Summary
Background Approximately 20% of men and 15% of women remain childless at the end of their reproductive lifespan, with childlessness increasing over time, yet we lack a comprehensive understanding of the role and relative importance of diseases associated with childlessness, particularly among men.
Methods We examined all individuals born in Finland (n=1,035,928) and Sweden (n=1,509,092) between 1956 and 1968 (men) or 1956 and 1973 (women) and followed them up until the end of 2018. Socio-demographic, health, and reproductive information was obtained from nationwide registers. We assessed the association of 414 diseases across 16 categories with having no children by age 45 (women) and 50 (men) using a matched pair case-control design based on 71,524 pairs of full-sisters and 77,622 full-brothers who were discordant for childlessness as well as a population-based approach.
Findings Mental-behavioral, congenital anomalies, and endocrine-nutritional-metabolic disorders had the strongest associations with childlessness. Novel associations were discovered with inflammatory (eg. myocarditis) and autoimmune diseases (eg. juvenile idiopathic arthritis). Mental-behavioral disorders had stronger associations amongst men, particularly for schizophrenia and acute alcohol intoxication, while congenital anomalies, obesity-related diseases such as diabetes, and inflammatory diseases had stronger associations amongst women. Associations were dependent on the age at onset of the disease, with the strongest association observed earlier in women (21-25 years old) than men (26-30 years old). For most diseases, the association with childlessness was mediated by singlehood, especially in men. Some diseases, however, remained associated with childlessness among partnered individuals, including some mood- and endocrine-nutritional-metabolic disorders. All results can be explored in an interactive online dashboard.
Interpretation We provide evidence that disease burden across multiple domains is associated with childlessness, identifying modifiable mental-behavioral disorders and novel autoimmune and inflammatory diseases. Evidence can be used for targeted health interventions to counter decreasing fertility, reproductive health, involuntary childlessness, and shrinking populations.
Funding European Research Council (835079, 945733) and The Leverhulme Trust.
Evidence before this study The majority of research on infertility and childlessness has focused on socio-environmental factors, diseases related to reproduction, and examined predominantly women. Diseases are often considered separately, without a yardstick of their relative importance, and rarely examined within an entire population.
Added value of this study This is the first large-scale population study examining the association of 414 diseases across 16 broad categories with remaining childless, examining the entire reproductive and disease histories of 2.5 million men and women.
Implications of all the available evidence Our study provides evidence that childlessness is associated with multiple diseases that are potentially modifiable with targeted public health interventions, particularly mental-behavioral disorders such as alcohol dependence in men or endocrine-nutritional-metabolic disorders linked to obesity and diabetes. Our broader approach revealed hitherto unknown links of childlessness with autoimmune (eg. juvenile idiopathic arthritis, multiple sclerosis, systemic lupus erythematosus) and inflammatory diseases (eg. myocarditis), warranting future studies examining the mechanisms underlying these associations.
Introduction
Approximately 20% of men and 15% of women complete their reproductive lifespan without having children.1 The sharp increase in childlessness (eg. in Finland, increasing since the 1970s from 14% to 22% for women aged 40 years and 22% to 32% for men2,3) together with the postponement of parenthood is changing the age structure and population size of many countries,4 representing an enormous challenge for the long-term sustainability of multiple services. From the standpoint of public health, childlessness may impact individuals’ other health domains, with involuntarily childless women more likely to suffer from relationship dissolution, lower levels of self-esteem and isolation, and higher risks of clinical depression,5 while childbearing and breastfeeding can lower a woman’s risk of breast cancer.6
Diseases can also influence individuals’ chances of remaining childless over their lifetime (appendix p 2). First, diseases may directly influence childlessness through medical conditions that affect the fecundity and mortality of the individual as well as the risk of stillbirth.7 Second, diseases can impact selection into a partnership which in turn lowers the chance of being partnered and thus having children.8 Also, some may opt to remain childless due to concerns of transmission of certain known genetic conditions.9 Finally, many diseases are associated with lower socioeconomic status, unemployment, and economic uncertainty, which may amplify the psychosocial impact of diseases on childlessness.10 There are multiple non-disease related reasons that individuals postpone or remain childless related to factors such as individual choice, work-life reconciliation, housing and economic uncertainty, or inability to find a partner.4,5
Previous medical studies largely only examined a limited selection of diseases directly associated with reproductive biology such as recurrent miscarriage,11 polycystic ovary syndrome,12 and endometriosis,13 or studied fecundity in patients with major psychiatric disorders.14 Current research lacks a systematic assessment of the role of multiple diseases on childlessness and their relative importance. Moreover, the association of diseases can be sex-specific, given that men and women differ regarding both reproductive patterns (men have a higher chance of being childless, higher parity, and a longer reproductive lifespan than women15) and the prevalence and severity of diseases. For men, comparatively limited information has been collected on the causes and consequences of infertility and having children,16 despite infertility being a major public health concern. Finally, previous studies have been limited in the time horizon to follow study participants and the methodological tools to control for unmeasured familial confounding, either genetic or environmental.
