Abstract
Background:
Adverse reproductive outcomes (AROs) in women can lead to the occurrence of a variety of diseases later in life. However, research on AROs and dementia risk in women has not been reported.
Objective:
This study explored the effects of miscarriage and stillbirth on future dementia risk in women.
Methods:
The Cox proportional hazards model was used to clarify the association between miscarriage, stillbirth, and dementia risk. In this cohort, only women with a history of miscarriage and stillbirth were selected. A genetic risk score for dementia was constructed, and the combined effect of miscarriage, stillbirth, and the genetic risk score for dementia on the future risk of incident dementia was determined.
Results:
For each increase in the number of miscarriages and stillbirths, the risk of dementia increased by 5% and 22%, respectively. Compared to women who had a low genetic risk score, no miscarriages and at least one live-born infant, women with more than 3 miscarriages and a high genetic risk score had a significantly increased risk of dementia.
Conclusions:
Our results indicate that miscarriage and stillbirth are associated with an increased risk of dementia, especially in women with a high genetic risk score.
Keywords: Alzheimer's disease, dementia, genetic susceptibility, miscarriage, stillbirth, UK biobank
Introduction
Dementia is a neurodegenerative disease characterized by progressive deterioration of memory, thinking skills, behavior, and mood. 1 Due to the increase in the percentage of the aged population, dementia has become a major global challenge and places a heavy burden on families and society. Approximately 50 million people worldwide are currently living with dementia. This number is projected to surge to 152 million by 2050. 2 After the age of 45 years, women are reportedly at least twice as likely to develop dementia as men.3,4 Furthermore, women are at greater risk of developing dementia, not only because they have a longer life expectancy than men. Chronic stress, metabolic health, and reproductive history may be the reasons,5–7 but there are few relevant studies.
Unlike men's hormones, women's hormones change drastically throughout their life cycle. The female reproductive period begins with the onset of menstruation and ends with a dramatic decrease in estrogen levels during menopause. Estrogen is known to have both neuroprotective and neurodamaging properties. 8 Epidemiological studies have shown that decreasing estrogen levels may lead to an increased risk of cognitive impairment in women. 9 However, most previous studies have focused on limited female reproductive factors, such as menopause status 10 and estrogen use, 11 in relation to cognitive function and dementia. Few studies have focused on the long-term effects of adverse reproductive outcomes (AROs), such as miscarriage and stillbirth, on dementia risk.
Some studies have shown that experiencing multiple pregnancies is associated with a high risk of dementia, and that having fewer pregnancies is associated with a lower risk of dementia later in life. 12 Another study showed that stillbirth was associated with an increased risk of dementia. 13 However, other studies have shown that stillbirth and miscarriage are not associated with dementia risk. 14 Therefore, conclusions about the relationship between AROs and future dementia risk have yet to be fully characterized. In addition, studies on the combined and interaction effects of AROs and dementia risk are scarce.
In this study, we utilized data from the UK Biobank (UKBB) to examine the association between AROs and future dementia risk in women. In addition, we assessed the association of the combination of AROs and genetic factors with the risk of dementia.
Methods
Study population
Women were drawn from the UKBB cohort. All participants in the study provided informed consent, and ethical approval for the UKBB research study was obtained from the North West Multi-Center Research Ethics Committee, the National Information Governance Board for Health and Social Care, and the Community Health Index Advisory Group. Participants provided various health-related information through touchscreen questionnaires, underwent physical measurements, and provided biological samples at baseline and during longitudinal follow-up. The inclusion criteria for this study were female sex, available reproductive history information, and follow-up data. The exclusion criterion was a diagnosis of dementia before enrolment. Figure 1 shows the detailed flow chart of participant inclusion.
Figure 1.
Flow diagram for inclusion of participants. ARO: adverse reproductive outcome; PRS: polygenic risk score.
ARO assessment
In the present study, information on AROs, including miscarriages and stillbirths, was obtained via a questionnaire. The specific questionnaires used were from data fields 3839 and 3829, and the questions were “How many spontaneous miscarriages have you had?” and “How many stillbirths have you had?”, respectively. We set at least one live birth and no miscarriages or stillbirths as the control setting; thus, we also included information on live births (data field 2734). The specific question was “How many children have you given birth to? (please include live-born children only)”. For the miscarriage exposure variables, we set at least one or more live births and no miscarriages as the reference (0), one miscarriage was set as 1, two miscarriages were set as 2, and three or more miscarriages were set as 3. For the stillbirth exposure variables, at least one or more live births and no stillbirths were set as the reference (0), one stillbirth was set as 1, and two or more stillbirths were set as 2. Supplemental Tables 1 and 2 show the numbers of women with miscarriages, women with stillbirths, and controls.
