Understanding Antenatal Care Needs through Co-Creation with Roma Women to Inform the Design of mHealth Technologies

Women from the Roma community experience significant health disparities during their pregnancy. Whilst Mobile Health (mHealth) technologies have the potential to improve antenatal care experiences and health outcomes, research on women from ethnically marginalised backgrounds in developed countries remains limited. We report on a series of Co-creation Workshops with 11 Roma women who have settled in the North of England. In this paper, we present thematic insights about their experiences and needs during pregnancy, and their perceptions and attitudes towards digital technologies to inform the design of culturally sensitive mHealth. We contribute to Human Computer Interaction (HCI) with new empirical research to discourses on Critical Digital Health, Intersectional HCI and Women-centred Design, highlight implications for design and encourage a more critical and intersectional design approach to accommodate better the experiences of ethnically marginalised groups whose needs arguably tend to be overlooked and stereotyped.


INTRODUCTION
The first 1,001 days of life are globally recognised as critical for lifelong health and wellbeing, and birth preparedness through antenatal care promotes safe motherhood and a reduction in adverse pregnancy and childbirth outcomes [75].In the United Kingdom (UK), free antenatal care services are provided by the National Health Service (NHS) to all residents and can be described as the care that women receive whilst they are pregnant and is usually delivered through appointments with a General Practitioner (GP) and a midwife.Most recently, there has been a trend for the digitisation of antenatal care services through Mobile Health (mHealth) technologies [40], including the use of an NHS-endorsed mobile application (app) for improving access and routine care during a woman's pregnancy [81].However, health inequalities and social injustice in antenatal care are well-evidenced in the UK: women from minority ethnic groups and those from lower socio-economic backgrounds have reported poorer health outcomes and experiences of antenatal care [5,10,34,39].This is particularly true for migrant women residing in high-income countries like the UK who suffer structural and institutional barriers to antenatal care, which reinforce racial and ethnic disparities [6,36,49,77].Care pathways and communications with Healthcare Professionals (HCPs) in a new host country may be complex and confusing for many migrant women, who lack information in their own languages [17].Research in the field of Human Computer Interaction (HCI) has investigated the potential of mHealth technologies for supporting antenatal care in many developing countries (e.g.[33]), but mHealth HCI research on women from marginalised groups in high-income countries remains sparse [5,11,13,80].Women from migrant and lower socio-economic backgrounds have complex health and wellness needs, which should be better understood and celebrated through intersectional approaches [11,32,43,44,53,61,67].
Against this backdrop, we respond to a previous call in HCI for more research with immigrant populations [11] by investigating the antenatal care needs of women from the Roma community who have settled in the UK.The Roma community is culturally rich and diverse, with dozens of subgroups that have distinctive characteristics and identities.They are the largest ethnic minority group in Europe [76], and public health research has reported significant barriers to healthcare, particularly with regard to racial stereotyping and stigma experiences in their original countries and across Europe [78], contributing to a lack of engagement with health services [54].In the UK, people from the Roma community are collapsed into the category of 'Gypsies, Roma, and Travellers' (GRT); and Government-led strategies have been developed to tackle health inequalities faced by this group [26].However, the racial profiling used by service providers can arguably lead to ethnic stereotyping and discrimination.Indeed, individuals from the Roma community may not accept the GRT category because of its negative connotations [24] and prefer not to identify themselves as travelling or itinerant like the Gypsy and Traveller communities.Nevertheless, those classed as 'GRT' are recognised as among the most disadvantaged in British society, experiencing significant barriers to healthcare due to cultural norms, discrimination, and limited familiarity with the healthcare system [20].Statistics also show significant disparities such as high infant and maternal mortality rates [21] and Roma women are often unaware of the benefits they are entitled to receive (i.e.Healthy Start vouchers) [25].To the best of our knowledge, there is no research on this group in HCI despite being the largest European ethnic minority.
We report on empirical findings from eight Co-creation Workshops conducted with 11 Roma women from Czech Republic and Slovakia who have settled in the UK.The aim of the workshop series was two-fold; first, to investigate women's pregnancy journeys to understand their experiences of healthcare services and needs during pregnancy; and second, to explore design opportunities for supportive tools and resources that meet the needs of the community.We build on previous HCI research that urges the need for a more in-depth phenomenological investigation for developing richer and more nuanced understandings of women's experiences from marginalised groups [44]; and which encourages a shift from a 'one-size fits all' approach to adopt a more holistic and intersectional perspective of women's health [32].With this paper, we contribute to HCI discourses on Critical Digital Health [37], Intersectional HCI [53,61,67] and Women-centred Design [4] with new qualitative insights and design implications for improving health equity through co-creation based on the experiences of marginalised groups whose needs tend to be overlooked and generalised.This paper is structured as follows: We first review relevant work on self-management during pregnancy, including previous HCI research that demonstrates the potential of mHealth to support women from minority groups and advocate for a more critical, intersectional and women-centred design approach.We then describe our co-creation approach and the workshops before reporting thematic insights on participants' experience of pregnancy and their attitudes or perceptions towards mHealth technologies.Finally, we discuss our insights and design implications for developing future mHealth technologies that can support the antenatal care needs of women from minority backgrounds.

