Childbearing intention and its associated factors: A systematic review

Abstract Aim This study aimed to provide comprehensive information about the core determinants of fertility intentions. Design Systematic review. Methods Ovid, MEDLINE, EMBASE, PsycINFO, CINAHL, Web of Science, SCOPUS and GOOGLE SCHOLAR were searched for the relevant articles published from 1946–December 2017. We updated our records by searching three computerized databases (Ovid MEDLINE, SCOPUS and WOS) from 2018–January 2021. Results 53 studies included in the qualitative synthesis. The results of some studies indicated the impact of demographic factors, physical and psychological health, happiness and child desire. The most frequent variables in a couple's mesosystem were marital status, parity, partnership satisfaction and gender role attitude. The mesosystem of childbearing intention also included family and peers network. The EXEO system of the ECSM includes certain variables, such as job characteristics, urban residence, housing condition. The macrosystem comprises cultural and societal principles with broader influences on the couple's system.


| INTRODUC TI ON
Today, contraception use has changed the prospect of parenthood to a personal choice (Mills, 2011). In the early 19th century, the modern fertility transition began in France and the United States. (Gutmann & Fliess, 1993;Westoff, ) and quickly expanded across Europe (Westoff, 2015). The regulation of fertility was achieved by widely using contraception and abortion. In every case, the small family size represented the intent of the individual couple (Watkins, 1987). The practice of family limitation started to spread among the population in Europe (Knodel, 1977;Watkins, 1987). In this respect, the modern fertility transition appeared to result from the spread of innovative behaviour and not an adjustment to new socio-economic circumstances. (Watkins, 1987). This European pattern quickly became popular in developed Asian and American societies (Beaujolai& Sobotka, 2019;Caldwell & Caldwell, 2005;Knodel, 1977;Sobotka, 2017). At the same time, the childbearing rate among less developed countries showed a significant reduction (Bongaarts, 2008;Timaeus & Moultrie, 2020).
Low fertility and population ageing in most European countries (Fahlén, 2013) have attracted researchers' and policymakers' attention. Factors that build fertility intentions can explain variations in fertility changes cross-nationally and over time (Harknett et al., 2014).

HASHEMZADEH Et Al.
A large body of research explains possible reasons for low fertility, such as the economic situation, child-rearing costs, women's education and employment (Cooke, ;McDonald, 2000;Mills et al., 2008). Since the 1990s, some recent studies have tried to explain fertility levels through gender (Yu & Kuo, 2017).
The literature suggests that childbearing decision-making is a complex process involving many social, economic, political and individual factors. These include the availability of qualified and affordable childcare support, cultural norms, individual beliefs and partner suitability (Abma & Martinez, 2006;Clarke & Hammarberg, 2005;Cooke et al., 2010;Mills et al., 2011;Proudfoot et al., 2009).
Bronfenbrenner provides an ecological model that envisages the existence of several environments or contexts that may be analysed from four levels, which are all part of the same reality.
• The "microsystem" includes the roles, relationships, and activity patterns developed by a person in their relationship with their environment.
• The "mesosystem" or the relation between two or more microsystems in which the person is actively involved.
• The "exosystem" or those environments in which the person in the process of becoming is not so actively involved in but do affect his/her development.
• The "macrosystem" as the relationships, both in form and content of the lower order systems (micro-, meso-and exo-) that exist or may exist at the sub-culture level or the culture as a whole, together with any belief system or ideology that supports these correlations (Rothbaum et al., 2002).
Applying ecological approaches is particularly suitable for factors associated with childbearing intentions (Balbo et al., 2013). The decision to have a child is a complex process involving many social, economic, political and individual factors. So it is better to describe it as a new system (Bronfenbrenner, 1994).
There are several reviews conducted about fertility (Balbo et al., 2013;Butler, 2004;Caldwell & Schindlmayr, 2003;Mills et al., 2011;Morgan & Taylor, 2006;Sobotka, ) that provide important views. Some of these studies have focused on specific aspects or geographical areas. As the predictors of fertility intentions are similar to the predictors of actual births (Harknett et al., 2014;Kuhnt & Trappe, 2016). A couple is the most important context for investigating fertility decision-making (Spéder & Kapitány, 2009).
Our study focuses on the determinants of childbearing plans to provide a lens for understanding fertility behaviour in couples.

