Abstract
Introduction:
With later marriage in South Korea, infertility is a growing concern. We assessed whether everyday health behaviors are associated with infertility among women of reproductive age.
Materials and Methods:
We conducted an online survey of women aged 20–49 years who had experienced infertility or childbirth since January 2020. Data were collected from October 5 to 10, 2023. Of 398 participants, 80 (20.1%) reported infertility and 318 (79.9%) reported childbirth. Group characteristics were compared, and the multivariate binary logistic regression identified factors associated with infertility.
Results:
Higher odds of infertility were observed for women over 35 compared with women of 20–34 years (adjusted odds ratio [aOR]: 2.57, 95% confidence interval [CI]: 1.33–4.99); for vegetable consumption on three days per week or fewer compared with four days or more (aOR: 2.86, 95% CI: 1.43–5.72); for no moderate-intensity physical activity compared with at least 1 day per week (aOR: 2.01, 95% CI: 1.00–4.02); and for anxiety (aOR: 11.48, 95% CI: 2.94–44.79). Current drinking (aOR: 2.65, 95% CI: 1.37–5.12) was positively associated with infertility. Each additional year of cohabitation with a spouse was associated with lower odds of infertility (aOR: 0.78, 95% CI: 0.66–0.92).
Conclusions:
These findings provide preliminary evidence that supporting health-related daily lifestyles—regular moderate-intensity activity, adequate vegetable intake, and prudent alcohol use—may help mitigate infertility risk.
Introduction
Infertility affects approximately one in six people worldwide as of 2023, with the prevalence rising by 0.37% every year. 1 The prevalence of infertility is particularly increasing in developing countries in South Asia, the Middle East, and North Africa. 2 Infertility is defined as a reproductive disease in women or men who fail to conceive despite regular sexual intercourse for more than 12 months. 3 As of 2022, approximately 153,000 women sought care for infertility in South Korea. 4 Considering those who did not undergo assisted reproductive technologies or utilize healthcare services, the true number may be higher.
The low birth rate has become a serious social problem in South Korea. In 2023, South Korea’s total fertility rate was 0.72, markedly lower than the OECD average of 1.59. 5 As the age at marriage and childbearing has increased, the public health importance of infertility has grown. Accordingly, infertility treatments have been covered by the National Health Insurance of Korea since 2017, with about 120,000 beneficiaries annually. Since July 2019, more infertile couples have benefited from coverage because age restrictions were removed and the number of reimbursed treatment cycles increased. 6
Many women in Korea remain unaware of their fertility status, due to low recognition, postponement of pregnancy for work or financial reasons, and difficulty obtaining assessment. 7 Clinically, causes of female infertility include sexual dysfunction, ovulation disorders, fallopian tube obstruction, abnormalities of the uterine body and cervix, and irregular menstrual cycles.8–10 In addition, lifestyle-related factors—diet, psychological stress, alcohol consumption, smoking, and obesity—have been implicated in infertility,11,12 with age, smoking, heavy drinking, obesity, and excessive physical or emotional stress frequently linked to higher risk.13,14
Although there have been studies on risk factors for infertility in Korea, few have included women who have given birth as a comparison group.15,16 There is a need for empirical evidence focusing on everyday health-related behaviors among women with infertility and those who have experienced childbirth. Therefore, this study aimed to develop a survey to measure the health conditions and behaviors of women of childbearing age and to explore associations between lifestyle factors and self-reported infertility through an online survey, using women with a recent live birth as the comparison group.
Materials and Methods
Questionnaire development
A survey was developed to identify factors related to infertility in women of childbearing age. We developed survey items by reviewing literature on dietary habits, physical activity, and sleep, which were reported to be associated with infertility. The major categories that consisted of the common items were “health-related characteristics” and “general status.” Health-related characteristics included subjective health status, chronic disease, diet and nutrition, physical activity, sleep, perceived body size (obesity), drinking, smoking, stress, and mental health (depression, anxiety). The questionnaire consisted of 31 items for those who experienced infertility and 26 items for those who experienced pregnancy and childbirth. There were 21 common items, with 10 items specific to infertility and 5 items specific to childbirth (Supplementary Data).
