Home visiting intervention over pregnancy and the first two years of life to promote healthy feeding practices and energy balance-related behaviors and prevent obesity risk in children from socially disadvantaged families: protocol for a participant-blinded, two-arm, parallel-group randomized controlled trial—the ECAIL study

Abstract

Background

Social inequalities in health emerge early in life. The French Malin program, a multi-behavioral and multilevel intervention targeting both individual and structural factors, was co-created by participatory approaches involving diverse stakeholders. These stakeholders include end-users, the French Red Cross, non-profit associations, and two French pediatric societies. The program aims to promote healthy feeding practices (such as breastfeeding) and foster optimal lifestyle behaviors to prevent obesity among children from socially disadvantaged families. The ECAIL study aims to evaluate its effectiveness.

Methods

ECAIL is an ongoing single-blind, two-arm, parallel-group randomized controlled trial, with balanced (1:1) randomization at the individual level. Co-designed with the stakeholders of the Malin program, the trial has been implemented since 2017 in Northern France. Pregnant women are screened for social disadvantages at the main hospital maternity wards in Lille and Valenciennes with the assistance of healthcare providers. Eligible participants are then enrolled for follow-up by trained dietitians, who provide home visits until the child’s second birthday. Participants in the intervention arm receive the two components of the Malin program: (1) personalized support, including the promotion of breastfeeding, advice, and practical tips for a balanced diet and active lifestyle and (2) incentives to access kitchen utensils and healthy foods at reduced prices. Participants in the control arm receive usual care. Primary and secondary outcomes include rate and duration of breastfeeding and complementary feeding practices, dietary intake and lifestyle patterns in children at age 1 and 2 years, and the risk of obesity at age 2. Quantitative data are primarily collected with face-to-face questionnaires and anthropometric measurements. Additionally, semi-structured interviews with study participants and stakeholders will contribute to the process evaluation of this complex intervention.

Discussion

The co-creation of ECAIL, utilizing a bottom-up approach with key stakeholders of the existing Malin program, is a meaningful step toward integrating experiential and academic knowledge. By leveraging mixed data, the findings from this action-research are expected to provide critical insights into the intervention: what it entailed, how it worked, for whom, and to what extent. Amid rising food insecurity, this knowledge will be pivotal in refining the Malin program as it scales up.

Trial registration

ClinicalTrials.gov: NCT03003117.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-25004-0.

Background

According to the World Health Organization (WHO), overweight and obesity (OW/OB) have reached concerning levels globally, affecting about 37 million children under 5 years old in 2022 [1]. More than 25% of preschool children are affected in Southern Europe (e.g., Spain, Portugal) [2, 3], the United States [4] and Australia [5]. This high and early prevalence indicates that risk factors are at play from an even earlier age. According to the Developmental Origins of Health and Disease paradigm, there is great plasticity in children’s early development in response to the nutritional, physicochemical, or stressful characteristics of their environment, both intrauterine and during the first years after birth [6, 7]. Such exposure to early stressors could leave a long-lasting imprint on the control of the energy balance, storage capacities, and the distribution of energy between different body systems and organs [8]. Hence, the WHO Commission on Ending Childhood Obesity emphasized the need to address obesity risk factors during the “first 1000 days”, which refers to the period from conception to age 2 years [9].

At the individual level, early risk factors for childhood OW/OB include parental pre-pregnancy OW/OB, unhealthy parental lifestyle during pregnancy (e.g., smoking and poor diet), fetal macrosomia, rapid weight gain in the early months of life, poor feeding practices, and the child’s adoption of suboptimal energy balance-related behaviors (EBRBs). These EBRBs encompass energy-dense, nutrient-poor dietary intake, excessive screen time, low physical activity levels and inadequate sleep [3, 7, 10–15]. These potentially modifiable behavioral risk factors are socially patterned [16–18] and explain in part the inverse relation between parents’ socio-economic position and childhood OW/OB observed in high-income countries [12, 19, 20]. In fact, at the societal level, childhood OW/OB disproportionally affects populations with socio-economic disadvantage as well as ethnic minorities [3, 4, 21–26].

Social inequalities in obesity risk not only are increasing [27] but also emerge early in life: evidence shows a social gradient in body mass index (BMI) as early as 1 month after birth, driven by a complex interplay of disparities in weight and height growth [28]. Children with OW/OB are more likely to grow into adults with obesity [3]. Furthermore, OW/OB negatively affects physical and mental health during childhood [3, 29, 30] and also human capital development, influencing cognitive performance, educational attainment, and social trajectories later in life [31]. As a result, social inequalities are perpetuated across generations, with health inequities beginning at birth, thus highlighting the broader issue of social injustice [24, 32, 33].

Epidemiologic studies on families with young children facing social disadvantage remain scarce in Europe [34]. This paucity is largely due to underprivileged populations being known to be “hard to survey”; they are not only difficult to sample but also challenging to identify, reach, persuade, interview, and follow up [35, 36]. Professionals working with underserved families and researchers have expressed concerns about the use of cow’s milk as a substitute for breastfeeding or infant formula during the first year of life because of its excessive content of protein, sodium, potassium, phosphorus, and calcium for human infant needs while lacking sufficient iron, vitamin C, and linoleic acid [37, 38]. Limited dietary surveys within these populations, primarily adults, have also revealed less structured eating behaviors, such as reduced food variety and meal skipping, both indicators of food insecurity [34, 39–43]. In fact, food insecurity is associated with increased consumption of discretionary, energy-dense and ultra-processed products, often at the expense of nutrient-dense foods such as vegetables, fruits, dairies, and fish—foods that are generally less affordable. Food insecurity disproportionally affects recipients of social benefits, single-parent families, immigrants and homeless individuals [40–44]. This imbalance contributes to the “double burden of malnutrition”—the coexistence of undernutrition and OW/OB within the same population—which is recognized in low- and middle-income as well as high-income countries [45].

Although the first 1,000 days represent a critical period of vulnerability, they also represent a unique window of opportunity to support parents—primary caregivers and role models—in fostering healthy family lifestyles and preventing obesity in their children [46–48]. EBRBs are set early and tend to track over the life course, which stresses the importance of developing healthy behaviors from early life [15, 17, 49–51]. However, universal and downstream educational programs require high levels of agency at the individual level, produce relatively limited effects among people living in socially disadvantaged situations and tend to increase social inequalities in health [52]. Therefore, evidence-based programs adapted to the most vulnerable population groups are needed. According to our recent systematic review [53], there is some promise for multi-behavioral interventions targeted and tailored to underserved families when started from pregnancy and pursued at least 2 years after birth, as long as they are co-created with stakeholders, including parents; adhere to theoretical frameworks; and engage professional delivery agents, multi-disciplinary teams and/or peer groups. Importantly, families with social vulnerabilities are often the least able to access health resources and support services. To enhance outreach, sustained home visiting has become an increasingly popular service delivery model, offering personalized support to families, fostering trust, enhancing maternal health and psychosocial skills, and strengthening the mother–child bond [53–56]. However, there is a pressing need for interventional research and public health action to provide more comprehensive support to disadvantaged families: this requires implementing structural facilitators and system-wide changes across multiple sectors and settings to synergistically empower parents to engage in healthy behaviors [33, 53, 57, 58]. Only such multilevel and upstream interventions are likely to more effectively and sustainably address the issue of social inequities [24, 59].

In 2021, 14.5% of the population in France lived below the poverty line [60, 61]. Of these individuals, 366,000 children under the age of 3 (19.7%) were affected, with higher rates in households in which only one parent was employed (32.0%) or in which a single parent was jobless (80.2%). In recent years, France has been among the European countries experiencing a substantial increase in income poverty [60], illustrated by recent case reports of scurvy and malnutrition [62]. Financial hardships often force households to make trade-offs, food budgets being one of the first areas affected [34, 42, 63, 64]. Although food aid plays a crucial role in mitigating acute food insecurity, it fails to address broader inequalities in access to healthy foods across the population [43]. Moreover, its capacity to improve overall diet is limited because it primarily provides long-shelf-life, pre-packaged, and ultra-processed foods, which are rarely suitable for young children’s nutritional needs [42, 65, 66]. Access to food aid also depends on factors such as social support, stable housing, and income thresholds [42]. Additionally, food aid can carry a stigma, requiring recipients to queue at specific locations during restricted hours. Consequently, failure to apply for, or use, food aid, is a major issue: only 25% of food-insecure adults actually use it (4.8 million in 2015 increasing to approximately 7 million after the COVID-19 pandemic and cost-of-living crisis) [42, 63, 67, 68]. A report on nutritional conditions of food-aid users [39, 69] called for new programs that combine the facilitation of access to healthy foods and nutritional education respectful of the culture, knowledge, and specific barriers of socially disadvantaged populations [39, 69]. Indeed, although the cost of food accounts for a portion of the social inequalities, financial resources do not explain all of it. Factors related to the environmental (e.g., availability, accessibility and affordability of food supplies) and socio-cultural (e.g., knowledge, social isolation and support, attitudes, norms, and beliefs) settings also play an important, although understudied, role [34, 70, 71].