| ME THODS
This systematic review was carried out following PRISMA guidelines.
Note that PRISMA is an evidence-based minimum set of items for reporting in systematic reviews and meta-analyses (For more information, see: www.prism a-state ment.org).

| Search strategy
This systematic literature review was performed using electronic databases such as Ovid MEDLINE, EMBASE, PsycINFO, CINAHL, Web of Science, SCOPUS and GOOGLE SCHOLAR. The article search was performed with an alternate combination (and/or) of these search terms: "fertility", "desire", "intention", "childbearing", and "reproductive decision making". The search was conducted from 1946 to December 2017 (see Appendix S1). Systematic reviews and meta-analyses were manually checked to distinguish the related studies missed by electronic databases search.

Figure 1
This study was the first step of a Ph.D. thesis, conducted in 2017. We updated our records by searching three computerized databases (Ovid MEDLINE, SCOPUS and WOS) from 2018 to January 2021.

| Inclusion and exclusion criteria
We included the original studies in English, published in peerreviewed journals. As we wanted to study fertility in a family structure, we excluded adolescent fertility and included studies in which their participants were men and/or women of reproductive age (men aged 18-55 and women aged 18-45). Family planning methods were available in the reference country. The intention of fertility makes sense when people have access to family planning methods and have voluntary childbearing. Studies attributed to China were excluded due to child-restricting national laws.

| Screening
Two authors independently searched and screened studies for the inclusion criteria. A total number of 7,440 articles were identified and imported to Endnote X8. We removed 2,640 duplicated articles and screened titles and abstracts of the 4,795 ones. 467 relevant articles were fully assessed for more screening. We evaluated the eligibility of these articles and, finally, 45 studies were included. Any disagreements between the two authors were discussed until consensus was reached.
A total number of 1594 articles were identified through an updated search in 2021. After removing duplicated articles, titles, and abstracts of 1,041, articles screened. 72 relevant articles were fully assessed and 8 studies were included. Finally, 53 studies were included in this review. Table 1 provides information about the included studies.

| Quality assessment
We used the STROBE checklist, which consisted of 22 items with a maximum score of 30. Studies with more than 23 total points were considered as high quality and those with less than 16 points were considered as low quality. Studies that received a total point between 16 and 23 were considered as medium quality. Among 53 studies, 46 (87%) were classified as high and 7 (13%) were classified as moderate. So, all the studies were included in our review. Among the 7 articles with medium quality, five got 23 points, one got 22 points, and the last one got 20 points. None of the articles had the main weakness in the methodological or result area. Appendix S2 provides a quality assessment of the included studies.

| Data extraction
The following variables were extracted from the included studies: author's name, year of publication, country, study-design, sample-size, statistical analysis method, scale and primary results (variable associating childbearing).

| Applying Bronfenbrenner's ecological model to the childbearing intention
A new concept was developed in which a couple's system acts like a union system in dyadic exclusivity. Based on this hypothesis and the overview of primary studies, a union system comprising two microsystems (individual and his or her partner) was developed, The ecological couple system model (ECSM) is a conceptual model of associated factors with fertility intention as a union system that interacts with all different systems in ecological systems theory. Other systems (mesosystem, exosystem, macrosystem and chronosystem) affect the ECSM (Figure 2). We next describe the factors associated with childbearing intention through Bronfenbrenner's ecological systems theory.

| Microsystems
The closest level to the individual is the microsystem. Numerous microsystems form throughout an individual's life. The microsystem of the ECSM includes information about socio-demographic characteristics, physical and psychological health, happiness and child desire.

| Socio-Demographic Characteristics
Socio-demographic characteristics include age, education and financial situation.