A pilot survey was conducted to determine the appropriateness of the questionnaire, and those who experienced childbirth or infertility reviewed it to see if there were any items they could not understand or found difficult when viewing it online. Feedback from the pilot was used to refine wording and response options. The final questionnaire was posted on a social media platform (Instagram) that many infertile and pregnant women used. In accordance with the Declaration of Helsinki, the survey landing page described the study purpose, use of data, and potential benefits and risks. The participants were provided with voluntary informed consent by clicking an electronic consent button. This study was approved by the Public Institutional Bioethics Committee designated by the Ministry of Health and Welfare (Korea National Institute for Bioethics Policy, KONIBP, IRB No. P01-202309–01-037).
Participants and recruitment
This study targeted women aged 20–49, considering the marriageable age in Korea, even though the World Health Organization (WHO) defines a woman’s reproductive age as 15–49 years old. 17 Eligible participants were women aged 20–49 who reported either infertility or pregnancy/childbirth occurring on or after January 2020. The inclusion criteria were (1) women aged 20–49 years and (2) women who experienced infertility or pregnancy and childbirth after January 2020. The exclusion criteria were (1) those who experienced pregnancy and childbirth before January 2020 and (2) those who were pregnant but had not yet given birth.
To recruit study participants, the survey was promoted and conducted through a social media platform (Instagram) that many infertile and pregnant women use in South Korea. Data were collected from those who voluntarily participated in the study between October 5 and October 10, 2023. A total of 434 women responded, including 354 women who experienced pregnancy and childbirth and 80 women who experienced infertility. We excluded 36 respondents who reported childbirth before January 2020 or provided invalid key responses (e.g., missing the item on childbirth/infertility status). Finally, 398 women—318 with childbirth experience and 80 with infertility—were included in the analysis (Fig. 1).
Flowchart of study participants.
Variables and measurements
Outcome variable
WHO defines infertility as a condition in which a healthy couple of reproductive age cannot conceive for more than a year despite engaging in regular sexual activity without using contraception. 3 Therefore, in this study, participants were classified as infertile if they self-reported being unable to conceive with their current spouse or partner for at least 12 month within the 3-year recall window preceding the survey (September 2020 to the survey date). Receipt of infertility evaluation or treatment did not alter classification; individuals meeting the ≥ 12-month criterion were considered infertile regardless of treatment history. Women who reported a live birth during the same period served as the comparison group.
Explanatory variables
The participants’ demographic characteristics, spouse’s characteristics, lifestyle factors, and abnormal reproductive system symptoms were investigated. Regarding demographic characteristics, the ages of the participants were divided into 20–34 years and 35–49 years based on the definition of advanced maternal age of 35 years. 18 The areas of residence were categorized as metropolitan areas (Seoul, Gyeonggi-do, and Incheon) and nonmetropolitan areas. The status of employment was categorized as unemployed, regular employee, and self-employed/temporary/day workers/unpaid family workers. For married or partnered participants, spouse age (20–34 vs. ≥ 35 years) and period of cohabitation (years; continuous) were collected.
To measure health-related daily lifestyle, we assessed diet, physical activity, sleep, body perception, alcohol use, tobacco use, mental health, reproductive-system symptoms, and chronic conditions. For diet, frequencies of eating breakfast and consuming protein (meat, fish, eggs), vegetables (including kimchi and leafy or green vegetables such as seasoned vegetables and salads), dairy products, and desserts were each categorized as three days per week or fewer versus four days per week or more. Similarly, the frequency of eating out (delivery food, convenience store meals, instant food, excluding meals provided at work or school) and late-night snacking were categorized as less than three days or more than four days a week.
