bMajority denoted Han Chinese in CMEC and White in UKB

^cIn UKB, education level was defined according to education qualifications: college or above (college or university degree); high school or equivalent (A levels, AS levels, or equivalent; O levels, GCSEs, or equivalent; CSEs or equivalent; NVQ, HND, HNC, or equivalent; other professional qualifications); less than high school (none of the above)

^dTownsend deprivation index was an area level variable of socioeconomic status; a higher number of TDI denotes lower area level SES

^eHealthy diet denoted the top 1/2 of healthy diet scores in the two cohorts. In CMEC, a Dietary Approaches to Stop Hypertension (DASH) score was used. In UKB, healthy diet score was calculated based on consumption of 7 dietary components (fruits, vegetables, fish, processed meats, unprocessed red meats, whole grains, and refined grains)

^fIn UKB, due to the high rate of missing data for physical activity measured by metabolic equivalent tasks, we defined physical activity as the number of days per week participants engaged in moderate or vigorous activity for more than 10 min per day

^gHistory of miscarriage included stillbirths, spontaneous miscarriages, and terminations

Associations of three menopausal factors with comprehensive and organ-specific BA acceleration

Menopausal status and BA acceleration

As shown in Table 2, in both cohorts, post-menopausal women and those who had undergone hysterectomy or oophorectomy exhibited greater comprehensive BA acceleration compared to pre-menopausal women. The most pronounced associations with comprehensive BA were found in the post-menopausal group in CMEC (β = 3.30, 95% CI = 3.14, 3.45) and the oophorectomy group in UKB (β = 3.14, 95% CI = 3.00, 3.28). Additionally, in CMEC, peri-menopausal women also exhibited increased BA acceleration, with an estimate of 1.87 (1.69, 2.05).

Table 2.

Associations between menopausal status and BA acceleration

BA acceleration	Menopausal status	CMEC		UKB
		n	β (95% CI)	n	β (95% CI)
Comprehensive BA	Pre-menopause (ref)	14,398	Ref	37,285	Ref
	Peri-menopause	3408	1.87 (1.69, 2.05)	–	–
	Post-menopause	17,108	3.30 (3.14, 3.45)	62,108	2.76 (2.67, 2.86)
	Hysterectomy	967	2.58 (2.26, 2.90)	11,149	2.73 (2.60, 2.85)
	Oophorectomy	795	1.91 (1.57, 2.25)	7841	3.14 (3.00, 3.28)
Cardiopulmonary BA	Pre-menopause (ref)	14,398	Ref	37,285	Ref
	Peri-menopause	3408	0.04 (−0.01, 0.09)	–	–
	Post-menopause	17,108	−0.03 (−0.08, 0.01)	62,108	−0.02 (−0.04, −0.01)
	Hysterectomy	967	−0.05 (−0.13, 0.03)	11,149	−0.04 (−0.06, −0.02)
	Oophorectomy	795	−0.15 (−0.24, −0.06)	7841	−0.03 (−0.05, −0.01)
Liver BA	Pre-menopause (ref)	14,398	Ref	37,285	Ref
	Peri-menopause	3408	3.97 (3.57, 4.36)	–	–
	Post-menopause	17,108	7.21 (6.87, 7.55)	62,108	4.97 (4.82, 5.12)
	Hysterectomy	967	5.53 (4.83, 6.22)	11,149	4.29 (4.10, 4.49)
	Oophorectomy	795	4.98 (4.23, 5.73)	7841	4.94 (4.71, 5.16)
Metabolic BA	Pre-menopause (ref)	14,398	Ref	37,285	Ref
	Peri-menopause	3408	1.17 (0.96, 1.39)	–	–
	Post-menopause	17,108	2.24 (2.05, 2.43)	62,108	2.00 (1.86, 2.14)
	Hysterectomy	967	1.91 (1.53, 2.29)	11,149	2.37 (2.18, 2.55)
	Oophorectomy	795	1.35 (0.95, 1.76)	7841	2.58 (2.37, 2.79)
Kidney BA	Pre-menopause (ref)	14,398	Ref	37,285	Ref
	Peri-menopause	3408	0.73 (0.56, 0.89)	–	–
	Post-menopause	17,108	1.35 (1.21, 1.49)	62,108	2.05 (1.94, 2.15)
	Hysterectomy	967	0.91 (0.62, 1.20)	11,149	1.82 (1.68, 1.96)
	Oophorectomy	795	0.90 (0.59, 1.21)	7841	2.02 (1.86, 2.17)