Dementia assessment
According to previous literature, 15 women were identified using International Classification of Diseases, Ninth Revision (ICD-9) and International Classification of Diseases, Tenth Revision (ICD-10) codes for dementia (ICD-9: 290.2, 290.3, 291.2, 294.1, 331.1, 331.2, 331.5; ICD-10: A81.0, F02, F02.0, F02.1, F02.2, F02.3, F02.4, F02.8, F03, F05.1, F10.6, G31.0, G31.1, G31.8). We calculated the follow-up times for subjects as the number of days from the assessment date until the diagnosis of dementia or the censorship date, according to the date of hospital admission.
Definition of the polygenic risk score
The dementia polygenic risk score (PRS) captures an individual's load of common genetic variants associated with Alzheimer's disease. According to a previous genome-wide association study (GWAS) on Alzheimer's disease, 16 we selected SNPs with p values less than 5 × 10−8 and their beta (β) or odd ratio values to construct the PRS for dementia. The constructed formula was as follows: PRS = β1 × SNP1 + β2 × SNP2 + … + βK × SNPk + βN × SNPn, where n is the total number of SNPs and β is the value of the natural logarithm of the odds ratio for Alzheimer's disease associated with the SNPs. PRSs generated by p values were classified into tertiles: low (lowest tertile), medium (middle tertile), and high (highest tertile).
Definitions of covariates
Four types of covariates, including sociodemographic factors, lifestyle factors, diseases, and genetic factors, were adjusted for in the statistical analysis. Among the sociodemographic factors, age and the Townsend Deprivation Index were continuous variables, race was divided into white and nonwhite, and body mass index (BMI) was divided into less than 25 kg/m2, 25–30 kg/m2, greater than or equal to 30 kg/m2, and missing. For the lifestyle factors, diet was divided into healthy and unhealthy. Drinking and smoking status were classified as never, previous, current, and missing. The number of metabolic equivalent task (MET) minutes, adopted to assess physical activity levels, was a continuous variable. For the disease variable, systolic blood pressure (SBP) and total cholesterol level were continuous variables. Diabetes status was determined based on participant reports of doctor-diagnosed diabetes at the initial assessment visit. The use of antihypertensive drugs or lipid-lowering drugs was a categorical variable for patients who were regularly taking medication for hypertension or hyperlipidemia, respectively. For the genetic factor variable, the PRS was a continuous variable, the genotyping batch was a categorical variable, and the first 10 genetic principal components were continuous variables. In further statistical analysis, according to previous literature, 17 we used a variety of variables to construct a score for predicting dementia based on the risk of miscarriage or stillbirth. The predictors of dementia scores included age, education level, blood pressure, total cholesterol level, BMI, and the number of MET minutes. The relevant variable descriptions and weight coefficients are shown in Supplemental Table 3.
Statistical analysis
Continuous data are presented as the means ± standard deviations. Categorical data are expressed as percentages. Because of the large sample size, a t-test was used to compare the continuous variables. The chi-square test was used to compare the categorical variables. We generated Kaplan‒Meier plots to visualize survival curves to determine the effect of different numbers of miscarriages and stillbirths on dementia risk. After adjusting for other covariates, a Cox proportional hazards model was used to examine the association between the number of miscarriages or stillbirths and the risk of dementia. We compared different numbers of miscarriages or stillbirths (as a reference for more than one live birth with no miscarriages or stillbirths), as well as the per miscarriage and per stillbirth rates. For the association between the PRS for dementia and the risk of dementia, we also used the Cox proportional hazard model. We added the genotype batch and the first ten genetic principal components in the covariate adjustment. In addition, we also evaluated the effect of the number of miscarriages or stillbirths combined with the PRS for dementia on the risk of dementia. Furthermore, we used indicators (relative excess risk due to interaction [RERI]; attributable proportion due to interaction [AP]) to explore additive interactions. We used several variables to construct the score for incident dementia and added the variables of the number of miscarriages and stillbirths to compare the different predictive efficacy. We also conducted a sensitivity analysis to ensure the robustness of the results. R software (version 4.30) was used for all analyses. Two-sided p values less than 0.05 were considered to indicate statistical significance.