RELATED WORK 2.1 Managing health and wellbeing during the pregnancy
Pregnancy is a challenging period of life often associated with health stressors such as hormonal changes, weight gain and inadequate sleep [42,46,68].For example, managing weight gain can be challenging with impact beyond birth [46] and there is a lack of guidance from healthcare providers as well as common misconceptions including the popular term 'eating for two' [66].Pregnancy is also a transformational journey towards parenthood [68], causing women to experience emotional distress and anxiety that can result in poorer self-care [42] and increased risks for postpartum depression [27].Indeed, mental health issues during pregnancy have increasingly been documented but these are still neglected in antenatal care [19,23,44] despite affecting a large number of pregnant women (i.e. 10 to 20% in the UK [52]).In the case of the GRT group, women have higher birth rates but are more likely to experience poorer health outcomes compared to the majority of the population including higher rates of anxiety and depression [25].Roma communities have also lower levels of access to mental health services due to the taboo and lack of vocabulary available for describing mental illnesses in the Roma culture [56].Psychosocial challenges during pregnancy are exacerbated for women from marginalised communities [44]; particularly for migrant women who are settling into a new host country.Migration can significantly increase life and health stressors due to readjustments and becoming familiar with health services in a new country [5,6,11].Examples include increased stress when starting over and dealing with unemployment, financial insecurity, mental distress, changes in lifestyle impacting dietary choices, and dissatisfaction with health care services, which are unprepared to cater to the distinctive needs and cultures of newly arrived women [11,50].
Indeed, each woman's pregnancy experience is unique and influenced by a range of physical, emotional, and social factors [46].Frameworks have been developed to help HCI researchers understand women's changing needs and identify psychosocial factors that may impact women's health during pregnancy [28,46].These include a stage-based model [28] and the Pregnancy Ecology framework [46], which show touchpoints for support seeking during pregnancy (i.e.self-guided information seeking in the first trimester).Indeed, we know that pregnancy presents a 'teachable moment' for health management [48] with high information seeking needs and behaviours [47] but HCI researchers have identified gaps in women's pregnancy journeys related to information needs, which require more investigation [28,47,62].With our research, we contribute to such a work by exploring Roma women's experiences and needs during pregnancy and identify what information is important to them through co-creation.

Mobile Health to support pregnant women from ethnic minorities and low-income backgrounds
Research in high-income countries shows that ethnic minorities and communities with low socio-economic status are less likely to receive appropriate antenatal care [5,13,45].In the UK, there are significant disparities for pregnant women depending on their ethnicity and socio-economic status [74].Black women are 3.7 times more likely to die during their pregnancy than White women and the highest maternal mortality rate was found among women who live in deprived areas [34].Women from immigrant communities are also facing challenges in receiving antenatal care because they may not be familiar with the care pathways in their new host country [5].More broadly, immigrant women have unique health and wellness needs, which should be better understood and translated into implications for technology design but research with this user group remains sparse in HCI [11].
Prior research has reported increased use of mobile apps by pregnant women [12,30,73] and showing how mHealth for pregnancy can support information seeking [33,70], self-tracking for emotional wellbeing [68], antenatal mental health screening [19] and social support via online communities [28].Indeed, there is an opportunity for mHealth to address barriers to antenatal care for women who may be disadvantaged because of their socio-economic backgrounds [13,45], ethnicity [44] and migrant identities [5,69].
However, limitations of existing mHealth interventions for supporting pregnant women were also highlighted [13,46].One of the main challenges is around user engagement and usability of apps, which can vary according to the user's ethnic background [19].More research is needed to ensure equitable digital health innovation [44] whereby efforts are placed on designing digital tools that address cultural differences and meet the particular needs and values of women from minority groups [19,32,64].With this paper, we contribute to this research gap in HCI by reporting on qualitative insights from an understudied ethnic minority group to inform implications for designing inclusive and culturally sensitive mHealth technologies.

Critical and Women-centred approach of mHealth technologies in HCI
A recent review [67] shows that there is limited contribution from HCI on mHealth for vulnerable groups and that there are tendencies in our community to overlook intersectionality and neglect societal and cultural factors, which are directly influencing health disparities.There is a growing discourse on Intersectional HCI [53,61], which highlights the pervasive nature of racism in socio-technical systems [9,43], urging the need for researchers to be more attuned to issues of race and to deepen their engagement with identity complexities [61].Intersectionality provides a framework to unpack structures of social inequalities and challenge the status-quo for how to design technology for vulnerable populations [53].Recent research in Digital Health with perinatal black women [44] shows the importance of gaining intersectional understandings to 'disrupt monolithic thinking' [22] and avoid exacerbating inequities through design.In their study, Oguamanam et al. [44] show that black women are not a homogenous group by offering a nuanced understanding into their different practices and attitudes towards mHealth, which varied depending on their levels of income and education.
A number of recent HCI work with women from marginalised groups has challenged the status-quo for tracking and sharing data [13,44,67]: Chaundry et al. [13] report on the deployment of an app with low-income women to track health data for sharing with health providers during pregnancy.Insights from their work suggest a need to rethink approaches for data tracking and sharing, and a greater understanding of context so that future tools can better fit in with the mundane realities of pregnancy.An alternative design paradigm was also suggested to consider how we may shift focus from designing mHealth for the individual to the community [67]; 'towards collective action and systems, interpersonal care and understanding yourself as a part of a pluralistic, diverse group of people, of society, of humankind as a whole' [65 p.11].This was explored with immigrant women who co-envisioned how mHealth may support community resilience by facilitating the rebuilding of mutual support groups [11].Research with transmigrant women has also considered Prosocial Designs for pregnancy mHealth interventions to facilitate both local and long-distance caregiving [5].
Finally, mHealth including self-tracking technologies can play a supportive role during pregnancy but research has criticised the techno-centric paradigm of such technologies, which can reinforce the medicalisation of the body and further diminish women's embodied experiences of pregnancy [4,28,68].There are also issues with the visual design and representations used in pregnancy apps: these tend to convey an idealised image of pregnancy and health outcomes grounded in heteronormative and gendered stereotypes [30,71].Moreover, research on female-oriented technologies (FemTech) including menstruation and pregnancy apps has highlighted risks for the security and privacy of oneself and others caused by the ever-extending collection of intimate data [3].Therefore, a critical perspective [37] is needed to challenge mHealth's Western and techno-centric paradigms.We align with recent HCI research that advocates for a Woman-centred Design approach rooted in Feminist theories and grounded in humanistic, activist, and inclusive principles [4].