Models were adjusted for age, ethnicity, education level, household income (CMEC only), Townsend deprivation index (UKB only), marital status (CMEC only), employment status, dietary score, total energy intake (CMEC only), smoking status, alcohol consumption, tea consumption, physical activity, body mass index, insomnia, depressive symptoms, anxiety symptoms, self-reported chronic diseases (cardiovascular disease, diabetes, cancer, chronic obstructive pulmonary disease), age at menarche, number of live births, ever taken oral contraceptive pill, and ever used hormone-replacement therapy (UKB only)

Abbreviations CMEC China Multi-Ethnic Cohort, UKB UK Biobank, BA biological age

For organ-specific BAs, both cohorts showed that menopausal status was associated with increased BA acceleration in the liver, metabolic, and kidney systems, with the strongest associations observed in liver BA. Specifically, the liver BA estimates for post-menopausal women were 7.21 (6.87, 7.55) in CMEC and 4.97 (4.82, 5.12) in UKB.

Menopausal transition and BA acceleration

Compared to pre-menopausal women, those who underwent the menopausal transition between baseline and follow-up exhibited accelerated comprehensive, liver, metabolic, and kidney BA in both cohorts, with stronger associations observed in UKB than in CMEC (Fig. 2). The estimated associations for comprehensive BA acceleration were 1.33 (0.89, 1.76) in CMEC and 2.60 (1.91, 3.30) in UKB. Among organ-specific BAs, menopausal transition was most strongly associated with increased liver BA, followed by metabolic and kidney BA.

Fig. 2.

Associations between change in menopausal status and change in BA acceleration. Models were adjusted for baseline time-invariant covariates as well as baseline values and concurrent changes of time-variant covariates. Time-invariant covariates included demographics, reproductive history, and self-reported chronic diseases: ethnicity, education level, household income (CMEC only), Townsend deprivation index (UKB only), cardiovascular disease, diabetes, cancer, chronic obstructive pulmonary disease, age at menarche, number of live births, ever use of oral contraceptives, and ever use of hormone-replacement therapy (UKB only). Time-variant covariates included age, marital status (CMEC only), employment status, dietary score, total energy intake (CMEC only), smoking status, alcohol consumption, tea consumption, physical activity, body mass index, insomnia, depressive symptoms, and anxiety symptoms, accounting for both baseline values and concurrent changes. The pre-pre group served as the reference group

In the CMEC analysis of six menopausal transition categories (Additional file 1: Table S5), those experiencing menopausal transition (including pre-post, peri-peri, peri-post, and pre-post) showed increased BA acceleration in comprehensive and most organ-specific measures, though some estimates did not reach statistical significance. The strongest associations were observed in the pre-post transition group, with estimates of 1.58 (0.97, 2.19) for comprehensive BA and 5.01 (3.77, 6.24) for liver BA.

Age at menopause and BA acceleration

Among women who experienced natural menopause, an earlier age at menopause (< 45 years) was generally associated with accelerated comprehensive BA, while a later age at menopause linked to a slower comprehensive BA. However, most estimates were not statistically significant in CMEC (Table 3).

Table 3.