Results
A total of 223,174 nondementia patients and 4048 dementia patients were included in this study. The median follow-up time was 13.7 years. The results in Table 1 show that the age of dementia patients was greater than that of nondementia patients (64.3 ± 4.5 and 56.6 ± 8.0 years, respectively; p < 0.001). In the dementia group, compared with those in the nondementia group, the number of women with obesity, diabetes, high systolic blood pressure, and high total cholesterol levels were greater, and the number of women using antihypertensive and lipid-lowering drugs was greater. Supplemental Tables 1 and 2 show the numbers of miscarriages and stillbirths, respectively. The results showed that the number of women with one miscarriage was 39,897, the number of women with two miscarriages was 10,189, and the number of women with three or more miscarriages was 5641. The number of women with one stillbirth was 6,053, and the number of women with two or more stillbirths was 959.
Table 1.
Descriptive characteristics of participants in the UK Biobank study by dementia.
| Incident dementia | |||
|---|---|---|---|
| No | Yes | P* | |
| No. | 223,174 | 4048 | |
| Age (years) | 56.6 ± 8.0 | 64.3 ± 4.5 | <0.001 |
| Race, white (%) | 91.0 | 92.7 | 0.040 |
| BMI (kg/m2) | <0.001 | ||
| <25 | 38.7 | 34.9 | |
| 25–30 | 37.2 | 39.3 | |
| ≥30 | 23.6 | 25.7 | |
| missing | 0.5 | 1.0 | |
| Townsend Deprivation Index | −1.4 ± 3.0 | −1.0 ± 3.3 | <0.001 |
| Smoke, (%) | <0.001 | ||
| Never | 58.6 | 53.3 | |
| Previous | 32.0 | 36.2 | |
| Current | 9.0 | 9.8 | |
| Missing | 0.4 | 0.7 | |
| Drink, (%) | <0.001 | ||
| Never | 5.7 | 10.0 | |
| Previous | 3.6 | 6.7 | |
| Current | 90.5 | 83.2 | |
| Missing | 0.2 | 0.1 | |
| MET, (min/week) | 2562.4 ± 2475.5 | 2799.5 ± 2774.0 | <0.001 |
| Diet, health (%) | 88.7 | 89.5 | 0.101 |
| Prevalent diabetes (%) | 3.8 | 10.4 | <0.001 |
| Systolic blood pressure (mmHg) | 137.5 ± 20.3 | 145.6 ± 20.7 | <0.001 |
| Total cholesterol (mmol/L) | 5.9 ± 1.1 | 5.9 ± 1.3 | 0.001 |
| Antihypertensive drugs | <0.001 | ||
| Used, % | 17.8 | 35.4 | |
| Lipid-lowering drugs | <0.001 | ||
| Used, % | 12.8 | 30.1 | |
Data are presented as the mean ± standard deviation (SD) and percentages.
Abbreviations: BMI, body mass index; MET, Metabolic Equivalent Task.
*t-tests or chi-square tests obtained P values.
Table 2 shows the number of women with miscarriages, stillbirths, and dementia. After adjusting for age and race, the risk of dementia increased with each increase in the number of miscarriages. The hazard ratio (HR) (95% confidence interval (CI)) was 1.06 (1.01, 1.12), and the p value was 0.010. The risk of dementia was 29% (95% CI, 7% to 56%) greater than that in the women with three or more miscarriages and at least one live birth. There was no significant difference in the risk of dementia in the women with two miscarriages. After additional adjustment for other covariates, the risk of dementia was similarly increased for each increase in the number of miscarriages. The HR (95% CI) was 1.05 (1.01, 1.10), and the p value was 0.028. After adjusting for age and race, the risk of dementia increased by 29% for each increase in the number of stillbirths. The risk of dementia in women with two or more stillbirths was greater than that in women with no stillbirths and at least one live birth. The HR (95% CI) was 2.12 (1.55, 2.90), and the p value was less than 0.001. After additional adjustment for other covariates, the risk of dementia increased by 22% for each additional stillbirth. The risk of dementia was greater in women with two or more stillbirths than in women with no stillbirths and at least one live birth. The HR (95% CI) was 1.74 (1.28, 2.39), and the p value was less than 0.001. Supplemental Figures 1 and 2 show the Kaplan–Meier survival plots for the number of women with miscarriages, stillbirths, and dementia, respectively.