We address these limitations of previous research by studying the entire reproductive and health history of a cohort of Finnish and Swedish men and women until the end of their reproductive period using high-quality nationwide population registers. The study determines to what extent diseases are associated with childlessness, whether such associations vary between sexes or differ by the age at disease onset, and the role that partnership formation plays in mediating the results. To answer these questions, we used a disease-agnostic family-based approach which allows us to robustly evaluate the relative importance of 414 disease diagnoses in relation to remaining childless, while providing public-health-relevant metrics to interpret the relative importance and contextualize the results of our study.
Methods
Study population
We considered all individuals born in Finland and Sweden between 1956 and 1968 (men) or between 1956 and 1973 (women) as index persons, for whom the vast majority completed their reproductive lifespan (45 years old for women, 50 for men) by 31 December 2018 (figure 1A). Individuals who emigrated during the study period were excluded to avoid incomplete follow-up for disease diagnoses and reproductive information. We further excluded individuals who died before the age of 16 to eliminate the impact of diseases on pre-reproductive survival. In total, we considered 1,425,640 index women (572,518 Finns and 853,122 Swedes) and 1,119,380 men (463,410 Finns and 655,970 Swedes). For these index individuals, we also obtained information for parents, spouses, siblings, and children for a total of 9,305,692 individuals (3,640,464 Finns and 5,665,228 Swedes).
Panel A shows the birth cohorts included in the analyses, the follow-up period, and the number of individuals included across three generations. Panel B shows the statistical approach used in the main analyses, where only families with same-sex siblings discordant for childlessness are included. Within a sibling pair, disease diagnoses are only considered if they occurred at least one year before the birth of the first child or within the corresponding age at which the sibling was childless, since sibling ages will differ.
Childlessness (main outcome)
The primary reproductive outcome of this study is childlessness, defined as having no live-born children by the age 45 for women and 50 for men. For every index person, we extracted demographic information of all biological children born before 31 December 2018 in Finland from the Population Information System and in Sweden from the Multi-Generation Register. Medical birth information including stillbirth and the use of assisted reproductive techniques was obtained from the Medical Birth Register available since 1987 in Finland and 1973 in Sweden. We excluded children conceived by assisted reproductive techniques (0·3% Finland, 0·8% Sweden) to control for potential confounding from social inequalities in medical help-seeking for infertility, especially in the observational period.
Marriage and partnership (secondary outcome)
For index individuals, we obtained their longitudinal marriage and partnership information (appendix p 36) in Finland between 11 April 1971 and 31 December 2018 from the Population Information System and in Sweden between 1 January 1977 and 31 December 2017 from Statistics Sweden. We defined individuals as partnerless if they did not have any partnership registered by age 45 (women) or 50 (men).
Disease diagnosis (exposure)
A wide variety of disease endpoints were defined by clinical expert groups17, through International Classification of Diseases (ICD) codes of version 8 (1969-1986), 9 (1987-1995 in Finland, 1987-1996 in Sweden), and 10 (1996-2018 in Finland, 1997-2018 in Sweden) (appendix p 2). We collected main diagnoses (ICD codes) and admission dates for secondary healthcare inpatient hospital treatments (since 1969, but in Sweden, nation-wide coverage began in 1973 for psychiatric diagnoses and 1987 for others18) and specialist outpatient visits in Finland from the Care Register for Health Care (since 1998) and in Sweden from the National Patient Register (since 2001). In Finland, we obtained cancer diagnoses (International Classification of Diseases for Oncology, 3rd Edition) and corresponding dates of diagnoses from the Finnish Cancer Registry, available since 1953. Additionally, for all individuals who died before the end of follow-up, we collected the date of death and basic, immediate, and contributing causes of death from Statistics Finland and Statistics Sweden to capture additional disease diagnoses that were missing from hospital inpatient and outpatient registers. The age of onset was defined as the first record of a disease diagnosis.
