1The promotion of breastfeeding forms part of many countries’ public health programmes based on the World Health Organization guidelines and epidemiological research on the beneficial effects of breastfeeding on both the mother’s and child’s health (Victora et al., 2016). [1] In France, the National Nutrition and Health Programme (Programme national nutrition santé, PNNS) recommends breastfeeding “exclusively, if possible, until the age of six months and at least until the age of four months for a health benefit” (Hercberg et al., 2008). But these official recommendations are not always put into practice. Although in 2016 more than two-thirds of infants were breastfed at birth, the incidence and duration of breastfeeding in France are among the lowest in the Western world, with a median breastfeeding duration of four months in 2011 (Blondel et al., 2017; Victora et al., 2016). This can be explained by various factors, such as mothers’ high participation in the labour market, the culturally ingrained use of breast milk substitutes (Rollet, 2006), and limited public investment in the promotion of breastfeeding compared with other European countries (Bosi et al., 2016). Professional practices, family customs, and social representations of breastfeeding seem to have contributed to the relatively low rates of breastfeeding in France (Di Manno et al., 2015). But an individual’s sociocultural environment also influences the decision to start and continue breastfeeding, with migration status appearing to be one of the principal differentiating factors (Kersuzan et al., 2014; Wagner et al., 2015).

2This article compares breastfeeding practices in France based on the parents’ geographical origin. It focuses on the effect of migration status compared with that of origin and duration of residence in France. Qualitative studies have identified several factors likely to cause immigrant mothers to modify their practices and to adopt those of their host country. First, the geographical distancing involved in migration isolates mothers from the advice of their friends and family and from the customary practices of their country of origin (Higginbottom, 2000). Secondly, the host country’s guidelines and pre- and postnatal medical monitoring may influence their behaviours, also distancing them from their original family and cultural customs. Although breastfeeding is generally encouraged by health professionals, several studies show that inadequate professional practices may generate conflicting norms for immigrants regarding how to feed their child (Nacu, 2011; Schmied et al., 2012). Furthermore, the frequent use in France of off-the-shelf products, such as breast milk substitutes, may appear as a sign of integration (Côté, 2003). Social background or class may also play a part, with breastfeeding less common in those working-class categories in which immigrants and second generations are over-represented (Kersuzan et al., 2014; Wagner et al., 2015). Despite these factors, in most Western countries, research shows that immigrants retain breastfeeding practices similar to those of their country of origin. The few English-language studies available on the specific effect of migration status (rather than race or ethnicity) conclude that breastfeeding is more prevalent among immigrants than among natives of any origin (Gibson-Davis and Brooks-Gun, 2006; Singh et al., 2007). However, interpreting immigrants’ practices depends on the context in which they are compared. In the European countries with the lowest rates of breastfeeding, such as Ireland, natives also breastfeed less than immigrants regardless of their country of birth (Nolan and Layte, 2015). In contrast, in countries with a high incidence and long duration of breastfeeding, particularly the Nordic countries such as Denmark, immigrants breastfeed less than natives (Busck-Rasmussen et al., 2014). But the data do not generally enable us to measure the extent to which these behaviours are maintained in the following generation among natives of immigrant origin. Some studies note the importance of paternal influence and measure a positive effect of the father’s immigrant status on breastfeeding at birth among natives (Gibson-Davis and Brooks-Gun, 2006; Nolan and Layte, 2015). In France, only a few studies are able to follow and compare breastfeeding practices based on the geographical origin of both parents, and even fewer for the following generation.

3This article therefore seeks to evaluate the extent to which the breastfeeding practices of immigrant and second-generation mothers converge with those of the host country, France, by taking paternal origin into account and investigating duration. After describing breastfeeding practices based on geographical origin and migration status (majority population, immigrant population, and second generations), from birth until the infant is six months old, this analysis will examine the effect of origin on the probabilities of breastfeeding, net of the influence of the sociodemographic characteristics, and investigate the role of the union’s parental diversity.