In this context, a multilevel and multi-behavioral program called Malin was developed in France on the basis of participatory principles to promote healthy feeding practices (including breastfeeding) and foster optimal EBRBs among children from socially disadvantaged families. The spectrum of social vulnerability is broad and includes families in a situation of fragility who do not receive social, medical or food aid because they are just above the monetary poverty thresholds. The Malin program has been implemented since 2012 in four districts; it is based on an innovative and sustainable partnership between the French Red Cross, non-profit associations (NPAs under French law, 1901 Act; including the Malin program NPA), and two French pediatric societies: the French society of pediatrics (SFP) and the French association for ambulatory pediatrics (AFPA). This program combines nutritional support from pregnancy to the child’s second birthday with incentives to access kitchen utensils, cooking devices, and baskets of fruit and vegetables at reduced prices as well as healthy foods from age 6 months onward. The Malin program’s components and tool-box were co-designed in accordance with the National nutrition and health program (Programme national nutrition santé [PNNS]) guidelines [72] at the cutting-edge of social innovation, health promotion and solidarity with social, health, and early childhood stakeholders as well as end-users (i.e., parents) [73]. Malin complies with and is supported by on-going national public policies: it was highlighted as a major action in the National strategy to fight against poverty in 2018 (then in the Solidarity pact, which took over in 2024); the PNNS since 2019; and the 1000 days National strategy launched in 2021. In the context of the COVID-19 pandemic and the cost-of-living crisis, along with the subsequent increase in food insecurity, Malin has been progressively scaled up in mainland France (except in the “Nord” district) since May 2021 through close collaboration with the National family allowances office (Caisse nationale des allocations familiales [CNAF])while following the principles of proportionate universalism [74]. Engagement of >200 stakeholders and enrolment of >260,000 young children and their families confirmed the program’s feasibility and acceptability. A thorough description of the Malin program principles and components is provided in Additional file 1.

With a bottom-up approach, the overarching aim of the Pregnancy and early childhood nutrition trial (ECAIL) led by the French national institute for health and medical research (Inserm) is to assess the Malin program’s effectiveness, when implemented via home visiting over pregnancy and the first 2 years of life. The general hypotheses tested with ECAIL is that a multi-behavioral and multilevel intervention simultaneously combining nutritional support and facilitation of access to appropriate foods and products has an effect on the rate and duration of breastfeeding, complementary feeding practices, dietary intake and lifestyle patterns in children at age 1 and 2 years and the risk of obesity in 2-year-old children from socially disadvantaged families. The protocol of the ECAIL trial is described hereafter according to the SPIRIT 2013 statement [75].

Methods and design

Overall study design

ECAIL is an on-going, single-blind, two-arm, parallel-group randomized controlled trial (RCT) that began in 2017 in Northern France, with balanced (1:1) randomization at the individual level (Table 1). Women are included during the third trimester of pregnancy, and families are followed up at home by trained dietitians until the child reaches age 2 years (Fig. 1). The intervention group receives the two complementary components of the Malin program, and the control group receives usual care. Participants remain blinded to their intervention status.

Table 1.

Detailed time schedule of enrolment, assessments, interventions, and visits for participants, highlighting those shared and unique to each trial arm

Time point	Location	Intervention	Control	Study period
V1: Screening of eligible women and provision of information (face-to-face when possible), from 4 to 7 months of gestation, during prenatal care.	Maternity ward	A		Enrolment
V2: Baseline visit conducted by the trained dietitian (in person) during the third trimester of pregnancy.	Home	B
		C
		D
		E		Allocation
		F	G
V3: Visit conducted by the trained dietitian (in person) after the child’s birth.	Maternity ward	F		Post allocation
		H
V4: Visit conducted by the trained dietitian (in person) when the child is 3 months old.	Home	I
		F
Child aged 6 months	By mail	J
V5: Visit conducted by the trained dietitian (in person) when the child is 6 months old.	Home	I
		F
Child aged 9 months	By mail	J
Child aged 12 months	By mail	J
V6: Visit conducted by the trained dietitian (in person) when the child is 12 months old.	Home	I
		K
		F
Child aged 16 months	By mail	J
V7: Visit conducted by the trained dietitian (in person) when the child is 18 months old.	Home	I
		F
Child aged 20 months	By mail	J
V8: Visit conducted by the trained dietitian (in person) when the child is 24 months old.	Home	I		Close-out
		K

A. Screening of pregnant women by health care providers using the eligibility questionnaire. Afterward, the ECAIL dietitians provide study information to the participants, face-to-face (when possible) or by phone

B. Re-verification of all inclusion and exclusion criteria with the participant (pregnant woman)

C. Signed informed consent obtained from the woman, with the father-to-be given the right to object to the child’s participation

D. Administration of the baseline V2 questionnaire (face-to-face with the participant). Completion of the implementation section of the questionnaire by the dietitian after the visit

E. Randomization

F. Provision of personalized nutritional support through the Malin program to the participant and her partner (if present), along with the option to subscribe to fresh fruit and vegetable baskets and access triannual online sales for cooking equipment and kitchen utensils at discounted prices

G. Provision and brief presentation of the two national nutrition and health program (PNNS) guides: “Nutrition for Pregnant and Breastfeeding Women” and “Guide from Birth to Age 3” (updated in 2021 as “Step by Step your Child Eats Like a Grown-Up”)

H. Completion of both the V3 questionnaire by the dietitian after the visit (reporting implementation information only) and the Case Report File with information on the mother–child dyad extracted from the hospital medical records

I. Administration of the V4-V8 questionnaires (face-to-face with the participant) and anthropometric measurements of the child. Completion of the V4-V8 questionnaires by the dietitian after the visit (reporting implementation information)

J. Discount vouchers from the Malin program, aimed at improving access to healthy foods, are sent to families’ homes when the child reaches 6, 9, 12, 16, and 20 months old

K. Administration of an additional questionnaire to assess the participants’ appreciation of and use of the various components and tools of the Malin program (face-to-face with the participant)

Fig. 1.

SPIRIT flow diagram: Summary of the study schedule of enrolment, interventions, and assessments

ECAIL is also a “participatory action-research” project [73, 76] because it was co-created by Inserm in collaboration with the Malin NPA and its long-standing stakeholders—the French Red Cross, SFP, and AFPA— alongside investigators from Lille University Hospital. These partners bring expertise from complementary disciplines, including social and nutritional epidemiology, public health, sociology, interventional research, clinical research, dietetics, obstetrics, midwifery, and pediatrics, as well as the expertise of professionals in social and solidarity action.

The ECAIL trial was registered at ClinicalTrials.gov (NCT03003117) in December 2016, and the first participant enrolled on April 3, 2017.

General hypotheses and outcomes

Primary outcome

As compared with children in the control group, children in the intervention group at age 24 months are expected to exhibit a higher frequency of vegetable consumption. Consistent with other RCTs [77–79], vegetable consumption at post-intervention was chosen as the principal outcome because encouraging vegetable intake is more difficult than encouraging fruit intake and it is good indicator of the overall diet quality [51].

Secondary outcomes

As compared with children in the control group, children in the intervention group are expected to demonstrate the following [79]:

• Higher rates of breastfeeding initiation and longer breastfeeding duration;

• A more appropriate age for the introduction of complementary foods, specifically between age 4 and 6 months;

• An older age at the introduction of processed food not intended for infant nutrition;

• Higher frequency of vegetable consumption at 12 months;

• Improved overall diet quality at 12 and 24 months;

• Healthier lifestyle patterns at 12 and 24 months;

• Slower velocity of weight gain in the first 6 months of life;

• More balanced growth patterns for weight and height over the first 2 years;

• Lower prevalence of OW/OB at 24 months.

Setting

The ECAIL RCT has been implemented in Lille and Valenciennes conurbations (in the Nord district) since 2017, where the poverty rate is among the highest in mainland France [80, 81]. Healthcare providers (HCPs; e.g. midwives, nurses, and obstetricians) identify eligible pregnant women during their prenatal care at the two main public hospital maternity wards of these two towns, namely: Jeanne de Flandre maternity at Lille University Hospital and Monaco maternity at Valenciennes Hospital (Table 1; Fig. 1). Trained dietitians then inform the women about the study and recruit them during their third trimester of pregnancy. The relatively slow pace of inclusions in ECAIL is partly due to underserved populations being hard to survey [35, 36] and the challenge of keeping HCPs engaged in screening despite their heavy workload, challenges that became even more pronounced after the onset of the COVID-19 pandemic. Also, the outbreak of the pandemic resulted in a 6-month suspension of screening and inclusion in the trial from March to September 2020. These factors prompted the decision to open a second investigation center in Valenciennes in 2021 to enhance the study’s inclusion potential.

As illustrated in Table 1; Fig. 1, the inclusion and follow-up of participants from both the control and intervention arms take essentially the form of home visiting starting at the third trimester of pregnancy (baseline, V2 visit), then with five other visits through to the child’s second birthday (i.e., at 3, 6, 12, 18, and 24 months; V4-V8 visits); if requested by participants, they are given the possibility to choose another place than their home to meet with the dietitian. An additional visit (V3) occurs at the maternity ward just after the delivery. Whenever possible, the same dietitian handles all interactions with a given family from the screening phase to the end of follow-up to build and maintain a trusting relationship.