To measure physical activities, the average number of days per week engaging in high-intensity physical activities (exercise that can reach maximum heart rate such as sprint running, burpees, and swimming), moderate-intensity physical activities (exercise that is difficult to do while singing), and walking or cycling for more than 10 minutes was categorized as being more than one day per week versus none. Sleep behavior was assessed as going to bed before midnight and sleeping 7–8 hours, each categorized as three days per week or fewer versus four days a week or more. Body perception referred to self-perceived body size and was categorized as underweight (very or slightly underweight), normal, or obese (slightly or very obese).
Mental health was measured by self-rated daily stress (high vs. low), depression screened using the Patient Health Questionnaire-9 (PHQ-9), and anxiety screened using the Generalized Anxiety Disorder-7 (GAD-7). A PHQ-9 score of 10 or more (range, 0–27) was defined as depression. 19 A GAD-7 score of 10 or more was defined as high-risk for generalized anxiety disorder. 20
Smoking was assessed as current smoking (yes or no) across all tobacco types, including combustible cigarettes, heated tobacco products, and nicotine-containing e-cigarettes. Alcohol use was assessed as current drinking (yes/no), defined as any alcohol consumption within the past 12 months.
As indicators of health status, reproductive system symptoms were recorded as the presence (yes or no) of irregular menstruation, dysmenorrhea, or abnormal vaginal discharge in the past year. Perceived health was self-rated as good or bad. For 12 common chronic diseases (hypertension, diabetes, dyslipidemia, osteoarthritis, rheumatoid arthritis, thyroid disease, stroke, myocardial infarction, angina, pulmonary tuberculosis, asthma, and cancer), participants indicated whether they were currently receiving treatment or taking medication; responses were recorded as yes or no.
Statistical analysis
We first summarized sample characteristics and compared the infertility and childbirth groups using frequencies (percentages), means and standard deviations, the chi-square test for categorical variables, and the t-test for continuous variables. Candidate predictors included sociodemographic characteristics, spouse/partner characteristics, lifestyle indicators, and abnormal reproductive system symptoms. Binary logistic regression was used to estimate adjusted odds ratios and 95% CIs. Model 1 included only women’s variables; Model 2 additionally included spouse/partner age and years of cohabitation. To select a parsimonious set of predictors while prioritizing clinically relevant lifestyle variables, we applied backward elimination as a secondary check; the variables retained for the final models were women’s age group, vegetable consumption, dessert consumption, moderate-intensity physical activity, sleep, body perception, current drinking, stress, anxiety, abnormal reproductive-system symptoms, chronic diseases, spouse/partner age, and period of cohabitation. Employment status was treated as an adjustment variable rather than a lifestyle exposure. Current smoking differed markedly between groups at the descriptive level; however, because it was rare in the childbirth group, the variable was adjusted for in the model but not presented. All tests were two-sided with α = 0.05, and analyses were performed using SPSS version 28.0 (IBM Corp., Armonk, NY, USA).
Results
General characteristics of the study participants
Of the 398 participants, 80 (20.1%) reported infertility and 318 (79.9%) had given birth. The mean age was 34.9 years in the infertility group and 32.1 years in the childbirth group. The proportion of women over 35 years old was higher in the infertility group than in the childbirth group (56.3% vs. 27.0%). Unemployment was reported by 43.8% of women with infertility and 83.6% of women who had given birth. Residential distribution was similar, with approximately 70% of both groups living in metropolitan areas. The mean duration of cohabitation with a spouse was 4.7 years among women with infertility and 4.9 years among women who have given birth. Spouses aged ≥35 years were more common in the infertility group than in the childbirth group (73.6% vs. 46.0%).
Among women meeting the WHO infertility criteria, 61 out of 80 (76.3%) had undergone hospital-based infertility evaluation, most often in pursuit of a first birth. Overall, 52.5% had experienced infertility procedures, and among the most recent procedures, 90.5% were in vitro fertilization (IVF). Among women who gave birth after 2020, the mean number of children was 1.19 (Table 1).