Table 2.
Adjusted hazard ratio and 95% confidence intervals of dementia by miscarriages and stillbirths.
| Model 1 | Model 2 | ||||
|---|---|---|---|---|---|
| Cases/Total | HR (95%CI) | P | HR (95%CI) | P | |
| Number of miscarriages | |||||
| 0# | 3108/162,423 | Reference | Reference | ||
| 1 | 614/39,283 | 0.90 (0.83, 0.98) | 0.019 | 0.90 (0.83, 0.98) | 0.019 |
| 2 | 166/10,023 | 1.01 (0.86, 1.18) | 0.935 | 0.98 (0.84, 1.15) | 0.805 |
| ≥3 | 113/5528 | 1.29 (1.07, 1.56) | 0.008 | 1.17 (0.97, 1.41) | 0.100 |
| Per miscarriage | 1.06 (1.01, 1.12) | 0.010 | 1.05 (1.01, 1.10) | 0.028 | |
| Number of stillbirths | |||||
| 0# | 3773/206,865 | Reference | Reference | ||
| 1 | 159/5894 | 1.19 (1.01, 1.39) | 0.034 | 1.09 (0.93, 1.28) | 0.296 |
| ≥2 | 40/919 | 2.12 (1.55, 2.90) | <0.001 | 1.74 (1.28, 2.39) | <0.001 |
| Per stillbirth | 1.29 (1.19, 1.41) | <0.001 | 1.22 (1.11, 1.34) | <0.001 |
Model 1 was adjusted for age (continuous) and race (white and non-white).
Model 2 was adjusted as age (continuous), race (white and non-white), body mass index (<25 kg/m2, 25 to29.9 kg/m2, ≥30 kg/m2, and missing), smoke status (never, previous, current and missing), drink status (never, previous, current and missing), physical activity (continuous, MET-min/week), diet condition (healthy and unhealthy), diabetes (yes and no), coronary artery disease (yes and no), systolic blood pressure (continuous), total cholesterol (continuous), antihypertensive drugs used (yes and no), and lipid-lowering medications used (yes and no).
The case group means participants who had dementia.
# At least one live birth without stillbirth or miscarriage.
HR, hazard ratio. CI, confidence interval.
Considering the contribution of genetic factors to the risk of dementia, we further evaluated the risk of dementia associated with the combined genetic factors of miscarriage and stillbirth. Compared with a low dementia genetic risk score, moderate and high dementia genetic risk scores increased the risk of dementia by 20% and 48%, respectively, and similar results were found in the two different adjusted models. In addition, the risk of dementia increased by 87% for each additional genetic risk score (Supplemental Table 4). The combined results of miscarriage and genetic risk showed that compared with that in women with low genetic risk, no miscarriages and at least one live birth, the risk of dementia in women with moderate genetic risk and three or more miscarriages was 1.80 (1.32,2.46), while the risk of dementia in women with high genetic risk and one, two and three or more miscarriages increased by 34%, 71% and 57%, respectively (Figure 2). The combined results of stillbirth and genetic risk showed that, compared with that in women with low genetic risk, no stillbirths and at least one live birth, the risk of dementia in women with moderate genetic risk and two or more stillbirths was 2.66 (1.50, 4.71). The risk of dementia in women with high genetic risk and one or two or more stillbirths increased by 97% and 80%, respectively (Figure 3). Furthermore, we evaluated the additive effect of miscarriage, stillbirth, and PRSs on the occurrence of dementia. Supplemental Table 5 shows the results of the additive interactions after adjusting for covariates. The results showed that the additive interaction effect of miscarriage, stillbirth, and genetic risk scores for dementia was greater than 0, but there was no significant difference.
Figure 2.
The joint association of miscarriage and genetic risk score with the incidence of dementia (participants with no miscarriages and a low genetic risk score as a reference). No. Of the events, participants who had dementia were included. Adjusted for age, Townsend Deprivation index, alcohol consumption status (current, former, never, missing), smoking status (current, former, never, missing), body mass index (<25, 25–30, ≥30 kg/m2, missing), physical activity level (MET-min/week), healthy diet classification (healthy/unhealthy), diabetes status (yes/no), systolic blood pressure (mmHg), total cholesterol level (continuous), antihypertensive drug use (yes and no), lipid-lowering drug use(yes and no), genotyping batch, and the first ten genetic principal components.
