Abstract
Background
Osteoporosis is a major public health concern worldwide. Although female reproductive factors have been suggested to influence bone health, their associations with osteoporosis remain inconclusive.
Objective
This study aimed to investigate the associations between reproductive characteristics and the odds of osteoporosis in Taiwanese women.
Methods
We conducted a cross-sectional analysis using data from women randomly selected from the Taiwan Biobank. Bone mineral density (BMD) status was classified into three categories according to the World Health Organization (WHO) criteria: normal BMD (T-score > − 1.0), osteopenia (− 2.5 < T-score ≤ − 1.0), and osteoporosis (T-score ≤ − 2.5). Univariable and multivariable logistic regression analyses were performed to assess the association between reproductive factors and the odds of BMD status.
Results
Among the 4,854 participants, 3,107 (64.0%) were classified as having normal BMD, 1,374 (28.3%) as having osteopenia, and 373 (7.7%) as having osteoporosis. Univariate analysis showed that a history of pregnancy was associated with a higher odd of osteopenia (odds ratio [OR] = 1.946, 95% CI: 1.589–2.382, p < 0.001), osteoporosis (OR = 1.825, 95% CI: 1.288–2.586, p = 0.001), and low BMD level (T-score < -0.1) (OR = 1.919, 95% CI: 1.597–2.306, p < 0.001). After adjusting for known confounders, pregnancy history was associated with a lower odd of osteoporosis (adjusted OR = 0.644, 95% CI: 0.422–0.981, p = 0.041), but not significantly associated with the odds of osteopenia or low BMD level.
Conclusions
Pregnancy history was independently associated with lower odds of osteoporosis among Taiwanese women after adjustment for known confounding factors. However, given the cross-sectional nature of this study, causal relationships cannot be established. Future longitudinal studies are warranted to validate this finding.
Keywords: Osteoporosis, Pregnancy, Secondhand smoke, Tea consumption, Menopause
Background
Osteoporosis is a systemic skeletal disorder marked by a decrease in bone mineral density (BMD) and the loss of bone microarchitecture, leading to decreased bone strength and an increased likelihood of fractures [1], with fragility fractures being the primary clinical symptom of osteoporosis, accounting for approximately 80% of all fracture cases [2]. As such fractures are linked to decreased quality of life, increased rates of premature mortality, long-term disability and considerable economic burden [3], identifying the risk factors associated with osteoporosis and developing effective strategies for the prevention of fracture occurrence are imperative.
Osteoporosis has a significant impact on postmenopausal women, affecting more than 200 million individuals globally, with approximately 30% of women in the United States and Europe diagnosed with the condition [4]. The predominant underlying factor is the decline in estrogen production following menopause, which activates bone remodeling pathways and accelerates bone resorption [5, 6]. This hormonal shift disrupts the bone turnover balance, ultimately leading to diminished bone strength [5]. Besides postmenopausal hormonal changes, reproductive factors throughout a woman’s life course may also influence bone health and contribute to the risk of osteoporosis in later life.
Multiple births during adolescence and frequent pregnancies throughout the reproductive years can negatively affect postmenopausal BMD [7]. Consistent with previous studies, longer interpregnancy intervals and later onset of menopause appear to reduce the risk of osteoporosis [7], while multiparity and breastfeeding may offer protective effects, particularly among women aged 49–54 [8]. However, findings have been inconsistent; for instance, Sioka et al. (2007) reported no significant association between obstetric history—including childbirths and miscarriages—and the risk of osteopenia or osteoporosis [9]. Although osteoporosis is a major health concern for postmenopausal women, the influence of female reproductive factors on its development remains inconclusive. Therefore, it is necessary to conduct in-depth research in specific populations (such as Taiwanese women) to clarify these contradictions.
This study aimed to investigate the association between female reproductive characteristics and the odds of osteoporosis among women in Taiwan. By identifying reproductive-related factors within this specific population, the study seeks to enhance understanding of osteoporosis etiology and support the development of tailored prevention and screening strategies.
Methods
Ethics statement
The Taiwan Biobank obtained ethical approval from both the Institutional Review Board on Biomedical Science Research (IRB-BM) at Academia Sinica and the Ethics and Governance Council (EGC) of the Taiwan Biobank. All participants provided written informed consent in compliance with ethical requirements and the principles outlined in the Declaration of Helsinki. Additionally, this study was approved by the Institutional Review Board of Ditmanson Medical Foundation Chiayi Christian hospital (approval number: IRB2024124) and by the Taiwan Biobank (approval number: TWBR11404-06).
Data source
As part of a community-based cohort research project, by May 31, 2025, the Taiwan Biobank had recruited nearly 200,000 healthy volunteers aged between 30 and 70 years. A structured, face-to-face interview was conducted using a standardized questionnaire to obtain detailed personal information, including dietary habits, health behaviors, and medical history. Additional sections of the questionnaire addressed women-specific factors, such as gynecological conditions and reproductive history, for female participants.
Study population
This study randomly selected 7,000 female participants from the Taiwan Biobank database who had recorded data on gynecological conditions and reproductive history. The participants were further screened for eligibility, and those without available BMD T-score measurements were excluded from the analysis.
Ascertainment of the BMD status
BMD was assessed using quantitative ultrasound of the calcaneus with an Achilles InSight ultrasound system (GE Healthcare, Madison, WI, USA) [10]. Measurements were obtained from the non-dominant foot, and T-scores were derived by standardizing individual values against a young-adult reference population [11]. Participants were categorized into three groups according to the World Health Organization (WHO) criteria based on BMD T-scores. Normal BMD was defined as a T-score > − 1.0, osteopenia as a T-score between − 1.0 and − 2.5, and osteoporosis as a T-score ≤ − 2.5. For additional analyses, low BMD was defined as a T-score < − 1.0, including both osteopenia and osteoporosis. BMD status was categorized into three clinically meaningful groups representing different stages of skeletal health, ranging from normal BMD to osteopenia and osteoporosis. Osteopenia is generally considered an intermediate stage between normal bone mass and osteoporosis, reflecting progressive bone loss and increasing fracture risk.
Prior known confounding factors
We adjusted for known confounding factors associated with BMD status, which were selected a priori based on clinical knowledge and evidence from previous studies. These factors included age, BMI, and menopausal status [12–14]. In addition, any history of using female-focused supplements, traditional Chinese medicines for gynecologic regulation or hormonal medications for extended periods (three or six months or more respectively) were also recorded [15, 16]. Lifestyle-related factors were also assessed, including alcohol consumption status (no or occasional drinking versus former or current drinking) [17], exposure to secondhand smoke [18], exercise habits (never, former, or current) [19], and tea consumption habits [20], all of which have been reported to be associated with BMD.
Reproductive factors
Several reproductive factors were obtained from the questionnaires. These data included reproductive and hormonal information; specifically, whether the participant had ever been pregnant, her age at menarche, her age at menopause, and current menopause status. Additional information was also recorded, including number of pregnancies, age at first or last pregnancy, delivery history, number of deliveries, age at first and last childbirth, abortion history, and breastfeeding experience.