In total, we considered 16 disease categories, with infectious-parasitic diseases and congenital anomalies defined only in Finland and all remaining categories defined in both countries. After removing highly correlated disease diagnoses (appendix p 3), we considered 414 diseases for which we had more than 30 affected individuals in the sibling-based analysis for each sex, in Finland or Sweden.
Sibling-based analysis (main analysis)
There were 274,205 pairs of full-sisters (118,978 in Finland and 155,227 in Sweden) and 212,849 full-brothers (97,849 in Finland and 115,000 in Sweden) among index persons (figure 1A), which allowed us to control for potential confounding from genetic and environmental factors by performing a matched pair case-control study design (figure 1B). First, we kept only families having at least two same-sex full-siblings discordant for childlessness (at least one sibling with children and one childless). Then, within each family, we randomly selected one childless sibling as a case, and, as control, the sibling with children that was the closest for birth order with the case. Childless siblings were of similar age as siblings with children (absolute mean age difference=0·3 (SD=4·0) and 0·5 (4·4) for full-sisters and full-brothers, respectively). Finally, within each sibling pair, for both the case and the control, we only considered disease diagnoses that had their onset at least one year before the age at first birth for the sibling with children, to eliminate the impact of diseases triggered by pregnancy. We used a conditional logistic regression model to investigate the association between a disease and remaining childless. Full siblings were likely to share environmental factors, especially during the pre-reproductive period. Therefore, we only adjusted for birth year effects. A similar model was used to calculate the association between diseases and singlehood.
Additionally, for disease diagnoses that were significantly associated with childlessness, we assessed whether there were any age-of-onset-dependent effects by considering the age of onset (8 groups <=15, 16-20, 21-25, 26-30, 31-35, 36-40, 41-45, or unaffected for both sexes and an additional group (46-50) for men) as fixed effects.
Population-based analysis (secondary analysis)
To estimate the population-level effect of diseases on childlessness we performed a nested incident-matched case-control design which matched each childless case to one control based on sex, birth year, municipality of birth (545 from Finland and 1,091 from Sweden), and the highest parental education level (International Standard Classification of Education 1997) (appendix p 41). Parental education level was used instead of the index person education level to avoid confounding between an individual’s educational attainment and medical conditions14. In total, we used 226,860 matched pairs of women (107,375 in Finland and 119,485 in Sweden) and 274,941 of men (127,689 in Finland and 147,252 in Sweden) for the population-based analysis.
Number of unborn children and completed fertility rate reduction related to disease diagnoses
To quantify the total fertility deficit related to diseases, for the studied cohort, we calculated the total number of unborn children per 100,000 people and partitioned the total fertility rate (TFR) as the sum of disease and non-disease age-specific fertility rates for each sex (appendix p 30).
Mediation effect by singlehood
To assess the role that partnership formation plays in mediating the observed associations, we performed a causal mediation analysis19 to partition the association between disease diagnosis and childlessness into indirect (mediated by singlehood) and direct components (appendix p 38).
Sensitivity analysis
First, to eliminate the impact of diseases on reproductive age mortality, we used the sibling design framework described above but restricted the analysis to 66,255 full-sister pairs (33,556 in Finland and 32,699 in Sweden) alive by age 45 and 66,009 full-brother pairs (32,996 in Finland and 33,013 in Sweden) alive by age 50. Second, we used the sibling design framework described above but instead of conditional logistic regression we performed a stratified Cox proportional hazards regression model considering chronological age as the time scale and disease status as a time (age) varying exposure (unaffected until disease onset and affected afterward). Follow-up started at age 16, the estimated start of the reproductive lifespan, and was censored at death or one year before the age at first birth of the control, whichever occurs first.
