Skip to main content
Download PDF

Number of children and dementia risk: a causal mediation analysis using data from the HUNT study linked with national registries in Norway

BMC Neurology volume 25, Article number: 39 (2025) Cite this article

Abstract

Background

Childlessness, as well as having a high number of children, has been reported to be associated with an elevated risk of dementia compared to having 2–3 children. The mechanisms underlying these relationships are not well understood and may be mediated by different midlife risk factors. We examined the mediating role of various factors on the relationship between the number of children and dementia risk. These factors include socioeconomic factors (e.g., occupational complexity), psychosocial (e.g.., social activities, loneliness, life satisfaction), lifestyle (e.g., smoking, physical inactivity, alcohol intake), and chronic diseases (e.g., obesity, diabetes, depression, hearing impairment and hypertension).

Methods

Using a historic cohort design, we included 9,745 participants born between 1931–48, with a mean age of 78.2 (SD = 6.4) years at the time of cognitive testing in the HUNT4 70 + sub-study (2017–2019). Further measures were obtained through data linkage between information from Statistics Norway and the HUNT1(1984–86), and HUNT2 (1995–97) Surveys. Causal mediation analyses using an inverse odd weighting approach were conducted to decompose the total effect of the number of children (0, 1, or 4 + children vs. 2–3) on the risk of dementia at age 70 + years into direct and indirect effects with mediators assessed at a mean age of 50.7 (SD = 6.4) years. The analyses were adjusted for age, sex, marital status at age 25 years, educational status, and religion assessed during HUNT3 (2006–2008).

Results

Overall, 15.7% were diagnosed with dementia. The proportions with dementia by the number of children were 22.3% among those with no children, 21.4% for those with one child, 13% for those with 2–3 children (specifically, 12.6% for those with 2 children and 13.4% for those with 3 children), and 19.9% for those with 4 + children. Compared to the reference group of individuals with 2–3 children, the dementia risk was higher among the groups with no children (relative risk (RR): 1.30, 95% confidence interval (CI) (1.12, 1.51)), those with one child (RR: 1.30, 95% CI (1.14, 1.47)) and those with 4 + children (RR: 1.12, 95% CI (1.01, 1.24)). The elevated risks of dementia were not mediated by the socioeconomic, psychosocial, lifestyle, or chronic diseases related factors that we tested. Sex-stratified analysis showed higher dementia risk for men without children and women with one or 4 + children compared to those with 2–3 children, with similar patterns across sexes. None of the mediators contributed to mediation in either group. None of the mediators appeared to contribute through mediation in either group.

Conclusions

Our findings suggest that the number of children—specifically being childless, having one child, or having four or more children—may influence the risk of dementia. These relationships were not mediated by psychosocial, lifestyle, and socioeconomic factors, or markers of chronic diseases in adulthood considered in this study.

Peer Review reports

Background

Dementia, a condition characterized by symptoms such as memory loss, cognitive deficiencies, and behavioral changes that significantly interferes with a person’s ability to perform daily activities [1]. It is among the most significant public health concerns worldwide, affecting over 55 million people globally [2]. The number of people affected by dementia is estimated to reach 153 million by 2050, in line with the aging global population [3]. Dementia accounts for 11.9% of years lived with disability due to noncommunicable diseases [4] and has adverse effects on national economies [5]. Therefore, public health measures aimed at reducing the incidence/prevalence of dementia are crucial. The 2020 Lancet commission on dementia prevention, intervention, and care [6], highlights that 40% of late-onset dementia cases could be prevented or delayed by targeting modifiable factors. These factors include lower level of education in early life, hearing loss, traumatic brain injury, hypertension, alcohol intake, and obesity in midlife, and smoking, depression, social isolation, physical inactivity, air pollution, and diabetes later in life [6].

Norway is experiencing low fertility rates, with the total fertility rate dropping to 1.4 in 2022 [7,8,9]. In addition, recent years have seen a shift in parity distributions [10], marked by a trend towards increasing childlessness in newer birth cohorts. For instance, at age 40, about 30% of men and 16% of women in Norway are childless. Given the combination of an aging population and declining fertility rates, it is important to elucidate the specific mechanisms that link the number of children to dementia risk in later-life.

Several studies have identified a U-shaped risk curve, indicating an increased risk of dementia among individuals with 0, 1, or 4 + children compared to those with 2–3 children [11,12,13].

The specific mechanisms linking the number of children to dementia risk remain unclear; however, factors influenced by whether an individual has no children, too few children, or many children may contribute to the risk of developing dementia in later life among certain groups. For example, childless individuals are often less engaged in social interactions [14, 15], experience loneliness [16, 17], have lower psychological wellbeing [18], and more likely to engage in unfavorable behaviors, such as smoking, alcohol consumption and poor physical activity, [14, 19,20,21,22], all of which are associated with increased risk of dementia [6]. On the other hand, having children can promote better social interactions due to child-rearing responsibilities and may encourage favorable lifestyle changes, such as quitting smoking, reducing alcohol consumption [23, 24], achieving greater life satisfaction [25, 26], all of which are protective against dementia [6, 27]. Individuals with many children may also face challenges such as economic strain, reduced working hours, and limited leisure time [28,29,30], leading to conditions like stress and hypertension that could increase dementia risk. Furthermore, unhealthy lifestyle factors than can be influenced by the number of children an individual has could also increase the risk of chronic diseases (e.g., obesity, diabetes, hypertension, high LDL) [14, 27, 31, 32], which in turn contributes to the risk of dementia later in life [6].

The factors that vary depending on whether an individual has too few or too many children could potentially play a mediating role for the relationship between the number of children and later-life dementia risk. Assessing if, and to what extent, factors influenced by the number of children an individual mediates the relationship between the number of children and dementia risk later in life using formal mediation analysis can help decompose the effect of the number of children on dementia into direct and indirect effects (i.e., through mediating pathways) [33,34,35,36]. Understanding how much of the effect is mediated through potential mediating factors is crucial for identifying modifiable risk factors for interventions aimed at reducing overall dementia risk and related differences by number of children. In this regard, prior studies have mainly examined the relationship between the number of children and dementia risk using standard regression techniques [11,12,13, 37, 38], primarily focusing on direct effects without quantifying the mediating pathways. This approach provides only a partial understanding of the relationship, as it does not reveal the indirect contributions of mediating factors that may be influenced by the number of children an individual has, or how these factors affect their risk of dementia later in life. Studies exploring such mediators in this context are currently lacking.

