==== Front Front Public Health Front Public Health Front. Public Health Frontiers in Public Health 2296-2565 Frontiers Media S.A. 10.3389/fpubh.2023.1179416 Public Health Original Research Psychosocial factors associated with health behaviors in pregnant women of advanced maternal age in Korea Jeon Songi 1 Noh Wonjung 2 * 1Department of Nursing, Catholic Kwandong University, Gangneung-si, Gangwon-do, Republic of Korea 2College of Nursing, Gachon University, Incheon, Republic of Korea Edited by: Rubén González-Rodríguez, University of Vigo, Spain Reviewed by: Rita Barros, Piaget Institute, Portugal; Forough Mortazavi, Sabzevar University of Medical Sciences, Iran *Correspondence: Wonjung Noh wjnoh@gachon.ac.kr 15 6 2023 2023 11 117941604 3 2023 12 5 2023 Copyright © 2023 Jeon and Noh. 2023 Jeon and Noh https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Objectives To determine the association between psychosocial factors and health-promoting and health-impairing behaviors in pregnant women of advanced maternal age (AMA) in Korea. Design A cross-sectional survey study. Setting Online survey. Samples A total of 217 pregnant women aged 35 and older agreed to participate in the study, with 207 participants completing the self-report questionnaires. Methods We collected self-reported data on demographic, obstetric, and psychosocial factors and prenatal health behaviors using standardized measures. We conducted a descriptive analysis of the collected data and a linear regression to identify significant associations with health-promoting and health-impairing behaviors. Results We found that maternal–fetal attachment (β = 0.43, p < 0.001) and “social atmosphere” of pregnancy stress (β = 0.13, p = 0.047) were positively associated with prenatal health-promoting behaviors. We found that artificial conception (β =-0.16, p = 0.011) was negatively associated with prenatal health-impairing behaviors and that multiparity (β = 0.23, p = 0.001) and “maternal role” of pregnancy stress (β = 0.27, p = 0.003) positively associated with prenatal health-impairing behaviors. Conclusion Health-impairing behaviors of pregnant AMA women need assessment and the importance of health-promoting behaviors for maternal and infant health need reinforcing. We recommend pregnancy stress assessments at prenatal checkups and stress relief interventions that consider cultural differences and contexts rather than standardized interventions. advanced maternal age health-impairing behavior health-promoting behavior pregnant women psychosocial factors This study was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2019S1A5B5A07087228). section-at-acceptancePublic Health Education and Promotion ==== Body pmc1. Introduction Women are delaying childbirth in advanced countries (1). In Korea, the average age of women giving birth was 32.6 years, the highest among Organization for Economic Cooperation and Development (OECD) countries in 2017 (2). Korean women aged 35 and older accounted for approximately 33.8% of all births in 2020 (3). Poor pregnancy and childbirth outcomes, including gestational diabetes mellitus (GDM), gestational hypertensive disorders (GHDs), placenta previa, congenital malformations, miscarriage, stillbirth, premature birth, low birth weight, and postpartum hemorrhage, are common among pregnant women of advanced maternal age (AMA) (4–6). The rate of premature births and the incidence of low birth weights are on the rise in Korea (3). The increased proportion of pregnant women of AMA, and the resulting detrimental pregnancy and childbirth outcomes not only have a negative impact on maternal and child health but also contribute to personal and national economic losses because of increased healthcare costs; thus, management and support for pregnant women of AMA are required (7). Lin et al. (8) reported that pregnant women of AMA could avoid negative childbirth outcomes through management, such as proper diet and physical activity, and many studies have shown that prenatal health behaviors lead to healthy birth outcomes (9–11). Therefore, it is necessary to explore which characteristics relate to the health behavior of pregnant women of AMA. However, most studies have focused on pregnant women of all ages, and studies on pregnant AMA women are limited. Pregnant AMA women perceived