Associations between age at menopause and BA acceleration

BA acceleration	Age at menopause (years)	CMEC		UKB
		n	β (95% CI)	n	β (95% CI)
Comprehensive BA	Continuous	17,108	−0.01 (−0.03, 0.01)	59,528	−0.05 (−0.06, −0.04)
	<40 (premature)	286	0.31 (−0.23, 0.85)	999	0.69 (0.39, 0.98)
	40–44 (early)	1795	0.10 (−0.14, 0.33)	4394	0.24 (0.09, 0.40)
	45–49 (ref)	6969	Ref	14,051	Ref
	50–54 (normal)	7168	−0.06 (−0.22, 0.09)	30,974	−0.31 (−0.41, −0.22)
	≥55 (late)	890	−0.03 (−0.35, 0.30)	9110	−0.34 (−0.47, −0.22)
Cardiopulmonary BA	Continuous	17,108	0.01 (0.00, 0.01)	59,528	−0.00 (−0.00, −0.00)
	<40 (premature)	286	−0.18 (−0.32, −0.03)	999	0.06 (0.01, 0.11)
	40–44 (early)	1795	−0.04 (−0.10, 0.03)	4394	0.03 (0.00, 0.06)
	45–49 (ref)	6969	Ref	14,051	Ref
	50–54 (normal)	7168	−0.00 (−0.04, 0.04)	30,974	−0.01 (−0.03, 0.00)
	≥55 (late)	890	0.01 (−0.08, 0.10)	9110	−0.02 (−0.04, 0.01)
Liver BA	Continuous	17,108	−0.04 (−0.08, −0.00)	59,528	0.01 (−0.01, 0.02)
	<40 (premature)	286	1.04 (−0.10, 2.18)	999	−0.11 (−0.57, 0.36)
	40–44 (early)	1795	0.43 (−0.07, 0.93)	4394	−0.15 (−0.39, 0.10)
	45–49 (ref)	6969	Ref	14,051	Ref
	50–54 (normal)	7168	−0.00 (−0.33, 0.32)	30,974	−0.09 (−0.23, 0.06)
	≥55 (late)	890	−0.12 (−0.82, 0.57)	9110	0.05 (−0.15, 0.25)
Metabolic BA	Continuous	17,108	0.01 (−0.01, 0.03)	59,528	−0.02 (−0.03, −0.00)
	<40 (premature)	286	0.43 (−0.23, 1.10)	999	0.21 (−0.23, 0.65)
	40–44 (early)	1795	0.11 (−0.19, 0.40)	4394	0.36 (0.13, 0.59)
	45–49 (ref)	6969	Ref	14,051	Ref
	50–54 (normal)	7168	0.15 (−0.04, 0.33)	30,974	−0.06 (−0.20, 0.08)
	≥55 (late)	890	0.17 (−0.24, 0.57)	9110	0.11 (−0.08, 0.29)
Kidney BA	Continuous	17,108	−0.02 (−0.03, 0.00)	59,528	−0.03 (−0.04, −0.02)
	<40 (premature)	286	0.24 (−0.27, 0.74)	999	0.42 (0.08, 0.75)
	40–44 (early)	1795	−0.01 (−0.23, 0.21)	4394	0.05 (−0.12, 0.23)
	45–49 (ref)	6969	Ref	14,051	Ref
	50–54 (normal)	7168	−0.15 (−0.29, −0.01)	30,974	−0.20 (−0.31, −0.09)
	≥55 (late)	890	−0.09 (−0.40, 0.21)	9110	−0.19 (−0.33, −0.05)

Models were adjusted for age, ethnicity, education level, household income (CMEC only), Townsend deprivation index (UKB only), marital status (CMEC only), employment status, dietary score, total energy intake (CMEC only), smoking status, alcohol consumption, tea consumption, physical activity, body mass index, insomnia, depressive symptoms, anxiety symptoms, self-reported chronic diseases (cardiovascular disease, diabetes, cancer, chronic obstructive pulmonary disease), age at menarche, number of live births, ever taken oral contraceptive pill, and ever used hormone-replacement therapy (UKB only)

Abbreviations CMEC China Multi-Ethnic Cohort, UKB UK Biobank, BA biological age

For organ-specific BAs, while the associations varied somewhat between the two cohorts, earlier menopause—particularly premature menopause—tended to be linked to greater BA acceleration across several organ systems. For instance, premature menopause was associated with accelerated kidney BA, with estimates of 0.24 (− 0.27, 0.74) in CMEC and 0.42 (0.08, 0.75) in UKB.