Additionally, the pathways linking the number of children to dementia risk may vary by sex due to different social roles, responsibilities, and stressors associated with parenthood. For example, among women, increased caregiving responsibilities can make it more challenging to maintain regular physical activity, due to compounded time constraints [19, 20]. They may also be at higher risk of hypertension [39, 40], likely due to cumulative physiological changes from multiple pregnancies and the ongoing stress of raising a large family. The risk of type 2 diabetes is also elevated, especially for those having gestational diabetes [41, 42]. Women with more children often face greater demands on their time [43, 44], which can lead to career interruptions or shifts to more flexible jobs to better manage childcare responsibilities. For men, fatherhood can impact health in ways less directly tied to physical demands. Men’s health risks related to fatherhood may stem from the stress of providing for a family and balancing work and family life. Societal expectations often position men as primary providers, and this socioeconomic pressure can affect their mental health and increase risk factors related to dementia, such as hypertension or depression [45]. However, there is a lack of knowledge regarding the mechanisms for the differential effect of number of children an individual has on dementia risk later in life for men and women.

In this study, we are investigating the idea that the number of children a person has—whether they have no children, one child, or many children (4 or more)—could influence their risk of developing dementia in later life. We think this effect might be mediated by various dementia risk factors. These include socioeconomic factors, such as the complexity of one’s occupation which can indicate the level of cognitive stimulation they are exposed to; psychosocial factors such as social activities, feelings of loneliness, and overall life satisfaction; lifestyle factors such as smoking, physical inactivity, and alcohol consumption; and chronic diseases such as obesity, diabetes, depression, hearing impairment, and hypertension. We propose that these factors could be influenced by whether a person is childless or has too many or too few children. In other words, the number of children a person has could indirectly affect their dementia risk by influencing these factors. In addition, we further explored whether there is sex specific mediating pathways for the relationship between number of children and dementia risk later in life. To address the gap in the literature, we have examined the joint mediating role of socioeconomic, psychosocial, lifestyle, and chronic disease risk factors during midlife on the relationship between the number of children and dementia risk in individuals aged 70 years and older. This was achieved through causal mediation analysis, using high-quality data from the Trøndelag Health Study (HUNT), and linked to data from nationwide registries in Norway.

Methods

Study population

The current study used a historical cohort design linking data from older adults aged 70 + years who underwent clinical cognitive assessment in the HUNT4 70 + Study in 2017–19 with administrative prospective data from Statistics Norway and with data from earlier HUNT surveys, HUNT1 (1984–86), HUNT2 (1995–97) and HUNT3 (2006–08). HUNT is a large ongoing general population study initiated in 1984 with the latest survey, HUNT4, completed in 2019 [46]. All adult inhabitants residing in the Nord-Trøndelag County of Norway were invited to participate in all four surveys [46, 47]. The average age of the participants was 45.1 (SD = 6.4) years during HUNT1 and 56.2 (SD = 6.4) years during HUNT2.

Among the 9930 participants in the HUNT4 70 + study, 185 were excluded due to insufficient information about cognitive diagnosis or the presence of conditions other than mild cognitive impairment or dementia. This resulted in a final study population of 9745 participants born between 1931–1949 (Fig. 1).

Fig. 1
figure 1

Overview of the sampling scheme: the HUNT Study, Norway

Full size image

Main exposure variable: number of children

The main exposure variable in our study was the number of children each participant had, categorized into four categories: no children, one child, two to three children, and four or more children. The risk of dementia was similar between the groups who had two vs. three children, and these two categories were thus merged into a single reference group.

Outcome variable: dementia

The main outcome variable in HUNT4 70 + study was the dementia status of participants, categorized as “yes” vs. “no”. All participants in the study underwent a thorough clinical examination, which included the assessment of cognitive function, daily-life function, neuropsychiatric symptoms, and subjective cognitive decline; interviews with next-of-kin were also conducted. Two medical doctors, from a pool of nine, with expertise in geriatrics, old-age psychiatry, or neurology, used the DSM-5 criteria to classify dementia status. The classifications were as follows: 0) no cognitive impairment, 1) amnestic mild cognitive impairment, 2) non-amnestic mild cognitive impairment, 3) dementia, and 4) either lack of information about cognitive diagnosis or presence of other types of cognitive illnesses [47]. For this study, a dichotomous variable was created: participants in group 3 were classified as “having dementia,” while those in groups 0, 1 and 2 were categorized as “not having dementia”. Group 4 was excluded from the study.

Potential mediators

We considered potential mediators assessed during HUNT1 and/or HUNT2 to avoid the risk of reverse causality for the mediators assessed during HUNT3. The potential mediators in our study were categorized into four main groups: socioeconomic factors, psychosocial factors, lifestyle factors, and factors related to chronic diseases.

For socioeconomic factors, we considered occupational complexity as a mediator. It was defined into four categories: high complexity, intermediate complexity, low complexity, and not working, based on information from Statistics Norway during HUNT1 and HUNT2 (Table 1).

Table 1 Source and definitions of mediating and covariate/confounding variables
Full size table

Regarding psychosocial factors, information was gathered from self-reported data in HUNT1 and/or HUNT2. These factors included participation in in social activities (dichotomized as never or a few times per year vs. more than once per week to 1–2 times per month), loneliness (dichotomized as feeling lonely very often or sometimes vs. never or very rarely), and life satisfaction (rated on a Likert scale from 1, very dissatisfied, to 7, very satisfied).

Lifestyle factors assessed in HUNT1 and/or HUNT2 included daily smoking (ever vs. never), physical inactivity (yes vs. no), and alcohol consumption (consumed alcohol five times or more times vs. less than five times in the past two weeks).

Chronic disease factors comprised obesity (yes vs. no), hypertension (yes vs. no), diabetes (yes vs. no), hearing impairment (yes vs. no), and depression and anxiety symptoms, which were assessed using a 14-item hospital anxiety depression score. The sources and the definitions of psychosocial factors, lifestyle factors, and chronic disease factors are described in Table 1.

