The importance of maternal nutrition for optimal fetal development and lifelong population health is increasingly recognized. A healthy dietary intake during pregnancy supports the physical and mental development of the fetus and may prevent congenital malformations, premature birth, and low birth weight or small-for-gestational-age babies [1-4].

Pregnancy provides opportunities to improve dietary intake, as it is a critical transition in the life course during which women consider nutrition important [5]. Nulliparous pregnant women, in particular, show an increased interest in nutrition, although their nutrition-related information-seeking behaviors depend on the time at which they start to feel like a mother [6]. Unfortunately, pregnant women’s adherence to dietary guidelines still appears to be limited [7-9], especially in low–socioeconomic status (SES) populations [10]. Low-SES pregnant women face additional barriers to healthy eating compared with higher-SES groups and experience multiple stressors that may prohibit the instigation and maintenance of a healthy dietary intake [11]. This warrants further study of contextual, behavioral, and psychosocial characteristics of low-SES pregnant women and the associations of these characteristics with dietary intake [11].

The World Health Organization (WHO) claims (pg xii) that interventions improving maternal nutritional status in low-SES populations are among the most effective and sustainable means to achieve positive impacts on health and to reduce health inequalities across generations [4]. In such interventions, diet quality in terms of adherence to dietary recommendations, including supplement use, should be considered, as well as energy balance in relation to gestational weight gain. Successful adoption of a healthy dietary intake depends not only on pregnant women’s nutrition knowledge, but even more so on their ability to decide on, act toward, and sustain a healthy dietary intake given the often stressful or disempowering contexts in which they live [11]. The WHO advocates an inclusive approach to public health and nutrition, such as through support for practitioners to ensure that they understand low-SES pregnant women’s circumstances without stigmatizing them when discussing diet and physical activity [4].

Midwives are an important source of nutrition-related information for most pregnant women in the Netherlands and could potentially play a significant role in improving the dietary intake of low-SES pregnant women [6,12]. Szwajcer et al showed that 80% of Dutch pregnant women were more interested in nutrition information during the first trimester than they were before [6]. Moreover, this study showed that 28% of Dutch women in the first trimester of their pregnancy considered the midwife to be an important channel for pregnancy-related information, including nutrition [6]. Dutch pregnant women considered the midwife to be a trusted source of nutrition information, valued the interactive character of consultations, and perceived the ambiance as pleasant [13]. Dutch midwives independently provide care during normal pregnancy, childbirth, and the early postpartum period. In 2016, 86.8% of pregnant women started consultations in primary prenatal care, and 30.0% of women eventually delivered their child with their primary care midwife. Referral from primary to secondary care during pregnancy or during birth took place in 35.2% and 21.5% of all cases, respectively [14].

Baron et al [12] suggested that a certain amount of proactivity from midwives in providing information may be justified, to increase awareness of beneficial health behaviors and shape positive health behaviors. Although Dutch midwives receive some training in nutrition as part of their education and appear motivated to discuss nutrition in their consultations, their actual nutrition communication appeared limited, was often general in nature, and focused primarily on risks (eg, food safety) and problems (eg, nausea) [12,13]. Midwives seemed to lack essential resources such as expertise, self-efficacy, and time [15,16]. To be more effective, midwives should be able to provide tailored nutrition communication throughout pregnancy and take into account women’s family situation and culture, as well as their current dietary intake [17-19].

Tools could provide support in tailored nutrition communication. Although a variety of tools informing pregnant women about healthy nutrition are available, no evidence-based tools are being used routinely by Dutch midwives to improve pregnant women’s dietary intake. This lack of tools has also been acknowledged by the Royal Dutch Organization of Midwives (Koninkelijke Nederlandse Organisatie van Versloskundigen; KNOV), the Dutch Association of Dietitians (Nederlandse Vereniging van Diëtisten; NVD), and the Netherlands Nutrition Centre (NNC), and in several scientific papers [20-22]. Tools that could be used may range from pocketbooks to educational videos and workshops [23]. Dietary assessment instruments, including digital ones, might also be helpful for increasing awareness of dietary intake, motivating people to adopt healthy eating habits, or providing support in dietary self-monitoring [24].