Participants

Eligible participants are pregnant women attending the prenatal consultations at the maternity wards of Lille and Valenciennes hospitals who complete the eligibility questionnaire covering all inclusion and exclusion criteria (Table 2). Like the Malin program, we used a wide variety of criteria to characterize social disadvantage in the questionnaire.

Table 2.

Inclusion and exclusion criteria of the ECAIL study, which the eligibility questionnaire screened for. The ECAIL trial

Inclusion criteria

Exclusion criteria

• Pregnant woman at 3rd trimester of pregnancy, with health insurance coverage (national health insurance fund or state medical assistance); • Age ≥ 18 years; • Lives in Lille, Valenciennes, or the surrounding areas*; • Willing and able to sign informed consent; • Who, in the Investigator’s judgment, is able to comply with the protocol and study procedures (including speaking French); • Who currently faces social disadvantages, defined by the presence of ≥ 1 of the following social vulnerabilities: - receives social or medical benefits; - the household’s main source of income is not employment (but rather social benefits, friends/family, or none); - perceived financial hardship, i.e., experiences real financial hardship in meeting her needs, such as food, medication, rent, and bills; - precarious housing (i.e., sheltered in a long-term rehabilitation center or in a social hotel); or lack of any of the following facilities: shower or bath, toilet, running water or heating system; - social isolation (i.e., planning to look after the baby on her own after birth, without a partner or other adult to help)

• Pregnant woman under guardianship; • Participation in any other interventional clinical trial that is not compatible with the participation into the ECAIL study; • No permanent mailing addresses; housed in emergency accommodation centers, portable dwellings or homeless; • Unstable housing (has moved several times in the past 12 months or planning to move outside the study area before the child’s first birthday); • Serious illness during pregnancy that could interfere with optimal participation in the ECAIL study; • Hard drug use; • Multiple pregnancy (≥ 3 fetuses).

A prescreening phase was implemented by the ECAIL dietitians to avoid the distribution of the eligibility questionnaire to pregnant women living outside the study area; aged < 18 years; before 4 and after 7 months of gestation; or enrolled in another trial not compatible with participation into the ECAIL study

* Surrounding areas in Lille include Lomme, Loos, Faches-Thumesnil, Ronchin, Hellemmes, Haubourdin, La Madeleine, Lambersart, Mons-en-Baroeul, Saint-André-Lez-Lille, Wattignies, Villeneuve-d’Ascq, Marcq-en-Barœul, and Wasquehal. In Valenciennes, the area is within a 30-minute drive from the maternity hospital

Sample size calculation

We based the calculation of the sample size on the ability to detect a difference between the intervention and control groups for the principal outcome measure. The sample size was calculated from the validated food frequency questionnaire (FFQ) data used in the EDEN mother–child cohort at the 2-year follow-up [51, 82]. The same tool is being used to collect the frequency of daily vegetable consumption (as well as other food groups) in the ECAIL study in children at age 12 and 24 months.

Because the intervention is individually adapted to each participant (personalized support at the mother’s home, baskets of fruit and vegetables and on-line sales for kitchen utensils in the family name, and discount vouchers in the child’s name), we considered simple randomization to be appropriate. To calculate the sample size with a power of 80% and a P value of 5%, we assumed a minimum difference of 20% in the frequency of daily vegetable consumption by children at age 24 months between the intervention and control groups. Data from the EDEN study indicated that 2-year-olds ate vegetables (excluding potatoes) an average of 1.10 times a day (SD 0.70). We accordingly expected a difference of 0.22 times daily (or 1.54 times weekly), which we deemed both reasonable and meaningful. For context, in the EDEN study, the mean frequency of vegetable consumption among 2-year-olds differed by 0.47 times daily between children born to mothers with an educational level below high school and those whose mothers had completed more than 2 years of university. The total number of participants necessary for the analysis at the intervention conclusion was set at 320 (160 in each arm). According to previous intervention research involving socially disadvantaged families [35, 53, 83], the estimated sample size was then adjusted to apply a conservative estimate of attrition (60%), which yielded a total sample size of 800 pregnant women at inclusion (400 in each group).

Implementation team at the two study sites and data monitoring

The implementation of the trial in the field—encompassing the pre-screening phase, study information, recruitment, follow-up, intervention delivery and data collection—relies on dietitians. These dietitians are employed by the Clinical investigation center (CIC-1403) at Lille University Hospital and undergo an 8-day training program provided by a multidisciplinary, multi-partner team. As of this article’s publication, 11 dietitians have been hired and trained since the study launch, with 3 to 5 engaged at a given time at the two study sites, depending on the workload. Their fieldwork is supervised by a senior dietitian from the CIC-1403. In her additional role as a clinical research assistant, the senior dietitian also monitors the overall quality of the data as it is collected. A clinical research engineer was involved in the co-creation and testing of the screening process with HCPs at the Lille University Hospital. He has also contributed to adapting the ECAIL trial protocol to comply with both CIC-1403 and Lille University Hospital standards and has ensured that communication flows smoothly between all stakeholders at Lille and Valenciennes hospitals. At Valenciennes Hospital, another clinical research assistant monitors the data collection and interacts with various stakeholders.

Screening, eligibility, information about the study (V1); consent and recruitment (V2)

At the two maternity wards, all pregnant women from the catchment areas receive the eligibility questionnaire (Additional file 2) from the reception staff and complete it while waiting for their appointment. During their appointment, HCPs use this questionnaire to assess women’s eligibility for the ECAIL trial (V1, Table 1; Fig. 1). When a woman is deemed eligible, the HCP is expected to briefly introduce the study using the provided script, then contact a dietitian from the ECAIL implementation team. The dietitian subsequently informs the woman and her partner (if present) about the study in person, immediately following the HCP appointment. Whenever possible, this information is provided in a dedicated room to ensure greater discretion and privacy; otherwise, it takes place in the maternity outpatient waiting room. The dietitian also gives a 5-page detailed information letter about the study, in compliance with General Data Protection Regulation guidelines. The following week, the same dietitian calls the eligible woman to answer any additional questions and ask whether she is willing to participate in the study. If she agrees, an appointment is scheduled at her home during the third trimester of her pregnancy for the baseline visit. If the woman declines participation, the dietitian records her reasons verbatim in the eligibility questionnaire. If the HCP fails to contact an ECAIL dietitian during the appointment, the questionnaire is collected and forwarded to a dietitian, who then calls the pregnant woman about the study. In these cases, the information letter is sent by mail to the woman’s home. Parents are given 1 week to consider participation, followed by an additional phone call to discuss willingness to participate, after which the same steps outlined above are followed.

During the V2 baseline home visit, the dietitian first re-verifies the inclusion and exclusion criteria to ensure that no changes have occurred in the pregnant woman’s situation since the V1 screening and information phase and also addresses any remaining questions about the study. Second, the mother-to-be provides signed written informed consent for both her own participation and that of her child-to-be on a dedicated form (with the father-to-be being able to object to the child’s participation). Third, the dietitian conducts the baseline face-to-face questionnaire with the woman; only at that point is the woman formally considered recruited into the ECAIL trial (Table 1; Fig. 1).

Randomization

The random 1:1 allocation to the intervention or control group was concealed by sequentially numbered, sealed opaque envelopes containing the group allocation, which was determined by a computer-generated random number. The block sizes will not be disclosed, to ensure concealment. Randomization was stratified by hospital (Lille and Valenciennes hospitals). A research assistant from Inserm, who had no direct contact with participating mothers or the implementation team, was responsible for generating the random numbers and preparing the envelopes.

To ensure baseline allocation concealment, dietitians discreetly open the sealed envelopes immediately after women are officially recruited (with the three steps of the baseline home visit described above actually implemented) to discover the participants’ assigned group, without sharing this information with them because they remain blinded to the group allocation. Only at that point of the baseline home visit do the dietitians implement the Malin program (intervention group) or usual care (control group), accordingly (Table 1).

Intervention (intervention arm)

The intervention implemented in the intervention arm takes place from the third trimester of pregnancy until the child’s second birthday (Table 1; Fig. 1, V2-V8). It is rooted in behavioral change theories [84] and operates at multiple levels of the socio-ecological model [15, 85]—both individual and structural—delivering the two complementary components of the Malin program (see Additional file 1 for further details on the Malin program’s components, tools, and the training of delivery agents). Notably, in accordance with the definition provided by the PNNS, the term “nutrition” hereafter encompasses all EBRBs, including diet, physical activity, sedentary behavior, and sleep.