Summary: Opportunities for HCI research
Previous work in HCI investigated how mHealth can support healthier pregnancy management through the design of supportive technologies for weight management [46] and emotional wellbeing [68] or mental health [44].Temporal and multi-layered perspectives have informed frameworks for supporting HCI researchers to understand women's complex needs and identify touchpoints during pregnancy [28,46].For instance, gaps in women's pregnancy journeys related to information needs were highlighted [28,47,62] and challenges related to mental health during pregnancy were documented [19,44] further pointing to the potential of mHealth for self-management during pregnancy.However, current HCI research in developed countries with marginalised groups is limited [13,46,67], particularly with immigrant women who are more likely to suffer from mental health issues and worse health outcomes [11].There is a call for HCI researchers to design meaningful tools and interactions that account for more diverse pregnancy experiences [68] whilst addressing cultural differences [44].Critical to this is for researchers to be grounded in community contexts and capture nuanced understandings through longitudinal studies [67].Our work follows such recommendations and contributes to an increasing body of work in Intersectional HCI [13,44,53,61,67] by engaging a group of Roma women in co-creation research to inform the design of pregnancy mHealth technology that is culturally sensitive and inclusive.

RESEARCH THROUGH CO-CREATION 3.1 Study Design
We conducted eight Co-creation Workshops to explore women's needs, experiences of healthcare services and cultural beliefs about pregnancy and childbirth.A subsequent aim was to explore available resources and tools including examples of pregnancy apps to assess the potential of mHealth and inform future work for supporting antenatal care provision in the Roma community -an ethnic minority group that suffers from significant health disparities [20,21,49,77].
Co-creation was scaffolded by design-based methods [58]: informed by women's stories, a fictional character named 'Matylda' by the participants was created to encourage conversation during the sessions (Figure 1); a pregnancy map was collectively created to visually represent and understand key moments in the pregnancy  Feedback and reflection on the workshops during a two-hour debrief session journey (Figure 1), and Zine making [29] was used as a method to synthesise participants' contributions to the workshops and highlight what key messages the women wanted to share about pregnancy with others in their community (Figure 2).To investigate women's views about mHealth, a number of pregnancy app examples collated in a PowerPoint presentation were also presented.Next, we provide more details on each of those steps and Table 1 provides an overview of the workshops.

Encouraging research co-creation.
Inspired by women's stories shared in the first session, the first author developed the fictional character 'Matylda'.The character was presented as an illustration, portraying a young woman who was pregnant and new to the country.At each session, different scenarios featuring Matylda were presented to encourage sharing and discussion.The women were invited to share whether they would identify with Matylda and if there was anything they wanted to comment on.
In Workshop 4, we shared a translated summary of insights from the previous sessions for discussion and then, we invited the women to co-create a map of the pregnancy journey based on their common experiences of being pregnant, which were discussed in workshops 1-3.The fictional character (Matylda) was used to contextualise the activity and help the women identify what was relevant for a young woman like Matylda.Participants were guided to make a collage, by drawing or writing their thoughts on three large sheets of paper (one for each trimester), which were joined together as one journey map at the end of the session to discuss everyone's contributions to the map.
In Workshop 5 we discussed and rated the importance of different information, and in Workshop 6, we used a walkthrough method [57] to introduce the women to different app features.This was preceded by a mini review of pregnancy apps conducted by Author 1 who also downloaded five of the most popular apps from the Google Play store.Key features were identified, and screenshots of the following features were collated into a PowerPoint presentation: onboarding, sharing pregnancy information with partner (i.e.creating a joined account), personalised information about pregnancy (i.e.week-by-week updates), visualisation of the baby's development, self-reporting and sharing with other mums-to-be in an online community.The screenshots were translated into Czech and a scenario featuring Matylda was presented alongside each feature to help participants identify opportunities and challenges for women from their community.
In Workshop 7, we brainstormed ideas for an antenatal care resource and the women settled on creating a booklet about pregnancy in the UK to share with other women in their community.We used Matylda to set the scene, and to inspire women's contributions, we shared a list of consolidated themes with associated quotes from the workshops.Ideas for contributions to the booklet ranged from creating a poem to sharing stories or tips inspired by women's personal experiences.Participants' contributions were collated at the end of Workshop 8 and assembled together into a Zine.The co-created output was printed in both English and Czech, and shared at a de-briefing session with the women for feedback and further discussion.The women agreed to be acknowledged by having their first name included on the cover of the Zine, which was published via an online publishing platform free of access1 and further disseminated for feedback at stakeholder and community events.
3.1.2Procedure.The workshops were co-developed with an experienced community worker (Zaneta) who received training through the project as a peer researcher -i.e. a member of the research population with lived experience conducting research.Zaneta was also our community gatekeeper and supported the organisation and delivery of the workshops by liaising with the women to attend each session.Recruitment was down to our gatekeeper (Zaneta) who recruited women by drawing on her networks to gather a group of Roma women with experience of motherhood from a range of ages and backgrounds.However, there were challenges in recruiting younger participants who were too busy with work and childcare responsibilities.
Participants in the group differentiated themselves from other subcommunities in the GRT group and self-identified as Czech or Slovak Roma.Six of the 11 women were previously involved in a community engagement project led by Author 3 in collaboration with Author 1, which consisted of two focus groups aimed at exploring their varied migration stories2 .Nine of the women recruited were from the Czech Republic, with two from Slovakia (Table 2).All the women were first-generation migrants.Three have lived in the UK for more than 20 years, whilst one moved to the UK three years ago.The number and ages of their children, 29 in total, ranged from six adults to a baby boy who attended each workshop with his mum.Four participants had partners from different ethnic minority groups, and one described her partner as white European therefore making the picture more complex than expected.
Most of the women were able to speak in English but had various levels of fluency, so they were given the choice of speaking in Czech or Slovak languages.We also worked closely with our interpreter Irma who helped to ensure that all participants had their voices heard by directing questions to individuals, clarifying any misunderstanding and relaying information to us.Both Zaneta and Irma contributed to the sessions as well by sharing their own experience of pregnancy and a relative of one of the participants who was not Roma, joined at the last workshop.
In total, there were 12 participants (11 Roma women), with between five to nine attending at any one time.Women who attended were given a £10 shopping voucher and breakfast was provided before each session.This further helped establish a rapport with them whilst informing the management of the project and its insights reported in this paper.