Therefore, we will address several omissions in current research and practice in this study. First, we aim to gain insight into the dietary intake of low-SES pregnant women in the Netherlands, including the dynamic interplay between pregnant women and their sociocultural environment. Second, we will study midwives’ current practice and their capability and willingness to provide nutrition communication. Third, we will provide an overview of tools that could support midwives in optimizing pregnant women’s dietary intake. Based on findings on these information gaps, we will either select and adapt or newly develop 1 or more tools fitting best with pregnant women’s and midwives’ needs for implementation in concurrent prenatal practice.

The objective of this study is to develop a tool or toolbox for midwives to promote healthy nutrition among pregnant women, taking into account insights into factors influencing the dietary intake of Dutch low-SES pregnant women. To explore the needs of both pregnant women and midwives and to identify best practices, the main research question is 2-fold:

(1) What is low-SES pregnant women’s contextual dietary intake?

In the first part of the project (2.5 years), we will answer 3 research questions and, based on the results, we will develop a tool(box):

(1) What individual (eg, food preferences, information-seeking behavior), interpersonal, and sociocultural (eg, family, social networks, social media) factors drive low-SES pregnant women’s dietary intake? (RQ1)

In the second part of the project (1.5 years), we will implement a newly developed or adapted tool and address a fourth research question will be addressed:

We will conduct interdisciplinary, mixed methods research to comprehensively address the research questions. We will collect data in the first part of the research (RQs 1-3) through systematic literature research, interviews with midwives and pregnant women, a diet history questionnaire, focus groups, and expert consultations. In the second part of the study, we will conduct a process evaluation of the implementation of the tool(box) throughout the Netherlands.

Stakeholder involvement is key in this study in order to develop a tool(box) that fits the needs of low-SES pregnant women and is feasible for midwives to use. Midwives, pregnant women, and relevant stakeholders identified in the process will therefore be actively engaged in the research activities. Furthermore, we will involve project partners KNOV, NVD, and NNC in all stages of the research.

The design of this study is focused on what creates health and well-being rather than on preventing disease. This is encompassed by the concept of salutogenesis, used to explore sources of adaptability and resilience [25]. In this study, we will use 3 building blocks for salutogenic research [26]: (1) taking a holistic orientation to food (nutrition), including physical, mental, and social dimensions of health; (2) supporting a healthful life orientation; and (3) facilitating health-directed learning processes through positive interactions and experiences with food [26].

The first building block is addressed by means of a socioecological model (Figure 1) and an integral model (Figure 2). Both models provide frameworks for a holistic approach, as they help to elucidate how multiple levels of influence shape a person’s dietary intake [27,28] and to unify multidisciplinary thinking, practice, and evidence gathering [29,30]. The socioecological model allows for categorization of personal, cultural, and environmental factors [31,32], whereas the integral model distinguishes between subjective and objective factors of influence on both the individual and the collective level [30].

The second building block advocates an orientation toward health rather than disease, as conceptualized by the salutogenic model of health. The core constructs of the salutogenic model are sense of coherence and general resistance resources. Sense of coherence comprises the 3 constructs comprehensibility, manageability, and meaningfulness. Comprehensibility encompasses a feeling of confidence that stimuli deriving from one’s internal and external environments are structured, predictable, and explicable; manageability, that resources to meet the demands posed by these stimuli are available; and meaningfulness, that demands are challenges, worthy of investment and engagement [33].

Sense of coherence is inherently related to general resistance resources: resources within an individual or the environment that can be used to counter the stressors of everyday life [34]. From the salutogenic model perspective, it is argued that health promotion activities should focus not only on changing beliefs, knowledge, or intentions, but also on empowering people to mobilize and reflect on resources already available to them [35]. It should be noted that it is not only pregnant women who need to be empowered, but also their midwives. To enable the empowerment process, their relationship should resemble a partnership rather than a traditional hierarchical relationship [36]. In our study, we will use the salutogenic model as a guiding perspective. We will address related concepts of empowerment and reflection in research activities with professionals and pregnant women and in the development or adjustment of a tool(box).

The third and final building block describes health-directed learning processes. Swan’s recommendations include engaging participants and taking into account participants’ social environment or changes in their environment [26]. The best practice framework by Ng and De Colombani comprises these aspects and provides guidance on defining criteria for tools and methods to be used in midwifery practice, with regard to context, process, and outcomes [37]. The framework addresses relevance, community participation, stakeholder collaboration, ethical considerations, and replicability, as well as effectiveness, efficiency, and sustainability. The practice-based evidence available in this framework helps to build on existing tools and practices.