pregnancy as more threatening than young women did (12), and they tended to engage in healthier behaviors (13). Hence, applying the findings of studies on pregnant women of all ages to pregnant women of AMA can make understanding the factors influencing prenatal health behaviors difficult. Furthermore, compared with other countries, pregnant women in Korea are under a lot of social pressure, which is natural given their role as mothers to behave for the health of their fetus (14). Previous studies have explored which factors relate to prenatal health behaviors. A meta-analysis study on pregnant women's health behavior reported age, employment, income, education, parity, maternal–fetal attachment, stress, depression, and social support as predictive factors (15). Based on this, we divided prenatal health behavior-related factors into demographic, obstetric, and psychosocial factors. As a demographic factor, the higher the level of education and income, the healthier the behaviors tended to be (16, 17). Obstetric characteristics, such as current conception type, gestational age, parity, and experience of abortion, also relate to prenatal health behavior: Pregnant women in the third trimester than those in the second trimester, and younger pregnant women than older pregnant women engaged in less healthy behaviors (13, 18). Meanwhile, some studies have considered psychosocial factors as they could otherwise enhance these components through interventions when compared with demographic and obstetric characteristics. According to previous research, psychosocial factors influence prenatal health behaviors: Pregnant women's elevated self-esteem levels, fetal attachment, and social support lead to more healthy behaviors, whereas high levels of depression and stress lower healthy behaviors (19–21). Prenatal health behaviors are classified as either health-promoting (e.g., exercise, adequate sleep, and nutrition) or health-impairing behaviors (e.g., inappropriate physical activity, smoking, drinking alcohol, and caffeine intake). Health-promoting behaviors necessitate consistent efforts, whereas health-impairing behaviors are reactive to situations and mood; thus, factors influencing health-promoting and health-impairing behaviors may differ (22). Accordingly, some previous research has identified psychosocial factors influencing both prenatal health-promoting behaviors (23, 24) and prenatal health-impairing behaviors (25). One study even simultaneously reported factors influencing prenatal health-promoting and health-impairing behaviors (26). These previous studies, however, have limitations. For example, they limited the measures of prenatal health-impairing behaviors to smoking and drinking while they limited prenatal health-promoting behaviors to physical activity and exercise. Therefore, the influencing factors should be identified by categorizing health behaviors, including nutrition and eating habits, physical activity, and exposure to hazardous substances, into prenatal health-promoting and health-impairing behaviors. Figure 1 shows the theoretical framework of this study. Figure 1 Theoretical framework. This study aimed to determine the association between psychosocial factors and health-promoting and health-impairing behaviors in pregnant women of AMA in Korea. The findings of this study may aid in the development of interventions for prenatal health behavior reinforcement based on the psychosocial factors of pregnant women of AMA in Korea, where AMA rates are high. 2. Materials and methods 2.1. Design This was a cross-sectional survey study design. 2.2. Sample Participants in this study were women in low-risk pregnancies aged 35 years or older. The inclusion criteria applied to women who could comprehend and answer the questionnaires in Korean, whereas the exclusion criteria applied to women with multiple pregnancies (e.g., twin and triplet pregnancy) and pregnancy complications [e.g., gestational diabetes mellitus (GDM), gestational hypertensive disorders (GHDs), and placenta previa]. The criteria for calculating the sample size were a level of significance of 0.05, a power of 0.90, and an effect size of 0.15. The sample size was calculated as 202 people using the G*Power program 3.1.9.7 (27). 