Role of the funding source
The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Results
Study population
Our study population included 1,425,640 women and 1,119,380 men, born after 1956 and alive by age 16, did not emigrate, and largely completed their reproductive period at the end of 2018 (figure 1A). Among them, 230,198 women and 279,454 men were childless. The proportion of childless individuals – those with no live-born children – was higher in men (25·4%) than women (16·6%), higher amongst Finns (23·3%) than Swedes (18·7%), and slightly higher in the general population than in individuals who had same-sex full-siblings (0·9% and 0·5% higher in women and men, respectively). For the entire study population, we have information from nationwide registers covering 414 disease diagnoses across 16 main categories. Disease prevalence across the entire follow-up was highly comparable between Finland and Sweden (appendix p 26) and we presented meta-analysis results between the two countries. Our main analysis focused on 71,524 full-sister and 77,622 full-brother pairs who were discordant on childlessness.
Disease diagnoses associated with childlessness
Eleven rare diseases resulted in an almost complete lack of children (eg. severe intellectual disability, childhood leukemia, muscular dystrophy) (appendix p 25) were not included in all the remaining analyses. Out of the remaining 403 diseases, 74 were significantly associated with childlessness in at least one sex (P<1·5×10−4, after multiple-testing correction), with 33 disease diagnoses shared among women and men (figure 2). A full list of the results can be found in appendix p 27 or in an interactive dashboard at https://dsgelrs.shinyapps.io/DiseaseSpecificLRS/. More than half of the significant associations were mental-behavioral disorders (26/53 in women and 30/54 in men), which was, together with congenital anomalies, the disease categories with the strongest associations with childlessness (OR=3·1 [95% CI: 2·6-3·7] in women compared to men (OR=3·2 [CI: 2·6-4·0]), averaging over all mental-behavioral disorders (figure 2 and appendix p 28). There was substantial heterogeneity among different mental-behavioral disorders (P=4·5×10−71 in women and 7·9×10−128 in men in a heterogeneity test). For example, mild intellectual disability was the condition with the strongest association with childlessness (eg. OR=21·7 [10·8-43·5] in men), while smaller effects were observed for mood disorders (eg. 2·1 [1·8-2·3] in men). Severe diseases of the brain such as cerebral palsy (OR=13·4 [7·9-22·7] in women and 12·0 [6·5-22·2] in men) and malignant neoplasm of the brain (5·5 [2·8-10·8] in women and 5·0 [2·8-8·9] in men), which impair health and functioning in several ways but can also result in behavioral and personality disorders,20,21 were also strongly associated with childlessness. For individuals with severe mental disorders or physical disabilities (eg. intellectual disability), especially amongst women, involuntary sterilization was historically carried out in Finland (from 1935 to 1970) and Sweden (1934 to 1976).22
Odds ratios are computed for each disease diagnosis (left figures of each panel) and averaged over disease categories (right figures of each panel). Only disease diagnoses that are significantly associated with childlessness after multiple-testing correction (P <1·5×10−4) are colored. Labels are assigned only for certain disease diagnoses that are described in the text.
Another disease category strongly associated with childlessness, in both sexes, encompassed the endocrine-nutritional-metabolic disorders (OR=1·4 [1·1-1·8] in women and 2·0 [1·6-2·6] in men, averaging over all endocrine-nutritional-metabolic disorders). For example, a diagnosis of obesity (OR=1·8 [1·5-2·4] in women and OR=2·7 [2·0-3·6] in men), which is normally recorded in the secondary healthcare registers only in severe cases, and type 1 and type 2 diabetes (eg. in women, OR=2·3 [2·0-2·6] for type 1 diabetes and 4·2 [2·2-8·1] for type 2 diabetes) were among the diseases with the strongest associations with childlessness.
We found several inflammatory diseases to be significantly associated with childlessness across multiple organ systems including respiratory, circulatory, genitourinary, digestive, nervous, and musculoskeletal systems. For example, a diagnosis of pneumonia (OR=1·4 [1·3-1·6]), myocarditis (3·7 [2·3-6·1]), chronic tubulo-interstitial nephritis (1·8 [1·3-2·3]), multiple sclerosis (2·1 [1·5-3·1]), systemic lupus erythematosus (2·7 [1·6-4·6]), and juvenile idiopathic arthritis (3·5 [2·6-4·9]) significantly increased subsequent childlessness in women. Puzzlingly, but consistently with previous findings23, we observed that chronic diseases of tonsils and adenoids and acute appendicitis, were associated with reduced, rather than an increased risk of childlessness for women and men (eg. for chronic diseases of tonsils and adenoids, OR=0·84 [0·80-0·88] in women and 0·84 [0·80-0·89] in men), with similar effect sizes in Finns and Swedes (P for heterogeneity=0·60 between two countries in both sexes).