I – Data

4The data are taken from the French Longitudinal Study of Children (Étude longitudinale française depuis l’enfance, ELFE) in which more than 18,300 children were included at the time of their birth in 2011, to be followed up until adulthood. Eligible mothers had to be of legal age, have delivered after at least 33 weeks of amenorrhoea, [2] have delivered a single child or twins, and not be planning to live outside metropolitan France during the following three years. A little over half (51%) of the parents approached agreed to be followed in one of the languages offered (French, English, Arabic, or Turkish). The representativeness of the results across metropolitan France is ensured by weighting to redress the design effect and non-response bias and to adjust the structure of the sample on key calibration variables. [3]

5The information about starting breastfeeding (breastfeeding at birth) was collected from the mother during a face-to-face interview in the maternity hospital, and the information on the duration of breastfeeding was collected through monthly self-administered questionnaires between three and ten months after the birth, concluded by a retrospective telephone report when the infant reached one year of age. Around 66% of parents completed at least one self-administered questionnaire, and 80% responded to the one-year questionnaire.

6The parents’ migration status, recorded in surveys carried out in maternity hospitals and when the child was two months old, differentiates native parents (those born in France) and immigrant parents (born as foreign nationals abroad). The two-month questionnaire also indicates the birth nationality and country of the child’s grandparents and can be used to differentiate natives based on whether they have at least one immigrant parent (descendant) or no immigrant parents (majority group). However, without oversampling the populations of foreign origin in the ELFE study, the results cannot be broken down by country of origin, only by broad geographical zone. This approach probably masks disparities concerning values from the country of origin and social integration in the host country. In the following analyses, we will therefore distinguish three groups based on geographical origin, nationality at birth, and migration status: the “majority” group composed of natives born to native parents; the “immigrant” group composed of individuals born as foreign nationals abroad; and the “descendant” group, composed of natives born to at least one immigrant parent, thus representing the second generation.

II – Method and study population

7This article investigates the incidence and duration of breastfeeding, whether exclusive (breast milk only) or partial (supplemented by water, fruit juice, or off-the-shelf milk products). Bivariate analyses were performed on a population of 14,110 mothers for whom the information on breastfeeding practices [4] and migration status was available. [5] Due to missing data for the control variables, the multivariate analyses were conducted on a population of 13,166 mothers for the study of breastfeeding at birth and 9,390 mothers for breastfeeding at six months. Tests enabled us to ensure the absence of any selection effect on the results obtained for the mother’s migration status. [6]

8The analyses systematically take into account the weightings and structure of the multistage survey design, [7] performed to correct the estimators of the effects of the unequal inclusion probability of the infants (selection of maternity hospitals and survey days) and to consider potential selective attrition. [8]

9Based on established breastfeeding factors, various sociodemographic characteristics were incorporated into the multivariate analysis as control variables. The literature has shown that breastfeeding is more frequently initiated and maintained for longer among mothers who are aged 25 and over, married, have a high level of education, and are in skilled employment (Bonet et al., 2013; Brand et al., 2011; Girard et al., 2016). Excess maternal weight and smoking during pregnancy reduce the probability of breastfeeding, while attending birth-preparation sessions, breastfeeding previously, and even having been breastfed by their own mother increase it (Bonet et al., 2007; Gojard, 2000; Negin et al., 2016). Some studies have also shown that the infant’s sex has an influence (Shafer and Hawkins, 2017). Additionally, in France, the first descriptive analyses of the ELFE cohort show that the father’s characteristics, irrespective of migration status, are associated with infant feeding practices, especially the impact of his involvement in the family unit, starting with his presence at the birth (Kersuzan et al., 2014). Qualitative research also emphasizes the positive impact on breastfeeding of the father’s involvement in household chores (Tohotoa et al., 2009).

10The control variables used are progressively added (see Appendix Table A.1) by group of variables to an initial empty model comprising only the origin of the mothers (Model 1). They are divided into three major groups. The first group (Model 2) comprises the infant’s sex and the mother’s sociodemographic characteristics: age, level of education, occupation and social class (OSC), body mass index (BMI) before pregnancy, marital situation when questioned in the maternity hospital, and net monthly household income per consumption unit (CU). It also includes a variable combining information about the mother’s activity status at the time of the survey conducted two months after the child’s birth (not working, on parental leave, working) and about her intentions relating to a return to work (parental leave versus delayed return to work). The second group (Model 3) refers to factors describing the mother’s education about breastfeeding, such as attendance of at least one birth-preparation session, a variable combining birth order and breastfeeding experience for multiparous women, and a variable combining the fact of having been breastfed themselves and having taken advice on childcare from their own mother. Finally, the third group (Model 4) supplements the previous model with the characteristics of the father: OSC, presence at the birth, geographical origin, and migration status. [9] It also includes a synthetic indicator of the extent to which domestic tasks are shared within the couple based on the mother’s report of each spouse’s involvement in various activities.