Figure 3.
The joint association of stillbirth and genetic risk score with the incidence of dementia (participants without stillbirths and low genetic risk score as a reference). The number of events in question involved participants who had dementia. Adjusted for age, Townsend Deprivation index, alcohol consumption status (current, former, never, missing), smoking status (current, former, never, missing), body mass index (<25, 25–30, ≥30 kg/m2, missing), physical activity level (MET-min/week), healthy diet classification (healthy/unhealthy), diabetes status (yes/no), systolic blood pressure (mmHg), total cholesterol level (continuous), antihypertensive drug use (yes and no), lipid-lowering drug use (yes and no), genotyping batch, and the first ten genetic principal components.
We calculated dementia risk scores. According to our results in the present study, the AUC was 0.675. After adding the factors of miscarriage and stillbirth, the AUC was 0.688 (Figure 4). After adding the factors of miscarriage and stillbirth alone, the predicted and actual AUCs were 0.672 and 0.689, respectively (Supplemental Figures 3–4).
Figure 4.
Receiver operating characteristic curves for different models in cohort analyses of 227,222 participants over a mean of 13.7 years of follow-up with 4048 incident dementia events.
To check the robustness of the results, we conducted a sensitivity analysis. We excluded participants with a follow-up time of less than 2 years, included different covariates, and replaced the covariates with the dementia risk score; the control group included women who had no miscarriages or stillbirths only (Supplemental Tables 6–9). The results of the above five cases are similar to our above results, indicating the robustness of our results.
Discussion
In this women-based study, we found an association between AROs and the risk of dementia later in life, regardless of whether covariates were adjusted for. The risk of dementia was greater in women who had three or more miscarriages than among those who had no miscarriages and at least one live birth. Similarly, the risk of dementia was greater in women with two or more stillbirths, regardless of whether covariates were adjusted for. Further inclusion of miscarriage and stillbirth in the preexisting dementia risk score 17 was found to improve the ability of the score to predict the risk of developing dementia.
In our study, we observed an association between a history of miscarriage and stillbirth and dementia risk: we found an association between repeated miscarriages (≥3) and stillbirths (≥ 1) and the future risk of developing dementia. This finding is contrary to the recently published results of Gong et al., 14 who reported that stillbirths and spontaneous abortions were not associated with dementia risk. All women without a history of miscarriage or stillbirth, including nonpregnant women, were included in the reference group in their study. 14 Another women-based study showed that miscarriage (either single or recurrent) was not associated with the risk of dementia later in life, and stillbirth was associated with an increased risk of dementia. Further typing of dementia revealed that miscarriage was associated with the risk of vascular dementia. 13 This finding corroborates our results. In our study, the effect of pregnancy itself on the risk of dementia was excluded, and we included women who were pregnant and had regular deliveries as the reference group. However, the number of women affected in these analyses was small, so these results should be interpreted with caution, and further experimental evidence is needed.
Miscarriage and stillbirth may contribute to the risk of maternal dementia later in life through biological processes such as vascular endothelial dysfunction. 18 Endothelial dysfunction may contribute to pregnancy loss by causing defects in placental formation, persisting after recurrent miscarriages, and potentially leading to late-life dementia by damaging the vasculature and increasing the risk of small vessel disease lesions later in life. 19 This is one of the reasons why there is a link between cardiovascular disease and dementia, as many studies have suggested.20,21 Previous studies have also shown that recurrent miscarriage is associated with a future risk of metabolic syndrome. 22 Chronic systemic inflammation due to systemic metabolic disorders affects amyloid distribution homeostasis and neuroinflammation, enhances pathological brain states, and promotes the development of dementia. 23 In addition, miscarriages and stillbirths have been linked to other factors associated with dementia, such as immune responses and genetic polymorphisms.24,25 All these factors may be associated with a further increased risk of dementia.26,27
This study has the following strengths. This study is the first to find an association between adverse pregnancy outcomes and dementia risk. When combined with genetic factors, we found that adverse pregnancy outcomes and genetic factors can jointly increase the risk of dementia, suggesting that early dementia prevention measures should be taken in women with a history of adverse pregnancy outcomes and a high genetic risk of dementia. Our study has the following limitations. First, although this study revealed an increased risk of dementia due to adverse pregnancy outcomes based on a cohort study design, research evidence of Mendelian randomization is still needed to determine the causal relationship between adverse pregnancy outcomes and dementia. However, there are no reliable genetic instrumental variables associated with miscarriage or stillbirth to allow for Mendelian randomization analysis. Second, the fertility outcome was retrospective, which may lead to recall bias. Third, the inability to distinguish between spontaneous and induced miscarriages interfered with the analysis of the association between miscarriage and dementia. More accurate data on birth outcomes are needed for further studies.