Consent process.
Before the start of the workshops, the women attended an introductory breakfast meeting in a local cafe with authors 1 and 3 together with Zaneta.Each woman was provided with a translated participant information sheet, had their questions answered, and completed two consent forms including a copy they could keep for their own records.The women were also given a version of the workshop topics (Table 1) with suggested dates.These were then modified according to women's availability due to school holidays or other commitments they had.At the first workshop, ground rules for participation were agreed with the women to create a safe space for discussion: for instance, participants agreed that each person should get a chance to speak, one at a time and that personal stories should stay in the group.Women were reminded about the rules in subsequent sessions.
3.1.4Data collection and analysis.Data collection lasted over four months: it started with the first workshop on 25 th January 2023 and concluded with the debriefing session on 3 rd May 2023.The workshops took place on Wednesday mornings in a community space in Gateshead.Conversations during the sessions were digitally audio-recorded and anonymous transcripts were typed up by Author 3. Transcripts were then analysed thematically [16] with the help of NVIVO qualitative data management software.Each transcript was first coded remotely by both Authors 1 and 3 using the commenting tool in Word to annotate provisional codes and reflections.Transcripts from the first three workshops were then coded and analysed thematically [16] by Author 3 with the help of NVIVO qualitative data management software to produce a summary report.The coding frame was then further developed and used by both Author 1 and Author 3 to code data from the remaining workshops.Provisional themes derived from data coded in NVIVO were iteratively refined as we progressed through data collection.This process involved moving back and forth between coding and defining themes, to weave together a thematic narrative with accounts of participants' experiences.This reflexive approach to thematic analysis [16] was facilitated by the following: (i) Weekly meetings between Authors 1 and 3 to make sense of the sessions and plan the following one.(ii) Zaneta regularly joined those meetings to offer feedback and support sense making of the data.(iii) Insights and progress were also presented during the monthly project advisory board meetings, which Author 2 was part of.In those meetings, the advisory group interrogated the themes and provided their expertise for further interpretation and sense-checking.
The analytical process was supported by the co-creation methods introduced in section 3.1.1.From one session to the next, visual materials (i.e.character-driven scenarios) were created by Author 1 as a way to synthesise the conversation and highlight the experiences and issues voiced by the women.These were shared in subsequent sessions with the women as narrative prompts to focus the conversation and further discuss and build on what was previously shared.Thematic insights were consolidated throughout and shared with the women through summaries, once as the report translated into Czech, and in discussion with women.
A debriefing session was also organised with opportunities to reflect on the wider process and discuss the next steps including a community event scheduled in Autumn 2023.Four women agreed for their reflections to be video-recorded and assembled into a short film, which was shared at a stakeholder event.The study was approved by the the Faculty of Medical Sciences Research Ethics Committee at Newcastle University in December 2022.

FINDINGS
This section is structured in two parts: in Part 1, we report on three overarching themes from our thematic analysis that communicate women's experiences of healthcare services and needs during pregnancy.We focus on participants' lived experiences including the challenges for women to familiarise themselves with a new health system (5.1.1);their experiences of managing mental and emotional wellbeing (5.1.2),and other experiences related to self-management and self-care during pregnancy (5.1.3).Each section is introduced by a vignette of the Zine co-created in Workshop 8 (see Table 1 for details) to show how thematic insights were consolidated through co-creation.In Part 2, we provide insights into women's views and attitudes towards mHealth technology (i.e.pregnancy apps).We identify opportunities for mHealth (5.2.1) and challenges (5.2.2), particularly with regard to privacy and identity-related concerns when sharing personal and health information via digital technologies.It was important to consider women's stigmatised ethnicity when interpreting their accounts.Older women referred to degrading treatment experienced in their EU countries of origin, where compassion appeared to be lacking.The question of whether similar attitudes were displayed towards them as Roma in the UK was raised by two younger participants who gave birth in the UK.For example, Sofia who had suffered a twin miscarriage and was six weeks into another pregnancy recalled the doctor being dismissive of her symptoms: "He says, oh, well, for the sickness is nothing.And for the bleeding, well we will see.Some of the pregnancies have bleeding".Ruzena also recalled her requests being dismissed, and not being given any additional pain relief during birth: "[the nurse] came on to me and said, did you never broke your bones or anything?Is this your first pain that you are experiencing?Just take it as it is".Ruzena then questioned if she had experienced racism and reflected on how women of colour may be disadvantaged compared to white women: "I think if you're a woman of colour, you always think, is it just because she had a bad day or is it because, I mean the way I look".Participants reported differences in the way statutory services were provided in their home country and in the UK.Younger women (Sofia, Petra and Lucka) were more positive about the healthcare system in their countries.They described antenatal clinics and family planning services being available from puberty onwards, and women having direct access to specialist care, whilst GPs are only seen for minor ailments.In the context of pregnancy, women were unaccustomed to having GPs as their first point of contact in the UK, whereas they were used to directly accessible antenatal care specialists.Primary care in the UK was a point of frustration because every time, they saw a different doctor or midwife, which did not help build a trusted relationship: "So in Czech Republic you have one doctor, and he knows everything for years and years. ... So you coming here every time, 'what's your name?" (Petra); "Every time I went to see my midwife, it was always a different one [. ..] so you can't build the relationship. .." (Zaneta).
Being new to the UK and learning to navigate the different antenatal care pathways without proficiency in English made women's pregnancy more challenging.Petra explained that 'even though the doctors are friendlier here, she still had that worry-ness, she wasn't fully understanding everything, giving birth here' (Interpreter).Two other women described particularly traumatic experiences of care with no known recourse to complain or further assistance.One participant reflected on how assertive she had to be during her pregnancies to speak up for her rights -something she realised in the workshops: "[The workshops] just made me realise some things [. ..]That I was mistreated in my first pregnancy.Yeah, but in my third pregnancy I sort of speak up for myself, and when I didn't get the scans that I needed, [. ..]I was like, yeah, good on you, you asked for, you speak up for yourself and you knew it wasn't right that you didn't get so many scans that you needed, because my pregnancy was a high risk" (Ruzena).
Women discussed how HCPs can make assumptions about Roma women based on stereotypes that depict them as starting their families young and having several children.For example, one participant shared that she was not given much information (e.g., about parenting classes) when she was pregnant with her first child: "I was nearly 30, and the way they talk to me they thought I'm having already five children, but it was my first one.They wouldn't give me enough information" (Irma).It was also surprising that their entitlement to free prescriptions regardless of earnings, was not provided to the women by healthcare staff and at least three women reported being told they were not eligible for wrong reasons.
However, the majority of the women reported that healthcare in their home countries was more prescriptive in contrast to the UK where mothers have more choices, for example in their birth plans.This was the case of Denise (aged 30) who specifically moved to the UK to benefit from the greater choice available in particular more natural birth methods.