III – Effects of geographical origins

11The analyses performed on all the mothers (n = 14,110) reveal significant disparities according to origin in the proportion of children breastfed at birth and at six months. Whereas only 67% of the mothers in the majority population breastfed their child at birth, 75% of the descendants of immigrants did so, and 88% of the immigrants (all differences significant with a 95% confidence interval) (Figure 1). The effect of origin persists at six months, when 45% of immigrants breastfed versus 20% of natives, but there is no longer any significant difference between the majority group and that of the descendants.

Figure 1

Breastfeeding rate (and 95% confidence interval) by infant’s age and mother’s origin (n = 14,110)

Breastfeeding rate (and 95% confidence interval) by infant’s age and mother’s origin (n = 14,110)

12Breastfeeding at birth is more frequent among immigrants from the Maghreb (92%) than among those from the European Union (87%), with those from sub-Saharan Africa (85%) positioned in the middle (Figure 2A). Immigrant mothers also breastfeed for longer than natives, especially those from sub-Saharan Africa, 55% of whom breastfed until the infant was six months old, versus around 40% of immigrants from all other countries (Figure 2B).

13Behaviours also vary between the descendants of immigrants, depending on whether only one of their parents is an immigrant (mixed marriage) or both (endogamous). Breastfeeding at birth is more common among natives with parents who were both born in the Maghreb or sub-Saharan Africa (82% and 92%, respectively), while descendants with only one parent from these countries and the other from another European Union country are no more likely to breastfeed than natives in the majority group (Figure 3A). Likewise, continued breastfeeding until six months reveals no great difference between natives based on the geographical origin of their parents. Only mothers with two parents born in a sub-Saharan African country breastfed more at six months (40%) than the mothers in the majority group. Among descendants whose parents both immigrated from sub-Saharan Africa, breastfeeding practices at birth and six months are most similar to those of immigrants of the same origin, with other descendants all breastfeeding less than immigrants of the same origin (comparison of Figures 2 and 3).

Figure 2

Incidence of breastfeeding (and 95% confidence interval) among immigrants based on region of birth (n = 14,110)

Incidence of breastfeeding (and 95% confidence interval) among immigrants based on region of birth (n = 14,110)

Figure 3

Incidence of breastfeeding (and 95% confidence intervals) among descendants of immigrants based on the combined region of birth of their parents (n = 14,110)

Incidence of breastfeeding (and 95% confidence intervals) among descendants of immigrants based on the combined region of birth of their parents (n = 14,110)

14These gross differences may be explained by the sociodemographic specificities of the three groups compared. Applying logistic regression models allows us to measure the effect of origin on breastfeeding, net of other observable characteristics.

15Immigrant mothers and mothers who are the descendants of immigrants have different characteristics from those of the majority population for the set of control variables, except infant’s sex and mother’s age at birth (Appendix Table A.2). Considering these structural effects changes the amplitude of the breastfeeding disparities initially observed based on origin, without modifying either the direction or the significance thereof (Table 1, Model 4).

16Progressively introducing the control variables into the models provides a more in-depth analysis and allows us to identify some of the structural factors that contribute to the origin-based differences in the mothers’ breastfeeding practices. Certain specificities of immigrants and descendants of immigrants are thus negatively associated with breastfeeding. Compared with the mothers in the majority group, the immigrants and descendants of immigrants were more likely to be overweight and have lower levels of education and income, and were less likely to occupy high positions in the social hierarchy. Their rates of attendance at birth-preparation sessions were lower; they were more likely to report the absence of their spouse at the time of the birth; and their spouse was less likely to be in a higher-level occupation (cadre) or in an intermediate occupation (professions intermédiaires). Including these properties in the model, all other things being equal, therefore serves to amplify or reduce the breastfeeding differences initially observed, compared with mothers in the majority population. The difference in the net probabilities of breastfeeding among immigrants from sub-Saharan Africa and the Maghreb becomes greater with identical sociodemographic characteristics (particularly level of education, socioprofessional category, and net monthly household income), between Model 1 (controlling only for the respondents’ origin) and Model 2. An opposite variation is observed for breastfeeding until the age of six months, where introducing sociodemographic variables reduces the value of the odds ratios calculated for the migrants. Planning an early return to work (less than 16 weeks after the birth) has more of a negative impact on the continuation of breastfeeding than on the choice of how to feed the infant. Breastfeeding mothers in the majority population were more likely to be planning an early return to work than immigrants and descendants of immigrants from the Maghreb and sub-Saharan Africa, who were more likely not to have worked during pregnancy. Furthermore, migrants share other characteristics positively correlated with breastfeeding. Applying an “all other things being equal” reasoning regarding the variables describing these properties reduces breastfeeding disparities (between Models 2 and 3, and then between Models 3 and 4).

