The father’s attachment to the fetus and related factors in late pregnancy: a longitudinal study using a path analysis model Authors: Marie Tabayashi, Tomoharu Sato, Tae Kawahara, Akemi Yamazaki Published: 2025 ABSTRACT AbstractBackgroundIt is important for fathers to develop attachment to the fetus during pregnancy because paternal attachment to the fetus during pregnancy affects attachment to the child after birth. A continuum of paternal attachment exists from early to late pregnancy. However, few studies have longitudinally investigated the development of paternal attachment to the fetus during pregnancy. This study aimed to explore longitudinally factors related to the father’s attachment to the fetus in late pregnancy.MethodsA longitudinal study was conducted using an anonymous self-administered questionnaire. The target population consisted of pregnant women (“mothers”) and their husbands (“fathers”) who were visiting a regional perinatal maternal and child health-care center in an urban area of Japan. The questionnaire was distributed to 300 pairs, among 195 who provided valid responses were included in the analysis. The questionnaire items for fathers included demographic characteristics, father’s attachment to the fetus, family functioning, and nurturing experience. The mother’s questionnaire included items on demographic characteristics and family functioning. The analysis used descriptive statistics, a corresponding t-test, and path analysis (p < .05).ResultsThe directly associated factors for increasing fathers’ attachment to the fetus in late pregnancy were higher attachment to the fetus in early pregnancy, higher family functioning of the father in late pregnancy, older age of the father, and first birth. The mother’s perception of family functions in early and late pregnancy, the father’s own experience of nurturing from both parents, and the financial condition of the father and mother each had an indirect effect on the father’s attachment in late pregnancy. The final modified path model yielded a good fit for the father’s attachment to the fetus in late pregnancy.ConclusionThe present findings suggest that to enhance the father’s attachment to the fetus in late pregnancy, it is necessary to support the father in enhancing attachment to the fetus and family functioning from early pregnancy. These findings are expected to help develop effective programs aimed at enhancing fathers’ attachment to the fetus in late pregnancy.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12884-025-08359-5.Keywords: Fathers, Fetus, Attachment, Family function, Path analysis, Conceptual model Background The father’s attachment to the child is highly continuous from pregnancy to the first year of life. This attachment is not solely derived from actual interactions with the newborn or infant, but rather, is activated in fantasy before the baby is even born [1]. Parental affection and warmth already begin to develop during pregnancy [2–4]. The attachment of the father to the fetus during pregnancy is important for the infant’s mental health and psychological development [5, 6], as well as the father’s attachment to the child after birth [7, 8]. Cognitive, emotional, psychological, and social development begins before birth and continues throughout life [9]. Attachment to the fetus during pregnancy and weeks of gestation have been shown to be positively correlated [2, 10]. Based on these findings, it can be inferred that the continuum of attachment to the child is not only prenatal and postpartum, but also highly continuous from early to late pregnancy. Therefore, it is important for fathers to develop attachment to the fetus during pregnancy. Regarding factors related to the father’s attachment to the fetus during pregnancy, previous studies have shown negative correlations with depression [1, 7, 10, 11] and anxiety [10–12]. It has also been verified that marital closeness, quality, and satisfaction are positively correlated with attachment to the fetus during pregnancy [1, 2, 10, 11, 13]. Still, only a few studies have explored factors related to the father’s attachment to the fetus. During pregnancy, fathers must adjust to their new role, in addition to their work and family responsibilities. These responsibilities can lead to stress, increase irritability toward problems, and impair family functioning [14]. For pregnant women, family functioning also refers to the family’s emotional connection, family rules, communication within the family, and efficiency in dealing with external events during pregnancy [15]. Pregnant women face multiple role changes in their lives, anxiety due to various physical discomforts, and psychological changes during pregnancy [16]. The pregnancy period is a time of forming new family connections while dealing with the restructuring of previous family relationships and changes in family functioning [17]. Therefore, it can be considered that each couple’s perception of family functions, including life and emotional aspects, differs and changes over time from early to late pregnancy. However, few previous