General Characteristics of the Study Participants
No response: 20.
No response: 13.
SD, standard deviation.
Comparison of lifestyle and health conditions of women with infertility and women with childbirth
The lifestyle-related factors analyzed were diet, physical activity, sleep, body perception, current drinking and smoking status, mental health, and health conditions. By contrast, consuming desserts on four days per week or more was markedly higher in the infertility group (42.5% vs. 28.6%). There were no significant differences between groups for eating breakfast, consumption of vegetables, protein intake, fruit intake, eating out, or late-night snacking.
For physical activity, the proportion of engaging in high-intensity or moderate-intensity physical activity and walking or cycling at least one day per week was higher in the childbirth group than in the infertility group. There was no significant difference between groups in going to bed before midnight and sleeping 7–8 hours on four days per week or more. Regarding body perception, normal weight was more frequent in the childbirth group (56.9%), whereas both underweight (20.0%) and obesity (45.0%) were more frequent in the infertility group; those differences were statistically significant. Current drinking was more common among women with infertility than with childbirth (67.5% vs. 43.4%), and the prevalence of current smoking was also higher in the infertility group.
High perceived stress was reported by 55.0% of women with infertility compared with 75.2% in the childbirth group. Depression and anxiety levels were higher among women with infertility. Health conditions were assessed using abnormal reproductive system symptoms, perceived health, and chronic diseases. Experiencing abnormal reproductive-system symptoms in the past year was more common in the infertility group (66.3% vs. 56.0%), whereas perceived health did not differ between groups. Chronic disease prevalence was about twice as high in the infertility group as in the childbirth group (26.3% vs. 13.8%); thyroid disease specifically was more frequent among women with infertility (10.0% vs. 4.7%) (Table 2).
Lifestyle and Health Conditions of Women with Infertility and Women with Childbirth
Factors associated with infertility
Using multivariate binary logistic regression, the adjusted odds of infertility were higher among women aged 35 years or older than among those aged 20–34 years (aOR: 2.57, 95% CI: 1.33–4.99). Compared with being unemployed, both regular employment (aOR: 6.17, 95% CI: 3.01–12.64) and self-employment or other employment (aOR: 17.93, 95% CI: 6.13–52.39) were associated with higher odds. With respect to diet, vegetable consumption on 3 days per week or fewer was associated with higher odds of infertility than consumption on 4 days per week or more (aOR: 2.86, 95% CI: 1.43–5.72), whereas dessert frequency showed no significant association. Women who reported no moderate-intensity physical activity had higher odds than those active on at least 1 day per week (aOR: 2.01, 95% CI: 1.00–4.02). Anxiety was strongly associated with higher odds compared with no anxiety (aOR: 11.48, 95% CI: 2.94–44.79). In contrast, reporting 7–8 hours of sleep on fewer than 3 days per week was associated with lower odds of infertility (aOR: 0.37, 95% CI: 0.18–0.75). After additionally including spouse or partner variables in Model 2, each additional year of cohabitation with a spouse was associated with lower odds (aOR: 0.78, 95% CI: 0.66–0.92) (Table 3).
Factors Associated with Infertility
The employment status of women was adjusted in each regression model.
The adjusted coefficient for current smoking was not estimable due to quasi-complete separation caused by the rarity of events in the childbirth group; therefore, the variable was adjusted for in the model but not presented.
aOR, adjusted odds ratio; CI, confidence interval; Ref., reference category.
Discussion
In this study, we developed and administered an online survey to assess daily lifestyle and health conditions among Korean women aged 20–49 years who had experienced either infertility or childbirth. Of the 398 participants, 318 women had given birth, and 80 reported infertility. In multivariate binary logistic regression, infertility was associated with older female age, spouse or partner’s age, less frequent vegetable consumption, absence of moderate-intensity physical activity, current drinking, and anxiety. We also observed an inverse association for sleeping seven to eight hours on three days per week or fewer, and employment status showed positive associations; however, these findings require cautious interpretation due to the possibility of reverse causation (i.e., continued employment because of not having children).