Statistical analysis
All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, North Carolina, USA). Continuous variables were expressed as mean ± standard deviation (SD), whereas categorical variables were presented as frequencies and percentages. The means of continuous variables were compared among the three groups using one-way analysis of variance (ANOVA). When the overall ANOVA was statistically significant, post hoc pairwise comparisons were conducted using Tukey’s test. Categorical variables were compared among the groups using the chi-square test. Univariable logistic regression was conducted to examine associations between individual factors and BMD status. Multivariable logistic regression was further used to examine these associations while adjusting for known potential confounders. Logistic regression analyses were conducted using complete-case data; participants with missing values in any variables included in the model were excluded from the analysis. All p-values were two-tailed, with statistical significance defined as p < 0.05.
Result
Major characteristics
Of the 7,000 women randomly sampled from the Taiwan Biobank database, 2,147 without a measured T-score were excluded, leaving 4,854 eligible participants (Fig. 1). The remaining participants were classified according to their T-scores into three groups: a normal BMD group (n = 3,107), an osteopenia group (n = 1,374), and an osteoporosis group (n = 373). The major characteristics of the three groups are presented in Table 1. The women with osteoporosis were significantly older and had lower BMI than those with osteopenia and normal BMD (both p < 0.001). Regarding reproductive characteristics, pregnancy history and delivery history were more common among women with osteopenia and osteoporosis than among those with normal BMD. In addition, women with osteopenia and osteoporosis had significantly higher numbers of pregnancies and deliveries (p < 0.001 for both). Women with osteoporosis also tended to have younger ages at first pregnancy, last pregnancy, first childbirth, and last childbirth compared with women with normal BMD. However, age at menopause, abortion history, and breastfeeding history did not differ significantly among the three groups.
Fig. 1.
Flowchart of participant selection and grouping from the Taiwan Biobank database
Table 1.
Demographic characteristics, reproductive history, lifestyle factors, and BMD-related variables according to BMD status
| Variable | Sample size | subgroup | Normal BMD group | Osteopenia group | Osteoporosis | p-value |
|---|---|---|---|---|---|---|
| Age (year) | 4,854 | 46.67 ± 10.42a | 53.80 ± 10.20b | 58.54 ± 8.76c | < .001 | |
| BMI (kg/m2) | 4,854 | 23.76 ± 3.82a | 23.32 ± 3.23b | 22.95 ± 3.41b | < .001 | |
| Current menopausal status | 4,854 | Yes | 1,135 (36.53%) | 953 (69.36%) | 328 (87.94%) | < .001 |
| No | 1,972 (63.47%) | 421 (30.64%) | 45 (12.06%) | |||
| Pregnancy history | 4,822 | No | 532 (17.24%) | 132 (9.67%) | 38 (10.24%) | < .001 |
| Yes | 2,554 (82.76%) | 1,233 (90.33%) | 333 (89.76%) | |||
| Delivery history | 4,118 | No | 96 (3.76%) | 27 (2.19%) | 6 (1.8%) | 0.012 |
| Yes | 2,456 (96.24%) | 1,206 (97.81%) | 327 (98.2%) | |||
| Abortion history | 4,105 | No | 965 (38%) | 454 (36.9%) | 113 (33.9%) | 0.332 |
| Yes | 1,576 (62%) | 777 (63.1%) | 220 (66.1%) | |||
| Breastfeeding history | 4,119 | No | 854 (33.5%) | 390 (31.6%) | 113 (33.9%) | 0.495 |
| Yes | 1,699 (66.5%) | 843 (68.4%) | 220 (66.1%) | |||
| Age at menopause (year) | 2,338 | 49.23 ± 5.20a | 49.40 ± 4.98a | 49.51 ± 4.88a | 0.621 | |
| Number of pregnancies | 4,788 | 2.70 ± 1.84a | 3.15 ± 1.80b | 3.22 ± 1.79b | < .001 | |
| Age at first pregnancy (year) | 4,054 | 25.28 ± 4.28a | 25.13 ± 4.12a | 24.63 ± 3.83b | 0.027 | |
| Age at last pregnancy (year) | 3,947 | 30.61 ± 4.65a | 30.49 ± 4.68a | 29.91 ± 4.39b | 0.039 | |
| Number of deliveries | 4,118 | 2.21 ± 0.95a | 2.39 ± 0.99b | 2.44 ± 0.96b | < .001 | |
| Age at first childbirth (year) | 3,986 | 26.36 ± 4.20a | 26.08 ± 4.05a | 25.38 ± 3.81b | < .001 | |
| Age at last childbirth (year) | 3,980 | 30.13 ± 4.21a | 30.00 ± 4.14a | 29.36 ± 4.16b | 0.008 | |
| Exercise habits | 4,854 | No | 1,914 (61.6%) | 720 (52.4%) | 190 (50.94%) | < .001 |
| Yes | 1,193 (38.4%) | 654 (47.6%) | 183 (49.06%) | |||
| Tea consumption habits | 4,854 | No | 2,148 (69.13%) | 997 (72.56%) | 299 (80.16%) | < .001 |
| Yes | 959 (30.87%) | 377 (27.44%) | 74 (19.84%) | |||
| Current alcohol consumption habits | 4,852 | No | 3,024 (97.36%) | 1,345 (97.89%) | 363 (97.58%) | 0.842 |
| Occasion-al | 22 (0.71%) | 9 (0.66%) | 2 (0.54%) | |||
| Regular | 60 (1.93%) | 20 (1.46%) | 7 (1.88%) | |||
| Hormonal medication use | 4,848 | No | 2,672 (86.08%) | 1,146 (83.59%) | 291 (78.02%) | < .001 |
| Yes | 432 (13.92%) | 225 (16.41%) | 82 (21.98%) | |||
| Use of female-regulating traditional chinese medicine | 4,853 | No | 2,725 (87.73%) | 1,242 (90.39%) | 341 (91.42%) | 0.008 |
| Yes | 381 (12.27%) | 132 (9.61%) | 32 (8.58%) | |||
| Use of female supplements | 4,851 | No | 2,930 (94.39%) | 1,290 (93.89%) | 344 (92.23%) | 0.229 |
| Yes | 174 (5.61%) | 84 (6.11%) | 29 (7.77%) | |||
| Exposure to secondhand smoke | 4,853 | No | 2,780 (89.48%) | 1,255 (91.34%) | 332 (89.25%) | 0.141 |
| Yes | 327 (10.52%) | 119 (8.66%) | 40 (10.75%) |
Data are presented as mean ± standard deviation or n (%)
Abbreviations: BMD Bone mineral density, BMI Body mass index
Different superscript letters (a, b, c) indicate statistically significant differences between groups based on post hoc comparisons
Statistically significant associations (p < 0.05) are shown in bold
For lifestyle-related factors, regular exercise and tea consumption were less common among women with osteoporosis, whereas hormone medication use was more frequent in this group. Significant differences were also observed in the use of female-regulating traditional Chinese medicine. In contrast, alcohol consumption, use of female supplements, and exposure to secondhand smoke did not differ significantly across BMD categories.