Women with repeated spontaneous miscarriages (≥3) and stillbirths (≥1) are at greater risk of developing dementia in the future. Repeated miscarriages and stillbirths should therefore be considered risk factors for dementia. When the incidence of dementia is reduced by selecting high-risk groups, long-term outcomes can improve.
Supplemental Material
Supplemental material, sj-docx-1-alr-10.1177_25424823251370717 for Adverse reproductive outcomes and future risk of incident dementia: The modifying effect of genetic susceptibility by Qiaoqiao Xu, Chengzhe Tao, Sijia Dai, Zhixi Lu, Michael Aschner, Guangfeng Long, Shaojun Li and Cheng Xu in Journal of Alzheimer's Disease Reports
Acknowledgements
The authors would like to thank the UK Biobank participants for dedicating their time toward this project and the UK Biobank team for collecting, preparing, and providing data used in this work. This research was conducted using the UK Biobank resource under application number 140851.
Footnotes
ORCID iD: Cheng Xu https://orcid.org/0000-0003-2605-2100
Author contribution(s): Qiaoqiao Xu: Conceptualization; Data curation; Supervision; Writing – review & editing.
Chengzhe Tao: Data curation; Formal analysis; Writing – original draft.
Sijia Dai: Investigation; Writing – review & editing.
Zhixi Lu: Investigation; Writing – review & editing.
Michael Aschner: Methodology; Writing – review & editing.
Guangfeng Long: Conceptualization; Formal analysis.
Shaojun Li: Data curation; Methodology; Writing – review & editing.
Cheng Xu: Conceptualization; Data curation; Supervision; Writing – review & editing.
Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (82100778), the Guangxi Key Laboratory of Precision Medicine in Cardio-cerebrovascular Diseases Control and Prevention (22-035-18), and the Guangxi Clinical Research Center for Cardio-cerebrovascular Diseases (AD17129014).
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement: After receiving permission under accession number 140851, the data from UK biobank was downloaded via the UKB access management system, as outlined at available via https://www.ukbiobank.ac.uk.
Supplemental material: Supplemental material for this article is available online.
References
- 1.Maloney B, Lahiri DK. Epigenetics of dementia: understanding the disease as a transformation rather than a state. Lancet Neurol 2016; 15: 760–774. [DOI] [PubMed] [Google Scholar]
- 2.Patterson C . World Alzheimer Report 2018. The state of the art of dementia research: New frontiers. London: Alzheimer's Disease International, 2018. [Google Scholar]
- 3.Andersen K, Launer LJ, Dewey ME, et al. Gender differences in the incidence of AD and vascular dementia: the EURODEM studies. EURODEM incidence research group. Neurology 1999; 53: 1992–1997. [DOI] [PubMed] [Google Scholar]
- 4.Li R, Singh M. Sex differences in cognitive impairment and Alzheimer's disease. Front Neuroendocrinol 2014; 35: 385–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lau Y, Bansal A, Palix C, et al. Sex differences in the association between repetitive negative thinking and neurofilament light. NPJ Ment Health Res 2024; 3: 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bae JB, Lipnicki DM, Han JW, et al. Does parity matter in women's risk of dementia? A COSMIC collaboration cohort study. BMC Med 2020; 18: 210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Luo R, Zeraatkar D, Glymour M, et al. Specification curve analysis to identify heterogeneity in risk factors for dementia: findings from the UK biobank. BMC Med 2024; 22: 216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Strom JO, Theodorsson A, Theodorsson E. Mechanisms of estrogens’ dose-dependent neuroprotective and neurodamaging effects in experimental models of cerebral ischemia. Int J Mol Sci 2011; 12: 1533–1562. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Song X, Wu J, Zhou Y, et al. Reproductive and hormonal factors and risk of cognitive impairment among Singapore Chinese women. Am J Obstet Gynecol 2020; 223: 410 e411–410 e423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sandyk R. Estrogens and the pathophysiology of Parkinson's disease. Int J Neurosci 1989; 45: 119–122. [DOI] [PubMed] [Google Scholar]