Sex-specific effects
The disease category encompassing congenital anomalies showed markedly different effects between sexes (OR=3·5 [95% CI: 2·6-4·9] in women and 1·8 [1·3-2·5] in men, P for sex difference=3·7×10−3). Overall, malformations of the digestive system (4·8 [2·2-10·3] in women and 1·7 [1·0-3·0] in men, P=0·03) and musculoskeletal system (3·5 [2·0-6·1] in women and 1·2 [0·7-1·9] in men, P=3·7×10−3) were more strongly associated with childlessness in women than men.
Significant sex-dependent effects were also observed for several mental-behavioral and endocrine-nutritional-metabolic disorders (figure 3A). For example, diagnoses of schizophrenia (OR=11·6 [9·1-14·9] in women and 20·8 [16·1-27·0] in men, P for sex difference=1·5×10−3) and acute alcohol intoxication (2·2 [1·6-2·9] in women and 4·7 [3·8-5·7] in men, P=4·0×10−5) had stronger associations with childlessness in men compared to women. Diagnosis of type 1 diabetes showed stronger association in women than in men (2·3 [2·0-2·6] in women and 1·7 [1·5-2·0] in men, P=3·9×10−3). The stronger association in women might be due to the fact that, in the observational period, women with type 1 diabetes were recommended against pregnancies if under poor glycemic control.24
In panel A, we compare the odds ratio for 60 disease diagnoses that significantly increased the odds of childlessness either in men or women. Only disease diagnoses that are significantly different between sexes at a nominal P value are colored. In Panel B we report, for each age category, the average effect associated with childlessness across 30 disease diagnoses in women and 31 in men as obtained from an age-of-onset-stratified analysis. In Panel C, we report age-of-onset-stratified odds ratios associated with childlessness for four major diseases/disorders for which we observe a significant trend. Estimates for each age category are computed only if the number of individuals with this disease within the age group is more than five.
Age-of-onset of disease effects
We hypothesized that the age when the disease was first diagnosed, a proxy for the age of onset, would influence the chances of being childless by either capturing disease severity or directly impacting factors underlying individuals’ reproductive trajectory. We first evaluated if there was a general trend across disease diagnoses (30 in women and 31 in men) that were associated with childlessness and for which we had enough individuals to estimate age-of-onset-stratified effects (figure 3B and appendix p 29). We observed a non-linear effect, with the strongest association with childlessness occurring when the disease was first diagnosed between 21-25 years old in women (OR=3·1 [2·5-3·8]) and later, between 26-30 years old, in men (3·1 [2·4-3·9]). Smaller effects were observed at younger (eg. in women, OR=1·9 [1·6-2·3] for onset before 16 years old, P=7·1×10−4) and, especially, older ages of onset (eg. in women, 1·2 [0·5-2·8] for onset between 36-40 years old, P=0·04). Despite this broader trend, there was substantial heterogeneity between disease diagnoses (figure 3C and appendix p 29). For example, among women diagnosed with obesity, the group that received their first diagnoses between 16-20 years old had higher levels of childlessness than those diagnosed at later ages (OR=3·0 [1·6-5·6] for diagnosed between 16-20 years old and 1·1 [0·5-2·7] between 26-30 years old, P differences=0·08).
Number of unborn children and completed fertility rate reduction related to disease diagnoses
Having identified a set of disease diagnoses that conferred a larger relative risk of childlessness, we therefore estimated the number of children not born in the observational period due to the disease diagnoses by combining relative risk, disease prevalence, and parity. We estimated that 2,728 [2,586-2,871] children per 100,000 men and 1,489 [1,376-1,605] per 100,000 women were not born due to their parents having at least one mental-behavioral disorder (appendix p 33). Accordingly, this disease category was associated with 68% of total unborn children among males and 59% among female participants. Disease-specific analyses (appendix p 31) revealed that schizophrenia was linked to the largest number of children unborn between both sexes (893 [848-930] children per 100,000 men, 612 [549-669] per 100,000 women) followed by anxiety disorders among men (569 [475-665]) and mood disorders among women (389 [313-470]).