Managing mental and emotional wellbeing during pregnancy.
Women spoke about feeling distressed and upset at the start of their pregnancy journey for different reasons i.e. teenage pregnancy and whether or not to keep the baby, trauma linked to a previous miscarriage and general feelings of anxiety related to possible pregnancy complications.Regarding mental wellbeing, the women were uncomfortable sharing emotional feelings and discussing intimate information related to their pregnancy with their own mothers.This is not to say that practical help and care was not available (e.g.looking after the new baby) but feelings of shame and embarrassment were highlighted: "When [the women] were pregnant and they had questions, they wouldn't go ask their mothers saying oh, because of the respect and being embarrassed and shy.Just they couldn't have the conversation, 'how's the pregnancy', 'what's the expectation', 'can I ask you', 'how was your pregnancy' -it was never that relationship, because of the respect and shame, shame-ness and embarrassment" (Interpreter for the group).
The women spoke of beliefs and attitudes in their community about mental health that can discourage pregnant women from getting help, for example counselling, talking therapies or medical treatment for depressive illnesses.Mental health was considered as 'a big taboo' in the community, which was associated to fear of being judged by professionals and others in the community.Indeed, the women discussed their fear of hearsay in the community: for instance, if one of them was to open up to the community, people would talk behind her back, "and then the word would spread you know, [. ..] people would say like she's crazy, she's mental, you know" (Interpreter for Helena).Sharing about mental health was also associated with the involvement of social services and the risk of child removal.One participant later reflected on how the workshops encouraged the women to open up about their mental health: "I was very surprised with this project, that everybody opened up about their mental health.It was not taboo, which in our community, it's a big taboo.But everybody was so supportive of each other [. ..]And every woman, different ages in the group we had, said, it's important to talk about mental health" (Irma, Debriefing session).
This was the case with Denise (aged 30) who shared with the group how medication and attending counselling sessions helped  her address her depression and panic attacks, which she experienced during her pregnancy.With her story, she believed it was important for her community to talk more about mental health and that women should not feel ashamed to ask for help: "People should know they shouldn't be ashamed of it, that you might have anxiety or panic attacks, it's, today it's good to talk about it, to realise that it's good to have therapies or counselling, to get over your fears or the panic attacks or anxiety".
4.1.3Self-management and Self-care in the pregnancy journey.Selfmanaging sickness and diet were important topics.The women reported that extreme sickness was challenging and discussed remedies, which included dry biscuits, pickles, ginger lollies, and chewing gum with the most quoted one being Gaviscon (a type of medicine to treat acid reflux).They also highlighted the difficulties of eating well during the early stages of pregnancy.For example, Denise shared that she was losing too much weight because of being sick with vomiting in the first three months, which impacted her energy and mood.She further described how she felt like a 'prisoner' stuck at home because of her fear of fainting.
Overall, women were generally relaxed about their food intake: Gabriela was pleased she could eat more during pregnancy, Ivana believed that "you need to eat what the baby is telling you to eat!", and Irma who was craving fizzy drinks related that her doctor echoed that same advice to her.Eating what makes you happy was thought to make the baby happier: "I'm happy then the baby's happy and then, after the baby, I'll go to gym and then, exercise, which I did" (Zaneta).Ruzena challenged this reasoning and highlighted the need for younger mothers to pay attention to eating healthily, further observing how weight gain can increase risks of postpartum depression: "Once the baby's out, the weight is going to stay, but that's going to make you depressed. ..".
The women discussed different ways of looking after themselves.When creating the pregnancy journey, they chose to include images of nature and physical activities for self-care including walking, dancing and swimming.They also described the importance of having 'me time': "take time to get ready, put make-up on" (Ruzena).They shared compassionate messages and highlighted the importance of conveying a positive body image to help women accept that their body will change (i.e.stretch marks).Self-management was also described in relation to developing chronic conditions (i.e.migraines, high blood pressure, heart disease) that some believed were a result of their pregnancy, which they were still trying to manage.Smoking was recognised as detrimental to health but reportedly difficult to give up.