The proportion of participants reporting infertility (20.1%) is broadly consistent with estimates from Korea. The National Family and Childbirth Survey reported a prevalence of 17.2% among Korean adult women in 2021. 21 In a survey conducted in Seoul in 2019, 19.5% of women were reported to be experiencing infertility, which is similar to the results of this study. 16 Globally, the prevalence of infertility is 12.9%, 22 but estimates vary substantially across countries. 23 For example, point estimates of about 12.5% have been reported in the United Kingdom in 2012 8 and in the United States during 2013–2016, 24 whereas higher values are observed across Central, South, and Southeast Asia, ranging from 12% to 25.9%. 25
Female age demonstrated a robust association, with higher odds among women aged ≥ 35 years, concordant with evidence linking delayed marriage and childbearing to reduced fecundity.8,21 Associations with employment status were also observed; however, causal interpretation is not justified in a cross-sectional design. Reverse causation (e.g., continued employment because of not having children) and residual confounding by work hours, job stress, or occupational exposures are plausible. Consequently, inferences such as “resignation increases pregnancy chances” are unwarranted, notwithstanding policies intended to support leave for infertility treatment in Korea.26–29
Regarding eating habits, vegetable consumption on three days per week or fewer was associated with higher odds of infertility relative to consumption on four days per week or more, aligning with literature that links healthier dietary patterns to improved reproductive outcomes and higher in vitro fertilization success rates.30,31 Dessert frequency was not significantly associated with infertility in the present models. Prior findings on carbohydrate-rich foods are mixed—some studies suggest adverse metabolic effects on ovulation, 32 whereas others report lower infertility with moderate sweet-food intake, 8 highlighting the need for more granular dietary measurement.
Women who reported no moderate-intensity physical activity had higher odds of infertility than those active on at least one day per week, consistent with reports that moderate activity may support fertility and treatment success. 33 By contrast, reporting 7–8 hours of sleep on fewer than three days per week was associated with lower odds of infertility. This direction conflicts with reviews indicating that short sleep may impair reproductive function and that 7–8 hours of sleep is associated with better outcomes, including in assisted reproduction.34,35 The contrasting results of this study may be due to the control group consisting of women with childbirth within the last three years, among whom infant care plausibly reduces sleep duration and increases stress. Such postpartum circumstances can induce reverse causation and collider bias, and the observed inverse associations should therefore be interpreted with caution.
Although perceived underweight and obesity were not significantly associated with infertility in this study, prior literature has emphasized the importance of maintaining a normal weight: a body mass index below 17 has been linked to increased ovulatory infertility,14,36 and obesity has been associated with roughly a threefold higher risk of infertility compared with normal weight.37,38 We also noted a positive association between current drinking and infertility. Although heavy drinkers are more likely to undergo infertility evaluation, evidence for a direct association with infertility per se is mixed and remains inconclusive. 11
Anxiety was associated with higher odds of infertility in our analyses. Consistent with prior work, approximately 36% of women reported anxiety related to infertility, a higher proportion than among women without infertility,11,37,39 and lower anxiety has been linked to improved outcomes in assisted reproductive treatment. 40 By contrast, we did not detect a significant association between self-reported stress and infertility. Nonetheless, prior studies suggest that psychological stress may affect oocyte maturation and diminish ovarian reserve,41,42 and although cortisol, a prototypical stress hormone, can influence reproductive function, its relationship with IVF outcomes remains controversial. 43 Taken together, anxiety and infertility are likely bi-directionally related—infertility and its treatment may elevate anxiety, and anxiety may in turn influence reproductive outcomes—highlighting the need for longitudinal research to clarify causality.