studies have examined longitudinal changes in perceptions of family functioning throughout pregnancy among individual couples. Muller [18] reported that the attachment experience pregnant women received from their family of orientation during childhood leads to attachment to their own fetus. Condon et al. [1] found that fathers indicated that the nurturing experiences they received from their family of orientation during childhood were a correlative factor in their attachment to their children at age 1 year. In fathers, as in the previous study by Muller [18], it was hypothesized that the nurturing experiences received from their family of orientation in childhood may influence attachment to not only the infant, but also the fetus. However, to our knowledge, no longitudinal studies of the relationship between each couple’s perception of family functions in the family of procreation, the father’s experiences of nurturance by their family of orientation, or the father’s attachment to the fetus have been conducted during pregnancy. Fathers’ attachment to the fetus in late pregnancy is predicted to be significantly affected by some factors from early pregnancy. Attachment in late pregnancy is essential, as it has been suggested that fathers with higher attachment to the fetus in late pregnancy are better adapted to the parenting role, more supportive of the mother, and more active in caring for the child after birth [9]. However, to our knowledge, neither how the attachment of fathers changes from early to late pregnancy nor what factors influence this change has been sufficiently examined. Given this background, this study aimed to investigate factors associated with Japanese fathers’ attachment to the fetus in late pregnancy and the extent to which they may affect it longitudinally throughout the gestational period. The findings could be expected to lead to the development of more effective health guidance focusing on these factors at the appropriate time. Methods Study design This study was conducted as a longitudinal study (two time points: early pregnancy, defined as < 16 weeks of gestation, and late pregnancy, defined as ≥ 28 weeks of gestation). Participants The study participants were 300 couples consisting of pregnant women (hereafter referred to as “mothers”) and their husbands (“fathers”) who were visiting a regional perinatal mother–child medical center in an urban area in Japan. The inclusion criteria were that the mothers were pregnant and that both members of the couple could read and write Japanese, were at least 18 years of age, and had provided appropriate consent to participate in the study. The participants were recruited regardless of the number of fetuses. The exclusion criteria were that the couple did not live together at the beginning of pregnancy and that the mother or father had a current or previous history of mental illness. Data collection and procedure This study was conducted from January 2023 to March 2024. The researcher reviewed the medical records for a summary of mothers at < 16 weeks of gestation who were scheduled to visit the obstetric outpatient clinic at the hospital providing the study information and identified mothers who met the inclusion criteria. While waiting for the mothers to visit the outpatient obstetrics clinic, the researcher provided explanations to the mothers using a study description document. If the mothers verbally consented to participate in the study, the researcher distributed the self-administered, anonymous questionnaires for fathers and mothers to the mothers. It was confirmed by the researcher that both members of the couples returned the questionnaire in early pregnancy. Next, the researcher confirmed that the mothers were continuing their pregnancies at approximately 26 weeks of gestation from their medical records. If the couple’s continued pregnancy was confirmed, a letter confirming their willingness to cooperate in the study during late pregnancy was sent by postal mail. Only those couples who returned the letter confirming their willingness to cooperate in the study were sent home with self-administered, unsigned questionnaires for the father and mother. Each couple was asked to complete the questionnaire independently at home and return the completed questionnaire. Couples who cooperated in the study in both the early and late pregnancy periods were paid an honorarium of 1000 JPY in the form of a gift card. Measures Paternal antenatal attachment scale (PAAS-J) Paternal attachment to the fetus during pregnancy was measured using the Japanese version of the Paternal Antenatal Attachment Scale (PAAS-J) [19] originally developed by Condon [20]. The PAAS-J measures the father’s feelings, behaviors, and thoughts toward the fetus during the past 2 weeks. It is composed of 13 items rated on a five-point Likert-type scale ranging from 1 to 5, with higher total scores indicating greater prenatal attachment. The PAAS-J is classified into two subscales: qualitative attachment, which describes the emotional experiences that occur when thinking about the fetus (e.g., tenderness/frustration, positive/negative), and quantitative attachment, which describes the amount of time spent thinking about, talking about, dreaming about, and touching the fetus and the intensity of the fetal relationship. The reliability of the PAAS-J in this study was confirmed (Cronbach’s alpha coefficient: 0.95–0.99). Family functioning scale (FFS) Family functioning was measured using the Japanese version of the Family Functioning Scale (FFS) [21] originally developed by Lee et al. [22]. The FFS was developed in Korea as a new family functioning scale appropriate to the cultural background of the respondents’ own countries after pointing out the difficulties in applying family functioning scales developed in Western cultures. The FFS is composed of 24 items rated on a four-point Likert-type scale ranging from “1, not at all so” to “4, always so”, with higher total scores indicating higher family functioning. The FFS is categorized into six subscales: “Emotional bonding and closeness”, “External relationships (social resources)”, “Family rules”, “Roles and responsibilities”, “Financial resources”, and “Communication”. Because Japan and Korea have similar cultural backgrounds surrounding the family, such as the long history of patriarchy as a social practice, the Japanese version of the FFS was used in this study. The reliability of the FFS in this study was confirmed (Cronbach’s alpha coefficient: 0.89–0.90). Parental bonding instrument (PBI) Nurturing experience was measured using the Japanese version of the Parental Bonding Instrument (PBI) [23] originally developed by Parker et al. [24]. The PBI is a subjective rating scale of parental nurturing attitudes from the child’s perspective, in which the child responds by recalling his or her own mother’s and father’s nurturing attitudes up to age 16 years, respectively. This questionnaire is composed of 25 items rated on a four-point Likert-type scale ranging from “0, not at all different” to “3, very much so”, with higher scores indicating that children rate their parents as more loving and accepting. The PBI is categorized into two subscales: “Caring” and “Overprotective”. “Caring” measures the degree of attachment, warmth, empathy, and closeness, while “Overprotection” measures the degree of maneuvering, excessive contact, infantile treatment, and interference with independent behavior. The reliability of the PBI in this study was confirmed (Cronbach’s alpha coefficient: 0.86). Data analysis The contents of the anonymous questionnaire in the early pregnancy period were as follows: PAAS-J, FFS, and PBI scores, and demographic characteristics (age, number of births, pregnancy planning, marital status, education level, employment status, and financial condition) for the fathers, and FFS scores and demographic characteristics (number of weeks of pregnancy, expected date of delivery, age, education level, employment status, and financial condition) for the mothers. The contents of the anonymous questionnaire in the late pregnancy period were PAAS-J and FFS scores for the fathers, and FFS scores and demographic characteristics (number of weeks of pregnancy) for the mothers. The data were analyzed using JMP Pro 17 (SAS Institute Japan, Tokyo, Japan), with the level of statistical significance level set at 5%. All analyses were performed using descriptive statistics and a corresponding t-test, correlation coefficients, one-way analysis of variance (ANOVA), and path analysis. Before conducting the path analysis, an initial path model was constructed based on the conceptual model in this study, which was devised from previous studies (Fig. 1). This study utilized Condon’s theory of paternal development (Condon, 1993) as its theoretical framework. This theory suggests that the quality of the marital relationship influences the relationship between attachment to the fetus formed during pregnancy and attachment to the child after birth. In this study, we applied this framework and focused on the changes in attachment to the fetus from early pregnancy to later pregnancy, setting up a hypothetical model of the influence of family functions, including marital relationships, on these changes. The path analysis was carried out in an exploratory manner, and the initial path model was modified to result in the final path model. The goodness-of-fit of the path coefficients was considered acceptable within the following ranges: goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), comparative fit index (CFI), Tucker–Lewis index (TLI), and normative fit index (NFI) all >0.90, and a root mean square error of approximation (RMSEA) value < 0.08 [25]. Conceptual model in this study Results The self-administered anonymous questionnaire was distributed to 300 couples. Subsequently, completed questionnaires were collected from 234 couples in early pregnancy (response rate: 78%) and 201 in late pregnancy (response rate: 67%). The dropout rate from early to late pregnancy was 14.10%. The reasons for dropping out were that the respondents did not reply to the late pregnancy questionnaires (23 couples), or