Reproductive characteristics associated with BMD status
The associations between reproductive characteristics and BMD status were evaluated using univariate logistic regression analyses (Table 2). For osteopenia, women with a history of pregnancy had significantly higher odds of osteopenia than those without a pregnancy history (OR = 1.946, 95% CI: 1.589–2.382, p < 0.001). A greater number of pregnancies (OR = 1.141, 95% CI: 1.102–1.182, p < 0.001), delivery history (OR = 1.744, 95% CI: 1.132–2.688, p = 0.012), and a greater number of deliveries (OR = 1.201, 95% CI: 1.119–1.289, p < 0.001) were also associated with increased odds of osteopenia. No significant associations were observed for age at menopause, age at first pregnancy, age at last pregnancy, age at first childbirth, age at last childbirth, abortion history, or breastfeeding history.
Table 2.
Univariable logistic regression analysis of the associations between pregnancy history and BMD status (osteopenia, osteoporosis, and low BMD)
| Variable | OR (95% CI) | p-value | sample size |
|---|---|---|---|
| The odds of osteopenia | |||
| Pregnancy history (ref: No) | 1.946 (1.589–2.382) | < .001 | 4,451 |
| Age at menopause (year) | 1.006 (0.989–1.024) | 0.468 | 2,022 |
| Number of pregnancies | 1.141 (1.102–1.182) | < .001 | 4,419 |
| Age at first pregnancy (year) | 0.992 (0.976–1.008) | 0.310 | 3,728 |
| Age at last pregnancy (year) | 0.994 (0.979–1.009) | 0.433 | 3,636 |
| Delivery (ref: No) | 1.744 (1.132–2.688) | 0.012 | 3,785 |
| Number of deliveries | 1.201 (1.119–1.289) | < .001 | 3,785 |
| Age at first childbirth (year) | 0.984 (0.968–1.000) | 0.057 | 3,660 |
| Age at last childbirth (year) | 0.993 (0.976–1.009) | 0.376 | 3,655 |
| Abortion history (ref: No) | 1.048 (0.910–1.206) | 0.515 | 3,772 |
| Breastfeeding history (ref: No) | 1.086 (0.939–1.257) | 0.264 | 3,786 |
| The odds of osteoporosis | |||
| Pregnancy history (ref: No) | 1.825 (1.288–2.586) | 0.001 | 3,457 |
| Age at menopause (year) | 1.011 (0.986–1.036) | 0.403 | 1,410 |
| Number of pregnancies | 1.159 (1.095–1.228) | < .001 | 3,429 |
| Age at first pregnancy (year) | 0.963 (0.936–0.991) | 0.009 | 2,836 |
| Age at last pregnancy (year) | 0.967 (0.942–0.992) | 0.011 | 2,756 |
| Delivery (ref: No) | 2.130 (0.926–4.899) | 0.075 | 2,885 |
| Number of deliveries | 1.281 (1.139–1.441) | < .001 | 2,885 |
| Age at first childbirth (year) | 0.943 (0.916–0.971) | < .001 | 2,780 |
| Age at last childbirth (year) | 0.957 (0.930–0.984) | 0.002 | 2,776 |
| Abortion history (ref: No) | 1.192 (0.937–1.516) | 0.152 | 2,874 |
| Breastfeeding history (ref: No) | 0.979 (0.769–1.246) | 0.860 | 2,886 |
| The odds of low BMD level | |||
| Pregnancy history (ref: No) | 1.919 (1.597–2.306) | < .001 | 4,822 |
| Age at menopause (year) | 1.008 (0.992–1.024) | 0.358 | 2,338 |
| Number of pregnancies | 1.148 (1.111–1.186) | < .001 | 4,788 |
| Age at first pregnancy (year) | 0.986 (0.971–1.001) | 0.059 | 4,054 |
| Age at last pregnancy (year) | 0.988 (0.975–1.002) | 0.102 | 3,947 |
| Delivery (ref: No) | 1.815 (1.216–2.709) | 0.003 | 4,118 |
| Number of deliveries | 1.219 (1.141–1.302) | < .001 | 4,118 |
| Age at first childbirth (year) | 0.975 (0.960–0.990) | 0.002 | 3,986 |
| Age at last childbirth (year) | 0.985 (0.970–1.000) | 0.051 | 3,980 |
| Abortion history (ref: No) | 1.077 (0.945–1.227) | 0.267 | 4,105 |
| Breastfeeding history (ref: No) | 1.062 (0.929–1.215) | 0.379 | 4,119 |
Abbreviations: ref reference, OR Odds ratio, CI Confidence interval
Statistically significant associations (p < 0.05) are shown in bold
For osteoporosis, pregnancy history (OR = 1.825, 95% CI: 1.288–2.586, p = 0.001), number of pregnancies (OR = 1.159, 95% CI: 1.095–1.228, p < 0.001), number of deliveries (OR = 1.281, 95% CI: 1.139–1.441, p < 0.001), younger age at first pregnancy (OR = 0.963, 95% CI: 0.936–0.991, p = 0.009), younger age at last pregnancy (OR = 0.967, 95% CI: 0.942–0.992, p = 0.011), younger age at first childbirth (OR = 0.943, 95% CI: 0.916–0.971, p < 0.001), and younger age at last childbirth (OR = 0.957, 95% CI: 0.930–0.984, p = 0.002) were significantly associated with osteoporosis. However, age at menopause, delivery history, abortion history, and breastfeeding history were not significantly associated with osteoporosis.
When low BMD (osteopenia and osteoporosis combined) was used as the outcome, pregnancy history (OR = 1.919, 95% CI: 1.597–2.306, p < 0.001), number of pregnancies (OR = 1.148, 95% CI: 1.111–1.186, p < 0.001), delivery history (OR = 1.815, 95% CI: 1.216–2.709, p = 0.003), number of deliveries (OR = 1.219, 95% CI: 1.141–1.302, p < 0.001), and younger age at first childbirth (OR = 0.975, 95% CI: 0.960–0.990, p = 0.002) were significantly associated with increased odds of low BMD. In contrast, age at menopause, age at first pregnancy, age at last pregnancy, age at last childbirth, abortion history, and breastfeeding history were not significantly associated with low BMD.
Adjusted associations between reproductive characteristics and BMD status
Based on the univariate logistic regression analyses, reproductive factors that were significantly associated with BMD status were further evaluated in multivariable logistic regression models. The associations were further examined after adjustment for known potential confounding factors. Pregnancy history and delivery history were identified as strong reproductive factors associated with BMD status and were therefore included in the adjusted models.
Pregnancy history
Pregnancy history was significantly associated with osteopenia in the univariate analysis; however, this association was attenuated and no longer statistically significant after adjustment for known potential confounding factors (adjusted OR = 1.038, 95% CI: 0.827–1.302, p = 0.749) (Table 3). In contrast, pregnancy history remained independently associated with osteoporosis in the multivariable model, with women who had experienced pregnancy exhibiting 35.6% lower odds of osteoporosis compared with those without a pregnancy history (adjusted OR = 0.644, 95% CI: 0.422–0.981, p = 0.041). When low BMD (osteopenia and osteoporosis combined) was used as the outcome, the association between pregnancy history and low BMD was no longer statistically significant after adjustment (adjusted OR = 0.957, 95% CI: 0.773–1.183, p = 0.683).
Table 3.