Sensitivity analysis

The findings remained largely robust after excluding individuals with self-reported chronic diseases, using the multiple imputation datasets, or additionally adjusting for medication use (Additional file 1: Figs. S2–S4, Tables S6–S11). To further evaluate the robustness of the observed associations, E-values were calculated and presented in Additional file 1: Table S12. E-values for the associations between menopausal transition and comprehensive BA acceleration were 6.19 in CMEC and 20.89 in UKB, suggesting that the observed estimates are robust, as unmeasured confounders would have to be unrealistically strong to fully account for the observed results.

Associations between menopausal factors and comprehensive BA acceleration according to reproductive history and socioeconomic factors

Stratified analysis showed variations in the associations between menopausal factors and biological aging across reproductive history characteristics (Fig. 3, Additional file 1: Tables S13 and S14). Women with later ages at first or last live birth exhibited greater comprehensive BA acceleration in both post-menopausal populations and those undergoing the menopausal transition. However, the heterogeneity P value reached statistical significance only for age at last live birth in the CMEC analysis (P = 0.043). In contrast, the number of live births was found to differently influence the associations between menopausal transition and biological aging across the two cohorts (Fig. 3). In CMEC, accelerated biological aging linked to menopausal transition was more pronounced in women with ≥ 2 live births, whereas in UKB, it was stronger in women with 0–1 live birth. Moreover, in UKB, HRT use among post-menopausal women was associated with a notable attenuation in comprehensive BA acceleration compared to non-users (Additional file 1: Table S13).

Fig. 3.

Stratified analysis of associations between menopausal transition and change in comprehensive BA acceleration according to reproductive history. The stratified analysis specifically examines the association between menopausal transition (pre to post) and comprehensive BA acceleration, with the pre-pre group as the reference. Models were adjusted for baseline time-invariant covariates as well as baseline values and concurrent changes of time-variant covariates, excluding the stratified variable where applicable. Time-invariant covariates included demographics, reproductive history, and self-reported chronic diseases: ethnicity, education level, household income (CMEC only), Townsend deprivation index (UKB only), cardiovascular disease, diabetes, cancer, chronic obstructive pulmonary disease, age at menarche, number of live births, ever use of oral contraceptives, and ever use of hormone-replacement therapy (UKB only). Time-variant covariates included age, marital status (CMEC only), employment status, dietary score, total energy intake (CMEC only), smoking status, alcohol consumption, tea consumption, physical activity, body mass index, insomnia, depressive symptoms, and anxiety symptoms, accounting for both baseline values and concurrent changes. In CMEC, the heterogeneity P value was > 0.05 for all variables except age at last live birth (P = 0.043). In UKB, all heterogeneity P values were > 0.05

The stratified analysis by socioeconomic factors showed that the associations between menopausal factors and comprehensive BA acceleration remained largely robust across different education and income levels (Additional file 1: Tables S15–S17).

Discussion

Summary of main results

In this study, we comprehensively investigated the associations between three menopausal factors and biological aging acceleration using data from two large, independent cohorts (CMEC and UKB). Women at peri-menopause and post-menopause stages exhibited increased biological aging acceleration compared with pre-menopausal women. Furthermore, women undergoing the menopausal transition, as captured by longitudinal data, exhibited a higher increase in change in biological aging acceleration compared with those who remained pre-menopausal. An earlier age at menopause tended to be associated with accelerated biological aging. Across most organ-specific BA, liver BA showed the most pronounced associations with menopausal factors. Additionally, reproductive history like age at live birth and number of live births emerged as potential modifiers of these associations.