The completed TFR was markedly lower among individuals who carry any of the disease diagnoses significantly associated with childlessness (TFR=1·5 for women and 1·1 for men) compared to the unaffected individuals (2·0 for women and 1·8 for men), resulting in an overall TFR reduction of 38% in women and 21% in men (appendix p 35). Overall, affected women and men accounted for 5·4% and 4·6% of overall observed TFR (appendix p 35), respectively.
Mediation effect by singlehood
There were 83·0% women and 77·0% men having registered partners before age 45 and 50, respectively. Restricting to childless individuals, the proportion dropped to 36·3% in women and 29·4% in men. We therefore examined to what degree the associations between disease diagnoses and childlessness were mediated by singlehood. First, we estimated the effect of disease diagnoses on the risk of being without a partner by age 45 in women and 50 in men (appendix p 37). We then compared log ORs on singlehood with log ORs on childlessness and observed a high correlation between these two estimates in both women (R2=0·71, P=1·0×10−15) and men (R2=0·85, P=3·5×10−23) (appendix p 38). A causal mediation analysis performed for each disease also indicated the mediation role of partnership formation, with 29·3% (median) of the disease effects on childlessness in women and 37·9% in men mediated by singlehood (appendix p 38). Different patterns were observed across diseases. For example, singlehood was a significant mediator for women diagnosed with schizophrenia (OR=2·1 [2·0-2·2] for indirect and 6·8 [5·0-9·2] for direct effect) but not for women with hypertension (1·2 [1·1-1·4] for indirect and 2·4 [1·6-3·5] for direct effect).
Next, we investigated if some of the associations between disease diagnoses and childlessness remained significant among partnered individuals. We considered 133 disease diagnoses in women and 123 in men, which had more than 30 affected individuals with registered partners in the sibling-based analysis. Some diseases that were strongly associated with singlehood (eg. mild intellectual disability) could not be included in the analyses because they were too rare among partnered individuals. Nonetheless, we observed six diseases in women and eleven in men that remained associated with childlessness among partnered individuals (appendix p 39). Compared to the estimates obtained from both partnered and unpartnered individuals, the effects on childlessness were largely reduced for most disease diagnoses (appendix p 40), such as epilepsy (eg. from OR=3·7 [3·2-4·3] for all women to 1·8 [1·4-2·4] for partnered women) and acute alcohol intoxication in men (from 4·7 [3·8-5·7] for all men to 2·4 [1·5-3·8] for partnered men). Several endocrine-nutritional-metabolic diseases retained a strong association with childlessness among partnered individuals, such as obesity in men (OR=2·7 [2·0-3·6] for all men and 3·3 [1·9-5·7] for partnered men) and type 1 diabetes in women (2·3 [2·0-2·6] for all women and 2·5 [2·0-3·2] for partnered women).
Population-based and sensitivity analyses
We performed several sensitivity analyses to determine the robustness of our estimation. First, our main results were based on within-sibling analysis, which can account for unmeasured familial risk factors but also made several assumptions, including siblings being generalizable to the population. Thus, we also obtained population-based estimates from a matched case-control design controlling for several potential confounders (appendix p 42). Overall, the population-based design resulted in larger sample sizes, similar ORs, and smaller CIs compared to the sibling-based design (appendix p 43). However, stronger associations were observed from the population-based design in men for psychoactive substance abuse (eg. for alcohol dependence, OR=2·9 [2·5-3·2] from sibling-based design and 3·8 [3·5-4·0] from population-based design, P differences=1·4×10−4) and mood disorders (2·1 [1·8-2·3] from sibling-based design and 2·7 [2·5-2·9] from population-based design, P=2·0×10−4). Second, in the main analysis, we included individuals who died before the end of follow-up (affecting 7·4% full-sisters and 15·0% full-brothers) thus including reproductive-age mortality as one of the possible mechanisms explaining childlessness. To understand the overall effect of reproductive-age mortality, we also conducted an analysis only considering individuals alive by the end of follow-up (appendix p 44). Overall, the results were consistent (appendix p 45), but we observed two disease diagnoses in men for which the main effects on childlessness were partially explained by reproductive age mortality: acute alcohol intoxication (OR=4·7 [3·8-5·8] for all men and 3·9 [2·2-3·8] for men alive by age 50, P differences=0·006) and subarachnoid hemorrhage (2·9 [1·8-4·5] for all men and 0·9 [0·4-1·8] for men alive by age 50, P=0·005). Third, in the main analysis, we opted for conditional logistic regression because the length of follow-up would be identical to the sibling pairs if ignoring reproductive-age mortality (figure 1 and 2). When using a Cox proportional hazards model that captured both time-varying and death (appendix p 46), we observed similar results with reductions in some disease diagnoses (appendix p 47).