Multivariable logistic regression analysis of the associations between pregnancy history and BMD status (osteopenia, osteoporosis, and low BMD) after adjustment for potential confounding factors
| Variable | Adjusted OR (95% CI) | p-value |
|---|---|---|
| The odds of osteopenia (N = 4,440) | ||
| Pregnancy history (ref: No) | 1.038 (0.827–1.302) | 0.749 |
| Current menopausal status (ref: No) | 0.487 (0.389–0.609) | < 0.001 |
| Age (year) | 1.051 (1.039–1.064) | < 0.001 |
| BMI (kg/m²) | 0.923 (0.904–0.943) | < 0.001 |
| Exercise habits (ref: No) | 0.827 (0.714–0.958) | 0.011 |
| Tea consumption habits (ref: No) | 0.996 (0.855–1.160) | 0.957 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.799 (0.354–1.804) | 0.589 |
| Regular | 0.812 (0.473–1.394) | 0.451 |
| Hormone medication use (ref: No) | 0.780 (0.643–0.946) | 0.012 |
| Herbal medicine use (ref: No) | 1.133 (0.903–1.423) | 0.281 |
| Female supplement use (ref: No) | 1.047 (0.781–1.403) | 0.761 |
| Secondhand smoke exposure (ref: No) | 1.166 (0.918–1.481) | 0.208 |
| The odds of osteoporosis (N = 3,840) | ||
| Pregnancy history (ref: No) | 0.644 (0.422–0.981) | 0.041 |
| Current menopausal status (ref: No) | 3.240 (2.047–5.128) | < 0.001 |
| Age (year) | 1.116 (1.092–1.140) | < 0.001 |
| BMI (kg/m²) | 0.869 (0.836–0.904) | < 0.001 |
| Exercise habits (ref: No) | 0.652 (0.508–0.838) | < 0.001 |
| Tea consumption habits (ref: No) | 0.785 (0.586–1.051) | 0.104 |
| Alcohol consumption (ref: Never) | 1.031 (0.225–4.731) | |
| Occasional | 1.280 (0.541–3.028) | 0.969 |
| Regular | 0.827 (0.612–1.119) | 0.574 |
| Hormone medication use (ref: No) | 1.221 (0.788–1.894) | 0.218 |
| Herbal medicine use (ref: No) | 1.234 (0.764–1.992) | 0.372 |
| Female supplement use (ref: No) | 1.924 (1.270–2.917) | 0.391 |
| Secondhand smoke exposure (ref: No) | 0.644 (0.422–0.981) | 0.002 |
| The odds of low BMD level (N = 4,810) | ||
| Pregnancy history (ref: No) | 0.957 (0.773–1.183) | 0.683 |
| Current menopausal status (ref: No) | 2.226 (1.797–2.758) | < 0.001 |
| Age (year) | 1.060 (1.049–1.072) | < 0.001 |
| BMI (kg/m²) | 0.915 (0.897–0.933) | < 0.001 |
| Exercise habits (ref: No) | 0.791 (0.688–0.908) | < 0.001 |
| Tea consumption habits (ref: No) | 0.952 (0.823–1.101) | 0.509 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.832 (0.383–1.803) | 0.640 |
| Regular | 0.887 (0.541–1.453) | 0.633 |
| Hormone medication use (ref: No) | 0.797 (0.666–0.953) | 0.013 |
| Herbal medicine use (ref: No) | 1.159 (0.934–1.438) | 0.180 |
| Female supplement use (ref: No) | 1.098 (0.835–1.445) | 0.502 |
| Secondhand smoke exposure (ref: No) | 1.262 (1.010–1.578) | 0.040 |
Abbreviations: ref reference, OR Odds ratio, CI Confidence interval, BMI Body mass index
Statistically significant associations (p < 0.05) are shown in bold
Number of pregnancies
Although the number of pregnancies was significantly associated with osteopenia, osteoporosis, and low BMD in the univariate analyses, these associations were no longer statistically significant after adjustment for known potential confounding factors (Table 4). The adjusted odds ratios were 1.000 (95% CI: 0.959–1.042, p = 0.995) for osteopenia, 0.934 (95% CI: 0.867–1.005, p = 0.069) for osteoporosis, and 0.986 (95% CI: 0.948–1.025, p = 0.478) for low BMD. These findings suggest that the observed associations between the number of pregnancies and BMD status were largely explained by confounding factors rather than representing an independent effect.
Table 4.
Multivariable logistic regression analysis of the associations between number of pregnancies and BMD status (osteopenia, osteoporosis, and low BMD) after adjustment for potential confounding factors
| Variable | Adjusted OR (95% CI) | p-value |
|---|---|---|
| The odds of osteopenia (N = 4,408) | ||
| Number of pregnancies | 1.000 (0.959–1.042) | 0.995 |
| Current menopausal status (ref: No) | 2.072 (1.655–2.594) | < 0.001 |
| Age (year) | 1.051 (1.039–1.064) | < 0.001 |
| BMI (kg/m²) | 0.925 (0.905–0.945) | < 0.001 |
| Exercise habits (ref: No) | 0.833 (0.718–0.965) | 0.015 |
| Tea consumption habits (ref: No) | 0.999 (0.858–1.165) | 0.994 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.794 (0.352–1.794) | 0.580 |
| Regular | 0.865 (0.503–1.489) | 0.602 |
| Hormone medication use (ref: No) | 0.772 (0.635–0.937) | 0.009 |
| Herbal medicine use (ref: No) | 1.147 (0.913–1.441) | 0.239 |
| Female supplement use (ref: No) | 1.040 (0.774–1.398) | 0.795 |
| Secondhand smoke exposure (ref: No) | 1.169 (0.920–1.485) | 0.202 |
| The odds of osteoporosis (N = 3,420) | ||
| Number of pregnancies | 0.934 (0.867–1.005) | 0.069 |
| Current menopausal status (ref: No) | 3.222 (2.038–5.094) | < 0.001 |
| Age (year) | 1.116 (1.092–1.141) | < 0.001 |
| BMI (kg/m²) | 0.869 (0.835–0.904) | < 0.001 |
| Exercise habits (ref: No) | 0.644 (0.501–0.828) | < 0.001 |
| Tea consumption habits (ref: No) | 0.794 (0.593–1.065) | 0.123 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 1.019 (0.222–4.683) | 0.980 |
| Regular | 1.372 (0.574–3.277) | 0.477 |
| Hormone medication use (ref: No) | 0.834 (0.616–1.129) | 0.240 |
| Herbal medicine use (ref: No) | 1.246 (0.803–1.933) | 0.327 |
| Female supplement use (ref: No) | 1.284 (0.794–2.074) | 0.308 |
| Secondhand smoke exposure (ref: No) | 1.907 (1.252–2.905) | 0.003 |
| The odds of low BMD level (N = 4,776) | ||
| Number of pregnancies | 0.986 (0.948–1.025) | 0.478 |
| Current menopausal status (ref: No) | 2.248 (1.813–2.787) | < 0.001 |
| Age (year) | 1.060 (1.048–1.072) | < 0.001 |
| BMI (kg/m²) | 0.916 (0.898–0.934) | < 0.001 |
| Exercise habits (ref: No) | 0.794 (0.691–0.913) | 0.001 |
| Tea consumption habits (ref: No) | 0.957 (0.827–1.108) | 0.558 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.828 (0.382–1.797) | 0.634 |
| Regular | 0.946 (0.575–1.555) | 0.827 |
| Hormone medication use (ref: No) | 0.794 (0.663–0.951) | 0.012 |
| Herbal medicine use (ref: No) | 1.174 (0.945–1.457) | 0.147 |
| Female supplement use (ref: No) | 1.101 (0.835–1.451) | 0.496 |
| Secondhand smoke exposure (ref: No) | 1.262 (1.009–1.579) | 0.041 |
Abbreviations: ref reference, OR Odds ratio, CI Confidence interval, BMI Body mass index
Statistically significant associations (p < 0.05) are shown in bold
Delivery history
Although delivery history was significantly associated with osteopenia and low BMD in the univariate analyses, these associations were attenuated and became non-significant after adjustment for known potential confounding factors (Table 5). The adjusted odds ratios were 0.977 (95% CI: 0.613–1.556, p = 0.922) for osteopenia and 0.937 (95% CI: 0.603–1.456, p = 0.774) for low BMD. In addition, delivery history was not significantly associated with osteoporosis in either the univariate or multivariable analyses (adjusted OR = 0.728, 95% CI: 0.283–1.878, p = 0.512).