Individual component

Grounded within the framework of the social cognitive theory, the nutritional support component seeks to build knowledge, skills, self-efficacy and social support regarding feeding practices and EBRBs [86]. The dietitians have been trained in the Malin program and are therefore aware of the context of social disadvantage and the non-judgmental posture/attitudes most appropriate to empower parents in this context. Pregnancy and the first years of motherhood/parenthood are periods during which caregivers seek and are most receptive to advice about raising their child as well as possible [49, 87]. Conversations are structured around the key messages of infant nutrition from the PNNS, stressed by the Malin program, especially the promotion of breastfeeding and information about how and when to introduce complementary foods, while accounting for reflexivity and responsiveness to infant hunger and fullness cues. In 2019, additional topics from the PNNS were added to the Malin program, such as the promotion of movement/physical activity as early as possible in the baby’s life (e.g., tummy time), the prevention of sedentary behaviors (e.g., screen use and time spent restrained) and the setting of adequate sleep habits. These different messages are specified progressively over the home visits (Table 3), anticipating the next developmental milestones of the child (anticipatory guidance) [15, 88]. They serve as anchor points to focus discussions on these topics or to extend discussions to related themes. In fact, increasing evidence suggests that given that EBRBs often co-occur and co-vary (change in one may facilitate change in the other), multi-behavioral interventions have greater impacts on health outcomes than single-behavioral interventions; furthermore, depending on the specific context of living, changing one given lifestyle behavior may be easier than changing another [89–91]. Practical tips for smoking prevention and healthier EBRBs are also provided to parents themselves because they are the main role models for their child.

Table 3.

Key themes and tools from the Malin program used by dietitians to facilitate discussions on feeding practices and energy balanced-related behaviors in the intervention arm. The ECAIL trial

Baseline (V2) and follow-up visits (V3-V8)	V2	V3	V4	V5	V6	V7	V8
Age of the child (V3-V8)	Prenatal	Upon birth	3 months	6 months	12 months	18 months	24 months
Key themes*
1. It is important for the mother to maintain a balanced diet during pregnancy and breastfeeding.	X	X
2. It is important for the whole family to eat a balanced diet, prioritizing home-cooked meals made from raw ingredients whenever possible (meaning not pre-cooked, except in the case of specific foods for babies).	X		X	X	X	X
3. Breast milk alone meets all your baby’s nutritional needs up to 6 months of age.	X	X	X
4. Cow’s milk (instead of formula): - not before 12 months - ideally not before 36 months - if consumed, use whole milk, not semi-skimmed.	X	X	X	X	X	X
5. Vegetable juices (e.g., almond, chestnut, rice, soy, or oat milk) are not suitable for infants and should be avoided.	X	X	X	X	X
6. If the mother is not breast-feeding (or is using mixed breastfeeding): - infant formulas are appropriate for children under 6 months of age. - follow-on formulas are appropriate for children aged 6 to 12 months. - growing-up milks are appropriate for children aged 12 to 36 months.	X	X	X	X	X	X
7. After 6 months, breast milk remains the best option for the child and should be supplemented with a variety of complementary foods.			X	X	X	X
8. Complementary feeding: - not before 4 months old. - not beyond 6 months of age. - once complementary feeding has started, breastfeeding can continue if desired and feasible.		X	X	X
9. Never force a child to eat; instead, respect their refusal. Mealtime is a shared moment, filled with enjoyment and discovery.			X	X	X	X
10. During complementary feeding, offer your child a variety of flavors, including different fruit and vegetables, every day.			X	X	X	X
11. A food that is initially refused can be offered again, up to 10 times.			X	X	X	X
12. Limit added salt and sugar as much as possible.			X	X	X	X
13. Add a small amount of vegetable oil or raw butter to home-cooked dishes after cooking.				X	X	X
14. Water is the only beverage that should be given.			X	X	X	X
15. Sugar-sweetened beverages, including fruit juices (even in baby bottles), should be avoided until the child is 3 years old.	X	X	X	X	X	X
16. Ready-made meals for adults, sweets, pastries, desserts, processed meats, pizzas, hamburgers, chips, and other fast foods should be avoided before 12 months and given only occasionally after that.				X	X	X
17. If possible, eat at regular times, in a peaceful environment, with family, without screens, and avoid snacking between meals.	X		X	X	X	X
18. Encourage a regular daily routine for children, with rest periods and sleep times suited to their age.		X	X	X	X	X
19. Avoid exposing children to screens (television, mobile phones, tablets, etc.) as much as possible, both during mealtimes and at other times.			X	X	X	X
20. Encourage the child’s free movement and exploration, both at home and outside, from birth.		X	X	X	X	X
Handouts and booklets provided by the study dietitians during home visits
One-page handouts to present the Malin program and two of its budgetary components: incentives for baskets of fresh fruits and vegetables and for cooking equipment and kitchen utensils.	X
Four booklets: “I’m Pregnant or Breastfeeding: What Can I Eat for My Health and My Baby’s?”, “My Booklet for Learning More About Breastfeeding”, “My Booklet on Bottle-Feeding” (including tips for feeding breast milk through bottle), and “The Right Foods at the Right Times”.	X
Fact sheet detailing parent support activities available in the area.	X
Booklet entitled: “My Child’s Diet: From Birth to Age 3”.			X
One-page handout introducing the vouchers sent to families’ homes to support access to healthy foods once the child reaches 6 months of age.			X
Digitally-based toolbox: Malin program website, monthly newsletter emailed to the participants, social media updates on Facebook and Instagram.	Accessible throughout the follow-up period

*These messages serve as anchor points to focus discussions on these themes, or to extend discussions to related themes. Information, advice and practical tips for smoking prevention and healthier energy balance-related behaviors are also provided to parents themselves because they are the main role models for their child. A major principle of the Malin program is the setting of individualized, gradual and achievable goals: the aim of this support is to be positive, with the level of assistance tailored to the questions and needs expressed by parents

Social cognitive theory [86] posits that discussions about barriers and facilitators enable caregivers to reinforce their feelings of self-efficacy in appropriating and implementing these nutritional recommendations. A major principle of the Malin program is the setting of individualized, gradual and achievable goals. The pleasure and involvement of the five senses around the act of eating are another kind of inducement to behavior change, which the dietitians were trained for and apply in their support. Consistent with the principles of reciprocity and proportionate universalism [74], this support aims to be positive, with the level of assistance tailored to the questions and needs expressed by caregivers. It is not intended to be prescriptive, normative, blaming, or stigmatizing in any way. For example, the words “obesity”, “poverty” and “must do” are intentionally avoided in the discussions facilitated by dietitians.

The toolbox includes a website, a monthly newsletter sent to parents via email, and regular social media posts on Facebook and Instagram. These resources enable parents to access essential knowledge, advice, and tips, such as more than 400 healthy, simple and multicultural recipes; homemade menus tailored to children’s needs at different ages; lists of in-season fruits and vegetables; tips for accessing healthy foods at reduced prices; advice on breastfeeding; simple ways to move and play with the child, both indoor and outdoor; guidance on avoiding screen exposure; and recommendations for sleep routines. Five booklets are also provided to parents during home visits (which can also be downloaded from the website) to facilitate discussions between dietitians and families on key topics. All these communication tools are highly illustrated, with minimal text, and are designed to be practical. Additionally, dietitians provide parents with a fact sheet detailing parent support activities available in the area (e.g., lactation counsellors, cooking workshops in social centers, parent groups, and social workers for broader needs such as housing and administrative procedures) to encourage the use of local community services, social support and peer interactions.

Structural component: availability, accessibility and affordability

At a more upstream level, the second component employs a non-stigmatizing approach to help families access products and services at a reduced price (from 20% to 80%) under the same conditions as the general population. Parents are given the opportunity to subscribe to community-supported agriculture baskets of fresh, organic fruits and vegetables at a significantly reduced cost from the third trimester of pregnancy to the child’s second birthday. A family deciding to subscribe can choose to pick up a basket (different sizes available) each week or every other week at a particular distribution point (e.g., social centers and stores or warehouses serving as delivery sites). In the Valenciennes conurbation, families can select fruit baskets, vegetable baskets, or a combination of both. In Lille, each basket is accompanied by a leaflet that provides descriptions of the included fruits and vegetables, along with recipe ideas. To promote homemade meals, parents also have access to triannual online sales that allow them to purchase cooking equipment and kitchen utensils at reduced prices (from pots and saucepans to food processors). Finally, nominative discount vouchers to access healthy foods recommended by the PNNS are sent to the families’ homes when the child is 6, 9, 12, 16, and 20 months old, and vouchers can be used in all supermarkets. Foods are relevant for the baby (e.g., little jars and plates of cooked fruit or vegetables, baby cereals, follow-on formula for infants aged 6 to 12 months, and young child formula for toddlers aged 12 to 24 months) but also for all other family members (e.g., frozen and canned vegetables, poultry, unsweetened plain yogurt, and stewed fruit with no added sugar). With these incentives, the Malin program aims to make raw products more accessible to encourage home-made cooking for the whole family as well as foods that require little preparation. In fact, some of the families lack access to a kitchen or are poorly equipped to cook properly (they may only have a microwave, with no burner or oven). Vouchers start to be handed out when the child is 6 months old to ensure that they do not interfere with the promotion of breastfeeding, a core component of the Malin program.