Age
Age affects fertility intention both in men and women.
The older the woman is at the beginning of the cohabitations, the sooner the couple has their first child. For the second and the third child, the age of the woman has a negative influence (Rijken & Liefbroer, 2009).

Financial situation
The birth requires access to financial resources. Better career prospects increase childbearing desire in both men and women (Bühler & Fratczak, 2007;Kuhnt & Trappe, 2016;Sinyavskaya & Billingsley, 2015;Yu & Kuo, 2017). Household income plays a key role in childbearing intentions (Hanappi et al., 2017). The intention of the second child is lower among families with limited income (Fiori, 2011).

| Physical and psychological health
Health concerns can seriously affect the second child's intentions among couples (Boivin et al., 2018). Having a negative birth experience could adversely affect women's fertility intentions. (Preis et al., 2020). The mode of the first delivery can affect childbearing intention in women. The mode of delivery can affect the next child's intention. The tendency to have two or three children is less in women who had a caesarean delivery (CD) compared with those who had a vaginal delivery (Kjerulff et al., 2013).

| Happiness
Happier men and women prefer to become parents sooner (Spéder & Kapitány, 2009 are most satisfied with their life course and their prospects are more likely to realize their fertility intentions (Spéder & Kapitány, 2009).

| Child desire
Fertility preference is an important predicting factor for childbearing intentions. The possibility for an additional child is 83% higher when the wife's desire is higher and 48% lower when her desire is lower (Fan & Maitra, 2010). So, in predicting the birth outcomes, the wife's tendency has a higher effect (Fan & Maitra, 2010;Testa, 2012).
Among women, two positive motives, "feeling needed and connected" and "joys of pregnancy, birth, and infancy" were identified as being predictive of stronger childbearing desire. Among men, the significant predictors of stronger childbearing desire were "satisfaction of child-rearing" and "traditional parenthood" (Mynarska & Rytel, 2020). In societies like India, the sex composition of children is associated with the desire for an additional child. One aspect of this influence is the preference for sons. In India, relative to women with no daughters, women with no sons have significantly higher odds of progressing to the next birth (Rajan et al., 2018).

Mesosystems
The mesosystem comprises the reciprocal interactions that take place among the Microsystems within an individual's life. Although a couple is defined as a union system, the interaction between two partners is affected by other system interactions. This subsystem emphasizes the surrounding interpersonal relationships (Jones et al., 2011). The most frequent variables in a couple's mesosystem are marital status, parity, and partnership satisfaction. Gender role attitude (traditional/egalitarian) is another important component in this system. The mesosystem of childbearing intention also includes family and peers network which can influence couples' decisions.

| Parity
Parity is an important criterion in childbearing decision-making (Hanappi et al., 2017;Park et al., 2008). People with more chil-  Women reporting intimate partner violence are less likely to desire more children (Kuhlmann et al., 2019).
Traditional men are more likely to have a child (Bernhardt et al., 2016;Kaufman, 2000). Egalitarian men seem to partner to a great extent than traditional men (Bernhardt et al., 2016). Egalitarian men are more likely to enter unions while traditional women are more likely to have a child. These findings showed that gender role attitudes have a different impact on fertility intentions among men and women (Kaufman, 2000).

| Family and peers network
Interaction within networks of family and friends is a major component in childbearing decision-making (Kuhnt & Trappe, 2016).

| Exosystems
The exosystem includes a situation or institution that affects an individual's daily settings but is not part of the individual's immediate environment (Rothbaum et al., 2002). The exosystem of the ECSM includes certain variables, such as job characteristics, urban residence and housing conditions.

| Job characteristic
The pattern of employment effect varies by gender and parity. Fulltime work is a key factor for childless men and women to become a parent shortly. For the second child, full-time employment loses its positive influence on mothers (Neyer et al., 2013). Job insecurity could affect childbearing intention negatively. Job instability in low and medium-educated people is less responsive compared with highly educated ones (Hanappi et al., 2017;Vignoli et al., 2020). Fiori et al., (2013 revealed that job insecurity is more associated with short-term intentions, and the number of children intended by couples is less affected by it.