2.3. Data collection Pregnant women of AMA enrolled in response to recruitment notices in online communities (Naver Cafe and Kakao Open Chatting, with the former being the most famous among community-type social network services, and the latter among open chatting platforms in Korea) of pregnant women and at local obstetrics clinics, implying convenience sampling. We gathered data online because the Internet and mobile penetration rates in Korea are very high, so anyone, regardless of education or income level, can easily access the online survey. Pregnant women keen on this study accessed the online survey platform, SurveyMonkey via QR code or Uniform Resource Locator (URL). They reviewed the information (the objective of the study and process of participation, disadvantages/risks and rewards for participation, a statement that the collected data will not be used for any intent other than the purpose of this study, and a statement that participants can quit at any time if they wish to discontinue participation), and then, they agreed to participate in this study. We screened and recruited participants who met the study's selection criteria by inquiring about their age, pregnancy complications, and multiple pregnancies. Additionally, we verified whether the respondent's mobile number to receive the mobile coupon as a reward was duplicated; whether the response time to complete the survey was too short; and whether the estimated day of confinement (EDC) matched gestational weeks. A total of 217 pregnant women volunteered to partake in the study between September and November 2020, but only 207 participants completed the self-report questionnaires via the online survey platform. Our online poll did not include any missing information. 2.4. Measures 2.4.1. Prenatal health behaviors We assessed prenatal health-promoting behaviors and prenatal health-impairing behaviors using the Prenatal Health Behavior Scale (PHBS), including items on sleep, physical activity, nutrition, smoking, and alcohol consumption (28). Studies have validated this scale, originally designed for women in their mid- and late pregnancy, by applying it to women in the initial stages of pregnancy (29). The PHBS assesses health-promoting and health-impairing behavior on a five-point Likert-type scale (from 1 to 5) with 10 items each. The higher score for health-promoting behavior and the lower score for health-impairing behavior denote a higher degree of health behavior. Cronbach's alphas for the original PHBS were 0.75 and 0.59, respectively, for health-promoting and health-impairing behaviors (28). In an earlier study (30) that assessed 20 PHBS items in pregnant women at all pregnancy stages as in this research study, Cronbach's alphas were 0.72 and 0.55 for health-promoting and health-impairing behaviors, respectively. In this study, Cronbach's alphas were 0.76 for health-promoting and 0.70 for health-impairing behaviors. 2.4.2. Demographic and obstetric characteristics According to research, general, and socioeconomic statuses (e.g., age, education, and income) relate to health behaviors (16, 17). Therefore, we gathered information about participants' age, employment status, religion, household income, and education. We also included the current conception type, gestational age, parity, and the number of spontaneous abortions, all of which link to health behaviors (13, 18, 30). Table 2 contains information on demographic and obstetric characteristics. 2.4.3. Psychosocial factors We assessed self-esteem using the Rosenberg Self-esteem Scale (SES) (31). SES consists of 10 items, and a four-point Likert-type scale (from 1 to 4) measures the scores: the higher the score, the higher the self-esteem. Cronbach's alpha for internal consistency reliability of the original version of the SES was 0.85 (31), and it was 0.90, and 0.86 in the Korean version (32) and this study, respectively. We evaluated prenatal depression using the Edinburgh Postpartum Depression Scale (EPDS) (33). EPDS includes common depressive symptoms related to pregnancies, and it is a viable tool during pregnancy (34); although no scale can examine depression during pregnancy (35), EPDS has been used to measure pregnancy-related depression in numerous kinds of research, including in Korea (36). EPDS comprises 10 items that a four-point Likert-type scale (from 1 to 4) evaluates, where a higher score