This study includes low-SES pregnant women and midwives living and working in the Netherlands, as well as stakeholders identified in the process (eg, family members and health professionals, other than midwives). We will collect data on educational level, occupational status of both the participant and, if applicable, her partner, and individual and household income in a general questionnaire to capture the multidimensional nature of SES.

We will recruit midwives through a combination of convenience sampling and purposive sampling. First, we will approach midwives in the network of project partners, including a midwife, a gynecologist, and the KNOV for participation. Second, we will select midwifery practices located in disadvantaged neighborhoods (both rural and urban) based on postal code, as defined by the Dutch Healthcare Authority [38]. We will approach all midwives selected through the network or through postal code by telephone and email and invite them to participate in an interview and to recruit low-SES pregnant women, or they may opt for just 1 of these activities. We estimate that we will need 20 midwives to reach data saturation. In case we do not reach data saturation, we will conduct more interviews.

We will recruit pregnant women (n=50) through the midwives using a purposive sampling technique. Midwives will receive oral as well as written instructions for the recruitment of pregnant women. We will use highest educational attainment as the primary indicator of SES, as education is a relatively permanent aspect of SES [39] commonly used in nutrition and health research and often included in midwives’ intake questionnaires. Pregnant women will preferably be interviewed as early as possible in pregnancy, as the intervention to be developed will also have to be implemented in the first trimester. All participants taking part in interviews or focus groups should be proficient in Dutch and have a Western dietary pattern. Additionally, we will post flyers and posters in midwifery practice waiting areas to reach pregnant women directly. On these flyers, educational level is communicated positively: “Have you graduated from or are you currently following prevocational or vocational education?:

Ethics approval was given by the Social Sciences Ethics Committee of Wageningen University & Research, Wageningen, the Netherlands. The committee thereby declares that the proposal deals with ethical issues in a satisfactory way and that it complies with the Netherlands Code of Conduct for Scientific Practice. Informed consent will be obtained from each participant, after the nature and possible consequences of the study have been explained.

We will manage data according to Wageningen University’s research data policy, based on the Netherlands Code of Conduct for Scientific Practice and the FAIR (findability, accessibility, interoperability, and reusability) principles. All participants will be assigned a unique study identifier to store data anonymously. We will use the study identifiers to link the quantitative and qualitative data of those participating in both the DDHQ and the in-depth interviews, allowing for investigation of associations between the quality of dietary intake and individual, interpersonal, and sociocultural factors.

We will audiotape the qualitative research data from the interviews and focus groups with the interviewees’ permission through informed consent, transcribe the audio intelligent verbatim style, and analyze the data by means of thematic analysis [53]. The thematic framework used in the analyses allows for iterative use of both deductive and inductive approaches. Interviews will be coded independently by multiple researchers to reduce interobserver bias and thereby increase the internal validity of the method. We will use ATLAS.ti version 8 software (ATLAS.ti Scientific Software Development GmbH) for these analyses to manage data and optimize transparency.

Quantitative data derived from the DDHQ will be administered and stored in the online Dutch FFQ tool [44], then exported for analysis using SAS version 9.4 statistical software (SAS Institute Inc). Food items in the DDHQ are aggregations of food codes in the Dutch Food Composition Database [54]. Total energy and nutrient intakes will be calculated automatically in the tool by multiplying frequency of intakes by consumed amounts and nutrient composition per item using the same food composition database and standard Dutch food portion sizes. We will present all DDHQ data adjusted for energy to partially account for measurement errors.

This study relates to several concurrent health challenges and developments in the Netherlands, such as health inequalities, midwives’ ambition to be strengthened in their role as public health professionals, and translating the first Dutch national dietary guidelines for pregnant women (by an ad hoc committee of the Health Council of the Netherlands) into practice.

This is, to our knowledge, the first time that the dietary intake of specifically low-SES pregnant women in the Netherlands will be studied. Gaining insight into determinants of dietary intake in pregnant women, and specifically low-SES pregnant women, will help to elucidate the factors that contribute to unhealthy habits or—from a more assets-based perspective—the factors that facilitate a healthy dietary intake. By including those pregnant women who would benefit most from nutritional education, this study will address socioeconomic inequalities in health, which are considered unfair and avoidable by, among others, the WHO and governments, in an early stage of life.