Usual care (control arm)

The control group receives standard care and advice. At the baseline visit, parents are provided two PNNS guides from the dietitians, who present them briefly: Nutrition for Pregnant and Breastfeeding Women and Guide from Birth to Age 3 (updated in 2021 as Step by Step your Child Eats Like a Grown-Up) (72). They also provide the list of maternal and child protection centers in the area. During the seven visits (baseline, V2; and follow-up, V3-V8), dietitians are instructed not to initiate any discussion regarding feeding and EBRBs. If the mother spontaneously asks questions about her or her child’s nutrition, dietitians are trained to respond as briefly as possible, referring only to the relevant pages of the PNNS guides provided at the baseline visit. Nonetheless, this information is not supported by the conceptual framework of social cognitive theory that is at the heart of the Malin Program. Nor are there any incentives or discount vouchers to facilitate access to healthy food. However, measurements (face-to-face questionnaires and anthropometric measurements) are implemented as in the intervention group (Table 1; Fig. 1).

Training of dietitians

The newly hired dietitians undergo an 8-day training program for the ECAIL study. Co-designed and implemented by various stakeholders, including members from Inserm, the Malin NPA, and the CIC-1403 of Lille University Hospital, the program integrates expertise from multiple disciplines. The training includes theoretical knowledge, interactive group sessions and situation simulation techniques relevant to the various stages of the trial, from V1 to V8. Importantly, previously trained dietitians who are already implementing the study in the field participate in training the newly hired dietitians, sharing their knowledge and experience in the study. As part of their induction process, newly hired dietitians also accompany experienced dietitians in the field for non-participatory observations. The various modules are outlined in Table 4. Additionally, dietitians meet with their supervisor weekly to discuss any issues related to implementation.

Table 4.

Training modules that dietitians undergo before contribution to the ECAIL trial

	Title	Who and What
1	Introduction	Members from Inserm and the Malin non-profit association (NPA) provide an overview of the epidemiological, public health and social context of the Malin program and the ECAIL study, including their ecosystem, co-creation principles, and the roles of various stakeholders. This module is open to healthcare providers from both Lille and Valenciennes hospitals.
2	The many faces of social disadvantage: understanding the daily realities of families facing social adversities and how to avoid stereotypes	This module is opened to all stakeholders, including Inserm researchers interested in the topic and new staff at the Malin NPA, with the goal of fostering greater interaction among professionals from diverse backgrounds. Facilitated by a member of the Malin NPA, the session begins with a discussion aimed at deconstructing stereotypes about poverty (Additional file 1).
3	The Malin program delivery in the intervention group	A dietitian from the Malin NPA presents the latest guidelines from the National nutrition and health program (Programme national nutrition santé [PNNS]) and discusses strategies for effectively communicating them to families. The focus is on delivering tips and advice in a positive, non-judgmental, non-stigmatizing, and non-prescriptive manner while also tailoring the guidance to fit the realities of their daily lives. This module includes a half-day workshop titled “Five Senses Around Eating”, along with picture-based activities designed to deconstruct stereotypes about diet and explore the various dimensions of eating, including social, cultural, and emotional aspects. Notably, this 2-day module has been used since 2012 to train social, health, and early childhood professionals involved in the four districts where the Malin program has been implemented. More details are provided in Additional file 1.
4	Implementation of the ECAIL study: from screening (V1) and inclusion (V2) to follow-up (V3-V8)	This module is facilitated by members from Inserm, the Malin NPA and the Clinical investigation center (CIC-1403) at Lille University Hospital. It outlines the roles, responsibilities, and tasks of dietitians, including essential talking points to inform and motivate eligible women to participate in the study. The module also covers the specific support expected for both intervention and control families as well as the dietitians’ role in data collection. Tools for fieldwork are contextualized and applied through scenario-based simulation sessions. In particular, the various stages of the V2-V8 visits are meticulously detailed in two comprehensive guides, called “Visit notebooks”. One guide, with 152 pages, is designated for the intervention group and the other, with 27 pages, is tailored for the control group. These guides were initially co-designed by Inserm and the Malin NPA, then refined by dietitians already trained and experienced in the ECAIL trial to better align with the realities of the field. Additionally, a 14-page guide provides detailed instructions for conducting the various face-to-face questionnaires. Dietitians are expected to strictly adhere to these guidelines in the field to ensure standardization and maintain fidelity to the protocol.
5	Psychological distress	After outlining common psychosocial vulnerabilities in socially disadvantaged families, this module aims to equip dietitians with the skills to identify warning signs and access available support resources for managing potential critical situations they may encounter during home visits. This session is conducted by psychologists from the CIC-1403 at Lille University Hospital, who are also available to assist in any concerning situations encountered by dietitians in the field, providing support to both the dietitians and the mothers as needed.
6	Basics of clinical research	This module covers the clinical research environment, as well as the regulatory requirements and quality standards that dietitians must follow. The CIC-1403 is committed to adhering to the French Data Protection Authority (CNIL, Commission nationale de l’informatique et des libertés) methodology No. 1 (MR06 001). The quality assurance procedures ensure that both participants and data are managed under the safest conditions, fully complying with medical, regulatory, and good clinical practice requirements.
7	Anthropometric measurements	Following the theoretical components, dietitians participate in hands-on training with pediatricians to calibrate measurement scales and accurately measure infants’ length, weight, and head circumference. This training involves 10 babies aged 3 to 24 months, with each measurement being repeated twice to ensure accuracy.

Assessments

Quantitative component: data collection, measurements, auditing and data management

Most of the study quantitative data are collected face-to-face by dietitians using structured questionnaires, at the different times of follow-up (V2-V8, Table 1; Fig. 1). These questionnaires were previously tested (and adapted) with target populations (n = 30) in emergency shelters as well as in a French Red Cross Mother-Baby center within the Greater Paris area. We use face-to-face questionnaires instead of self-reported ones for several reasons. First, this approach helps to prevent additional stress for mothers who are already burdened with completing numerous forms to access various types of assistance and manage their administrative status. Second, it maximizes the completion rate of the questionnaires. Third, it allows us to include mothers who understand French but may not be able to read or write, thereby reducing selection bias. So far > 98% questionnaires have been completed in full during the home visits. Of note, while responding to the questionnaires, mothers have so far seemed more inclined to say what they really think and do because dietitians have built a trustful, non-blaming and respectful relationship with them. This reassuring observation has been consistently reported by all dietitians in the field and should limit social desirability bias. Also, these questionnaires are administered to the mother at the beginning of each home visit, before any support is provided, to further minimize response bias.

Primary and secondary outcomes

Feeding practices (e.g., the type of milk used to feed the child, consumption of processed foods, age of introduction of complementary foods, and frequency of consumption of different food groups), screen time, outdoor play time and sleep duration are collected in the face-to-face questionnaires (Table 5). These behaviors change rapidly during the successive developmental phases of early life, which is why data are collected at short intervals—three times in the first year of follow-up and twice in the second year—to minimize recall bias. The children’s anthropometric measurements at birth (length, weight, and head circumference) are collected from the hospital medical file. The children’s weight at discharge from the maternity ward and weight and length at age 2 months are collected from the child’s health booklet. Dietitians also measure the child’s weight to the nearest 10 g (without clothing and diaper, using a calibrated electronic baby scale [baby scale cartoon design, LBS Medical, Croissy-sur Seine, France]) and recumbent length to the nearest 1 mm using an infantometer (SECA-417, SECA Gmbh, Hamburg, Germany) at each home visit (V4-V8). Head circumference is measured at V6 and V8 only, with a tape measure, to the nearest mm. Each measurement is taken twice, and if the difference between the two measurements exceeds a given threshold, it is repeated a third time. If the mother objects to these measurements, the dietitians record at every visit the most recent measurements in the health booklet. All anthropometric data collected during home visits are recorded in a dedicated section of the face-to-face questionnaires.

Table 5.