Table 5.
Multivariable logistic regression analysis of the associations between delivery history and BMD status (osteopenia, osteoporosis, and low BMD) after adjustment for potential confounding factors
| Variable | Adjusted OR (95% CI) | p-value |
|---|---|---|
| The odds of osteopenia (N = 3,774) | ||
| Delivery history (ref: No) | 0.977 (0.613–1.556) | 0.922 |
| Current menopausal status (ref: No) | 1.942 (1.534–2.457) | < 0.001 |
| Age (year) | 1.056 (1.042–1.069) | < 0.001 |
| BMI (kg/m²) | 0.918 (0.897–0.939) | < 0.001 |
| Exercise habits (ref: No) | 0.819 (0.701–0.956) | 0.012 |
| Tea consumption habits (ref: No) | 0.964 (0.819–1.135) | 0.663 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.898 (0.389–2.073) | 0.800 |
| Regular | 0.880 (0.506–1.533) | 0.653 |
| Hormone medication use (ref: No) | 0.761 (0.621–0.932) | 0.008 |
| Herbal medicine use (ref: No) | 1.238 (0.963–1.592) | 0.096 |
| Female supplement use (ref: No) | 1.057 (0.771–1.449) | 0.731 |
| Secondhand smoke exposure (ref: No) | 1.317 (1.018–1.703) | 0.036 |
| The odds of osteoporosis (N = 2,876) | ||
| Delivery history (ref: No) | 0.728 (0.283–1.878) | 0.512 |
| Current menopausal status (ref: No) | 3.283 (2.009–5.367) | < 0.001 |
| Age (year) | 1.117 (1.092–1.143) | < 0.001 |
| BMI (kg/m²) | 0.869 (0.834–0.906) | < 0.001 |
| Exercise habits (ref: No) | 0.619 (0.476–0.806) | < 0.001 |
| Tea consumption habits (ref: No) | 0.808 (0.595–1.097) | 0.172 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 1.085 (0.231–5.101) | 0.917 |
| Regular | 1.113 (0.443–2.794) | 0.820 |
| Hormone medication use (ref: No) | 0.863 (0.632–1.178) | 0.35 |
| Herbal medicine use (ref: No) | 1.270 (0.783–2.060) | 0.332 |
| Female supplement use (ref: No) | 1.266 (0.761–2.107) | 0.363 |
| Secondhand smoke exposure (ref: No) | 1.883 (1.202–2.950) | 0.006 |
| The odds of low BMD level (N = 4,106) | ||
| Delivery history (ref: No) | 0.937 (0.603–1.456) | 0.774 |
| Current menopausal status (ref: No) | 2.101 (1.677–2.632) | < 0.001 |
| Age (year) | 1.064 (1.052–1.077) | < 0.001 |
| BMI (kg/m²) | 0.910 (0.891–0.930) | < 0.001 |
| Exercise habits (ref: No) | 0.775 (0.669–0.898) | < 0.001 |
| Tea consumption habits (ref: No) | 0.930 (0.796–1.087) | 0.361 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.937 (0.421–2.084) | 0.872 |
| Regular | 0.915 (0.547–1.532) | 0.737 |
| Hormone medication use (ref: No) | 0.788 (0.653–0.951) | 0.013 |
| Herbal medicine use (ref: No) | 1.261 (0.992–1.602) | 0.058 |
| Female supplement use (ref: No) | 1.103 (0.821–1.482) | 0.515 |
| Secondhand smoke exposure (ref: No) | 1.393 (1.094–1.775) | 0.007 |
Abbreviations: ref reference, OR Odds ratio, CI Confidence interval, BMI Body mass index
Statistically significant associations (p < 0.05) are shown in bold
Number of deliveries
Although the number of deliveries was significantly associated with osteopenia, osteoporosis, and low BMD in the univariate analyses, these associations were attenuated and no longer statistically significant after adjustment for known potential confounding factors (Table 6). The adjusted odds ratios were 0.988 (95% CI: 0.912–1.070, p = 0.763) for osteopenia, 0.943 (95% CI: 0.819–1.086, p = 0.417) for osteoporosis, and 0.975 (95% CI: 0.904–1.052, p = 0.514) for low BMD. These findings indicate that the number of deliveries was not independently associated with BMD status after controlling for established BMD-related risk factors.
Table 6.