Potential confounders/covariates

Information regarding the age of participants in 2018, sex (male vs. female), educational status (having completed secondary school or below vs. tertiary education and higher), and marital status at age 25 years (married vs. not married/widowed/divorced/separated) was sourced from Statistics Norway. Religious affiliation (official religious order vs. humanistic or atheistic beliefs) was obtained from HUNT3, was as this was the only wave of the study that included data on religion. Importantly, we assumed that there was no risk of reverse causation for the religion variable. Further details about the sources and definitions of these variables are described in Table 1.

Statistical analysis

Causal mediation analysis was conducted using an inverse odds weighting approach [36]. We examined the mediating role of various potential mediator groups, as depicted in the direct acyclic graph (DAG) presented in Fig. 2. The analysis focused on decomposing the total effects of having 0, 1, or 4 + children (compared to the reference group of 2–3 children) on dementia risk at age 70 + years into natural direct and natural indirect effects. This decomposition was achieved by using a generalized linear model with Poisson family and the log link function. To estimate the relative risk (RR) of dementia with 95% confidence intervals (CI), a bootstrap method with 200 replications was conducted. We estimated the joint mediating effects of each mediator group individually, as well as all the mediator groups combined. The natural direct effect captures the remaining effect of number of children (0, 1, or 4 + vs. 2–3) on dementia if we were to eliminate the pathway from exposure to the mediators. The natural indirect effect captures the difference between the counterfactual outcomes for an exposed individual (0, 1, or 4 + children) with the mediators set to the value it would normally take when an individual is exposed compared to the same exposed individual with the mediator set to the value it would normally take when the individual is unexposed (2–3 children). The total effect captures how much the outcome would change overall if the exposure status were altered from unexposed (2–3 children) to exposed (0, 1, or 4 + children). Details are found in Supplementary Table 2 and in the causal mediation analyses section of the supplementary file. Missing values were handled using multiple imputations via chained equations using 20 imputed datasets (see Supplementary Table 1). We examined sex-specific mediating pathways for the relationship between number of children dementia risk by running a separate mediation model for men and women. Additional sensitivity analyses were conducted to i) assess the impact of any unmeasured confounder on both mediators and outcome, ii) assess the robustness of the outcome definition, iii) explore if the mediating paths differed for individuals with 0 or 4 + children, and iv) examine if there were education-specific mediating paths for the effect of the number of children on dementia risk (for more details, see the sensitivity analyses section of the supplementary Table 2). In addition, we found no significant interaction effects between sex and the number of children. Hence, sex-specific analyses were not conducted. AGReMA guidelines for observational study was followed for reporting mediation analysis [48]. All analyses were performed using the STATA 16/MP statistical package [49].

Fig. 2
figure 2

Directed acyclic graph linking number of children to dementia risk

Full size image

Results

The mean age among the 9745 eligible participants was 70.2 (SD = 6.4) years at HUNT4 70 + (Table 2). Overall, 15.7% had dementia, 6.6% were childless, 7.5% had one child, 33.3% had two children, 32.2% had three children, and 20.5% had 4 + children. The proportion with dementia by number of children was 22.3% among those who were childless, 21.4% among those with one child, 13.0% among those with two–three children, and 19.9% among those with 4 + children. Social activity levels varied by the number of children. Specifically, participants with 0 and 1 child were less socially active, while those with 4 + children were more socially active than those with 2–3 children. Participants with 0–1 children reported more loneliness than those with 2–3 children. Participants with 0, 1, or 4 + children more often reported an unfavorable health lifestyle and a history of chronic diseases compared to those with 2–3 children. Those with dementia were less satisfied in life, less socially active, experienced more loneliness, demonstrated an unfavorable health lifestyle and had a history of chronic diseases outcomes compared to those without dementia (Table 3).

Table 2 Characteristics of study participants by number of children (N = 9745): The HUNT Study, Norway
Full size table
Table 3 Characteristics of study participants (N = 9745) given as mean, SD, frequencies (%) by dementia diagnosis or not
Full size table

Mediation analyses results

Compared to participants with 2–3 children, those who were childless had a higher dementia risk (total effect, RRTE: 1.30, 95% CI (1.12, 1.51)), as did those with one child (RRTE: 1.30, 95% CI (1.14, 1.47)) and those with 4 + children (RRTE: 1.12, 95% CI (1.01, 1.24)) (Fig. 3).

Fig. 3
figure 3

Total effect, natural direct effect, and natural indirect effect of number of children on later-life dementia

Full size image

Our analysis indicated that psychosocial factors did not mediate the increased dementia risk associated with having no children (natural indirect effect, RRNIE: 1.00, 95% CI (0.91, 1.09)), having one child (RRNIE: 0.96, 95% CI (0.92, 1.00)), or having 4 + children (RRNIE: 0.99, 95% CI (0.94, 1.02). The direct effect of having 0, 1, and 4 + children on dementia, after accounting for psychosocial factors, was (RRNDE: 1.30, 95% CI (1.09, 1.55), 1.35, 95% CI (1.17, 1.57), and 1.14, 95% CI (1.02, 1.27), respectively) (Fig. 3).

Similarly, occupational complexity did not mediate the increased dementia risk among participants who had no children (RRNIE: 0.99, 95% CI (0.90, 1.08)), had one child (RRNIE: 0.94, 95% CI (0.89, 1.00)), or had 4 + children (RRNIE: 1.00, 95% CI (0.96, 1.04)). The direct effect of having 0, 1, and 4 + children on dementia, after accounting for occupational complexity, was (RRNDE: 1.32 (1.11, 1.56), 1.38, 95% CI (1.17, 1.62), and 1.11, 95% CI (1.00, 1.24), respectively) (Fig. 3).

Moreover, lifestyle factors did not mediate the total effect on dementia risk among those who had no children (RRNIE: 0.98, 95% CI (0.90, 1.07)), had one child (RRNIE: 0.95, 95% CI (0.91, 1.00)), or had 4 + children (RRNIE: 1.00, 95% CI (0.96, 1.04)). The direct effect of having 0, 1, and 4 + children on dementia, after accounting for lifestyle factors, was (RRNDE: 1.31, 95% CI (1.12, 1.54), 1.35, 95% CI (1.16, 1.59), and 1.12, 95% CI (0.99, 1.25), respectively) (Fig. 3).