Data collected in the V2-V8 face-to-face questionnaires

Baseline (V2) and follow-up visits (V3-V8)	V2	V3	V4	V5	V6	V7	V8
Age of the child (V3-V8)	Prenatal	Upon birth	3 months	6 months	12 months	18 months	24 months
Socio-demographic factors
Household composition	X						X
Matrimonial status of the pregnant woman	X
Socio-cultural factors
Country of birth (woman and partner*)	X*
Time since arrival in France (woman)	X
Languages spoken at home	X
French literacy: reading and writing (woman)	X
Socio-economic factors
Level of education and diploma achieved (woman and partner*)	X*
Occupational category (woman)	X
Employment (woman and partner*)	X*		X	X	X	X	X
Level of income, main source of income	X
Receipt of medical and social allowances	X
Health coverage	X						X
Living conditions
Accommodation type	X						X
Appliances, electronics, storage, and other equipment	X
Access to a kitchen	X
Food insecurity (Household Food Security Survey Module [HFSSM])	X						X
Access to food assistance					X		X
Location for purchasing fresh produce							X
Childcare arrangements			X	X	X	X	X
Psychosocial factors
Feelings about pregnancy timing	X
Family and social support	X						X
Perceived standard of living	X						X
Economic barriers to accessing essential goods (food, clothing, diapers, toys) and services (childcare, healthcare, and medications) for the child					X		X
Economic barriers to accessing household goods (e.g., food, clothing, gasoline, transportation, and leisure activities), healthcare, medications, paying bills, and managing debt					X		X
Person responsible for preparing the child’s meal					X		X
Person in charge of grocery shopping					X		X
Pre-shopping food checklist							X
Nutrition and infant feeding knowledge	X			X		X
Intentions regarding breastfeeding after birth	X
Had the pregnant woman been breastfed as an infant by her mother?			X
Did she breastfeed her previous children?			X
Prenatal care details: location, frequency of consultations, hospitalizations, and social worker support			X
Number of childbirth preparation sessions			X
Breastfeeding support: timing and provider			X
Provision of information on (complementary) feeding: timing and source				X			X
Maternal self-efficacy regarding child feeding						X	X
Maternal lifestyle behaviors
Meal preparation and duration at home	X					X
Meal consumption frequency	X					X
Dietary intake (33-item FFQ)	X					X
Frequency of leisure physical activity	X					X
Screen time (weekdays and weekends)	X					X
Smoking habits before and during pregnancy			X
Current smoking habits						X
Exposure to passive smoking						X
Feeding practices
Type of milk fed to the infant at the maternity hospital and reasons for the decision			X
Type of milk provided to the child from maternity hospital discharge to the follow-up visit			X	X	X	X	X
Child’s age at cessation of breastfeeding and reasons for the decision (if applicable)			X	X	X	X	X
How the child is fed: on demand or at regular times			X	X
Frequency of inappropriate bottle-feeding practices (e.g., sugary drinks, bottle in bed)			X	X	X		X
Complementary feeding: initiation, textures used and reasons for introducing foods into the child’s diet			X	X	X	X
Beverages, foods, and ingredients introduced: frequency (not yet to daily) from a 32-item questionnaire, including age at repeated introduction			X	X
Age of introduction to processed foods not intended for infant nutrition			X	X	X		X
Other parental feeding practices (e.g., number of attempts before determining that the child dislikes a given food, adding ingredients to enhance food acceptability)			X	X	X		X
Lifestyle behaviors of the child
How often the screen is on in the child’s sleeping room							X
Sleep time (weekdays and weekends)							X
Supervised tummy time			X	X
Outdoor play time (weekdays and weekends)					X	X	X
Nature and frequency of play activities with parents/adults				X	X	X	X
Frequency of use of various screen types					X	X	X
Screen time (weekdays and weekends)				X	X	X	X
Screen use during mealtime				X	X	X	X
Meal context at home: participants, duration, types of beverages, and other details					X		X
Meal consumption frequency					X		X
Dietary intake (31-item FFQ)					X		X
Acceptance of new foods					X
Maternal health
Depression symptoms (Edinburgh Postnatal Depression Scale [EPDS])	X		X
Height, pre-pregnancy weight, and gestational weight gain			X
Serious health issues and hospital stays over 24 h since the last visit			X	X
Self-perception of sleep quality						X
Self-perception of health						X
Occurrence of another pregnancy since the start of the study							X
Child’s Health
Maternal perception of child’s health			X	X	X	X	X
Maternal perception of child’s weight			X	X	X	X
Serious health issues and hospital stays over 24 h since the last visit			X	X	X	X	X
Anthropometric measurements: weight, length and head circumference§			X	X	X§	X	X§
Emotional development (10 items from the Brief Infant-Toddler Social and Emotional Assessment [BITSEA])							X

*The partner’s information is available only at the V2 baseline visit

^§The dieticians measured the child’s head circumference only at the V6 and V7 follow-up visits. FFQ, food frequency questionnaire

The principal outcome is the mean daily frequency of vegetable consumption at age 24 months (including raw and cooked vegetables, but excluding potatoes); this will be assessed using the relevant items from the 31-item FFQ.

Secondary outcomes are the following:

• Breastfeeding initiation rate and duration, considering both predominant and any breastfeeding [92];

• Age at which complementary foods were introduced, defined as the first instance of any food or drink (other than water, breast milk or formula) being given at least twice;

• Age at introduction of processed products not intended for an infant diet;

• Mean daily frequency of vegetable consumption at age 12 months;

• Overall quality of the child’s diet at age 12 and 24 months, determined using global approaches to characterizing diet (dietary patterns resulting from principal component analysis), based on the 31-item FFQs administered at 12 and 24 months [51];

• Similarly, using integrative approaches developed with principal component analysis to evaluate multi-component programs [17, 90, 91], lifestyle patterns at 12 and 24 months will include dietary intake for key (un)healthy food and beverage groups and movement behaviors (including outdoor play time and screen time);

• BMI will be calculated from the child’s weight and length measured at various points during follow-up (weight[kg]/length²[m²]);

• Child overweight (including obesity, OW/OB) will be defined by BMI at 24 months according to the International Obesity Task Force criteria [93];

• Growth curves for weight and length in the first 2 years of life will be modelled using the Jenss-Bayley model [94].

Other variables of interest

Information about the participating woman, including her age, parity, the course of her current pregnancy (e.g., duration, hypertension, gestational diabetes, anemia) and birth outcomes (e.g., pre-eclampsia, hemorrhage, mode of delivery), is extracted from the hospital medical records by the implementation team and documented in the Case Report File. All other data—covering socio-demographic factors, socio-cultural background, living conditions, psychosocial factors, as well as maternal lifestyle behaviors and health—are collected by the V2-V4 face-to-face questionnaires administered by the dietitians (Table 4).

Adverse events and risks associated with the research

Participation in this study is not expected to involve any specific risks, and none have been observed thus far. However, dietitians are responsible for reporting any onset of serious health problems or hospitalizations of the mother or child that are identified during the home visit questionnaires (Table 4). If a serious adverse event is determined to be directly related to the research, the procedure established by the CIC-1403 will be followed to manage it.

Implementation

In addition to describing eligibility, participation and follow-up indicators (e.g., attrition, delayed, skipped or failed home visits), we will outline the actual implementation of the trial to contribute to the process evaluation of this complex intervention [76, 95, 96]. After each visit, the dietitians document its details in a dedicated section of the V2-V8 questionnaires (Table 1), including information such as the visit’s location and duration, the duration of nutritional support (when applicable), the presence or absence of the partner, and any non-nutritional requests, such as those related to psychosocial issues. For the intervention group, dietitians also document the extent to which key nutrition messages were addressed during the visit, whether not at all, briefly mentioned, or thoroughly discussed. Additionally, the Malin NPA tracks the families’ actual use of the fruit and vegetable baskets (frequency, size, duration, content), on-line sales for cooking devices and kitchen utensils (frequency, type of product), and the various vouchers (number of vouchers used per type of product). This will allow us to precisely measure participants’ actual adherence to each component of the intervention, capturing the “dose” (i.e., the amount of intervention actually used/received). When children from the intervention group reach age 12 and 24 months, dietitians administer additional face-to-face questionnaires to gather more detailed information on parents’ appreciation of and use of the various components and tools of the Malin program as well as any obstacles they may have encountered (Table 1).

Auditing

Medical and field observations, consent forms, and data collected by study questionnaires and the Case Report File are stored in the participant’s study file by the implementation team at each stage of management, in compliance with the CIC-1403 good clinical practices and the study protocol. Any deviations from the protocol, along with their reasons, must be reported by the implementation team. A clinical research assistant from the Research and Health Innovative Department of Lille University Hospital will ensure compliance with the protocol, good clinical practices, and applicable legislative and regulatory requirements: an auditing will be conducted during the study period, following a monitoring plan outlined in the protocol, at both study sites.

Data management and confidentiality

All participants and dietitians are assigned unique numerical identifiers (ID codes) to be used throughout the study. The implementation team at the CIC-1403 of Lille University Hospital ensures that all the study data are pseudonymized before being transmitted to Inserm for analysis, with a copy stored in the locked cabinets at both Lille and Valenciennes hospitals. Importantly, the blind will be lifted only after the last participant has completed the 24-month follow-up.

Qualitative component

A qualitative component has been developed by two researchers specializing in social sciences and humanities: a senior researcher in the sociology of health and migration and a postdoctoral researcher in psychosocial sciences. Its during-trial design aims to contribute to the process evaluation of this complex intervention [76, 96–98]. Since 2022, in addition to non-participatory observations of the dietitians’ V4-V7 home visits, the post-doctoral researcher has been conducting in-depth semi-structured interviews with participants at their home, between the V7 and V8 visits, in the two study sites. The questions in the interview guides were structured around the three components of the COM-B behavior change theoretical framework: capabilities, opportunity, motivation [84]. Capabilities refer to an individual’s psychological and physical capacity to engage in the behavior, encompassing both knowledge and skills. Opportunity involves external factors that facilitate or prompt the behavior. Motivation includes the cognitive processes that drive behavior, such as habits, emotional responses, and analytical decision-making. This model acknowledges that behavior is influenced by multiple factors and that changes in behavior result from modifying at least one of these components, all of which are integral to the Malin program. The interview guides address behaviors that align with the study’s primary and secondary outcomes. Participants in the ECAIL study will be interviewed until theoretical saturation is reached, meaning no new themes or insights emerge from additional interviews [99]. From previous research, we anticipate that about 20–30 interviews by study site will allow us to reach this objective [100]. In May 2023, the post-doctoral researcher conducted an in-person group interview with the entire ECAIL implementation team, consisting of six dietitians, including the coordinator, at Lille University Hospital. The purpose was to explore the impact of the COVID-19 pandemic on trial implementation. To capture their experiences with the study, semi-structured interviews were conducted with seven dietitians who had chosen to leave the project. Each interview is conducted on a voluntary basis, requires the consent of the interviewee and is, with their consent, recorded. We additionally intend to conduct focus groups and semi-structured interviews with various stakeholders involved in the ECAIL project in Lille and Valenciennes as well as the Malin program in its real-world context across the four districts. This process will include interviews with HCPs (including dietitians), social workers, and early childhood professionals (e.g., daycare providers).