that they were hospitalized or transferred because of complications during pregnancy and thus unable to respond (10 couples). No significant differences in demographic characteristics or the total scores on each scale between study completers and dropouts were observed, and the dropouts were considered random failures. Three couples whose responses could not be reliably, as well as three couples with missing data on the dependent variable of this study, paternal attachment to the fetus in late pregnancy (late pregnancy PAAS-J), were excluded from the analysis. Therefore, 195 couples who provided valid responses were included in the analysis. The reasons for removing the three pairs for which valid responses were not obtained from the analysis are as follows. When missing values of 20% or more existed in each scale, one pair was deleted because it was judged that no valid response was obtained. If the total score of the scale was perfect in both early and late pregnancy, two pairs were deleted because the reliability of the responses was considered to be low. In the case of a few missing values, the pairwise method was used and the available case analysis was performed. The mean weeks of response was 12.49 (standard deviation [SD] = 1.86) for early pregnancy and 29.67 (SD = 1.96) for late pregnancy. The mean age of the fathers was 34.04 ± 5.59 years (range, 23–50) and that of the mothers was 32.71 ± 4.47 years (range, 20–42). In total, 132 (67.69%) of the couples were primiparous and 63 (32.31%) were multiparous. Incidentally, four couples with multiple pregnancies were among the subjects. The other basic characteristics of the participants are shown in Table 1. Table 1Demographic characteristics of the participantsStratification factorsMean ± SD (range) or n (%) Fathers Mothers Age (years) 34.04 ± 5.59(23–50)32.71 ± 4.47(20–42)Parity Primiparous132(67.69) Multiparous63(32.31)Planning for pregnancy Yes162(83.08) No33(16.92)Marital status Married184(94.36) Unmarried11(5.64)Employment status Full-time, self-employed188(96.41)127(64.06) Part-time6(3.07)39(19.04) Unemployed1(0.51)29(16.90)Education level University degree or higher126(64.62)117(60.00) High school graduate or below69(35.38)78(40.00)Financial condition* Wealthy87(44.62)81(58.46) Below average108(55.38)114(41.54)SD Standard deviation*Financial conditions were categorized on a 5-point Likert scale ranging from “very wealthy living” to “very poor living” based on respondents' subjective judgments. When analyzing the data, the respondents were divided into two groups: those living “relatively wealthy or above” and those living “below average" Demographic characteristics of the participants SD Standard deviation *Financial conditions were categorized on a 5-point Likert scale ranging from “very wealthy living” to “very poor living” based on respondents' subjective judgments. When analyzing the data, the respondents were divided into two groups: those living “relatively wealthy or above” and those living “below average" A paired t-test was used to analyze the total score of each scale from early to late pregnancy. The results showed that PAAS-J scores were significantly higher in late pregnancy (48.81 ± 6.90) than in early pregnancy (46.63 ± 6.81 ). Additionally, FFS scores did not change between early (75.61 ± 9.79) and late pregnancy (75.33 ± 9.51) for fathers. On the other hand, for mothers, FFS scores were significantly lower in late pregnancy (72.67 ± 8.36) than in early pregnancy (77.49 ± 8.39). No apparent differences in the scores of couples with singleton and twin pregnancies on each scale were observed. A correlation analysis showed that total PAAS-J scores in late pregnancy correlated positively with total PAAS-J scores in early pregnancy, total FFS scores in early and late pregnancy for fathers, and total FFS scores in late pregnancy for mothers (p <.0001). The results of the independent t-test showed that the PAAS-J total score was higher for the primiparous group than for the multiparous group in late pregnancy (t = 2.352, p =.020). Using path analysis, an initial path model based on the conceptual model was drawn, but it did not fit well (GFI = 0.719, AGFI = 0.590, CFI = 0.366, TLI = 0.230, NFI = 0.343, RMSEA = 0.154). Therefore, nonsignificant paths and paths with path coefficients less than 0.1 were removed. Regarding the addition of paths, paths were added to model direct effects to allow for the nursing implications of the modified path model to pass through. The final modified path model was determined to be optimal because of its good numerical value considering the goodness-of-fit indices. The final modified path model is shown in Fig. 2. The standardized estimates of the direct, indirect, and total effects on the father’s attachment to the fetus in late pregnancy are shown in Table 2. Final modified path model Table 2Summary of direct, indirect, and total effects on father's attachment to fetus in late pregnancyVariablesEffectsSignificant factorsFather's attachment to the fetus in early pregnancyFamily functioning of the father in early pregnancyFamily