Discussion
The rich longitudinal Finnish and Swedish nationwide population registers provided a unique resource to comprehensively assess the role and relative importance of multiple diseases on childlessness. In addition to the results summarized here, all findings can be explored on the interactive online dashboard available at https://dsgelrs.shinyapps.io/DiseaseSpecificLRS/.
Our study provides a new hypothesis-free approach to estimate the burden of diseases through childlessness across the entire reproductive lifespan. Importantly, we identified several novel disease-childlessness associations such as autoimmune (eg. juvenile idiopathic arthritis, multiple sclerosis, systemic lupus erythematosus) and inflammatory diseases (eg. myocarditis). For certain diseases, treatments such as methotrexate (immunosuppressant, widely used for rheumatoid arthritis and multiple sclerosis, and occasionally used for systemic lupus erythematosus) were reported to have side effects on fertility in women.25
Amongst all disease categories, mental-behavioral disorders had the strongest associations with childlessness and were associated with 68% of the total unborn children among males and 59% females attributable to these diseases. Crucially, we provide evidence that childlessness is associated with multiple diseases that could be modifiable with targeted public health interventions. This is particularly the case for mental-behavioral disorders such as alcohol dependence in men or endocrine-nutritional-metabolic disorders linked to obesity and diabetes. Given that these diseases substantially lower lifetime fertility and ultimately the population size, targeted public health interventions could be intensified in these areas.
Previous literature has given much attention to the role that diseases play in childlessness among women, while men have been understudied. We observed substantial sex differences for disease-childlessness associations. For example, mental-behavioral disorders such as schizophrenia and acute alcohol intoxication had stronger associations with childlessness in men than women, while diabetes-related diseases and congenital anomalies had stronger associations amongst women. These sex-specific effects may potentially be explained by differences in disease severity, bias in diagnostic practices, partner choices, and direct biological effects on fertility. Sex differences were also observed in the impact that age-of-onset of disease has on childlessness. Overall, diseases diagnosed between 21-25 years old in women and, later, between 26-30 years old in men, tend to have the strongest association with childlessness. This is also consistent with the sex difference in age at union formation, which occurs later in men.
Besides describing the relative risk of several diseases on childlessness, we evaluated their impact on the completed TFR. We found that the observed contribution of childlessness-associated diseases to changes in TFR is on average larger than the contribution of assisted reproductive techniques.26
Our analyses highlight the important effects that many diseases have on partnership formation. Childless individuals are twice as likely to be single. Singlehood represents a major mediator for the risk of remaining childless, especially in men, and we estimated that 29·3% (median) of the disease effect on childlessness in women and 37·9% in men is mediated by partnership formation. Nonetheless, we also identify several diseases that impact childlessness among partnered individuals. Some of these diseases are more likely to exert a direct biological effect on fertility or pregnancy complications as in the case of diabetes and obesity, whereas others such as some mental-health disorders, might impact family stability or delay the age at family formation.