indicates more severe depression. Cronbach's alpha for EPDS was 0.92 (33), and Cronbach's alphas for the Korean version (37) and this study were 0.85 and 0.82, respectively. We assessed social support using the Multidimensional Scale of Perceived Social Support (MSPSS) (38) and the Spouse Supportive Behavior Scale (SSBS) (39). These scales were developed a long time ago, but numerous studies have used them as measures of social support until recently. MSPSS includes 12 items, including support from family, friends, and remarkable others. This study restricted family support to family members excluding spouses. As the spouse is intimate with a pregnant woman, we sought to distinguish them from other family members and accurately assess the spouse's support using SSBS. We used 10 items of SSBS, exempting the items of the spouse's supportive behaviors for physical convenience (e.g., “My husband comes home at the right time” and “My husband buys me comfortable clothes and shoes”), to identify the influence of psychosocial factors. We measured these two social support scales on a five-point Likert-type scale (from 1 to 5), with a higher score indicating more social support. Cronbach's alphas for original-version MSPSS (38) and the Korean version (40) were 0.88, and 0.90, respectively, and SSBS was 0.89 (39). This study determined them as 0.92 and 0.91. We assessed maternal–fetal attachment using the Maternal–Fetal Attachment Scale (MFAS) (41). Although this scale is many years old, recent research still extensively uses the scale. MFAS comprises 24 items. We assessed MFAS on a five-point Likert-type scale (from 1 to 5), with a higher score indicating greater fetal attachment. Cronbach's alphas were 0.85, 0.92, and 0.89 for the original, the Korean version (42), and this study, respectively. We assessed pregnancy stress using the Korean Pregnancy Stress Scale (PSS) (14). PSS includes 36 items split into seven dimensions: “physical and psychological changes (e.g., ‘I have difficulty breathing'),” “health of the mother and baby (e.g., ‘I am worried about having an abnormal fetus'),” “maternal role (e.g., ‘Becoming a mother is a burden'),” “family support (e.g., ‘I am disappointed that my husband is indifferent to me'),” “healthcare services (e.g., ‘I have difficulty determining prenatal tests'),” “social atmosphere (e.g., ‘If my baby has some problems, our society seems to believe that the mother is the main cause'),” and “coping in daily life (e.g., ‘I eat a balanced diet')”. We used 31 items and six dimensions of them in this study: “physical and psychological changes (eight items),” “health of the mother and baby (six items),” “maternal role (six items),” “family support (four items),” “healthcare services (three items),” “social atmosphere (four items)”. The five excluded items were questionnaires that overlapped with health behavior scales or related to postpartum and childcare; three items of “coping in daily life”; “financial burden about postpartum care”; and “social atmosphere about childcare facility”. PSS was calculated using a five-point Likert-type scale (from 1 to 5), with a higher score indicating greater stress. Cronbach's alphas for the original and this study were 0.85 (14) and 0.90, respectively. 2.5. Ethical considerations Before participant enrollment and data collection, the institutional review board of the author's institution approved this research work (Korea University; No. KUIRB-2020-0244-03). The institutional review board waived the requirement of written informed consent for participation. All participants provided online informed consent, and we obtained the data only from those who voluntarily participated. 2.6. Data analysis We evaluated data using the STATA 16.1 program and measured descriptive statistics for the demographic, obstetric, psychosocial variables, and prenatal health behaviors of samples. The t-test and analysis of variance confirmed the differences between pregnant AMA women's prenatal health-promoting and health-impairing behaviors on the basis of demographic, obstetric, and psychological characteristics. We used the Bonferroni method for the post-hoc test (43). Pearson's correlation analysis established the relationship between variables. Furthermore, we investigated the association between the factors and prenatal health behaviors using linear regression. We assessed multicollinearity by a variance inflation factor (VIF). 