| Urban residence
Living in districts characterized by some forms of unease (high rate of school dropout, many minors followed by social services etc.) decreases childbearing intention (Meggiolaro, 2011;Riederer & Buber, 2019).

| Housing condition
Housing type is associated with childbearing intention. Homeowners and couples living in single-family houses are significantly more likely to have their first child sooner (Kulu & Vikat, 2007;Park et al., 2008;Vignoli et al., 2013). Living in single-family houses is associated with a higher intention to have a first child compared with living in apartments. (Kulu & Vikat, 2007;Park et al., 2008).

| Macrosystem
The macrosystem of the ECSM comprises cultural and societal principles with broader influences on the couple's system.

| Family policies
A large body of fertility literature has concentrated on specific family policies and the degree of availability of childcare services (Fiori et al., 2013;Park & Cho, 2011;Wesolowski, 2015;Yu & Kuo, 2017).
In developed societies, higher-order births are likely to be more responsive to policy and environmental changes compared with the first births (Morgan & Taylor, 2006). Countries with family policies-as a part of their labour market policies, care policies and gender policies-seem to keep fertility above the lowest-low levels (Wesolowski, 2015).

| Religiosity
Religiosity also plays an important role in childbearing decisionmaking. Religious couples are more likely to become parents (Kuhnt & Trappe, 2016;Rosina & Testa, 2009). But, the religiousness of only one of the two partners will cause conflict between couples.
The results showed that when a woman wants a child but, the man does not, male religiosity increases the opposition. In contrast, when the man wants a child but the woman does not, women's religiosity decreases opposition between the couple (Rosina & Testa, 2009).
Belief in God is independently associated with fertility desires. At least some of the connections between religiosity and fertility is attributed to metaphysical beliefs-and not just traditional and institutional religiosity (Cranney, 2015).

| Child value
The value placed on having children can motivate fertility intentions.
Psychological benefits (e.g. comfort and paternal and maternal feelings) of the child show a greater preference for a second child's intention. Instrumental values (e.g. current economic support and in the elderly years, a continuation of the family line and social duty) of children are not significantly associated with the intentions to have a child (Miettinen et al., 2011;Park & Cho, 2011).

| Individualism attitudes
Self-realization is a predictor that affects childbearing intentions negatively. Nowadays, people believe that more children cannot increase their social esteem enough to take charge of raising a child (Wesolowski, 2015).

| Chronosystem
The chronosystem of the ECSM consists of changes relating to time or throughout the lifecycle that influence individuals and their environment (Bronfenbrenner, 1986).