- 11.Popat RA, Van Den Eeden SK, Tanner CM, et al. Effect of reproductive factors and postmenopausal hormone use on the risk of Parkinson disease. Neurology 2005; 65: 383–390. [DOI] [PubMed] [Google Scholar]
- 12.Jang H, Bae JB, Dardiotis E, et al. Differential effects of completed and incomplete pregnancies on the risk of Alzheimer disease. Neurology 2018; 91: e643–e651. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Basit S, Wohlfahrt J, Boyd HA. Pregnancy loss and risk of later dementia: a nationwide cohort study, Denmark, 1977–2017. Alzheimers Dement (N Y) 2019; 5: 146–153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Gong J, Harris K, Peters SAE, et al. Reproductive factors and the risk of incident dementia: a cohort study of UK biobank participants. PLoS Med 2022; 19: e1003955. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hu M, Zhang K, Su KJ, et al. Unveiling the link between physical activity levels and dementia risk: insights from the UK biobank study. Psychiatry Res 2024; 336: 115875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lambert JC, Ibrahim-Verbaas CA, Harold D, et al. Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease. Nat Genet 2013; 45: 1452–1458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Coley N, Hoevenaar-Blom MP, van Dalen JW, et al. Dementia risk scores as surrogate outcomes for lifestyle-based multidomain prevention trials-rationale, preliminary evidence and challenges. Alzheimers Dement 2020; 16: 1674–1685. [DOI] [PubMed] [Google Scholar]
- 18.Wolfova K, Miller EC. Impact of adverse pregnancy outcomes on brain vascular health and cognition. Res Pract Thromb Haemost 2024; 8: 102331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wardlaw JM, Smith C, Dichgans M. Small vessel disease: mechanisms and clinical implications. Lancet Neurol 2019; 18: 684–696. [DOI] [PubMed] [Google Scholar]
- 20.Nordestgaard LT, Christoffersen M, Frikke-Schmidt R. Shared risk factors between dementia and atherosclerotic cardiovascular disease. Int J Mol Sci 2022; 23: 9777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Leszek J, Mikhaylenko EV, Belousov DM, et al. The links between cardiovascular diseases and Alzheimer's disease. Curr Neuropharmacol 2021; 19: 152–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bae JH, Jung YM, Lee J, et al. Future risk of metabolic syndrome after recurrent pregnancy loss: a cohort study using UK biobank. Fertil Steril 2023; 120: 1227–1233. [DOI] [PubMed] [Google Scholar]
- 23.Wieckowska-Gacek A, Mietelska-Porowska A, Wydrych M, et al. Western diet as a trigger of Alzheimer's disease: from metabolic syndrome and systemic inflammation to neuroinflammation and neurodegeneration. Ageing Res Rev 2021; 70: 101397. [DOI] [PubMed] [Google Scholar]
- 24.Wotton CJ, Goldacre MJ. Associations between specific autoimmune diseases and subsequent dementia: retrospective record-linkage cohort study, UK. J Epidemiol Community Health 2017; 71: 576–583. [DOI] [PubMed] [Google Scholar]
- 25.Li J, Chen Y, Wu H, et al. Apolipoprotein E (Apo E) gene polymorphisms and recurrent pregnancy loss: a meta-analysis. J Assist Reprod Genet 2014; 31: 139–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Rohn TT. Is apolipoprotein E4 an important risk factor for vascular dementia? Int J Clin Exp Pathol 2014; 7: 3504–3511. [PMC free article] [PubMed] [Google Scholar]
- 27.Jevtic S, Sengar AS, Salter MW, et al. The role of the immune system in Alzheimer disease: etiology and treatment. Ageing Res Rev 2017; 40: 84–94. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-docx-1-alr-10.1177_25424823251370717 for Adverse reproductive outcomes and future risk of incident dementia: The modifying effect of genetic susceptibility by Qiaoqiao Xu, Chengzhe Tao, Sijia Dai, Zhixi Lu, Michael Aschner, Guangfeng Long, Shaojun Li and Cheng Xu in Journal of Alzheimer's Disease Reports