Similarly, exposure to markers of chronic diseases in midlife did not mediate the total effect on dementia risk for those who were childless (RRNIE: 1.00, 95% CI (0.92, 1.08)), had one child (RRNIE: 0.96, 95% CI (0.91, 1.00)), or had 4 + children (RRNIE: 0.99, 95% CI (0.96, 1.03)). The direct effect of having 0, 1, and 4 + children on dementia, after accounting for a history of chronic diseases during midlife, was (RRNDE: 1.30, 95% CI (1.10, 1.54), 1.35, 95% CI (1.17, 1.62), and 1.12, 95% CI (1.01, 1.24), respectively) (Fig. 3).

When evaluating the joint effect of all factors combined, there were no indirect effects on dementia risk among individuals who had no children (RRNIE: 1.00, 95% CI (0.91, 1.10)), had one child (RRNIE: 0.98, 95% CI (0.92, 1.03)), and had 4 + children (RRNIE: 1.01, 95% CI (0.97, 1.07)) compared to those with 2–3 children. The direct effect of having 0, 1, and 4 + children on dementia, after accounting for a history of chronic diseases during midlife, was (RRNDE: 1.30, 95% CI (1.10, 1.55), 1.32, 95% CI (1.14, 1.55), and 1.10, 95% CI (0.98, 1.24), respectively) (Fig. 3).

Mediation analysis results by sex

Compared to those with 2–3 children, childless men had a higher risk of dementia (total effect, RRTE: 1.41, 95% CI: 1.15–1.71), as did men with one child (RRTE: 1.26, 95% CI: 0.99–1.61) and men with four or more children (RRTE: 1.03, 95% CI: 0.86–1.22) (see Fig. 4). Similarly, in women, childless women showed a higher dementia risk (RRTE: 1.18, 95% CI: 0.96–1.44), as did those with one child (RRTE: 1.29, 95% CI: 1.07–1.57) and four or more children (RRTE: 1.17, 95% CI: 1.04–1.33) (see Fig. 4). None of the mediators demonstrated a mediated effect in either sex.

Fig. 4
figure 4

Total effect, natural direct effect, and natural indirect effect of number of children on later-life dementia by sex

Full size image

Sensitivity analyses results

In our analysis, only E-values for the natural direct effect are presented, as there were no observed indirect effects. The result showed that with an observed natural direct effect (RRNIE) of 1.30, 1.32, and 1.10 among those with 0, 1 child and 4 + children, respectively, in the model for all the mediators together, an unmeasured confounder that was associated with number of children and dementia by RRs of 1.92-fold each among childless, 1.97-fold each for those with 1 child and 1.43-fold each among 4 + children, conditional on the measured confounders, could fully explain the natural direct effect estimates, but weaker unmeasured confounder associations could not. The sensitivity analysis excluding those with mild cognitive impairment from the analysis and outcomes defined by including those with mild cognitive impairment to the groups with dementia showed a negligible effect on the mediation parameter estimates (Supplementary Tables 1 and 2). Analyses where data was split into having 0 child vs. 2–3 children and having 4 + children vs. 2–3 children revealed no significant differences in the mediating paths between those without children and those with 4 + children (data not shown). Further sensitivity analyses, in which we divided the data by educational level, also showed no significant differences in the mediation parameters regarding the effect of the number of children on late-onset dementia risk (Supplementary Table 3).

Discussion

In this large Norwegian population-based historic cohort study, individuals with 0, 1 or 4 + children were found to have a higher dementia risk compared to those with 2–3 children. Contrary to our initial hypothesis, this increased risk was not attributable to differences in midlife psychosocial factors, socioeconomic position, lifestyle, or chronic disease markers.

Our observation of a U-shaped association between the number of children and increased dementia risk aligns with findings from other studies exploring this relationship [11,12,13, 37, 38, 50, 51]. We initially hypothesized that the higher dementia risk associated with having 0,1 and 4 + children may be influenced by the combined effects of various factors in midlife: socioeconomic position (with occupational complexity as a proxy for cognitively stimulating environments), psychosocial factors (such as participation in social activity, loneliness, life satisfaction), lifestyle factors (including smoking, physical inactivity, alcohol intake), and markers of chronic diseases (obesity, diabetes, depression score, hearing impairment, and hypertension). However, our results did not support this hypothesis, as these factors did not mediate the observed increased risks of dementia associated with having 0, 1 or 4 + children. The absence of mediating effects in our study may partly be due to unmeasured and complex multifactorial mediators, which were unavailable for us, such as quality of life, diet, the amount and quality of sleep, and various forms of social support, which potentially vary by the number of children [52,53,54]. Given that there was a substantial direct effect on dementia risk of having no children or having one child, even after accounting for potential mediators, further research is needed to fully understand the underlying mechanisms.

The pathways through which having 0 or 4 + children influence later life dementia risk might differ. Childless individuals are likely to have smaller social networks and engage in unfavorable lifestyle behaviors, and they may be at a higher risk of chronic diseases outcomes [14, 55]. In contrast, those with 4 + children may face higher economic strain, contributing to adverse health conditions such as stress and hypertension [28,29,30]. Our separate analyses, comparing the mediating paths for individuals with 0 and 4 + children to those with 2–3 children, showed no differences in the mediation parameters. However, individuals with 0 or 1 child exhibited a 30% higher later life dementia risk compared to those with 2–3 children. Given the existing evidence of Norway’s lower fertility rates compared to some countries globally [7,8,9] and the rise in childlessness in recent cohorts [56], it is crucial to support individuals in fulfilling their reproductive preferences. Furthermore, providing assistance to those with a substantial number of children could help mitigate the impact of varying fertility rates on the risk of dementia later in life.

Our stratified analysis by sex indicated that the total effect of having 0, 1, or 4 + children on the risk of dementia, compared to those with 2–3 children, was comparable for both men and women with a higher risk among childless men and women with 1 and 4 + children. None of the considered in our study (i.e., socioeconomic status (e.g., occupational complexity), psychosocial aspects (e.g., social activities, loneliness, life satisfaction), lifestyle choices (e.g., smoking, physical inactivity, alcohol intake), and markers of chronic diseases (e.g., obesity, diabetes, depression, hearing impairment, and hypertension) did not show mediated effects in either sex. Further research is warranted to explore the potential underlying mechanisms.