3. Results 3.1. Demographic, obstetric, and psychosocial characteristics Table 1 summarizes the demographic, obstetric, and psychosocial characteristics. Most of them have bachelor's degrees. In terms of obstetric variables, more than three-quarters of the subjects had natural conception, and nearly half of them were in their third trimester. More than half of the women were nulliparous, and 62.32% did not have an abortion spontaneously. Among the psychosocial variables, the mean of self-esteem was 3.00 (SD 0.40) while the mean of prenatal depression was 1.76 (SD 0.42). Maternal–fetal attachment averaged 4.09 (SD 0.47). In terms of pregnancy stress, “social atmosphere” was the most significant (mean 3.84), while “family support” was the lowest (mean 1.91). Prenatal health-promoting behaviors averaged 3.50 (global score: 62.5), while health-impairing behaviors averaged 2.18 (global score: 29.5). Table 1 Demographic, obstetric, and psychosocial characteristics of participants (N = 207). Variables n % M (SD) Range Age (years) 35–39 189 91.30 36.72 (1.94) 35–44 Above 40 18 8.70 Employment status In office 80 38.65 Leave 37 17.87 Resignation 44 21.26 Never employed 46 22.22 Religion Buddhism 93 44.93 Christianity 20 9.66 Catholics 69 33.33 Nothingarian 25 12.08 Monthly household income ($*) Below 2,000 9 4.35 2,000~4,000 70 33.82 4,000~6,000 62 29.95 6,000~8,000 35 16.91 Above 8,000 31 14.97 Education High school 8 3.87 College (Associate degree) 27 13.04 University (Bachelor's degree) 123 59.42 Graduate (Master's and Doctoral degree) 49 23.67 Conception type Natural 160 77.29 Artificial 47 22.71 Gestational age (days) 1st trimester (-97) 36 17.40 173.85 (70.22) 32–275 2nd trimester (98–195) 76 36.71 3rd trimester (196-) 95 45.89 Parity Nulliparous (0) 142 68.60 0.39 (0.63) 0–3 Multiparous (≥1) 65 31.40 Spontaneous abortions 0 129 62.32 0.55 (0.85) 0–5 1–2 70 33.82 ≥3 8 3.86 Self-esteem 3.00 (0.40) 1.6–4 Prenatal depression 1.76 (0.42) 1–2.8 Social support Support of spouse 4.10 (0.62) 1.3–5 Support of family 4.22 (0.74) 1–5 Support of friends 3.74 (0.82) 1–5 Support of significant others 3.36 (1.14) 1–5 Maternal-fetal attachment 4.09 (0.47) 2.7–5 Pregnancy stress 3.02 (0.57) 1.7–4.5 PSS 1 3.56 (0.70) 1.5–5 PSS 2 3.13 (1.07) 1–5 PSS 3 2.68 (0.94) 1–5 PSS 4 1.91 (0.77) 1–4.8 PSS 5 2.37 (0.94) 1–5 PSS 6 3.84 (0.69) 2–5 Prenatal health-promoting behavior 3.50 (0.58) 1.3–4.8 Prenatal health-impairing behavior 2.18 (0.48) 1–4.4 *1$ = 1,000 Won (Korean), PSS 1, physical and psychological changes; PSS 2, health of the mother and baby; PSS 3, maternal role; PSS 4, family support; PSS 5, healthcare services; PSS 6, social atmosphere. 3.2. Prenatal health behavior according to demographic, obstetric, and psychosocial characteristics Table 2 shows the prenatal health behavior according to demographic, obstetric, and psychosocial characteristics. In demographic and obstetric variables, education (F = 4.57, p = 0.004), parity (t = 6.92, p = 0.009), and the number of spontaneous abortions (t = 3.40, p = 0.035) were significantly different in prenatal health-promoting behaviors. Meanwhile, conception type (t = 12.96, p < 0.001) and parity (t = 13.61, p < 0.001) differed significantly in prenatal health-impairing behaviors. A post-hoc analysis was performed using the Bonferroni method. Statistically different demographic and obstetric variables were used to adjust the regression model. Table 2 Prenatal health behavior according to demographic, obstetric, and psychosocial variables (N = 207). Variables Prenatal health-promoting behavior Prenatal health-impairing behavior M ±SD t/F r p M ±SD t/F r p Age 0.00 0.956 −0.13 0.062 Employment status In-office 3.4 ± 0.6 0.22 0.886 2.2 ± 0.5 0.22 0.886 Leave 3.5 ± 0.6 2.2 ± 0.6 Resignation 3.5 ± 0.6 2.2 ± 0.5 Never employed 3.6 ± 0.5 2.1 ± 0.4 Religion Buddhism 3.4 ± 0.6 1.55 0.203 2.2 ± 0.5 1.55 0.203 Christianity 3.4 ± 0.6 2.3 ± 0.6 Catholics 3.6 ± 0.5 2.1 ± 0.5 Nothingarian 3.6 ± 0.5 2.2 ± 0.4 Monthly household income ($*) Below 2,000 3.8 ± 0.6 1.17 0.326 1.9 ± 0.5 1.17 0.326 2,000~4,000 3.5 ± 0.7 2.2 ± 0.5 4,000~6,000 3.4 ± 0.6 2.2 ± 0.4 6,000~8,000 3.5 ± 0.5 2.1 ± 0.4 Above 8,000 3.6 ± 0.4 2.2 ± 0.4 Education† High schoola 3.2 ± 0.5 4.57b