Methods used to limit attrition

Populations facing social adversities are known to be challenging to reach and follow up [35, 36]. One contributing factor is housing instability, which we attempted to mitigate with the eligibility questionnaire by excluding participants who were likely to move outside the study area or who had experienced multiple moves in the previous year. Despite these measures, some participants have moved outside the study area and subsequently dropped out. The final dropout rate, along with the reasons for dropout (as recorded in the Case Report File when known), will be assessed at the end of the study. To promote retention in the trial, we have implemented various strategies to prevent other causes of dropout, detailed in Table 6.

Table 6.

Methods used to limit attrition. The ECAIL trial

Setting	Participants are followed up through home visits to encourage outreach and can choose the day and time. They also have the option to meet at another location of their convenience, with financial reimbursement for transport available, if applicable.
Alternate contacts	At the baseline visit, dietitians request participants to provide one or more alternate contacts in case they encounter difficulties reaching the mother directly.
Personalized participation cards	Participants receive two participation cards: one for the mother at the V2 baseline visit and one for the child at the V3 visit. These cards indicate their participation in the study, provide the name and contact details of the referring health care providers at the investigation site, and include a list of maternal and child protection centers in the study area. The child participation card also features a section for reporting of weight and length measurements by the study dietitians.
Mode of questionnaire delivery	The face-to-face administration of the questionnaires (instead of self-reporting) minimizes participants’ burden because they are often overwhelmed with completing administrative forms. We deliberately decided to rely solely on dietitians to administer the questionnaires, despite their awareness of the participants’ intervention status, to avoid doubling the number of home visits and the associated burden on participants. Additionally, we believe that the trusting relationship established between the mother and the dietitian over the course of the follow-up helps mitigate response bias. Also, dietitians build relationships not only with the mothers/parents but also with the infants, which has proven beneficial for conducting follow-up visits, particularly for anthropometric measurements.
Acknowledgment of time dedicated to the study	Small gifts (valued at €5–10) are offered at the end of each visit as a token of appreciation for the family’s participation. For example, a scented stick is given at the baseline visit, and baby toys are provided during follow-up visits.
Study newsletter	A quarterly, 4-page newsletter is mailed to families to foster a sense of belonging to the ECAIL research community. The first page provides updates on the study progress, including participant inclusions and follow-up visits, and highlights any changes in the implementation team (e.g., the departure or arrival of dietitians). Pages 2 and 3 feature family-oriented content unrelated to energy balance-related behaviors, such as tips on oral hygiene and summer sun protection, and fun craft activities, coloring pages and artwork created by participants’ children. The fourth page highlights family-friendly activities in the study area and includes photos and contact details of the ECAIL dietitians.
Dietitians’ availability and flexibility	Dietitians tailor their communication approach to reach families, including the timing, frequency, and method (phone calls and/or texts), to align with the mothers’ preferences. During each follow-up visit, dietitians also remind mothers that they are available to answer any questions between visits if needed. These processes are documented for each participant in dedicated files by the implementation team, ensuring that another dietitian can effectively take over if necessary.
Procedure for lost to follow-up	The study dietitians are responsible for implementing local standard operating procedures to optimize participant retention, including at least three reminders at different times and on different days of the week, via telephone, text message and mail. The study dieticians also use the alternate contacts provided by the participant at baseline.

Statistical analyses

Description of the sample at baseline by treatment group

We will present characteristics of the baseline sample by treatment group using proportions for categorical variables and means (SD) and medians (interquartile range) for continuous variables.

Assessment of the effect of the intervention on primary and secondary outcomes

In accordance with CONSORT [101] and SPIRIT [75] guidelines, the analysis of the effect of the intervention on primary and secondary outcomes will follow an intention-to-treat approach, including all participants as randomized (regardless of protocol adherence), with multiple imputation methods used to address missing data [102–104]. The comparative analyses will employ regression models tailored to the nature of the outcome. For continuous or semicontinuous outcomes, we will use linear regression and report mean differences with 95% confidence intervals (CIs). For binary outcomes, we will use logistic regression and present odds ratios with 95% CIs. The primary analysis will be unadjusted for confounders and will include only the treatment variable and study site to account for randomization stratification by site. Covariance analysis will not be applicable because no baseline data on the outcomes of interest are available (the child was not yet born). Two-sided tests will be used, with a significance level set at α ≤ 0.05.

Additional adjusted analyses will account for baseline factors that may not be evenly distributed despite randomization and that are potentially related to child outcomes. These factors include mothers’ socio-economic position, migration status, pre-pregnancy BMI, parity, age, and child sex. A complete case analysis will also be conducted as a sensitivity analysis.

This effectiveness analysis will be conducted after the trial concludes, and the results will be published in a single article.

Process evaluation of the intervention implemented in the ECAIL trial

Following the framework provided by the UK Medical Research Council for the process evaluation of complex interventions [76, 96], we will use mixed data (i.e., quantitative and qualitative) to describe and respond to the following questions: ∎ Implementation: what was implemented and how? ∎ Mechanisms of impact: how did the delivered intervention produce change? ∎ Context: how did context affect implementation and outcomes? This process evaluation will be published in separate articles from the one reporting the intervention’s effects on primary and secondary outcomes.

Description of the trial implementation

Data collected in the eligibility questionnaire used during the V1 visit at the maternity ward provide valuable information concerning reach [105]: it is defined as the number/rates of eligible and participating pregnant women, the reasons for refusal to participate (when applicable), and characteristics of social disadvantage associated with participation. The follow-up of the families, such as attrition/retention (and associated social factors), and characteristics of the V2-V8 visits will be described (see section Assessments/Implementation for the various indicators). This description of the study/intervention implementation will help assess fidelity and adaptations relative to the protocol as well as the engagement and interest of the families in the support provided by the dietitians [76, 95, 96].

Hypotheses generation regarding mechanisms of impact and context

We will conduct moderation analyses to explore whether the effect of the intervention varies according to baseline factors, particularly socio-demographics (e.g., first-time motherhood, maternal age), socio-cultural factors (e.g., migration status), or socio-economics (e.g., level of social disadvantage) [53, 75, 101, 106, 107]. Additionally, mediation analyses will be used to examine the intermediate mechanisms driving the intervention’s effect, focusing on the levers it aims to strengthen: maternal knowledge of feeding practices, self-efficacy, maternal diet quality, and budget trade-offs, all assessed after baseline [107, 108].

Strengthening of the process evaluation with the qualitative component

We will explore the dimensions of implementation, mechanisms, and contexts from the perspectives of the families, dietitians, and other stakeholders involved in this action-research project [105]. This will involve assessing whether the study was carried out as originally planned (fidelity), identifying obstacles and facilitators encountered, and understanding any adaptations made. We will also examine the extent of the intervention delivered (dose), including how actively participants and stakeholders were involved, their level of interaction, receptiveness, and utilization of the materials or resources provided, as well as the barriers and facilitators to their use. We will explore their satisfaction with the study, including their motivation to participate and their initial expectations. The mechanisms and contexts by which the intervention brings about change will be analyzed using the COM-B behavior change theoretical framework. Finally, the semi-structured interviews that will also be undertaken with about 20 professionals engaged in Malin in its four districts will provide complementary insights into how the program is delivered in other contexts and its acceptability among the various professionals. This qualitative analysis will support the evaluation of ECAIL’s external validity in terms of settings and delivery modes [105].

The post-doctoral fellow will transcribe all interviews and transfer them for treatment by using N’Vivo 11 QSR software for qualitative analysis. For this research, 25% of the interviews will also be coded and cross-analyzed by the senior researcher in social sciences. The data will be analyzed thematically according to the method of grounded theory [109]. Its underlying principle is the systematic inductive generation of a theory through predefined stages and continual movement back and forth (iterations) between data collection and analysis, consistent with current recommendations to analyze qualitative data within a wider process evaluation [76, 96, 98]. Once the floating reading of all interviews is done, the corpus will be structured into thematic units by using N’Vivo. A global thematic analysis will allow us to observe each theme across all interviews.

Economic evaluation

We will collaborate with economists to assess incremental cost-effectiveness ratios, evaluating the intervention’s additional costs from the Malin NPA perspective in relation to the primary outcome (vegetable consumption frequency) and BMI at post-intervention. This step will involve comparing costs accrued in the intervention arm to those in the control arm.