functioning of the father in late pregnancyAge of the father Father's financial births conditionNumber of birthsMother's careFather's careFamily functioning of the mother in early pregnancyFamily functioningof the mother in late pregnancyMother's financial conditionFather's attachmentto the fetus in late pregnancyDirect0.678**0.000**0.192**0.129*–0.099*0.0000.0000.0000.0000.0000.000Indirect0.0000.3780.0000.000 – 0.040 0.083**0.115 **0.073* 0.153 **0.040 *0.028 *Total0.678**0.378**0.192**0.129*–0.139** 0.083 **0.115 ** 0.0730.153 **0.040 *0.028 *Mother's care the fathers’ own experiences of nurturing from their own mothers (subscale of the PBI)Father's care the fathers’ own experiences of nurturing from their own fathers (subscale of the PBI)*p <.05, **p <.0001 Summary of direct, indirect, and total effects on father's attachment to fetus in late pregnancy Mother's care the fathers’ own experiences of nurturing from their own mothers (subscale of the PBI) Father's care the fathers’ own experiences of nurturing from their own fathers (subscale of the PBI) *p <.05, **p <.0001 The father’s attachment to the fetus in late pregnancy was highly influenced by that in early pregnancy, indicating a direct effect (β = 0.678, p <.0001). The father’s perception of family functioning in early pregnancy had an indirect effect on the father’s attachment to the fetus in late pregnancy (β = 0.378, p <.0001), and the father’s perception of family functioning in late pregnancy had a direct effect (β = 0.192, p <.0001). The older the father, the more direct the effect on the father’s attachment to the fetus in late pregnancy (β = 0.129, p =.0101). Regarding parity, primiparous fathers showed higher attachment to the fetus in late pregnancy directly (β = − 0.099, p =.0489) compared with multiparous fathers. Furthermore, the fathers’ own experiences of nurturing from their own mothers (β = 0.115, p <.0001) and the mothers’ perceptions of family functioning in early pregnancy (β = 0.153, p <.0001) each had indirect effects on the father’s attachment to the fetus in late pregnancy. In addition, the following factors showed weak but indirect effects on the father’s attachment to the fetus in late pregnancy: the fathers’ financial condition (β = 0.083, p =.0003), the fathers’ own experiences of nurturing from their own fathers (β = 0.073, p =.0050), the mothers’ perceptions of family functioning in late pregnancy (β = 0.040, p =.0076), and the mothers’ financial condition (β = 0.028, p =.0154). Neither the fathers’ and mothers’ education level, employment status, nor planning for pregnancy had any direct or indirect effects on the father’s attachment to the fetus. The final modified path model demonstrated a good fit for fathers’ attachment to the fetus in late pregnancy (GFI = 0.980, AGFI = 0.962, CFI = 0.965, TLI = 0.951, NFI = 0.915, RMSEA = 0.051). Discussion The present study established a path model by conducting a longitudinal exploratory study of factors related to the father’s attachment to the fetus from early to late pregnancy. The factor with the largest direct effect on the father’s attachment to the fetus in late pregnancy was the father’s attachment to the fetus in early pregnancy. This result is consistent with previous studies that have reported continuity of attachment, such as increased attachment to the fetus from early to late pregnancy [2, 10, 26–28] and suggests that attachment to the fetus develops progressively throughout pregnancy. Each couple’s perception of family function, financial condition, the father’s own experiences of nurturing his family of orientation, the father’s age, and primiparity affected the father’s attachment to the fetus in late pregnancy both directly and indirectly. Among these factors, the father’s family functioning in early pregnancy had the strongest indirect effect on his attachment to the fetus in late pregnancy. This finding suggests that the father’s family function in early pregnancy is an important predictor of his attachment to the fetus in late pregnancy. Previous studies have shown that stronger couple relationships and greater paternal responsibility are associated with higher fetal attachment [11, 13, 29]. In addition, fathers who feel supported by their partners and report no problems in their relationship have been shown to have higher attachment to the fetus [30]. In terms of having a good relationship, this finding agrees with the FFS subscales of “Emotional bonding and closeness” and “Communication”. Additionally, greater paternal responsibility has been linked to stronger attachment to the fetus [29], which aligns with the “Roles and responsibilities” subscale of the FFS. Thus, the results of this study are consistent with those of previous studies. The mother’s family functioning in early and late pregnancy was found to have a direct effect on the father’s family functioning in both early (β = 0.331) and late (β = 0.206) pregnancy, respectively. The father’s family functioning in early pregnancy had a direct effect the father’s attachment to the fetus in early pregnancy, whereas