There is some evidence to suggest that cognitive health and behavior, including educational attainment and socioeconomic status, both of which are related to dementia risk, might affect family planning decisions [57, 58]. Many studies cite length of education as a reason for why women with more education and higher socioeconomic status tend to have fewer children [58, 59]. Studies have shown that individuals with higher cognitive reserve, often reflected in higher educational attainment and socioeconomic status, tend to have different fertility patterns compared to those with lower cognitive reserve. One study found that in both men and women, those with two or three offspring had significantly better cognitive function compared to those without offspring [60]. Another study found that having children was associated with better cognition for men, but not for women [61]. However, these factors were only associated with cognitive functioning, as the individuals were not diagnosed with ADRD. Such studies would require much longer follow ups to fully understand the long-term impacts on ADRD. Additionally, underlying genetic predispositions to dementia could theoretically influence neurological development and subsequent life choices, including the number of children. However, we are not aware of any direct causal evidence linking dementia liability specifically to fertility decisions and this requires further study. Such studies would face several practical challenges. Although educational attainment and socioeconomic status are both associated with ADRD risk, the relationship between these factors and parity may not be directly influenced by the knowledge of one’s own ADRD risk as ADRD typically occurs much later in life, very few individuals are apt to do genetic testing for ADRD susceptibility unless there are several family members affected, and there are several competing causes of death.

The outcome variable in this study was defined by excluding those either lack of information about cognitive diagnosis or presence of other types of cognitive illnesses [47]. Excluding participants with incomplete outcome information could introduce selection bias, impacting both direct and indirect effects. If the excluded group has distinct characteristics affecting the mediator or outcome, this could bias mediator distribution and direct effects, potentially misrepresenting the true relationship between exposure and outcome. In our analysis, efforts were made by applying weights to adjust for potential nonresponse bias and multiple imputation to ensure that the remaining sample represents the larger population. Although, efforts have been made to account for non-response and missing data, it is difficult to quantify to what extent exclusion of these groups from the analysis influences our mediation parameters. Future research should consider using external data or other resources to perform a comprehensive quantitative bias analysis, addressing bias in complex multiple-mediator models like ours.

Furthermore, mediation analysis requires the identification of assumptions of no unmeasured confounding in the exposure-mediator, mediator-outcome, and exposure-outcome relationships [62,63,64]. However, the exposure variable (i.e., number of children), the outcome variable (i.e., dementia), and the potential mediators, such as lifestyle and markers of chronic diseases, could be affected by various unmeasured factors. This makes the assumptions of no unmeasured confounding challenging, as many of the residual variables are either inaccessible or unknown. In our analysis, we conducted a sensitivity analysis using mediational E-value to assess the possible influence of unmeasured confounder on direct and indirect effects [65]. The results from mediational E-value analysis indicated that to explain away an observed direct effect, an unmeasured confounder, would have to be relatively strongly associated with both dementia and the number of children.

Strength and limitations

The strengths of this study include the large sample size, population-based sampling standardized approach to diagnosis of dementia, use of high quality registry data for number of children, application of a life-course approach, reduction of missingness using multiple imputation and application of causal mediation analysis [35].

As with many other observational studies, ours is not without limitations. The mediation analysis method we have employed does not show the independent contributions of each mediator, which precludes the identification of a single mediator that is actionable in an intervention. Mediation analysis requires identification assumptions of no unmeasured confounding [62,63,64], yet the outcome variable, dementia, and the potential mediators, such as lifestyle and markers of chronic diseases, could be affected by various unmeasured factors. An additional limitation of the inverse odds weighting approach in mediation analysis is that this method does not account for time-varying exposures or mediators. This limitation means that the approach may not fully capture dynamic relationships where exposures or mediators change over time, potentially leading to biased estimates if these time-dependent variations significantly impact the outcome. Future studies could address this by employing models that can accommodate time-varying factors, thereby providing a more nuanced understanding of causal pathways in longitudinal data. Assuming that there is no confounder on the mediator-outcome relationships could be challenging as many of the residual variables are either inaccessible or unknown.

There is healthy selection bias in our study as participation in HUNT surveys depend on survival, socioeconomic status, and absence of chronic diseases [66]. Ignoring competing risk, as we have done, might introduce bias [67]. The indirect effect estimates for the factors such as lifestyle behaviors and chronic diseases might be underestimated [68, 69], if there is competing risk, as reported in smoking [67].

Conclusions

Our findings suggest that the number of children—specifically being childless, having one child, or having four or more children—may influence the risk of dementia. These relationships were not mediated by psychosocial, lifestyle, and socioeconomic factors, or markers of chronic diseases in adulthood considered in this study.

Data availability

The data used in the current study are available after approval by the Regional Committee for Medical and Health Research Ethics and HUNT's Data Access Committee.

References

  1. Weiner MF, Lipton AM. Clinical manual of Alzheimer disease and other dementias. Am Psychiatr Pub; 2012.

  2. World Health Organization.: Global action plan on the public health response to dementia 2017–2025. 2017. In.; 2020.

  3. Nichols E, Steinmetz JD, Vollset SE, Fukutaki K, Chalek J, Abd-Allah F, Abdoli A, Abualhasan A, Abu-Gharbieh E, Akram TT. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019. Lancet Public Health. 2022;7(2):e105–25.

    Article  Google Scholar 

  4. Prince M, Albanese E, Guerchet M, Prina M. World Alzheimer Report 2014. Dementia and Risk Reduction: An analysis of protective and modifiable risk factors. Alzheimer's Dis Int; 2014.

  5. Prince M, Wimo A, Guerchet M, Ali G-C, Wu Y-T, Prina M. World Alzheimer Report 2015. The Global Impact of Dementia: An analysis of prevalence, incidence, cost and trends. Alzheimer's Dis Int; 2015.

  6. Livingston G, Huntley J, Sommerlad A, Ames D, Ballard C, Banerjee S, Brayne C, Burns A, Cohen-Mansfield J, Cooper C. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396(10248):413–46.