17Compared with mothers in the majority population, migrants are much more likely to come from a family environment in which the preceding generation practised breastfeeding. More than 80% of the immigrants and 60% of the descendants of immigrants were themselves breastfed, compared with fewer than half of mothers in the majority population. The pervasiveness of the family culture, associated with the preference for using family as a source of advice on caring for infants, is stronger among migrants. More than 50% of them had sought advice from their mother about childcare, compared to 41% of mothers in the majority population; greater preference for family sourcing in working classes has already been shown (Gojard, 2003). The advice from grandmothers among immigrants certainly has a more positive impact on breastfeeding than the advice from the grandmothers among the majority population, the latter having been mothers in a context that placed much less value on breastfeeding than it does today. Furthermore, among multiparous migrants, the choice of how to feed a newborn follows on from previous breastfeeding, potentially initiated in their region of origin. More than 80% of multiparous immigrants breastfed all their previous children, compared with a little under 70% of descendants of immigrants and 60% of mothers in the majority population (Appendix Table A.2). Lastly, due to the endogamy of the marriages, immigrants and descendants of immigrants frequently have spouses of the same origin. Studies have shown that having an immigrant husband has a positive effect on breastfeeding at birth (Kersuzan et al., 2014). This model implicitly assumes that the effect of the father’s origin on breastfeeding is identical for all mothers, whether they belong to the majority population or not. It is therefore impossible to know whether marital diversity favours the alignment of the breastfeeding practices of immigrants or descendants of immigrants with those of the host country.

Table 1

Odds ratios and 95% confidence intervals calculated using logistic models^{(a),(b),(c),(d)}

Breastfeeding at birth (n = 13,166) Model 1 (a) Model 2 (b) Model 3 (c) Model 4 (d) Majority group (Ref.) 1.0 1.0 1.0 1.0 Mother’s migration status Descendant Immigrant 1.6 [1.3–1.9] *** 3.5 [2.7–4.6] *** 1.7 [1.4–2.1] *** 4.0 [3.0–5.2] *** 1.4 [1.1–1.7] ** 2.0 [1.5–2.8] *** 1.3 [1.1–1.6] * 1.6 [1.2–2.2] *** Mother’s migration status and geographical origin. Immigrant from: European Union Maghreb Sub-Saharan Africa Other countries 3.1 [2.1–4.8] *** 5.5 [3.2–9.2] *** 2.2 [1.5–3.4] *** 4.2 [2.2–8.0] *** 3.1 [1.9–4.9] *** 6.7 [3.9–11.5] *** 3.0 [2.0–4.5] *** 4.3 [2.3–8.2] *** 2.2 [1.3–3.9] ** 3.7 [2.1–6.7] *** 2.3 [1.5–3.5] ** 2.3 [1.2–4.3] ** 1.9 [1.1–3.4] * 3.0 [1.7–5.3] *** 1.9 [1.2–3.2] ** 2.0 [1.1–3.7] * Descendant of: European Union Endogamous marriage Mixed marriage Maghreb Endogamous marriage Mixed Sub-Saharan Africa Endogamous marriage Mixed marriage Other countries Endogamous marriage Mixed 1.1 [0.7–1.7] 1.1 [0.9–1.5] 3.0 [1.8–4.7] *** 1.1 [0.7–1.6] 4.9 [1.8–13.0] ** 1.9 [0.9–4.2] 3.0 [1.5–5.7] ** 3.4 [1.6–7.3] ** 1.1 [0.7–1.7] 1.1 [0.8–1.5] 3.6 [2.3–5.8] *** 1.1 [0.7–1.6] 5.7 [2.1–15.4] ** 2.4 [0.9–6.0] 3.6 [1.8–7.4] ** 3.4 [1.6–7.3] ** 1.0 [0.6–1.7] 1.1 [0.7–1.5] 2.0 [1.2–3.3] ** 1.1 [0.7–1.6] 3.6 [1.5–8.5] ** 1.4 [0.7–2.9] 2.6 [1.1–6.3] * 2.3 [1.1–4.7] * 1.0 [0.6–1.6] 1.1 [0.7–1.5] 1.8 [1.1–2.9] * 0.9 [0.6–1.4] 3.0 [1.2–7.2] ** 1.2 [0.6–2.6] 2.3 [1.1–4.9] * 2.1 [1.0–4.5] *