Part 2: Assessing women's views and attitudes towards mHealth
4.2.1 Perceived opportunities.The women perceived some value in using pregnancy apps: Sofia, Lucka and Gabriela commented on how the younger generation is more interested in technology and are digitally savvy: "I think it's good because the new generation, I mean everything is run by app these days, I think it's good, it's very handy" (Sofia).Ruzena who was introduced to an NHS-endorsed app (www.badgernotes.net)by her midwife for her third pregnancy felt that the app was a good alternative to the pregnancy book, which she did not have to carry around anymore.She also appreciated being able to monitor her baby's development, check scan and other test results and be reminded of appointments: "so maybe if I did not have the app, I definitely would have missed some".Another participant wished she could have used the app: "because, you could like track everything, all the information, instead of bringing the big massive book every time you go for your appointments" (Zaneta).
Women in our study reported using Google as their primary source of information on the internet.In our case, participants reported searching for information on the internet because they preferred not to ask their relatives about pregnancy and, they wanted to obtain the information promptly.For instance, Denise recalled using Google search to find out information about what tablets or pain killers she was allowed to take to manage her pain.She then observed how a pregnancy app would have helped her straight away: "In that moment you don't realise what you want to know or what you don't want to know until it happens.For example, she had a headache and then she did not know if she is allowed to take tablets or pain killer so she googled it and found out she was allowed but then again, that app would have helped her straight away" (Interpreter for Denise).However, the group reported that the quality of translation available online can be inadequate.
The women also discussed challenges related to their diet and accessing information recommending what foods to eat or avoid during pregnancy was found valuable.The visualisations of the baby developing and images of the changing body were also appreciated and were found to encourage some level of body awareness.For example, Helena commented on one app feature that visualises how the baby is growing week-by-week: "It's a miracle how you can create and see the vision of the baby already and how it's going to look".One older woman reflected that when she was pregnant, she was always wondering "how it looks like inside, so that interests her now knowing like the pictures, how would it look" (Interpreter for Helena).
All the women were comfortable communicating through their mobile phones and used Facebook Messenger to connect with each other.This was often led by Zaneta who would share practical information about events and organise informal chats (i.e.Facetime) with the group.The interpreter also shared that she was in charge of a Facebook page for the Roma community, which most of the participants were part of.Both Zaneta and Irma were trusted and acted as gatekeepers for their community.Overall, participants did not seem to be active in online communities, but they saw its potential for emotional support; for instance, two of the youngest women commented on one community feature: "It's good thing because you can support each other".When introduced to the chatbot feature, three women observed how it could provide mental health support, particularly for women with low mood or depression who may not feel like talking to someone.For instance, Gabriela commented that "sometimes people are shy to talk face to face, they'd rather type it in" and Helena reported: "Sometimes you don't want to talk to someone so talking to this app or robot and explain, and you know you can get the answer back, it's useful" (Interpreter for Helena).Sofia added that using the chatbot could be helpful if "you don't have the time to see somebody or it's just sometimes, you can't sleep in the middle of the night, so you can just go on the app and just you know. ..".However, two women were concerned about whether the chatbot could translate information instantly so that the women would not wait too long for a reply.Women also commented on safeguarding users and in the case of sharing suicidal thoughts, it was critical for the app to connect the woman with someone straightaway: "If the person writes down I have suicidal thoughts, would the robot connect to someone to call her.Would it?"(Irma) 4.2.2Perceived challenges.When stepping through the onboarding feature of one pregnancy app, women expressed mixed feelings about registering and sharing personal information about themselves and their family.Two of the women (Sofia and Marta) were positive towards sharing personal information if they knew it would help others or be beneficial for the baby.However, the group also expressed privacy concerns.Denise commented that she likes "having everything private" and with a pregnancy app, she would not want to share information about herself or family; "because it's very private information and it's not only about her, it involved other people" (Interpreter).When expanding on her privacy concerns, Denise shared that she never uses real information about herself on social media and created a fake profile on Facebook; "She hides her email everything because of the privacy because she knows her information will be shared" (Interpreter).Denise also recalled looking for women's groups online typing 'pregnant mummies' and making her decision for which group to follow based on the group with the most followers.She further said that she would follow the group but never post anything in it, re-emphasising her privacy concerns about not wanting to share information online about herself and her family.Throughout the workshops, we also learned that the women would seek help and share personal issues and experiences directly with the community worker.But it was a relatively new experience for them to share about a personal and sensitive topic like pregnancy in a peer group setting.
In the case of sharing demographic information related to identity or ethnicity, two women commented that they would need to assess the app first; for example, Sofia would not share information about her ethnicity with any type of app; "In here [this app] yes but in different types of app or different types of. .., probably not everywhere, it depends where".If the app was affiliated to or endorsed by a health organisation like the NHS, it was found more acceptable for the women to share their personal information; "If you see NHS, for me if it is the app and it's coming from NHS, it's all legal, there'll be nothing you should worry about" (Zaneta).However, there were several barriers to sharing information about identity or ethnicity.First, the options usually provided to define one's ethnicity are not representative or specific enough: "It's not very specific, with the ethnicity because it says always 'white European', it does not say the 'Roma European' so it's not really, real ethnicity, you never put the real ethnicity".(Interpreter for Denise) The women also talked about having unpleasant personal experiences in the past and two of them (Marta and Helena) shared their fear of being judged.For Marta, it was also not clear why information about her ethnicity was needed to be shared with the pregnancy app: "She thinks it's not relevant for them to know exactly about ethnicity, because she's taking it like we're all human so all the time she's wondering why they're asking about ethnicity" (Interpreter).Zaneta expanded on this by explaining that there were different types of Roma and in the UK, people who identify as Czech or Slovak Roma tend to be categorised under the term Gypsy-Roma-Traveller (GRT), which the women were uncomfortable with.As a result, the women would rather not share this information or preferred to indicate their ethnicity as Slovakian or Czech.
One of the app features we showed the women enabled the partner or husband to download the app and create a joint profile to access some information and follow the pregnancy.In this case, the women expressed mixed feelings: the women voiced privacy concerns and wishing to keep some things about their health and pregnancy private: "She thinks it's going behind her back checking on her or checking on the baby instead of checking with her directly to see if she is ok, if she needs any support.It's more about following something behind her back -because she sometimes wants to keep things private" (Interpreter for Denise).Ruzena and Denise both commented that it did not make sense for their partners to use the app because "it's not like he is going through it [pregnancy], it's the woman who is going through it" (Ruzena); "it's not like he is pregnant" (Denise).Helena agreed and both Ruzena and Denise felt that it was too much for their partner to download the app and instead, they would rather share the information themselves: "She would share all the information with him and he can check with her about all the information but to download the app she thinks it's too much" (Interpreter for Denise).Ruzena who used the NHS-endorsed app during her pregnancy further explained that this app did not have a sharing feature: "It was just for me, it's private, just for me and it's very simple".She also recalled showing the app to her partner, but he did not have any interest: "When I had my app, I said, look I am thirty-seven weeks and plus days, but he did not even look at it.[Laugh] There was no interest".These opinions contrasted with another woman who experienced fertility treatment and as a result of this, her partner was very involved in her pregnancy.In the case of using an app, she saw some potential for a sharing feature or joint account: "I have a different opinion.My partner would want to know exactly everything.Growing, what side effect, what they can support us with, he would want to know all the information" (Irma).Overall, there were mixed opinions to how much partners should be involved but it was important for the women to retain some level of control on their information and experiences: "I think you want to be more in control instead of like, oh, but then you do want him [partner] to be part of it [pregnancy]" (Zaneta).
Choice and control were important, and women had different views on what information they value and how often they should receive these.Lucka observed that daily notifications would be annoying: "I think it's annoying, like always something comes on your phone.If you want to check something, you will open your app and check".In the case of receiving personalised updates and information, women were unsure about the frequency, but Helena observed that in the case of a first-time pregnancy, women would want to receive personalised information on a daily basis.