    Article  PubMed  PubMed Central  Google Scholar 

  7. World Bank. Fertility rate, total (births per woman). World Bank 2019.

  8. Group ECW. Europe the continent with the lowest fertility. Hum Reprod Update. 2010;16(6):590–602.

    Article  Google Scholar 

  9. Skirbekk V. Decline and prosper!: changing global birth rates and the advantages of fewer children. Springer; 2022.

  10. Statistics Norway. Decline in fertility In.; 2023.

  11. Gong J, Harris K, Peters SA, Woodward M. Reproductive factors and the risk of incident dementia: A cohort study of UK Biobank participants. PLoS Med. 2022;19(4):e1003955.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Basit S, Wohlfahrt J, Boyd HA. Associations between parenthood and dementia in men and women: biology or confounding? BMC Neurol. 2023;23(1):90.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Bae J, Lipnicki D, Han J, Sachdev P, Kim T, Kwak K, Kim B, Kim S, Kim J, Moon S. Parity and the risk of incident dementia: a COSMIC study. Epidemiol Psychiatric Sci. 2020;29:e176.

    Article  CAS  Google Scholar 

  14. Kendig H, Dykstra PA, van Gaalen RI, Melkas T. Health of Aging Parents and Childless Individuals. J Fam Issues. 2007;28(11):1457–86.

    Article  Google Scholar 

  15. Umberson D, Pudrovska T, Reczek C. Parenthood, childlessness, and well-being: A life course perspective. J Marriage Fam. 2010;72(3):612–29.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Penning MJ, Wu Z, Hou F. Childlessness and social and emotional loneliness in middle and later life. Ageing & Society. 2022:1–28.

  17. Bachrach CA. Childlessness and social isolation among the elderly. J Marriage Family.1980:627–637.

  18. Huijts T, Kraaykamp G, Subramanian SV. Childlessness and psychological well-being in context: A multilevel study on 24 European countries. Eur Sociol Rev. 2013;29(1):32–47.

    Article  Google Scholar 

  19. Barnekow-Bergkvist M, Hedberg G, Janlert U, Jansson E. Physical activity pattern in men and women at the ages of 16 and 34 and development of physical activity from adolescence to adulthood. Scand J Med Sci Sports. 1996;6(6):359–70.

    Article  CAS  PubMed  Google Scholar 

  20. Brown WJ, Trost SG. Life transitions and changing physical activity patterns in young women. Am J Prev Med. 2003;25(2):140–3.

    Article  PubMed  Google Scholar 

  21. Rattay P, von Der Lippe E. Association between living with children and the health and health behavior of women and men. Are there differences by age? results of the “German health update”(GEDA) study. International journal of environmental research and public health. 2020;17(9):3180.

  22. Liu A, Akimova E, Ding X. The relationship of early-life diseases with lifetime childlessness: evidence from Finland and Sweden. medRxiv, 2022.03. 25.22272822, 2 April 2022. preprint

  23. Jarvis M. The association between having children, family size and smoking cessation in adults. Addiction. 1996;91(3):427–34.

    Article  CAS  PubMed  Google Scholar 

  24. Matusiewicz AK, Ilgen MA, Bohnert KM. Changes in alcohol use following the transition to motherhood: Findings from the National Epidemiologic Survey on Alcohol and Related Conditions. Drug Alcohol Depend. 2016;168:204–10.

    Article  PubMed  Google Scholar 

  25. Parr N. Satisfaction with life as an antecedent of fertility: Partner+ Happiness= Children? Demogr Res. 2010;22:635–62.

    Article  Google Scholar 

  26. Pollmann-Schult M. Parenthood and life satisfaction: Why don’t children make people happy? J Marriage Fam. 2014;76(2):319–36.

    Article  Google Scholar 

  27. Livingston G, Huntley J, Liu KY, Costafreda SG, Selbæk G, Alladi S, Ames D, Banerjee S, Burns A, Brayne C. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet. 2024;404(10452):572–628.

    Article  PubMed  Google Scholar 

  28. da Costa Fogaca Silva A. Investigating the difference in parental stress across different sized families. Dublin, National College of Ireland; 2019.

  29. Letablier M-T, Luci A, Math AN, Thévenon O: The costs of raising children and the effectiveness of policies to support parenthood in European countries: a Literature Review. 2009.

  30. Thévenon O, Luci A. Reconciling work, family and child outcomes: What implications for family support policies? Popul Res Policy Rev. 2012;31:855–82.

    Article  Google Scholar 

  31. Lawlor DA, Emberson JR, Ebrahim S, Whincup PH, Wannamethee SG, Walker M, Smith GD. Is the association between parity and coronary heart disease due to biological effects of pregnancy or adverse lifestyle risk factors associated with child-rearing? Findings from the British Women’s Heart and Health Study and the British Regional Heart Study. Circulation. 2003;107(9):1260–4.

    Article  PubMed  Google Scholar 

  32. Brisson C, Laflamme N, Moisan J, Milot A, Mâsse B, Vézina M. Effect of family responsibilities and job strain on ambulatory blood pressure among white-collar women. Psychosom Med. 1999;61(2):205–13.

    Article  CAS  PubMed  Google Scholar 

  33. Pearl J. Interpretable conditions for identifying direct and indirect effects. In.: CALIFORNIA UNIV LOS ANGELES DEPT OF COMPUTER SCIENCE; 2012.

    Book  Google Scholar 

  34. VanderWeele T. Explanation in causal inference: methods for mediation and interaction. Oxford University Press; 2015.

  35. Tchetgen Tchetgen EJ. Inverse odds ratio-weighted estimation for causal mediation analysis. Stat Med. 2013;32(26):4567–80.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Nguyen QC, Osypuk TL, Schmidt NM, Glymour MM, Tchetgen Tchetgen EJ. Practical guidance for conducting mediation analysis with multiple mediators using inverse odds ratio weighting. Am J Epidemiol. 2015;181(5):349–56.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Read SL, Grundy EM. Fertility history and cognition in later life. J Gerontol B Psychol Sci Soc Sci. 2017;72(6):1021–31.

    Article  PubMed  Google Scholar 

  38. Saenz JL, Díaz-Venegas C, Crimmins EM. Fertility history and cognitive function in late life: The case of Mexico. J Gerontol Series B. 2021;76(4):e140–52.