Breastfeeding at six months (n = 9,390) Model 1 (a) Model 2 (b) Model 3 (c) Model 4 (d) Majority group (Ref.) 1.0 1.0 1.0 1.0 Mother’s migration status Descendant Immigrant 1.1 [0.9–1.3] 2.7 [2.2–3.3] *** 1.1 [0.9–1.3] 2.2 [1.8–2.7] *** 1.0 [0.9–1.3] 2.0 [1.7–2.5] *** 1.0 [0.8–1.2] 1.7 [1.4–2.1] *** Mother’s migration status and geographical origin. Immigrant from: European Union Maghreb Sub-Saharan Africa Other countries 2.5 [1.7–3.5] *** 2.1 [1.4–2.9] *** 4.3 [3.0–6.2] *** 2.9 [1.9–4.2] *** 2.1 [1.4–2.9] *** 1.7 [1.3–2.4] ** 4.3 [2.9–6.5] *** 2.1 [1.4–3.2] *** 2.0 [1.4–2.0] *** 1.6 [1.1–2.2] * 3.8 [2.5–5.7] *** 2.0 [1.3–2.9] ** 1.8 [1.3–2.6] ** 1.5 [1.1–2.1] * 2.9 [1.9–4.5] *** 1.7 [1.2–2.5] ** Descendant of: European Union Endogamous marriage Mixed marriage Maghreb Endogamous marriage Mixed Sub-Saharan Africa Endogamous marriage Mixed marriage Other countries Endogamous marriage Mixed 1.1 [0.7–1.7] 0.9 [0.6–1.2] 1.1 [0.8–1.6] 1.0 [0.7–1.5] 1.5 [0.7–1.5] 0.9 [0.5–1.8] 1.8 [1.1–3.2] * 1.4 [0.7–2.7] 0.9 [0.6–1.5] 0.8 [0.6–1.1] 1.0 [0.7–1.4] 1.0 [0.6–1.7] 1.6 [0.7–1.6] 1.1 [0.5–2.3] 1.9 [1.1–3.4] * 1.7 [0.8–3.5] 0.9 [0.6–1.5] 0.8 [0.6–1.1] 1.0 [0.7–1.4] 1.1 [0.6–1.8] 1.6 [0.7–1.6] 0.9 [0.4–2.0] 1.9 [1.1–3.4] * 1.5 [0.7–3.2] 0.8 [0.6–1.5] 0.8 [0.6–1.1] 0.9 [0.6–1.3] 1.0 [0.6–1.7] 1.3 [0.5–1.3] 0.8 [0.4–1.8] 1.8 [1.0–3.2] * 1.5 [0.7–3.2]

Odds ratios and 95% confidence intervals calculated using logistic models^{(a),(b),(c),(d)}

Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001
Note: The odds ratios shown are adjusted for a set of control variables:
(a) Model 1: empty model, containing only the variable of interest.
(b) Model 2: infant’s sex and mother’s sociodemographic characteristics (age, level of education, marital situation at the time of the maternity hospital survey, OSC, pre-pregnancy BMI, net monthly household income, activity status, and professional plans for working mothers: return to work or parental leave).
(c) Model 3: variables from Model 2 + variables describing the mother’s relationship to breastfeeding (attendance at birth-preparation sessions, birth order, and breastfeeding experience among multiparous mothers, whether the mother was breastfed, and whether the mother has taken advice on childcare from the maternal grandmother).
(d) Model 4: variables from Model 3 + characteristics of the father (OSC, presence at the birth, geographical origin, migration status, and each spouse’s degree of involvement in domestic tasks).