Multivariable logistic regression analysis of the associations between number of deliveries and BMD status (osteopenia, osteoporosis, and low BMD) after adjustment for potential confounding factors
| Variable | Adjusted OR (95% CI) | p-value |
|---|---|---|
| The odds of osteopenia (N = 3,774) | ||
| Number of deliveries | 0.988 (0.912–1.070) | 0.763 |
| Current menopausal status (ref: No) | 1.941 (1.534–2.456) | < 0.001 |
| Age (year) | 1.056 (1.042–1.070) | < 0.001 |
| BMI (kg/m²) | 0.919 (0.898–0.940) | < 0.001 |
| Exercise habits (ref: No) | 0.819 (0.701–0.956) | 0.012 |
| Tea consumption habits (ref: No) | 0.964 (0.819–1.135) | 0.662 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.894 (0.387–2.065) | 0.793 |
| Regular | 0.880 (0.506–1.532) | 0.652 |
| Hormone medication use (ref: No) | 0.761 (0.621–0.932) | 0.008 |
| Herbal medicine use (ref: No) | 1.237 (0.962–1.590) | 0.097 |
| Female supplement use (Yes vs. No) | 1.057 (0.771–1.449) | 0.730 |
| Secondhand smoke exposure (ref: No) | 1.319 (1.020–1.705) | 0.035 |
| The odds of osteoporosis (N = 2,876) | ||
| Number of deliveries | 0.943 (0.819–1.086) | 0.417 |
| Current menopausal status (ref: No) | 3.260 (1.996–5.327) | < 0.001 |
| Age (year) | 1.118 (1.093–1.145) | < 0.001 |
| BMI (kg/m²) | 0.871 (0.835–0.909) | < 0.001 |
| Exercise habits (ref: No) | 0.618 (0.475–0.804) | < 0.001 |
| Tea consumption habits (ref: No) | 0.805 (0.593–1.093) | 0.164 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 1.082 (0.230–5.088) | 0.921 |
| Regular | 1.098 (0.435–2.766) | 0.843 |
| Hormone medication use (ref: No) | 0.867 (0.636–1.184) | 0.370 |
| Herbal medicine use (ref: No) | 1.266 (0.781–2.054) | 0.339 |
| Female supplement use (Yes vs. No) | 1.257 (0.756–2.092) | 0.378 |
| Secondhand smoke exposure (ref: No) | 1.894 (1.208–2.968) | 0.005 |
| The odds of low BMD level (N = 4,106) | ||
| Number of deliveries | 0.975 (0.904–1.052) | 0.514 |
| Current menopausal status (ref: No) | 2.099 (1.675–2.629) | < 0.001 |
| Age (year) | 1.065 (1.052–1.078) | < 0.001 |
| BMI (kg/m²) | 0.911 (0.891–0.931) | < 0.001 |
| Exercise habits (ref: No) | 0.775 (0.669–0.898) | < 0.001 |
| Tea consumption habits (ref: No) | 0.930 (0.796–1.086) | 0.358 |
| Alcohol consumption (ref: Never) | ||
| Occasional | 0.930 (0.418–2.069) | 0.858 |
| Regular | 0.915 (0.547–1.531) | 0.735 |
| Hormone medication use (ref: No) | 0.788 (0.653–0.951) | 0.010 |
| Herbal medicine use (ref: No) | 1.258 (0.990–1.599) | 0.060 |
| Female supplement use (Yes vs. No) | 1.102 (0.821–1.481) | 0.517 |
| Secondhand smoke exposure (ref: No) | 1.399 (1.098–1.781) | 0.007 |
Abbreviations: ref reference, OR Odds ratio, CI Confidence interval, BMI Body mass index
Statistically significant associations (p < 0.05) are shown in bold
Discussion
In this population-based study of Taiwanese women, several reproductive characteristics were associated with BMD status in the univariate analyses. However, after adjustment for known osteoporosis-related risk factors, most reproductive characteristics were no longer independently associated with osteopenia, osteoporosis, or low BMD. Interestingly, pregnancy history remained independently associated with osteoporosis, whereas the number of pregnancies, delivery history, and number of deliveries did not. These findings suggest that the relationship between reproductive history and bone health may be largely explained by established demographic and lifestyle factors rather than reproductive experience itself.
An interesting finding of this study was that pregnancy history showed a significant association with osteoporosis in both the univariable and multivariable analyses; however, the direction of the association changed after adjustment for potential confounders. In the unadjusted model, women with a history of pregnancy had higher odds of osteoporosis, whereas pregnancy history became inversely associated with osteoporosis after adjustment. This reversal suggests the presence of substantial confounding, particularly by menopausal status, which was strongly associated with the odds of osteoporosis in the adjusted model. Women with a history of pregnancy are generally older and more likely to be postmenopausal than women who have never been pregnant. Because both aging and menopause are strongly associated with bone loss, the crude association may have overestimated the detrimental effect of pregnancy on bone health. After controlling for these factors, pregnancy history appeared to be associated with a lower odd of osteoporosis. This finding may reflect long-term physiological adaptations following pregnancy or may indicate that women who successfully completed pregnancy represent a healthier subgroup with more favorable hormonal or metabolic profiles. Nevertheless, given the cross-sectional nature of the study, the underlying mechanisms remain unclear and warrant further investigation.
Interestingly, pregnancy history remained significantly associated with osteoporosis, whereas delivery history and the number of deliveries did not. One possible explanation is that the physiological adaptations affecting bone metabolism occur primarily during pregnancy rather than at the time of childbirth itself. Pregnancy is accompanied by substantial hormonal changes, increased calcium demands, and alterations in bone remodeling to support fetal skeletal development. These adaptations may influence long-term bone health regardless of whether the pregnancy ultimately results in childbirth.
We found that a history of pregnancy was independently associated with lower odds of osteoporosis after adjustment for potential confounding factors, regardless of childbirth history (Tables 3, 4, 5 and 6). This finding differed from the results of several previous studies investigating the relationship between reproductive factors and bone health. Previous studies have reported inconsistent findings regarding the effects of reproductive history on BMD and osteoporosis risk. Some studies have suggested that adolescent pregnancy, frequent pregnancies, and multiple births may adversely affect maternal bone health [21] and contribute to lower postmenopausal BMD [7]. In contrast, other studies have reported protective effects of multiparity and breastfeeding, particularly among women aged 49–54 years [8]. Furthermore, Sioka et al. found no significant association between obstetric history, including childbirths and miscarriages, and the risk of osteopenia or osteoporosis [9]. One possible explanation for these inconsistent findings is the influence of confounding factors, particularly age and menopausal status. In our study, pregnancy history was positively associated with osteoporosis in the univariable analysis but became inversely associated after multivariable adjustment. This reversal suggests that the crude association may have been substantially confounded by age-related factors. Women with a history of pregnancy are generally older and more likely to be postmenopausal than women who have never been pregnant. Because aging and menopause are among the strongest determinants of bone loss, failure to adequately control for these factors may overestimate the adverse effects of reproductive history on osteoporosis. Although the underlying biological mechanisms remain unclear, our findings suggest that pregnancy history itself may not adversely affect long-term bone health and may even be associated with a lower odd of osteoporosis after accounting for potential confounding factors. Further prospective studies are needed to clarify the causal relationship between reproductive history and osteoporosis.
In addition, several studies have examined changes in BMD during pregnancy. Pregnancy is characterized by increased bone turnover with a notable rise in bone resorption during the first and second trimesters followed by enhanced bone formation in the third trimester [22]. The study reported a significant reduction in BMD during pregnancy, with lumbar spine BMD decreasing by approximately 3.5% over the course of 9 months [22]. Another study observed an 8.1% lower femoral BMD in pregnant women than in non-pregnant controls [23]. Despite these changes, osteoporosis and fragility fractures remain relatively uncommon during pregnancy and lactation, and conservative management is typically sufficient [24]. While pregnancy and lactation can cause temporary bone loss, BMD usually recovers within 12 months after weaning [24]; nonetheless, if the observed reductions in BMD during pregnancy play a role in the prevention of osteoporosis, further studies are warranted to validate this potential protective association.
Several risk factors have been reported to be associated with osteoporosis risk. Vergatti et al. (2024a, 2024b) demonstrated that passive smokers face a risk of osteoporosis similar to that of current smokers where higher rates of fragility fractures and vitamin D deficiency were observed in both groups compared to non-smokers [25, 26]. Our results also showed that women exposed to secondhand smoke had higher odds of osteoporosis or low BMD level (Table 3). Recent epidemiological evidence indicates that tea consumption may confer protective benefits against osteoporosis and related fractures. For instance, a large-scale longitudinal study by Huang et al. (2023) demonstrated that habitual tea intake was associated with a reduced risk of osteoporosis and hip fractures, with particularly strong effects observed among women and middle-aged adults [27]. Consistently, Zhang et al. (2024) analyzed data from the UK Biobank and found that moderate tea consumption (approximately 3 to four cups per day) corresponded to a lower incidence of osteoporosis [28]. Moreover, non-fermented teas—especially green tea—appeared to offer notable advantages for bone health: Hsu et al. (2022) reported that elderly Chinese men who regularly consumed green tea exhibited higher BMD compared to non-drinkers [29], while Lee et al. (2021) observed that Korean postmenopausal women with consistent green tea intake had significantly lower odds of both osteopenia and osteoporosis relative to those who did not consume green tea [30]. Collectively, these findings suggest that tea—and in particular green tea—may play a role in maintaining skeletal integrity and reducing osteoporosis risk across diverse populations.