    Article  Google Scholar 

  39. Gillis EE, Sullivan JC. Sex differences in hypertension: recent advances. Hypertension. 2016;68(6):1322–7.

    Article  CAS  PubMed  Google Scholar 

  40. de Ritter R, de Jong M, Vos RC, Van der Kallen CJ, Sep SJ, Woodward M, Stehouwer CD, Bots ML, Peters SA. Sex differences in the risk of vascular disease associated with diabetes. Biol Sex Differ. 2020;11:1–11.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Kautzky-Willer A, Leutner M, Harreiter J. Sex differences in type 2 diabetes. Diabetologia. 2023;66(6):986–1002.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Huebschmann AG, Huxley RR, Kohrt WM, Zeitler P, Regensteiner JG, Reusch JE. Sex differences in the burden of type 2 diabetes and cardiovascular risk across the life course. Diabetologia. 2019;62:1761–72.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Evertsson M, Grunow D, Aisenbrey S. Work interruptions and young women’s career prospects in Germany, Sweden and the US. Work Employ Soc. 2016;30(2):291–308.

    Article  Google Scholar 

  44. Dodini S. The Gender Career Interruption Gap: Measuring the Cumulative Prevalence of Labor Force Dropout Due to Caregiving. In: PAA 2018 Annual Meeting: 2018: PAA; 2018.

  45. Read S, Grundy E, Wolf DA. Fertility history, health, and health changes in later life: A panel study of British women and men born 1923–49. Popul Stud. 2011;65(2):201–15.

    Article  Google Scholar 

  46. Åsvold BO, Langhammer A, Rehn TA, Kjelvik G, Grøntvedt TV, Sørgjerd EP, Fenstad JS, Heggland J, Holmen O, Stuifbergen MC. Cohort profile update: the HUNT study, Norway. Int J Epidemiol. 2023;52(1):e80–91.

    Article  PubMed  Google Scholar 

  47. Gjøra L, Strand BH, Bergh S, Borza T, Brækhus A, Engedal K, Johannessen A, Kvello-Alme M, Krokstad S, Livingston G. Current and future prevalence estimates of mild cognitive impairment, dementia, and its subtypes in a population-based sample of people 70 years and older in Norway: the HUNT study. J Alzheimer’s Dis. 2021;79(3):1213–26.

    Article  Google Scholar 

  48. Lee H, Cashin AG, Lamb SE, Hopewell S, Vansteelandt S, VanderWeele TJ, MacKinnon DP, Mansell G, Collins GS, Golub RM. A guideline for reporting mediation analyses of randomized trials and observational studies: the AGReMA statement. JAMA. 2021;326(11):1045–56.

    Article  PubMed  PubMed Central  Google Scholar 

  49. StataCorp: 2019. Stata Statistical Software: Release 16. In.: StataCorp LLC. College Station, Texas: StataCorp LLC College Station; 2019.

  50. Zhang Y. Fertility History and Risk of Cognitive Impairment Among Older Parents in the United States. J Gerontol Series B. 2022;77(12):2326–37.

    Article  Google Scholar 

  51. Wolfova K, Wu D, Weiss J, Cermakova P, Kohler H-P, Skirbekk VF, Stern Y, Gemmill A, Tom SE. Sons and parental cognition in mid-life and older adulthood. J Psychiatr Res. 2022;156:284–90.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Matricciani L, Fraysse F, Grobler AC, Muller J, Wake M, Olds T. Sleep: population epidemiology and concordance in Australian children aged 11–12 years and their parents. BMJ open. 2019;9(Suppl 3):127–35.

    Article  PubMed  PubMed Central  Google Scholar 

  53. Nugent CN, Black LI. Sleep duration, quality of sleep, and use of sleep medication, by sex and family type, 2013–2014: US Department of Health and Human Services, Centers for Disease Control and …; 2016.

  54. Hubens K, Arons AM, Krol M. Measurement and evaluation of quality of life and well-being in individuals having or having had fertility problems: a systematic review. Eur J Contracept Reprod Health Care. 2018;23(6):441–50.

    Article  PubMed  Google Scholar 

  55. Dykstra PA, Wagner M. Pathways to childlessness and late-life outcomes. J Fam Issues. 2007;28(11):1487–517.

    Article  Google Scholar 

  56. Population Statistics [www.ssb.no/befolkning].

  57. Jalovaara M, Andersson L, Miettinen A. Parity disparity: Educational differences in Nordic fertility across parities and number of reproductive partners. Popul Stud. 2022;76(1):119–36.

    Article  Google Scholar 

  58. Huber S, Bookstein FL, Fieder M. Socioeconomic status, education, and reproduction in modern women: an evolutionary perspective. Am J Hum Biol. 2010;22(5):578–87.

    Article  PubMed  Google Scholar 

  59. Kim J, Sung J. Socioeconomic status and number of children among Korean women: the healthy twin study. J Prev Med Public Health. 2013;46(1):50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Ning K, Zhao L, Franklin M, Matloff W, Batta I, Arzouni N, Sun F, Toga AW. Parity is associated with cognitive function and brain age in both females and males. Sci Rep. 2020;10(1):6100.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Zhang Y, Fletcher J, Lu Q, Song J. Gender differences in the association between parity and cognitive function: Evidence from the UK biobank. Soc Sci Med. 2023;320:115649.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Daniel RM, Cousens S, De Stavola B, Kenward MG, Sterne J. Methods for dealing with time-dependent confounding. Stat Med. 2013;32(9):1584–618.

    Article  CAS  PubMed  Google Scholar 

  63. Cole SR, Frangakis CE. The consistency statement in causal inference: a definition or an assumption? Epidemiology. 2009;20(1):3–5.

    Article  PubMed  Google Scholar 

  64. Westreich D, Cole SR. Invited commentary: positivity in practice. Am J Epidemiol. 2010;171(6):674–7.

    Article  PubMed  PubMed Central  Google Scholar 

  65. Smith LH, VanderWeele TJ. Mediational E-values: approximate sensitivity analysis for unmeasured mediator–outcome confounding. Epidemiology. 2019;30(6):835–7.

    Article  PubMed  PubMed Central  Google Scholar 

  66. Langhammer A, Krokstad S, Romundstad P, Heggland J, Holmen J. The HUNT study: participation is associated with survival and depends on socioeconomic status, diseases and symptoms. BMC Med Res Methodol. 2012;12(1):1–14.

    Article  Google Scholar 

  67. Rojas-Saunero LP, Young JG, Didelez V, Ikram MA, Swanson SA. Considering questions before methods in dementia research with competing events and causal goals. Am J Epidemiol. 2023;192(8):1415–23.