Study governance and management, roles and responsibilities

All ECAIL study partners (Inserm, French Red Cross, Malin NPA, SFP, AFPA, Lille University Hospital) signed a multiparty agreement drafted by the Inserm legal department, outlining roles and responsibilities of each partner while ensuring the scientific independence of the researchers involved in this study. The French Red Cross sponsors the ECAIL study and is legally responsible for it vis-à-vis the competent authorities, with some operational tasks delegated to the Malin NPA, which coordinates study implementation. This coordination takes place in partnership with the Inserm team, which holds scientific responsibility for the study, in close collaboration with Lille and Valenciennes hospitals. In 2021, Lille and Valenciennes hospitals signed a study agreement, including a delegation log for task division under CIC-1403 of Lille University Hospital.

Operational follow-up of the study implementation

A monthly meeting gathers representatives from HCPs at the maternity ward (including reception staff, midwives, nurses, and obstetricians) to discuss any challenges in the screening and patient information process and co-design solutions to overcome these barriers. Additionally, a weekly video conference involving members of the implementation team addresses any strategic or operational issue related to the implementation process, and collaboratively develops solutions.

Strategic follow-up of the study

A steering committee, meeting quarterly, includes project partners (French Red Cross, Inserm, Malin NPA, SFP, AFPA, Lille University Hospital, and Valenciennes Hospital), with voting rights limited to the French Red Cross, Inserm, SFP, and AFPA. Its tasks include monitoring progress, ensuring schedule compliance, addressing performance issues, and recommending modifications with financial estimates, subject to funder approval. The committee also facilitates communication and oversight.

Discussion

The ECAIL RCT will evaluate the effectiveness of a multi-level, multi-behavioral, and innovative program that leverages the complementarity of nutritional education and facilitated access to appropriate foods and products to promote healthy feeding practices and foster optimal EBRBs in children. Positioned at the intersection of disease prevention and health promotion, this program also aims to improve overall health and well-being among the most vulnerable families.

Perpetuation and generalization of the intervention: translational research

Adopting an intersectoral and bottom-up partnership approach, ECAIL is a complex intervention that brings together professionals from diverse backgrounds and languages, including those working on the frontlines to combat poverty and exclusion, HCPs, and researchers [43, 76]. This collaboration fosters the cross-fertilization of experiential and academic knowledge and perspectives [73]. Using an interdisciplinary approach, the integration of quantitative (epidemiology) and qualitative (sociology) methods will support the process evaluation embedded within the trial [76, 96, 98, 105]. This evaluation will assess protocol adherence (fidelity and adaptations), describe implementation, clarify causal mechanisms, and identify contextual factors associated with variations in outcomes. By addressing both internal and external validity, these comprehensive findings will provide critical insights into what the intervention entailed, how it worked (or why it did not), for whom, in which context, and to what extent. In the current context of rising food insecurity, this knowledge will be instrumental in refining the Malin program as it scales up. This refinement is especially important given that the Malin NPA, with the support of the CNAF, aims to reach and engage as many as possible of the 400,000 children aged 0–3 years experiencing social disadvantage, along with their families, nationwide. Consequently, this action-research is positioned within a continuum between efficacy and effectiveness, bridging explanation and pragmatism [110].

A high-risk research project

As is often the case with action-research, the development and implementation of the ECAIL RCT have been a complex and long-term undertaking. Securing all regulatory approvals alone took nearly 3 years, largely due to the project’s groundbreaking nature, which did not align with the traditional legal frameworks for clinical trial research regulation in France at the time. After 5 years of co-creation with various stakeholders—including end-users—the trial’s implementation began in 2017 and remains ongoing. The pace of participant inclusion in the trial has been slower than anticipated because of factors inherent to the targeted population, operational challenges within hospital settings, and disruptions caused by the COVID-19 pandemic. Notably, successfully including socially disadvantaged groups in research requires specific methodological adaptations, such as tailoring questionnaires and dedicating additional human resources to inform, recruit, and follow up participants. These adjustments demand more time and significantly higher budgets than studies conducted in the general population (36).

Precious knowledge for all other interventions targeting hard-to-survey populations

This research project will generate novel and critical insights into the social determinants and mechanisms shaping behavioral and health trajectories from pregnancy through early childhood in disadvantaged and under-researched families. We will analyze the barriers and facilitators affecting the adoption of healthy feeding practices and EBRBs as well as the risk factors for OW/OB that are more specific to this vulnerable, heterogeneous and socio-culturally diverse population. We believe this knowledge is essential for designing tailored, family-based interventions that effectively promote health while reducing social inequalities from the very beginning of life. More broadly, the unique experience gained from the ECAIL trial will enhance our understanding of how to identify, recruit, and follow up families experiencing social disadvantage.

Social justice and reduction of social inequities in health

By empowering mothers (or parents) with the practices they can use to feed and engage with their child, the Malin program also aims to positively influence parents’ EBRBs and health, potentially affecting both generations. Intervening during the critical 1,000-day period provides an opportunity to shape the child’s behavioral and health trajectory favorably across the life course and potentially even their future social trajectory [32, 47]. Finally, by providing support through both individual and structural facilitators, this intervention aligns with the principle of proportionate universalism in public health, with the goal of reducing social inequities in health [24, 74, 111].

Abbreviations

AFPA

Association française de pédiatrie ambulatoire (French association for ambulatory pediatrics)

BMI

Body mass index

CIC

Clinical investigation centre

CI

Confidence interval

CNAF

Caisse nationale des allocations familiales (National family allowances office)

EBRB

Energy balance-related behavior

ECAIL

Pregnancy and early childhood nutrition trial

FFQ

Food frequency questionnaire

HCP

Healthcare provider

Inserm

Institut national de la santé et de la recherche médicale (National institute for health and medical research)

NPA

Non–profit association

OW/OB

Overweight and obesity

PNNS

Programme national nutrition santé (National nutrition and health program)

SFP

Société française de pédiatrie (French society of pediatrics)

RCT

Randomized controlled trial

WHO

World Health Organization

Authors’ contributions

SL leads the ECAIL randomized controlled trial (principal investigator) and co-designed it with OS, BC, NP, DT, CS, DM, PH, LB, DD, DS, DL (principal investigator at Lille University Hospital), AD, PD (principal investigator at Valenciennes Hospital), BLG, BH, and MAC; OS is responsible for managing the implementation of the ECAIL trial with SL; at Valenciennes Hospital, AD monitors the data collection and interacts with the various stakeholders; PH monitors the overall quality of the data collection at the two study sites and supervises the implementation team; LB ensures that communication flows smoothly between all stakeholders at Lille and Valenciennes hospitals; SB, MC and EC implements the trial at the two study sites; DP implements the qualitative component under the co-supervision of PS and SL; CLG is involved in data collection and analysis; SL will supervise data analysis and has drafted the manuscript. All authors contributed to refinement of the study protocol, read and approved the final manuscript.

Funding

The ECAIL study has received funding from the Agence nationale de la recherche (ANR-12-DSSA-0001, SOFI and ANR-19-CE36-0006, NutPrev); the fonds Français pour l’alimentation et la santé (FFAS-12-A-010); Blédina; the Institut de recherche en santé publique (IReSP): as part of the call for projects launched by IReSP in 2015 “Programme prévention primaire 2015”, this research was supported by the Direction générale de la santé (DGS), the Caisse nationale de l’assurance maladie des travailleurs salariés (CNAMTS), the Mission interministérielle de la lutte contre les drogues et les conduites addictives (MILDECA) and the Observatoire national des Jeux (ODJ) (IReSP-15-Prévention-07); additionally, as part of the 2022 call for research projects on health-promoting services, interventions and policies (SIP), this research was supported the Caisse nationale de l’assurance maladie (CNAM), the DGS, Inserm and Santé publique france (AAP-2022-SIP-311921); the Guy Demarle Enfance & Bien Manger Foundation Price, under the umbrella of the Fondation de France in 2018; the Joint programming initiative ‘A Healthy Diet for a Healthy Life’ PREPHOBES (JPI-HDHL-INTIMIC 2020); the Projet fédératif hospitalo universitaire (FHU) 1000 days of the Lille University Hospital; the Fondation Roquette pour la santé, under the umbrella of the Fondation de France; the Agence régionale de santé des Hauts-de-France; the Health philanthropy program of Mutuelles d’assurance AXA; and the Délégation interministérielle à la prévention et à la lutte contre la pauvreté (DILP, Interministerial delegation for the prevention and fight against poverty) set up as part of the Solidarity pact national policy. While this study protocol has undergone independent peer-review by major public funding bodies, the funders had no role in the design or execution of the ECAIL RCT, data collection, or the writing of the current manuscript. They will also have no role in the design of future studies based on ECAIL, data analysis and interpretation, decision to submit results, or the writing of related manuscripts.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The ECAIL RCT has approval from the committee for protection of participants engaged in human research (Comité de protection des personnes Nord-Ouest 1 [CPP], 24/07/2014, N° RCB: 2014-A00849-38), the national advisory committee on information processing in health research (Comité consultatif sur le traitement de l’information en matière de recherche scientifique [CCTIRS], 04/09/2014), and the French data protection authority (Commission nationale de l’informatique et des libertés [CNIL], 15/06/2016). The protocol for the qualitative component was approved by the Nord-Ouest 1 CPP on 28/06/2021. Written and signed consent were obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.