it had an indirect effect on late attachment. The father’s family functioning remained stable throughout the pregnancy, whereas the mother’s family functioning declined from early pregnancy. The development of family functioning during pregnancy is considered an adaptive process to external tasks from biological, perceptual, cognitive, behavioral, and interpersonal perspectives [31]. This is thought to be a complex situation that requires adaptation by each family member. Researchers and health professionals have emphasized the importance of family functioning, especially in early pregnancy, which is considered a vulnerable time for the family, thereby reinforcing the importance of paying more attention to family functioning in early pregnancy [32]. The results of the present study indicate that family functioning influences the attachment to the fetus. Hence, supporting both parents to improve family functioning from early pregnancy may strengthen the father’s attachment to the fetus in late pregnancy. However, the influence of family functioning may vary by cultural background. In Japan and South Korea, the influence of patriarchy is strong, and fathers often play a supportive role in the family [22]. In Scandinavian countries, parental leave systems are well developed and fathers’ participation in childcare is encouraged [30]. In the Middle East and South Asia, the extended family exerts a strong influence, and family functions are formed not by the parental unit, but by the entire family, including the grandparents [14]. Wilson et al. [33] indicated that family cooperation is a factor that strengthens the father’s attachment. Given the environment of the increasingly nuclear family in Japan, the influence of family functioning may be different when compared with than in foreign countries. Further comparative multicultural research is needed to gain a more comprehensive understanding of the relationship between family functioning and attachment. The fathers’ nurturing experiences from their own parents directly influenced their family functioning in early pregnancy. Previous research has reported a positive correlation between the childhood memories of one’s own parental nurturing attitudes and the father’s attachment to the fetus in late pregnancy [7]. However, the present study found no direct effect of the father’s own nurturing experiences on fetal attachment. Instead, these experiences indirectly influenced attachment through family functioning, which is a novel finding. Furthermore, the nurturing experience received from the care of one’s own mother was found to have a stronger effect on attachment to the fetus than the nurturing experience received from that of one’s own father. In this regard, when the subjects of this study were children, they spent more time with their mothers than with their fathers because the number of households with full-time housewives exceeded those with both parents working in Japan [34]. As a result, it is thought that a stronger effect of care from one’s own mother emerged. Regarding the demographic characteristics of the participants in the present study, the most common age of childbearing for fathers and mothers in Japan is 30–34 years, which is consistent with the average age of the subjects in this study [35]. In Japan, 25.5% of individuals have a university or graduate school degree [36], and the fathers and mothers in this study were highly educated. Some studies have found no clear demographic effects on father–fetus attachment [37–39]. However, this study showed direct effects in that primiparous fathers had stronger attachment in late pregnancy, consistent with other studies [12, 30]. This may be because multiparous fathers are more concerned with the child they are currently caring for and therefore less concerned with the current pregnancy and fetus [37]. A direct effect of the father’s age was also found on attachment to the fetus in late pregnancy. As in the present study, previous research has found that older fathers have greater attachment to the fetus [29]. By contrast, some studies have argued that attachment to the fetus is higher when fathers are younger [28, 37–39]. Multiple factors, including research methodology and socioeconomic factors, may be behind the discrepancies in the results of previous studies. The participants in the present study were fathers in Japan aged 23–50 years; as a relatively wide age range was covered, it was relatively easy to detect age-related effects statistically. On the other hand, previous studies reporting that younger fathers are more attached to the fetus have consisted of a smaller proportion of older fathers and thus may not have been able to compare attachment trends in detail by age. Therefore, differences in sample characteristics and survey methods may have led to differences in the results from previous studies. In addition, cultural backgrounds may also have had an influence. Although the participants in the present study were Japanese fathers, most studies showing that younger fathers have higher attachment to the fetus have