Despite accumulating evidence suggesting a protective effect of tea consumption on bone health, we did not observe a significant association between tea consumption habits and osteoporosis risk after adjustment for potential confounders. Several explanations may account for this discrepancy. First, the Taiwan Biobank only collected information on tea consumption as a binary variable (yes/no) and did not provide detailed data regarding tea type, frequency, duration, or quantity consumed. Previous studies have suggested that the beneficial effects on bone health may vary according to tea type, with green tea appearing to provide the greatest protective effect [29]. Therefore, exposure misclassification may have attenuated any true association. Second, the protective effect of tea may depend on habitual and long-term consumption. Because information on cumulative intake was unavailable, occasional tea drinkers and regular tea consumers were analyzed together, potentially diluting the effect estimate. Finally, the influence of tea consumption on bone health may be relatively modest compared with major determinants of osteoporosis, such as age, menopausal status, and BMI. After adjustment for these established risk factors, the independent contribution of tea consumption may no longer be detectable.
In addition, exercise is essential for both prevention and management of osteoporosis. Engaging in regular physical activity helps strengthen bones, lowers the risk of falls, and improves quality of life for individuals affected by osteoporosis [31]. Consistent with these findings, our results also showed that women who engaged in regular exercise had a reduced odd of osteoporosis (Table 3). BMI is also associated with the odd of osteoporosis. A study involving the Taiwanese population reported that underweight individuals had a higher prevalence and incidence of osteoporosis compared to those with normal weight [32]. Consistently, our results demonstrated a negative association between BMI and the odds of osteoporosis (Table 3).
Strengths and limitations
The study’s main strength lies in its use of a well-characterized, community‐based cohort from the Taiwan Biobank, with several known confounding factors—allowing for rigorous adjustment for potential confounding factors. Despite the valuable insights provided by our study, several limitations should be acknowledged. Firstly, the cross-sectional design limited our ability to establish causal relationships between reproductive factors and osteoporosis risk; and secondly, although the Taiwan Biobank provides a large and well-characterized dataset, it comprises only community-dwelling volunteers, which might introduce selection bias and limit the generalizability of our findings to the broader population. Thirdly, the number of women with osteoporosis, particularly within certain pregnancy-history subgroups, was relatively small, which may have limited the statistical power and contributed to wider confidence intervals. Additionally, self-reported data on lifestyle behaviors and reproductive history are subject to recall bias and potential misclassification. Finally, although we adjusted for several known confounders, residual confounding by unmeasured variables (such as calcium or vitamin D intake, genetic factors or comorbidities) cannot be entirely excluded.
Conclusion
Among Taiwanese women, several reproductive characteristics were associated with BMD status in univariable analyses; however, most of these associations were attenuated after adjustment for known BMD-related factors. Pregnancy history remained independently associated with lower odds of osteoporosis, whereas delivery history, number of pregnancies, and number of deliveries were not independently associated with the odds of osteoporosis. These findings suggest that the relationship between reproductive history and bone health may be largely influenced by established demographic and lifestyle factors. Further longitudinal studies are needed to confirm the observed association between pregnancy history and osteoporosis.
Acknowledgements
None.
Abbreviations
- ANOVA
Analysis of variance
- aOR
adjusted odds ratio
- BMI
Body mass index
- CI
Confidence interval
- EGC
Ethics and Governance Council
- IRB
Institutional Review Board
- OR
Odds ratio
- SD
Standard deviation
- TBW
Taiwan Biobank
Authors’ contributions
Y.E.C. analyzed the data and wrote the first draft manuscript. C.Y.H. wrote the first draft manuscript. C.L.S. and S.S.H. manuscript revision and final approval.
Funding
This study was supported by a grant from Ditmanson Medical Foundation Chia-Yi Christian Hospital (Grant No. R114-010).
Data availability
The data that support the findings of this study were obtained from the Taiwan Biobank (TWB), a population-based research database that contains genomic, phenotypic, and lifestyle information from Taiwanese participants. Due to legal and ethical restrictions imposed by the Taiwan Biobank and the Personal Data Protection Act in Taiwan, the dataset cannot be made publicly available. Researchers who wish to access the data must submit a formal application to the Taiwan Biobank Data Release Committee. Detailed information regarding data access procedures, application requirements, and relevant regulations is available on the official Taiwan Biobank website (https://www.twbiobank.org.tw). Access to the data requires approval of the research proposal and the signing of a data use agreement with the Taiwan Biobank.
Declarations
Ethics approval and consent to participate
This study was approved by the Institutional Review Board of Ditmanson Medical Foundation Chiayi Christian hospital (approval number: IRB2024124) and by the Taiwan Biobank (approval number: TWBR11404-06). All participants provided written informed consent in compliance with ethical requirements and the principles outlined in the Declaration of Helsinki.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Yuen-En Chang and Song-Shan Huang contributed equally to this work and share first authorship.
References
- 1.Anam AK, Insogna K. Update on Osteoporosis Screening and Management. Med Clin North Am. 2021;105(6):1117–34. [DOI] [PubMed] [Google Scholar]
- 2.Bessette L, Ste-Marie LG, Jean S, Davison KS, Beaulieu M, Baranci M, Bessant J, Brown JP. The care gap in diagnosis and treatment of women with a fragility fracture. Osteoporos Int. 2008;19(1):79–86. [DOI] [PubMed] [Google Scholar]
- 3.Liu E. Hip fractures: mortality, economic burden, and organisational factors for improved patient outcomes. Lancet Healthy Longev. 2023;4(8):e360–1. [DOI] [PubMed] [Google Scholar]
- 4.Suvvari TK, Thomas V, Nazir A, Parvathy G, Thomas L. From fragility to resilience: Advancing early identification and management of osteoporosis in post-menopausal women. Womens Health (Lond). 2023;19:17455057231194140. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Charde SH, Joshi A, Raut J. A Comprehensive Review on Postmenopausal Osteoporosis in Women. Cureus. 2023;15(11):e48582. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Khalid N, Yousaf Q, Shehzadi L, Fatima S. Postmenopausal Osteoporosis: A Primary Women Health Concern. In: 2021; 2021.