Comparison with previous studies

Previous studies on menopause and biological aging have predominantly focused on age at menopause, consistently demonstrating harmful associations between early menopause (< 45 years) and various aging measures, including epigenetic age [14], frailty [16], telomere length [15, 23], and phenotypic age [17, 18]. Our results regarding early menopause and BA acceleration were largely consistent with previous findings. Although most estimates in CMEC were not statistically significant, their directions aligned with those observed in UKB. In contrast, late menopause (≥ 55 years) showed inconsistent associations with biological aging across studies. For example, while our study found lower comprehensive and kidney BA acceleration in women with late menopause, other studies reported no association with frailty index [16] or higher phenotypic age [17]. These discrepancies might be explained by differences in aging measures [39, 40], population characteristics, or organ-specific aging process [19, 20].

Our study also observed accelerated biological aging among peri-menopausal and post-menopausal women, as well as those who had undergone hysterectomy or oophorectomy, consistent with previous findings linking hysterectomy and oophorectomy to frailty and epigenetic aging [14, 16]. In CMEC, the weaker associations between oophorectomy and BA acceleration compared to bilateral oophorectomy in UKB may be due to the CMEC questionnaire not distinguishing between unilateral and bilateral oophorectomy. Besides, post-menopausal women using HRT exhibited reduced comprehensive BA acceleration compared to non-users. This finding aligns with a recent study showing that ever-users of HRT appeared biologically younger than never-users [41]. However, according to the updated guidelines from the National Institute for Health and Care Excellence (NICE) [42], there is no clear evidence that HRT provides long-term health benefits such as reduced cardiovascular risk or increased life expectancy, although no significant harm has been identified. In this context, the decision to initiate HRT for the management of menopausal symptoms should be made individually, considering the severity of symptoms, personal health risks, type and combination of hormones used, and dosage, as well as the timing and duration of therapy.

In our change-to-change analysis, we identified the longitudinal association between menopausal transition and accelerated biological aging. A key strength of this change-to-change analysis is its ability to capture within-individual changes from pre-menopause to post-menopause, thereby to some extent reducing the time-invariant unmeasured confounding [43, 44]. In our cross-sectional analysis of menopausal status, post-menopausal women exhibited greater biological aging acceleration compared to pre-menopausal women. However, in the longitudinal change-to-change analysis, no significant increase in BA acceleration was observed among post-menopausal women compared to their pre-menopausal counterparts. Instead, pronounced acceleration in biological aging was identified among women undergoing the menopausal transition. These findings suggest that the menopausal transition, rather than the post-menopausal state, is the critical period when biological aging accelerates most rapidly. This highlights the menopausal transition as a pivotal window for screening and intervention to improve women’s health and promote healthy aging [45].

The menopausal transition typically lasts 2–8 years [2]. In this study, the median follow-up duration was 2.01 years for CMEC and 4.50 years for UKB. The longer follow-up in UKB may have captured a higher proportion of women who completed the full transition process, potentially explaining the stronger associations observed in UKB. This interpretation is aligned with findings from the six transition groups in CMEC, where early (pre-peri), late (peri-post), and complete (pre-post) transitions were all associated with accelerated comprehensive BA, with the strongest associations observed in those undergoing the complete transition. Future research could explore the heterogeneity of biological aging among individuals with varying transition durations, investigate whether accelerated aging differs between the early and late phases of the transition, and identify the optimal timing for interventions.

Our study highlighted the associations between menopause and accelerated biological aging across multiple organ systems, identifying the most vulnerable systems as key targets for early intervention during the menopausal transition to reduce aging-related health risks. Among the four organ-specific BAs evaluated in this study, menopause was associated with accelerated liver, metabolic, and kidney BAs. These findings are consistent with existing evidence indicating increased risks of chronic diseases in multiple organ systems following menopause, such as nonalcoholic fatty liver disease [46], metabolic-related diseases [47, 48], and chronic kidney disease [49]. In contrast to the well-documented increased risks of cardiovascular diseases associated with menopause [2, 3], we did not observe a significant acceleration in cardiopulmonary BA, possibly due to the limited biomarkers used to construct cardiopulmonary BA in this study (SBP and PEF/FEV1), which may not fully reflect the complex aging processes of the cardiovascular and pulmonary systems.