DISCUSSION
Building on our insights and related work, we now turn to discuss the potential of mHealth for supporting women from marginalised communities.We frame a new design space and highlight a number of practical implications for preventing and addressing mental health issues (5.1), and supporting information seeking and sharing during pregnancy (5.2).We then argue for the need to adopt a critical and intersectional design approach for developing mHealth that best fulfils the needs of marginalised individuals (5.3).Summary of key takeaways are included in Table 3.

Preventing and supporting mental health issues during pregnancy
Workshop findings revealed how women struggled with their emotional and mental wellbeing during the antenatal period, whether their pregnancy was planned or not.This became more apparent as we progressed through the sessions, where women shared more about their experiences of anxiety, depression and helplessness.These insights were further consolidated with the Zine making activity where participants chose to include compassionate messages and prompts to encourage other women from their community to get help by sharing their worries or anxiety with HCPs and others (Figure 3)."I love this idea you know where you know Matylda can put this [Zine] in her, bag, in her pocket, little bit of information, little bit of everything, start with depression, how her body is going to change, mental health will change, it's okay to feel down, and to share this information with her professionals, doctors, nurses, midwife" (Zaneta, Debriefing session).The co-creation of the Zine prompted further discussion on the importance of addressing mental health during pregnancy, which was described as a 'big taboo'(Irma) in the community.Indeed, previous HCI research reports that mental health-related stigma is the primary barrier to care during pregnancy [31] but despite affecting a large number of women, mental health is still neglected in antenatal care [19,23,44].For women from non-Western or ethnic minority backgrounds, sharing about mental health issues may be less common because professional assistance can be perceived as inappropriate [60].There is also a 'cultural legacy' of being perceived as strong and resilient [11] and endorsing the 'superwoman persona' [44], which further prevents women from acknowledging and communicating about potential mental health issues.In our study, participants were likely to downplay their feelings because of being ashamed and embarrassed, and also, because of the perceived repercussions that mental health issues could have on their family; for example, the involvement of social services and the risk of child removal.Indeed, it was reported that Roma women in the UK are less likely to access maternity care because of their fear of social services [25] but despite this, our participants were keen for the perceptions of mental health in their community to change, particularly with regard to treatment for depression, counselling and talking therapies.Information resources co-created with the community can be a powerful way to develop new ways of conceptualising mental health and addressing stigma [56].Here we see the potential of the Zine method and resource as a first step toward raising awareness of mental health and wellbeing during pregnancy.This will be tested in future work where we will use the Zine at a community event with a wider group of Roma women, as a way to start the conversation on this subject and exploring ways to turn it into an information resource, which could be integrated into future design of supportive tools.
Next, we turn back to the potential of mHealth for providing mental and emotional support via interactive and web-based app features (i.e.chatbot) to facilitate self-reporting and sharing.Our participants expressed positive views towards using a chatbot for pregnant women who might experience mental health issues and who consequently do not feel like talking to someone directly or are unable to meet with their doctor.However, participants highlighted the need to be signposted or connected to a real person straight away if experiencing severe mental health issues (i.e.suicidal thoughts).More research is needed to understand how such features can be designed to support language needs and signposting to relevant resources.Here we point to the limitations of current features in pregnancy apps as it was reported that little information is provided about how, when or where to seek help and furthermore, this information is rarely up-to-date [23] and mainly in English, which poses challenges for women from non-English-speaking backgrounds [30].Furthermore, mental distress in mHealth technology tends to be portrayed as a personal issue [7] and building on this, recent work with refugee women advocates for reconsidering mental health as 'a social and collective experience and concern' [7 p.17].
On the other hand, we see an opportunity for further investigation to understand how remote and virtual means of communication can support or extend care to minority groups like the Roma community: for example, how web-based advice provided via mobile phones and online communication with HCPs (i.e.telehealth visits) may help reduce anxiety and stress for pregnant women [23,44].There is also an increasing interest in designing tools for facilitating the diagnosis of mental health issues early on in pregnancy.An example includes the deployment of a mobile app with women in the UK to facilitate remote screening and monitoring via selfreporting mood and depression throughout their pregnancy [19].Remote monitoring and self-reporting can arguably improve followup in routine care particularly if women are unable to see the same HCPs throughout their pregnancy (something that was described by our participants).Digital tools that support self-reporting and self-tracking can also empower women from minority groups to manage their mental health needs, including those who often mistrust and receive lower-quality care compared to White women [44].This is particularly relevant for our participants who have reported experiencing discrimination and the need to stand up for themselves and be assertive in their communication with HCPs.Further research with this group could focus on exploring the potential of designing self-reporting tools to support communication in routine care.Here we also see an opportunity to explore how such tools may support wider self-management and self-care practices during pregnancy; for example, via seeking and sharing information, which we describe next.