- 7.Kaya AE, Doğan O, Başbuğ A, Sönmez CI, Sungur MA, Ataoğlu S. An Evaluation of the Association of Reproductive History and Multiple Births during Adolescence with Postmenopausal Osteoporosis. Geburtshilfe Frauenheilkd. 2019;79(3):300–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Schnatz PF, Marakovits KA, O’Sullivan DM. Assessment of postmenopausal women and significant risk factors for osteoporosis. Obstet Gynecol Surv. 2010;65(9):591–6. [DOI] [PubMed] [Google Scholar]
- 9.Sioka C, Bougias C, Papadopoulos A, Fotopoulos A. Is osteoporosis in postmenopausal female patients related to previous pregnancies and/or miscarriages? Climacteric. 2007;10(5):381–5. [DOI] [PubMed] [Google Scholar]
- 10.Lin TY, Chen SC, Geng JH, Tsai HJ. Living alone decreased calcaneus ultrasound T-score in a large Taiwanese population follow-up study. Front Public Health. 2022;10:1004794. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Damilakis J, Perisinakis K, Gourtsoyiannis N. Imaging ultrasonometry of the calcaneus: optimum T-score thresholds for the identification of osteoporotic subjects. Calcif Tissue Int. 2001;68(4):219–24. [DOI] [PubMed] [Google Scholar]
- 12.Aspray TJ, Hill TR. Osteoporosis and the Ageing Skeleton. Subcell Biochem. 2019;91:453–76. [DOI] [PubMed] [Google Scholar]
- 13.Cai S, Gao J, Li C, Lu F, Hu D, Luo X, Zhao Z, Pan Y, Duan G. Association of body mass index and bone mineral density in postmenopausal women: a systematic review and meta-analysis. Ann Med. 2025;57(1):2561227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sheng Y-H, Chen J-H, Chiou J-M, Tsai K-S, Lee Y-Y, Tsao C-K, Chen Y-C. Association of Renal Function and Menopausal Status with Bone Mineral Density in Middle-aged Women. Sci Rep. 2015;5(1):14956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cong B, Zhang H. The effects of combined calcium and vitamin D supplementation on bone mineral density and fracture risk in postmenopausal women with osteoporosis: a systematic review and meta-analysis of randomized controlled trials. BMC Musculoskelet Disord. 2025;26(1):928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ping R, Quanwu W, Wei B, Miao S, Zheling L, Ming G, Liang W, Bo P, Liguang XU. Identifying the effective combination of acupuncture and traditional Chinese medicinal herbs for postmenopausal osteoporosis therapy through studies of their molecular regulation of bone homeostasis. J Tradit Chin Med. 2024;44(1):212–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Berg KM, Kunins HV, Jackson JL, Nahvi S, Chaudhry A, Harris KA Jr., Malik R, Arnsten JH. Association between alcohol consumption and both osteoporotic fracture and bone density. Am J Med. 2008;121(5):406–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chen L, Wang J, Wan D. Association between secondhand smoke exposure and osteoporosis risk in postmenopausal women: a cross-sectional analysis of NHANES data. J Obstet Gynaecol. 2025;45(1):2482708. [DOI] [PubMed] [Google Scholar]
- 19.Kemmler W, Shojaa M, Kohl M, von Stengel S. Effects of Different Types of Exercise on Bone Mineral Density in Postmenopausal Women: A Systematic Review and Meta-analysis. Calcif Tissue Int. 2020;107(5):409–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhou F, Wang T, Li L, Yu J, Liu Z, Zhang J, Wang G, Li J, Shao C, Wang P, et al. Tea consumption and risk of bone health: an updated systematic review and meta-analysis. J Bone Min Metab. 2024;42(1):99–114. [DOI] [PubMed] [Google Scholar]
- 21.Cho GJ, Shin JH, Yi KW, Park HT, Kim T, Hur JY, Kim SH. Adolescent pregnancy is associated with osteoporosis in postmenopausal women. Menopause. 2012;19(4):456–60. [DOI] [PubMed] [Google Scholar]
- 22.Black AJ, Topping J, Durham BH, Farquharson RG, Fraser WD. A detailed assessment of alterations in bone turnover, calcium homeostasis, and bone density in normal pregnancy. J Bone Miner Res. 2000;15(3):557–63. [DOI] [PubMed]
- 23.Valentina Anna D, Maria Luisa B, Greta C, Sergio C, Delia C, Francesco C, Pasquo Elvira D, Stefano G, Fiorella Anna L, Paola P, et al. First assessment of bone mineral density in healthy pregnant women by means of Radiofrequency Echographic Multi Spectrometry (REMS) technology. Eur J Obstet Gynecol Reprod Biol. 2021;263:44–9. [DOI] [PubMed] [Google Scholar]
- 24.Capozzi A, Scambia G, Lello S. Bone metabolism in pregnancy and lactation. Minerva obstetrics and gynecology. 2021;73(6):697–703. [DOI] [PubMed]
- 25.Vergatti A, Abate V, Giaquinto A, Altavilla N, D’Elia L, Evangelista M, De Filippo G, Piccinocchi G, Gennari L, Merlotti D, et al. Role of active and environmental tobacco smoke on susceptibility to osteoporosis in women undergoing dual-X-ray absorptiometry. J Endocrinol Invest. 2024;47(4):937–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Vergatti A, Abate V, D’Elia L, De Filippo G, Piccinocchi G, Gennari L, Merlotti D, Galletti F, Strazzullo P, Rendina D. Smoking habits and osteoporosis in community-dwelling men subjected to dual-X-ray absorptiometry: a cross-sectional study. J Endocrinol Invest. 2024;47(12):3129–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Huang YP, Chen LS, Feng SH, Liang YS, Pan SL. Tea consumption and the risks of osteoporosis and hip fracture: a population-based longitudinal follow-up study. Osteoporos Int. 2023;34(1):101–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Zhang S, Wu S, Xia B, He Q, Mi N, Zhao J, Hu L, Wang D, Zheng L, Sheng P, et al. Association of coffee and tea consumption with osteoporosis risk: A prospective study from the UK biobank. Bone. 2024;186:117135. [DOI] [PubMed] [Google Scholar]
- 29.Hsu CL, Huang WL, Chen HH, Cheng-Yen Lai J. Non-fermented tea consumption protects against osteoporosis among Chinese male elders using the Taiwan biobank database. Sci Rep. 2022;12(1):7382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lee DB, Song HJ, Paek YJ, Park KH, Seo YG, Noh HM. Relationship between regular green tea intake and osteoporosis in Korean postmenopausal women: a nationwide study. Nutrients. 2021;14(1):87. [DOI] [PMC free article] [PubMed]
- 31.Brooke-Wavell K, Skelton DA, Barker KL, Clark EM, De Biase S, Arnold S, Paskins Z, Robinson KR, Lewis RM, Tobias JH, et al. Strong, steady and straight: UK consensus statement on physical activity and exercise for osteoporosis. Br J Sports Med. 2022;56(15):837–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Chiu C-T, Lee J-I, Lu C-C, Huang S-P, Chen S-C, Geng J-H. The association between body mass index and osteoporosis in a Taiwanese population: a cross-sectional and longitudinal study. Sci Rep. 2024;14(1):8509. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study were obtained from the Taiwan Biobank (TWB), a population-based research database that contains genomic, phenotypic, and lifestyle information from Taiwanese participants. Due to legal and ethical restrictions imposed by the Taiwan Biobank and the Personal Data Protection Act in Taiwan, the dataset cannot be made publicly available. Researchers who wish to access the data must submit a formal application to the Taiwan Biobank Data Release Committee. Detailed information regarding data access procedures, application requirements, and relevant regulations is available on the official Taiwan Biobank website (https://www.twbiobank.org.tw). Access to the data requires approval of the research proposal and the signing of a data use agreement with the Taiwan Biobank.