IV – Significance of marital diversity in infant feeding practices

18Examining breastfeeding rates by parents’ origin shows that marital diversity has a distinct influence on infant feeding practices, depending on the migration status of the mothers (Figure 4). Natives, whether or not they have an immigrant background, are more likely to breastfeed at birth and at six months in marriages where the father is an immigrant. On the other hand, among native parents, having an immigrant background appears only to have a minor impact on the likelihood of breastfeeding. Among immigrant mothers, their spouse’s origin has little effect on breastfeeding. Breastfeeding rates remain higher when the mother is an immigrant, regardless of the type of marriage (endogamous or mixed). Several studies have demonstrated the importance of the father’s values and opinions for infant feeding decisions (Salanave et al., 2012), and the father’s attitude towards breastfeeding may depend on the context in which he grew up. Most immigrant fathers were born in countries where more value is placed on breastfeeding than it is in France and may therefore have more supportive attitudes towards this practice than native fathers. Having an immigrant husband therefore increases the likelihood that a mother born in France will breastfeed. However, we must assume from our results that, for immigrant mothers, the advice from the maternal grandmother is more important than that of the husband regarding the choices surrounding infant feeding. The husband’s attitude to breastfeeding is less of a determining factor among immigrant mothers. Because they are aligned with the public health guidelines, the positive attitudes of spouses may also have more impact on mothers than the attitudes of fathers who do not support breastfeeding.

Figure 4

Breastfeeding rate based on marital diversity (n = 14,110)

Breastfeeding rate based on marital diversity (n = 14,110)

Note: Significance of coefficients calculated for each type of parental configuration (Ref.: father and mother from the majority group) using a logistic model adjusted for all variables for analysis (n = 13,166 for breastfeeding at birth / n = 9,390 for continued breastfeeding to six months).
Significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001.

Conclusion

19Migration status is one of the principal factors in the social differentiation of breastfeeding practices in France (Kersuzan et al., 2014). The ELFE data have enabled us to measure how this migration effect acts, not only on the decision to breastfeed at birth but also on its duration. It has also enabled us to track whether specific breastfeeding practices are continued among the descendants of immigrants and to appreciate the impact of paternal immigrant origin on these behaviours.

20Our results show that, in France, immigrant mothers breastfeed more at birth and at six months than natives, regardless of their region of birth, which is consistent with other studies conducted in English-speaking countries. This observation shows the significance of migration status on breastfeeding practices. Whereas, among the general population, mothers from the most socioeconomically disadvantaged categories breastfeed the least in France (Kersuzan et al., 2014; Wagner et al., 2015), immigrants breastfeed more frequently and for longer, especially those originating from sub-Saharan Africa and the Maghreb, who are, however, over-represented in these social environments. The father’s migration status also exerts an influence on natives, who breastfeed more when the child’s father is an immigrant. Nevertheless, the association between origin and breastfeeding diminishes among the second and future generations, with parents who are born in France having similar practices, whether or not they have a migration background; only African descendants from an endogamous marriage breastfeed slightly more at birth. Overall, migration status has much more effect than origin on differences in breastfeeding. This finding reflects the progressive incorporation of French breastfeeding practices among the second and future generations. Regarding differences based on the immigrants’ origin, those from the Maghreb breastfeed the most at birth, and mothers from sub-Saharan Africa continue breastfeeding for longest. These specificities may arise from differences in the perception of breastfeeding in France and in the immigrants’ country of origin. Most African immigrants come from countries where breastfeeding is the dominant and expected method of infant feeding, whereas in France, the use of alternatives is common and well established. Family customs also favour breastfeeding more among immigrants than among natives. However, breastfeeding rates among immigrant mothers must be put into perspective, given the specific context of breastfeeding in France, where its prevalence at birth and duration are some of the lowest in the Western world.

21Although the size and wealth of variables from the ELFE cohort greatly contribute to the statistical study of the breastfeeding practices of mothers based on origin, other potential determinants may have been overlooked in the analysis because they were not collected in the survey, such as the role of midwives (Lind et al., 2014). In addition, the effect of mothers’ pre-migration socioeconomic status on origin-based breastfeeding differences still needs to be measured, given that the social gradient in breastfeeding in some of the immigrants’ countries of origin (where breastfeeding decreases as level of education increases) is the reverse of that which exists in France. This research could also be developed by studying methods of breastfeeding, especially non-exclusive breastfeeding, which is a common practice in some regions of parental origin.