Notably, liver aging emerged as the most vulnerable to menopause, followed by metabolic and kidney aging. The potential biological mechanisms underlying these findings are likely related to the decline in estrogen levels during menopause, which plays a critical role in regulating physiological functions across multiple organs and tissues [50, 51]. For example, estradiol plays a critical role in regulating lipid metabolism by reducing visceral fat accumulation, lowering cholesterol levels, modulating fatty acid metabolism, and protecting the liver from inflammatory damage [50, 52]. Consequently, the decline in estradiol levels in post-menopausal women may increase the risk of liver fibrosis. Epidemiology studies support the protective role of estrogen. Before menopause, women have a lower risk of liver-related conditions compared to men, but this advantage diminishes after menopause [46]. Following the menopausal transition, the prevalence and severity of nonalcoholic fatty liver disease (now termed as metabolic dysfunction-associated fatty liver disease) and hepatic steatosis significantly increase in women [53, 54]. Furthermore, a recent study has suggested that hepatocellular senescence could induce multi-organ senescence through the TGF-β pathway [55], providing a potential explanation for the prominent liver aging observed in our study.

Our study suggested that reproductive history, particularly maternal age, may modify the association between menopause and accelerated biological aging. Specifically, we observed stronger biological aging acceleration among women with a later age at first or last live birth in both cohorts. However, previous studies generally suggested that early maternal age (< 20 years) was associated with accelerated biological aging [17, 23]. Due to the relatively limited sample size of women undergoing menopausal transition in our study, we stratified maternal age into two broad categories and were unable to specifically examine the effects of early maternal age (< 20 years). Despite these findings, evidence on the relationship between maternal age and long-term health outcomes remains inconsistent. For instance, younger maternal age has been linked to an increased risk of incident dementia [56], while other studies found no significant associations between maternal age and cardiovascular disease mortality [57, 58]. While our findings indicate potential subgroups of women at higher risk for accelerated aging during the menopausal transition, whether these differences translate into long-term health consequences warrants further investigation.

Strength and limitations

To the best of our knowledge, this is the first study to comprehensively investigate the association of menopause with comprehensive and multi-organ biological aging. By leveraging longitudinal data and a change-to-change design, we identified accelerated biological aging from pre-menopausal to post-menopausal stages. Also, we explored the potential effect modification of reproductive history factors on the relationship between menopause and biological aging. Furthermore, we utilized data from two large cohorts with distinct genetic backgrounds, yielding consistent results that enhance the robustness and credibility of our findings.