The diet history method, applied in an interview by trained dietitians, is the best method for a low-SES target group. However, it is a burdensome method and not well standardized. Therefore, we developed a meal-based questionnaire with food items covering at least 95% of the intake of women of reproductive age. As we will ask additional open questions about the intake of other foods not included in the questionnaire, we expect to cover the complete dietary intake of the women like that captured by the diet history method.

All stakeholders’ perspectives need to be taken into account to generate an evidence base of what works and why in a real-life setting. The participation of multiple stakeholders (pregnant women, midwives, and other health professionals and experts) will be stimulated throughout the research, thereby generating context-sensitive and usable knowledge [55]. We will develop the tool(s) in close collaboration with pregnant women and midwives, KNOV, NVD, and NNC, taking into account their concurrent practices.

All research activities and project meetings conducted in the first phase of the research will contribute to the development of this tool(box) and its implementation and evaluation in the second phase. A limitation of this study is its inability to measure the effectiveness of the developed tool(box) within the scope of the study. An evaluation of the effects on dietary intake, as well as maternal and child health outcomes, could be part of a follow-up study. If the adapted or newly developed tool(box) improves pregnant women’s dietary intake, the impact on perinatal and postnatal outcomes may have health and social benefits, as well as economic benefits.

We will disseminate results to participants who express an interest in this, as inventoried at the end of the interviews. Project partners will be informed on progress, and results will be conveyed orally at least every 6 months by way of a group meeting or individual phone calls. All project partners have furthermore agreed to share the results of the research with their members, for example, through their websites or newsletters. To inform the scientific community, we will disseminate results in scientific journals, as well as at national and international conferences.

The researchers involved in this study have ample experience with the successful recruitment of low-SES pregnant groups, such as the MetSLIM lifestyle intervention [56], SLIMMER diabetes prevention intervention [57], and Communities on the Move [58]. From these experiences, we have learned that considerable efforts are required to recruit low-SES groups, that too-strict inclusion and exclusion criteria hinder recruitment, and that a personal approach and trust in the recruiter are success factors. Therefore, in this study, we allow for sufficient time for recruitment activities and have budgeted for incentives. We will ask midwives to be gatekeepers in the recruitment of pregnant women, as they have strong trust relationships with their clients. A flexible recruitment protocol will be based on the needs and desires of low-SES pregnant women and their midwives, and incentives, as stated, are budgeted for.

Compared with other countries, in the Netherlands midwives play a central role in maternity care. In general, they are interested in research that will strengthen their profession. Unfortunately, they often experience a lack of time and receive numerous requests to participate in research. We have addressed this problem by holding interviews at their location, by trying to minimize midwives’ time investment in recruitment, and by compensating them financially (based on regular hourly wages).

Low-SES pregnant women may be considered a vulnerable population for several reasons. First, pregnancy is a time during which women (and their unborn babies) are physically vulnerable. Second, people with a low SES may have problems understanding information. We will instruct interviewers to ensure that participants understand information correctly if the interviewers doubt comprehension and to inform the main researcher about reconsideration of participation. Researchers involved in this project are experienced with research including low-SES groups. The research as a whole is specifically responsive to the health needs and priorities of low-SES pregnant women and ultimately aims to reduce health inequalities.

We are aware of pregnant women’s dependence on midwives and shall therefore emphasize to midwives that all participants should enter the study freely. Participants will not run any physical, social, or political risk by participating in this research.

All participants will be informed about the aims of the research, duration of interviews, data preparation and anonymous data storage, the voluntary nature of participation, and their right to withdraw at any time prior to each research activity. After an opportunity to ask questions, written informed consent will be obtained from each study participant for each research activity. The term low SES, which sounds negative, will not be used in any communication with pregnant women.

To our knowledge, this will be the first study to address the dietary intake of low–socioeconomic status pregnant women in the Netherlands and to link dietary intake to contextual factors by using an ecological perspective on health. We hope the results obtained will inform multidisciplinary strategies to promote a healthy dietary intake in prenatal care, specifically in low–socioeconomic status populations in developed countries.