We did not find significant adjusted associations for abnormal symptoms of the reproductive system or chronic disease with infertility. However, previous studies have linked gynecological conditions and suboptimal genital hygiene with elevated infertility risk. 44 A U.S. cohort study reported that 31.6% of infertility cases involved ovulatory disorders, and women with infertility were less likely to report regular menstrual cycles. 9 In a national, population-based study in China, irregular cycles and menstrual flow abnormalities were associated with infertility. 8 Moreover, chronic diseases has been associated with a higher likelihood of an infertility diagnosis and adverse delivery outcomes. 45 Our null findings may reflect self-report measurement error, limited statistical power, or residual confounding.
Among spouse factors, longer cohabitation duration was associated with lower odds of infertility. Spouse factors—including age, stress, nutrition, physical activity, and environmental exposures—can affect fertility through changes in semen quality. 46 Because our study included only partner's age and cohabitation duration, interpretation is limited; future work should incorporate more comprehensive data on spouse’s lifestyles and clinical measures to inform couple-based prevention and counseling.
This study has several limitations. First, the online survey may have made items involving medical terminology (e.g., abnormal reproductive system symptoms such as ovulatory disorders) difficult for some participants. Second, convenience sampling and modest sample size limit generalizability. Although we restricted eligibility to experiences since 2020, recall bias remains possible owing to variable intervals from pregnancy, childbirth, or infertility. Third, while we identified specific associations with vegetable intake and alcohol use, these factors may collectively represent a generally healthy lifestyle rather than independent predictors. It is possible that the comparison group, having recently experienced pregnancy, maintained healthy behaviors adopted for maternal and fetal health. Thus, the observed differences might reflect the persistence of pregnancy-related lifestyle modifications rather than preconception behaviors, highlighting the potential for reverse causation. Fourth, detailed spouse/partner clinical information (e.g., male infertility diagnoses) was not collected. Fifth, the sample was accrued over six days, further limiting representativeness. Sixth, current smoking differed substantially between groups descriptively, but because it was very rare in the childbirth group, the adjusted coefficient could not be estimated (quasi complete separation); thus, we report only descriptive differences for smoking. Finally, due to the cross-sectional design, the results are limited to associations rather than causal effects.
Conclusions
Using survey data from 398 women aged 20–49 years (318 with recent childbirth and 80 reporting infertility), we found that infertility was associated with older female age (≥35 years) and several daily lifestyle indicators, including less frequent vegetable consumption, absence of moderate-intensity physical activity, current drinking, and anxiety. After accounting for spouse factors, longer cohabitation was inversely associated. These findings emphasize the importance of everyday health behaviors among women who have experienced infertility and suggest that for women of childbearing age, adopting healthy daily routines—regular moderate-intensity activity, adequate vegetable intake, and prudent alcohol use—may help mitigate the risk of infertility and can inform preventive interventions. Because this study used a cross-sectional, self-reported, convenience sample, the results should be interpreted as associations rather than causal effects. Future studies should employ longitudinal, couple-based designs incorporating comprehensive, standardized behavioral and clinical assessments for both women and their spouses to clarify causal pathways and to guide family-centered prevention.
Authors’ Contributions
Conceptualization: B.J. and J.S. Methodology: B.J. and J.S. Formal analysis: S.M. Data curation: B.J. and S.M. Writing—original draft: B.J., S.M., S.P., S.W.C., and S.Y. Writing—review and editing: B.J., S.P., S.W.C., S.Y., and J.S. Supervision and project administration: B.J. All authors have read and approved the final article.
Footnotes
Author Disclosure Statement
The authors declare no conflicts of interest.
Funding Information
Following are results of a study on the “Leaders in INdustry-university Cooperation 3.0” Project, supported by the Ministry of Education and National Research Foundation of Korea. The funders were not involved in study conceptualization and design, data analysis, interpretation of the results, or article submission.
Data Availability
All data underlying the findings are included in this article and its supplementary material. De-identified data and the survey instrument are available from the corresponding author upon reasonable request, subject to institutional and ethical approvals.
Supplemental Material
Abbreviations
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