| D ISCUSS I ON
Childbearing intentions are not always voluntary and are often influenced by actual and perceived circumstances (Holton et al., 2011).
Childbirth in the early years of the marriage sounds like a threat for couples to spend their time together. Losing their freedom, leisure time and traveling opportunities would lead them to delay childbearing. Over time in some couples, this attitude may change and the need for children is created (Qu & Weston, 2001). As one grows older, his/her time left for childbearing decreases. However, the biological limits are known for women, but social constraints apply to both men and women in childbearing decision-making (Spéder & Kapitány, 2009).
First-birth intentions are closely related to the wish of establishing a family and more influenced by normative pressure than the economic situation. Second-birth intentions are also governed by age but related to educational level, availability and cost of childcare and labour market situation (Wesolowski, 2015).
Education can simultaneously indicate the economic and cultural effect mechanisms. A variety of lifestyles and cultural resources are tied up with education. (Fiori et al., 2013;Spéder & Kapitány, 2009) Therefore, the interaction will be created not only within the couple but also between the couple and other systems.
Higher education level is positively associated with higher fertility intention. It could be due to a lack of resources among those with lower education (Brauner-Otto & Geist, 2018;Mills et al., 2008), as well as higher advantageous positions among high-level educated ones (Dommermuth, et al., 2017;Wesolowski, 2015). In countries where there are more opportunities for women to reach high levels of education, other structural circumstances affecting fertility are also available, for example life satisfaction, sense of well-being and levels of trust. Besides, policies can successfully combine work and family life for highly educated women in these countries.
Undoubtedly, the marriage market has also an important role in this regard. As highly educated women have more opportunities to marry and have a better-educated partner, so they can plan to have larger families (Testa et al., 2011).
But in some societies, educated women increasingly face a deficit of educated men with whom to pursue childbearing. This leads them to resort to elective egg freezing (EEF). Women may be resorting to EEF to pursue careers and achieve reproductive autonomy (Inhorn et al., 2018).
Although non-working women have enough time to raise children, lack of personal investments in non-working women may let them enjoy less fortunate conditions and also receive less support from their partner and family, and institutions. As a result, it may limit their intentions to have a child (Wesolowski, 2015). The psychological value of children, such as providing comfort during old age, is associated with higher intentions to have a second child among non-working mothers. Non-working mothers may be more dependent on their children for emotional support during their elderly years compared with employed women and even fathers.
The labour force participation of women may provide an alternative source of satisfaction, which can decrease the psychological advantages of children (Park & Cho, 2011). There is an indirect effect of satisfaction with job security on childbearing intention.
More satisfaction with job security may lead to fewer conflicts in the couple relationship. Hence, it may lead to satisfaction within a partnership and a higher intention to have a first child (Dommermuth et al., 2017).
The results showed that happier men and women prefer to become parents sooner. When both the woman and the man report a particularly high level of happiness, the probability of becoming parents for the first time increases more than when only one of the two partners is happier than usual. (Timaeus & Moultrie, 2020). Women with a low level of happiness might not have a positive experience with the first child while women with a high level of happiness may not want to change the positive status they live; So both prefer to limit their childbearing intentions (Aassve et al., 2016). Health concerns can seriously affect second child intentions among couples.
Poor health threatens childbirth and may compromise the health of both mother and child (Wesolowski, 2015).
In zero parity, a couple agreement or disagreement has the same outcome. As childlessness is not a norm in most countries, under the pressure of society, they will experience a birth almost as often as the same (Testa et al., 2011). The possibility for an additional child is associated with couples' desires. As parenthood is more related to women's lives, it seems that the female partner's opinion plays an essential role in childbirth decision-making (Bernhardt et al., 2016;Fan & Maitra, 2010). When a man wants to have a child but, his wife does not, he prefers to express negative or uncertain childbearing intention. It indicates that women have a stronger influence on short-term childbearing intentions (Bernhardt et al., 2016). It is undeniable that the man's active involvement in childcare duties turns disagreement more towards childbearing in higher parities (Testa et al., 2011).
Although a state of satisfactory partnerships is required to have a child (Fiori et al., 2013), both highly positive and highly negative interactions between couples have a significant negative influence on the rates of the first and subsequent births. A great deal of satisfaction means partners are happy with their current family situation so having another child may threaten their satisfactory condition.
Unsatisfactory relationships between couples disturb a suitable pre-condition for child-rearing. Thus, an additional child may face their life with a new challenge. Nevertheless, effective aspects of relationship quality and their mechanisms are still unknown (Park et al., 2008).