been conducted in Europe and the US [28, 37–39]. In those countries, there is an entrenched culture in which even young fathers are actively involved in child-rearing. In Japan, fathers still spend little time participating in childcare, regardless of age [40]. These cultural factors may account for the differences in the study results. Considering the socioeconomic aspect, older fathers are often more professionally and economically stable and more likely to have a psychological margin for family life and child-rearing. Previous studies have also reported that fathers’ economic stability is associated with their level of involvement in child-rearing [38]. On the other hand, young fathers are in the formative years of their careers and more likely to face work and economic challenges, which may make them more conscious of establishing a foundation for their lives than with their attachment to the fetus. The present results provide new insights into the higher level of attachment to the fetus among older fathers. Future research should include a comprehensive analysis of fathers’ psychosocial factors and cultural backgrounds. This study has some limitations. First, this study was conducted at only one institution, a regional perinatal medical center in an urban area of Japan. In the future, research should be conducted at multiple institutions, such as clinics, rural hospitals, and university hospitals, involving participants with a wider range of demographic characteristics. Second, the PBI was used in this study to assess the fathers’ nurturing experiences from their biological parents, but this was a retrospective study to age 16 years, so the possibility of recall bias cannot be denied, and biases that mix with events over the age of 16 years are unavoidable. It will be necessary to analyze what specific memories of nurturing experiences are being recalled in more detail in the future, including through interviews. Third, although the observed variables used in this study were input based on previous studies as deemed necessary, the possibility of confounding factors such as psychological and social factors other than the variables treated in this study cannot be ruled out and should be interpreted with caution. Future studies should consider identifying and correcting for further confounding factors. Conclusion We believe that this is the first longitudinal study to use path analysis to identify factors associated with the father’s attachment to the fetus in late pregnancy. The results showed that the father’s attachment to the fetus in late pregnancy was directly affected by the father’s attachment to the fetus in early pregnancy, the father’s perceptions of family functions in late pregnancy, primiparous birth, and the father’s age. Moreover, the family functioning of fathers and mothers in early pregnancy and the nurturing experiences of their own parents indirectly affected the father’s attachment to the fetus in late pregnancy. These findings emphasize the need to increase fathers’ and mothers’ perceptions of family functioning to enhance the father’s attachment to the fetus in late pregnancy. These findings are also expected to be useful in developing more effective interventions to enhance the father’s attachment to their fetus in late pregnancy. Supplementary Information Supplementary Material 1 Acknowledgements We would like to express our sincere gratitude to the study participants. We would also like to thank all the staff members of Yodogawa Christian Hospital of South Presbyterian Mission in Japan for their kind cooperation with our study. Abbreviations Japanese version of the Paternal Antenatal Attachment Scale Japanese version of the Family Functioning Scale Japanese version of the Parental Bonding Instrument Goodness-of-fit index Adjusted goodness-of-fit index Comparative fit index Tucker–Lewis fit index Normative fit index Root mean square error of approximation Standard deviation Authors’ contributions MT was responsible for the entire research process, including the study conception, data collection, data analysis and interpretation, and writing of the manuscript. TS advised on the analytical conception of the study, data analysis and interpretation, and suggestions for the manuscript. TK and AY provided advice on data analysis and interpretation, suggestions for the manuscript, and the entire research process. All authors gave final approval of the manuscript for publication. Funding This work was funded by the Fumiko Yamaji Professional Nursing Education and Research Grant Fund (No. 079-27-11). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate This study was approved by the Ethical Review Board of The University of Osaka Hospital and the Ethics Review Committee of Yodogawa Christian Hospital of South Presbyterian Mission in Japan (approval Nos.: 22383 and 2022-062). Written informed consent to participate in this study was obtained from all participants before the study began. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement No datasets were generated or analysed during the current study.