This study has several limitations. First, information on menstrual history and other reproductive variables was self-reported, which may have introduced recall bias, particularly for events that occurred long ago, such as age at menarche. Additionally, in UKB, menopausal status was only categorized as menopausal or non-menopausal during each survey, without distinguishing between pre-menopause and peri-menopause. This limited the precision of classification for transition groups in the longitudinal analysis. For instance, individuals in the pre-peri transition phase were misclassified as being in the pre-pre group. Second, the CMEC lacked data on HRT, and although HRT use was recorded in the UKB, it was not possible to determine whether or how it influenced the reporting menopausal status. Participants whose menopausal status may have been affected by HRT could not be identified, potentially leading to exposure misclassification. Furthermore, the CMEC did not differentiate between unilateral and bilateral oophorectomy. As a result, women with unilateral oophorectomy—who may still retain ovarian function—were excluded from the longitudinal analysis, which could also contribute to misclassification bias. Third, we constructed four organ-specific biological ages primarily based on clinical laboratory test indicators. However, due to data availability, we were unable to develop biological ages for other organ systems that may be influenced by menopause, such as skin, skeletal, and brain aging [51]. Fourth, as an observational study, the possibility of unmeasured confounding cannot be ruled out. However, the use of change-to-change analysis helps mitigate the impact of time-invariant unmeasured confounders, and the calculated E-values indicate that any unmeasured confounder would need to be exceptionally strong to fully account for our primary findings. Fifth, although the transition analysis employed a prospective design, the possibility of reverse causality remains. Accelerated biological aging or underlying morbidity may contribute to the loss of ovarian function and earlier menopause. This potential scenario emphasizes the need for healthcare providers to stay vigilant regarding possible health issues that may arise during the menopausal phase. Nonetheless, sensitivity analyses excluding participants with self-reported chronic diseases yielded robust results. Moreover, the follow-up periods were relatively short, especially in the CMEC (median 2.01 years), whereas the menopausal transition can span 2–8 years. Further studies with extended follow-up are warranted. Therefore, causal inferences should be drawn with caution. Last, although our findings were largely consistent between the Chinese and UK populations, caution is warranted when generalizing the results. Neither the CMEC nor the UKB are nationally representative cohorts [24, 59]. Moreover, potential selection bias due to loss to follow-up, exclusion criteria, and missing data may further limit the generalizability of our findings. Future studies involving more diverse and representative populations are needed for validation.

Conclusions

In this study incorporating data from CMEC and UKB, menopause, particularly the menopausal transition, was associated with accelerated comprehensive and organ-specific biological aging, with the most pronounced associations observed in liver aging. Additionally, reproductive history, such as age first or last live birth, emerged as potential modifiers of these associations. Our findings highlight the menopausal transition as a critical window for interventions aimed at promoting women’s health and longevity in post-menopausal stages. Further research with longer follow-up in more representative populations is needed to validate these findings.

Abbreviations

ALB

Albumin

ALP

Alkaline phosphatase

AST

Aspartate aminotransferase

BA

Biological age

BFP

Body fat percentage

BMI

Body mass index

CA

Chronological age

CMEC

China Multi-Ethnic Cohort

COPD

Chronic obstructive pulmonary disease

CVD

Cardiovascular disease

FEV1

Forced expiratory volume in 1 s

GGT

Gamma-glutamyl transpeptidase

HBA1C

Glycated hemoglobin

HDL-c

High-density lipoprotein cholesterol

HRT

Hormone-replacement therapy

IGF-1

Insulin-like growth factor 1

KDM

Klemera-Doubal method

LDL-c

Low-density lipoprotein cholesterol

MCV

Mean corpuscular volume

NICE

National Institute for Health and Care Excellence

MET

Metabolic equivalent tasks

PEF

Peak expiratory flow

PLT

Platelet count

SBP

Systolic blood pressure

TG

Triglyceride

UKB

UK Biobank

WHR

Waist-to-hip ratio

Authors’ contributions

Y.X., Z.H., and X.X. conceptualized the present study. Y.X. conducted data analysis, wrote and revised the manuscript. X.Z., X.X., Q.M., X.D., J.Y., B. and L.L. contributed to the data collection, data management, and data cleaning. X.X., X.Z., Q.M., Z.H., N.Z., and Y.Z. reviewed and commented on the data analysis, all drafts and the final paper. All authors read and approved the final manuscript.

Funding

This study was funded by the National Natural Foundation of China (Grant No. 82273740) and Sichuan Science and Technology Program (Natural Science Foundation of Sichuan Province, Grant No. 2024NSFSC0552). The study sponsor played no role in the collection, analysis, or interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Data availability

The CMEC data can be accessed by contacting the corresponding author. The UKB data are available to the public upon application through the UK Biobank website (https://www.ukbiobank.ac.uk).

Declarations

Ethics approval and consent to participate

The CMEC study was approved by the Sichuan University Medical Ethical Review Board (ID: K2016038, K2020022). The UKB was approved by the National Information Governance Board for Health and Social Care and the National Health Service North West Multi-Centre Research Ethics Committee. All participants provided informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.