Our study has several strengths. First, for the majority of the study participants, we have virtually complete coverage of health and reproductive information until the end of their reproductive period. This allows us to estimate public-health-relevant metrics such as the total number of unborn children related to diseases, which cannot be estimated with incomplete follow-up. Second, we considered more than 400 diseases across both men and women allowing us to compare relative effects on childlessness both between diseases and between sexes. The use of nationwide data from two countries provided a large sample of 2·5 million and allowed us to assess how robust our findings were to different healthcare systems and diagnostic practices. Finland has had a marked increase in childlessness since the 1970s of up to 22% for women and 32% for men,2,3 with our analyses revealing potential underlying mechanisms such as the wider detrimental effects of alcohol dependence in men and endocrine-nutritional-metabolic disorders. Third, by leveraging the information about family formation and singlehood, we could explore to what extent this information, which has been described as an important mediator of childlessness27, mediated the association between disease and childlessness. Fourth, we used a matched-sibling design to limit confounding from a shared familial environment. In our study, individuals with siblings had only a slightly lower proportion of childlessness compared to the general population. Sibling designs have been extensively used in analyses of Nordic registers and have several advantages compared to population-based estimates.28,29
Our study also has several limitations. First, disease diagnoses are largely obtained from registers covering secondary and tertiary healthcare in hospitals that capture more severe disease cases. Moreover, disease diagnoses have changed over time; we have defined disease outcomes by harmonizing three versions of ICD codes that might capture diseases with different accuracy. Our findings are based on individuals born between 1956 and 1973 and results may thus not be generalizable to more recent cohorts both because reproductive practices have changed and because better treatments might alleviate the effect of some diseases on childlessness. Second, we were unable to partition the impacts that diseases have into voluntary and involuntary childlessness due to a lack of data on reproductive preferences and intentions. Among factors that impact involuntary childlessness, the effect of long-term institutionalization and disease-related treatments are two aspects that future registry-based studies might be able to explore. For example, medications for epilepsy are known to be associated with pregnancy complications.30 Third, assortative mating may lead to an overestimation of the disease impact of childlessness among partnered couples. Fourth, despite our attempt to use a study design that minimizes confounding and reverse causality, these biases might challenge the interpretation of the results. Finally, it would be important to replicate this study beyond Nordic countries using population register data to explore differences in populations, practices, and treatments on childlessness.
In conclusion, we comprehensively describe the effects different diseases have on the risk of remaining childless over a lifetime. Public-health practitioners can use these results to inform how disease prevention might help reduce involuntary childlessness in the population.
Data Availability
All results (aggregated data) can be explored on the interactive online dashboard available at https://dsgelrs.shinyapps.io/DiseaseSpecificLRS/. Due to data protection regulations, we are not allowed to share individual-level data directly. However, all Finnish and Swedish register data used in this study can be applied from national data agencies listed in appendix p 2.
Contributors
AG, MCM, and AL designed the study and wrote the manuscript. AL performed data cleaning and conducted the statistical analysis. ETA calculated the number of unborn children related to diseases and total fertility rate and wrote the relevant paragraph. XD helped in performing sensitivity analysis and drafted figure S1. SJ, PV, and TK assisted in interpreting the results from a clinical standpoint. TK and ASH defined disease endpoints. SK, AG, and AL developed the R shiny application. TF and MG coordinated data application and extraction for Swedish and Finnish nationwide registers. SL contributed to the editing of socio-economic data and statistical modeling. All authors discussed the results, revised the manuscript, and had final responsibility for the decision to submit for publication.
Declaration of interests
We declare no competing interests.
Data sharing
All results (aggregated data) can be explored on the interactive online dashboard available at https://dsgelrs.shinyapps.io/DiseaseSpecificLRS/. Due to data protection regulations, we are not allowed to share individual-level data directly. However, all Finnish and Swedish register data used in this study can be applied from national data agencies listed in appendix p 2.
Acknowledgments
This work was supported by European Research Council (835079, 945733) and The Leverhulme Trust, Leverhulme Centre for Demographic Science. The computations for the Swedish data were enabled by resources in project sens2019018 provided by the Swedish National Infrastructure for Computing (SNIC) at UPPMAX, partially funded by the Swedish Research Council through grant agreement no. 2018-05973. We thank Therese Andersson (Karolinska Institutet), Amir Sariaslan (University of Helsinki), and Guosheng Su (Aarhus University) for valuable advice on statistical modeling, Gianmarco Mignogna (Massachusetts General Hospital and University of Helsinki) and David Brazel (University of Oxford) for useful discussions during project planning, and Vincent Llorens (University of Helsinki) for suggestions on disease coding and applying Cox proportional hazards regression.