  68. Valeri L, Proust-Lima C, Fan W, Chen JT, Jacqmin-Gadda H. A multistate approach for mediation analysis in the presence of semi-competing risks with application in cancer survival disparities. preprint 2021.

  69. Valeri L, Coull BA. Estimating causal contrasts involving intermediate variables in the presence of selection bias. Stat Med. 2016;35(26):4779–93.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

The Trøndelag Health Study (the HUNT Study) is a collaboration between the HUNT Research Centre (Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, NTNU), the Trøndelag County Council, the Central Norway Health Authority, and the Norwegian Institute of Public Health.

Funding

Open access funding provided by Norwegian Institute of Public Health (FHI) We are grateful for financial support from the National Institutes of Health (NIH), R01 grant no. R01AG069109-01. We acknowledge support from the Research Council of Norway, project numbers 288083 and 301010. This work was also partly supported by the Research Council of Norway through its Centres of Excellence funding scheme, project number 262700.

Author information

Authors and Affiliations

  1. Department for Physical Health and Aging, Norwegian Institute of Public Health, Oslo, Norway

    Teferi Mekonnen, Vegard Skirbekk, Ekaterina Zotcheva, Bo Engdahl, Asta Kristine Håberg & Bjørn Heine Strand

  2. Norwegian National Centre of Ageing and Health, Vestfold Hospital Trust, Tønsberg, Norway

    Vegard Skirbekk, Ekaterina Zotcheva, Geir Selbæk, Yehani Wedatilake & Bjørn Heine Strand

  3. Centre for Fertility and Health, Norwegian Institute of Public Health, Oslo, Norway

    Vegard Skirbekk, Bernt Bratsberg, Astanand Jugessur, Jennifer R. Harris & Dana Kristjansson

  4. Ragnar Frisch Center for Economic Research, Oslo, Norway

    Bernt Bratsberg

  5. Department of Geriatric Medicine, Oslo University Hospital, Oslo, Norway

    Geir Selbæk, Trine Holt Edwin & Bjørn Heine Strand

  6. Independent Researcher, Vienna, Austria

    Catherine Bowen

  7. Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway

    Astanand Jugessur

  8. Faculty of Medicine, University of Oslo, Oslo, Norway

    Geir Selbæk

  9. Population Aging Research Center and Department of Sociology, University of Pennsylvania, Philadelphia, PA, USA

    Hans-Peter Kohler

  10. Department of Neurology, Columbia University, Vagelos College of Physicians and Surgeons, New York, USA

    Sarah E. Tom, Katrin Wolfova & Yaakov Stern

  11. Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, USA

    Sarah E. Tom

  12. Department of Public Health and Nursing, Faculty of Medicine and Health Sciences, HUNT Research Centre, Norwegian University of Science and Technology, Trondheim, Norway

    Steinar Krokstad

  13. Levanger Hospital, Nord-Trøndelag Hospital Trust, Levanger, Norway

    Steinar Krokstad

  14. Department of Genetics and Bioinformatics, Norwegian Institute of Public Health, Oslo, Norway

    Dana Kristjansson

  15. Department of Neuromedicine and Movement Science, Faculty of Medicine and Health Sciences, Norwegian University of Science and Technology, Trondheim, Norway

    Asta Kristine Håberg

  16. Department of Epidemiology, Second Faculty of Medicine, Charles University, Prague, Czech Republic

    Katrin Wolfova

Authors
  1. Teferi Mekonnen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Vegard Skirbekk
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Ekaterina Zotcheva
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Bo Engdahl
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Bernt Bratsberg
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Astanand Jugessur
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Catherine Bowen
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Geir Selbæk
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Hans-Peter Kohler
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Jennifer R. Harris
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Sarah E. Tom
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Steinar Krokstad
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Trine Holt Edwin
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Yehani Wedatilake
    View author publications

    You can also search for this author inPubMed Google Scholar

  15. Katrin Wolfova
    View author publications

    You can also search for this author inPubMed Google Scholar

  16. Dana Kristjansson
    View author publications

    You can also search for this author inPubMed Google Scholar

  17. Yaakov Stern
    View author publications

    You can also search for this author inPubMed Google Scholar

  18. Asta Kristine Håberg
    View author publications

    You can also search for this author inPubMed Google Scholar

  19. Bjørn Heine Strand
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

TM: responsible for the original and final draft of the manuscript, data curation and formal analyses, contributed to the conceptualization, drafting, and reviewing of the final manuscript. VS: project administrator, contributed to the conceptualization, contributed to the original draft of the manuscript, acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. EZ: contributed to the conceptualization, drafting, and reviewing of the final manuscript. BE: contributed to the conceptualization, drafting, and reviewing of the final manuscript. BB: contributed to the conceptualization, drafting, and reviewing of the final manuscript. AJ: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. CB: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. GS: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. HPK: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. JRH: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. SET: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. SK: contributed to the conceptualization, drafting, and reviewing of the final manuscript. THE: contributed to the conceptualization, drafting, and reviewing of the final manuscript. YW: contributed to the conceptualization, drafting, and reviewing of the final manuscript. KW: contributed to the conceptualization, drafting, and reviewing of the final manuscript. DK: contributed to the conceptualization, drafting, and reviewing of the final manuscript. YS: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. AKH: acquired funding, contributed to the conceptualization, drafting, and reviewing of the final manuscript. BHS: project manager, acquired funding, contributed to the conceptualization, contributed to the conceptualization, drafting, and reviewing of the final manuscript.

Corresponding author

Correspondence to Teferi Mekonnen.

Ethics declarations

Ethics approval and consent to participate

This study was approved by the Regional Committee for Medical and Health Research Ethics (REC 2019/149 South East). HUNT1, HUNT2, HUNT3 and HUNT4 were approved by the Regional Committee for Medical and Health Research Ethics. The HUNT4 70 + data collection was approved by the Norwegian Data Protection Authority. Participation was based on an informed written consent. When a participant’s ability to consent was reduced, based on the judgment of health personnel in nursing homes and assessors, informed written consent was provided by the closest proxy [47].

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mekonnen, T., Skirbekk, V., Zotcheva, E. et al. Number of children and dementia risk: a causal mediation analysis using data from the HUNT study linked with national registries in Norway. BMC Neurol 25, 39 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12883-025-04044-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12883-025-04044-4

Keywords

BMC Neurology

ISSN: 1471-2377

Contact us