Gender role attitudes have an important effect on how men and women view parenthood (Yoon, 2016;Yu & Kuo, 2017). Traditional men are more likely to partner with traditional women. In comparison, egalitarian men might prefer to partner with egalitarian women who are considerably less fascinated by the benefits of motherhood.
Holding a traditional gender role attitude is positively related to childbearing (Bernhardt et al., 2016;Dommermuth, et al., 2017). In comparison, egalitarian men are more likely to partner and remain partnered (Bernhardt et al., 2016). It seems that they are more attractive to their partners as they share more equally with them.
Having less conflict with their wives leads to a happier marital relationship and may encourage them to plan to have a child. That is why most women prefer men who share in the responsibilities of the household (Kaufman, 2000). Childless couples are influenced by network partners and their family size. Unlike one-child peers, peers with two or more children have a negative influence on the intentions of the childless couples.
Women are more likely to have a/another child while they live with or near parents (in-law). Also, this behaviour is established among those who are not working or whose husbands work relatively long hours or have relatively low educational attainment (Raymo et al., 2010).
However, frequent intergeneration exchange and co-residence decrease intentions to have an extra child (Harknett et al., 2014).
Although strong extended family ties are expected to encourage higher levels of fertility, the generation of middle-aged adults (so-called, "sandwich generation") may face concurrent commitment to support elderly parents and dependent children (Grundy & Henretta, 2006;Testa et al., 2011). Therefore, they may limit their family size to limit their support obligations (Harknett et al., 2014). In another way, couples from large families have experiences of having probable problems related to a large number of siblings. Their parents may evaluate their own high fertility experiences negatively and lead them to limit their fertility intentions (Bühler & Fratczak, 2007).
Living in single-family houses is associated with a higher intention to have a first child compared with living in apartments. This result may be due to the impossibility of expanding an apartment space or living in crowded apartments. However, the exact reasons are unclear (Kulu & Vikat, 2007;Park et al., 2008). In countries with no government regulation in the house renting market, couples face a difficult housing regime (Kulu & Vikat, 2007;Vignoli et al., 2013).
Religion and family make up a large proportion of shaping personal identity. Higher fertility intention among religious people is related to family beliefs and values, including schemas about the importance of marriage and parenthood, and gender roles in families.
Fertility differentials are a part of a widespread association between religiosity and family behaviour (Hayford & Morgan, 2008).
According to ECSM, fertility behaviour largely depends on the characteristics and ability of individuals to communicate experiences and information (Bühler & Fratczak, 2007). Influential communication networks show reproductive planning and decision-making (Fan & Maitra, 2010;Kuhnt & Trappe, 2016;Testa, 2012;Vignoli et al., 2013). Kuhnt and Trappe, (2016) mentions that those perceived social pressure to have a child are more likely to have positive fertility intentions and vice versa. This may reflect the willingness of individuals to comply with norms.
Increasing female labour force participation in developed countries around the world was initially associated with a decline in the fertility rate. Gender inequality and the difficulty of combining work and family are possible explanations for this. However, the negative relationship between women's labour force participation and childbearing has turned positive since 1990 (Averett & Whittington, 2001;Bernhardt et al., 2016;Mills et al., 2008).
In developed societies, higher-order births are likely to be more responsive to policy and environmental changes compared with the first births (Morgan & Taylor, 2006). Labour market policies are expected to change the labour market so that couples can maintain their employment and income, even with young children (Neyer et al., 2013;Wesolowski, 2015). Countries with high national-level gender equality also need household-level equity to see an increase in fertility (Bernhardt et al., 2016;Dommermuth, et al., 2017;Mynarska et al., 2015).

| CON CLUS ION
Our analysis demonstrates that childbearing intention is affected by complex reciprocal interactions of multilevel ecological factors that can help policymakers expand their awareness of factors that affect fertility. Hence, it is not possible to separate the influence of one domain from another. To promote childbirth, it is essential to consider a multidimensional programme according to the features of each regional and geographic area.

| S TUDY S TRENG TH S AND LIMITATI ON S
53 studies met our inclusion criteria. 24 (45%) of studies were conducted on both men and women, 27 (51%) on women and just 2 (4%) on men. Most of the studies deal exclusively with women, and the theorizing regarding what leads to childbearing decision-making, particularly among men, is almost underdeveloped. There are limited instances of interdisciplinary studies, studying fertility would highly benefit by crossing disciplinary and geographic boundaries.
Developing comparable data collection in many countries will improve fertility research. The lack of actual instruments such as the collection of quantitative data for the network-based approach is another key factor in infertility studies.

ACK N OWLED G EM ENTS
The present study was supported by Shahroud University of medical sciences as a PhD Thesis. We hereby acknowledge the research deputy for grant No 9691.

CO N FLI C T O F I NTE R E S T S
Not applicable.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author, upon reasonable request.