Supporting information seeking and sharing
It is reported that many women use the internet or 'Dr Google' [47] as a primary source of information during their pregnancy [59] and this was also reported by our participants.Previous HCI research has identified an information gap in antenatal care, more specifically in the first trimester, which leads women to find support online or through the use of commercial mobile apps [47].In our case, participants reported searching for information on the Web because they did not want to ask their relatives about pregnancy.An important factor was also the ability to source information straight away, which has been highlighted in other research [30].However, despite being the primary information site, the internet is also the least trusted and can become the source of worries and anxiety for pregnant women [51,55] or a space of harassment for women from minority groups [44].Previous research reports that trust and the perceived reliability of web-based pregnancy information increase when it is recommended by HCPs [1,23] and a review shows that clinically endorsed mHealth apps and expert opinions are desired by women [30].In the case of sharing personal health information, our participants found it more acceptable if the app was affiliated with a clinical established institution (i.e.NHS).But currently, the majority of apps are 'for profit' meaning that personal data is often shared (i.e.sold) with commercial entities, something users may not be fully aware of [3,72].Moving forward, research advocates for more collaboration with the health sector to ensure apps follow standardised practice and up-to-date content but notes the challenge of keeping up with the fast-growing market of mHealth [23].A recent initiative in the UK is the NHS Apps Library to assist patients and the public in finding trusted apps.However, the platform has now closed and instead recommends apps throughout the NHS website [41] but more platforms like this should be developed and sustained over time.
Privacy concerns related to using pregnancy apps were shown as low or non-existent in research with women from marginalised groups, aligning with previous research in this space [13].We provide contrasting insights in our work where women expressed mixed feelings towards sharing their information via a pregnancy app.This included privacy and identity-related concerns: as mentioned before, people from the Roma community often experience racial stereotyping and stigma, and some of our participants were uncomfortable with being categorised under the GRT (Gypsy-Roma-Traveller) group.This group is culturally rich and diverse but current options for describing ethnicity were found not representative or specific enough, and some were fearful of being judged because of identifying as Roma.Future research should focus on designing platforms or systems that address the complex identity and representation needs of women from migrant communities and further investigate the impact that both public and self-stigma have on data sharing.Our participants were also concerned with their privacy: one participant reported using fake data and described a sense of responsibility when sharing information via an app as the shared information was not only about the individual user of that technology but also her family (i.e.baby, partner), which she felt protective of.Here we connect to developing critical discourses in Digital Health research raising security and privacy concerns for the processing of sensitive data via Self-care apps [65], highlighting the potential for constant digital surveillance [38].Indeed, there are risks related to privacy and surveillance when capturing intimate data that extend beyond the individual user, to their relatives and acquaintances [3].Almeida et al. [3] offer design-led suggestions to ensure that future female-oriented technologies (FemTech) is inclusive of marginalised and vulnerable users.In our study, another aspect relating to privacy and intimacy was the sharing of data or information about pregnancy with spouses or partners.Previous mHealth research observes that this group tends to be excluded from the pregnancy and suggestions were made to design a platform on which partners could also participate and follow women's experiences of pregnancy [47].Our participants expressed mixed feelings about this, and conversations pointed towards the need for women to retain agency over their pregnancy experience and control for sharing their information with others including their spouse or partner.
Finally, insights from our work contrast with the assumption that women are social sharers and likely open to share about their private lives with others (i.e.family, strangers) in online settings [46].For example, one participant preferred adopting the role of the viewer rather than sharing (i.e.posting) anything with others online.This echoes findings from another study with low-income immigrant women [5] where they reported not feeling comfortable sharing their information with strangers and instead adopted a more passive and lurking behaviour for seeking pregnancy-related information and encouragement online.Such findings call for mHealth designs that require 'no role interactions' [5], intended for selfsupport and pursuing wellbeing during pregnancy more discretely.On the other hand, HCI research [5,11] also points to the potential of Prosocial Designs whereby technology-mediated systems or platforms could support a more place-based approach via a community-building form of social sharing and local caregiving.Designing prosocial systems can help immigrant women rebuild their social and organisational support structures by connecting with local support and organisational resources (i.e.transportation, childcare, wellness) particularly as women from this group are likely to be marginalised or excluded by the healthcare system [5].

Towards a critical and intersectional design approach for mHealth
Most of our participants were more confident speaking in the language of their country of origin, and critical to our engagement was the role of the interpreter to translate conversations.Participants also reported numerous instances of using Google Translate and highlighted the limitations of such a tool in terms of its accuracy and practicality.Here we build on recent research that stresses the need to address language and health literacy barriers for the uptake of web-based digital technologies.Indeed, current mHealth apps for pregnancy do not fulfil the needs of women from culturally and linguistically diverse backgrounds [30].Women who have limited English proficiency are likely to experience lower health literacy levels [2,37] and be further disadvantaged by the digitisation of healthcare [30].Connected to this are cultural and language barriers for accessing and critically assessing information found online [35].Levels of income and education also play a part in digital exclusion [14] and more research is needed to understand the intersectional factors that influence app uptake, use and utility [44], and to develop resources and training for improving women's access and use of digital health management systems [30].However, maternal mHealth design has typically taken a one-size-fits-all approach and rarely included women from racial and minority groups in research [44], more particularly women from culturally and linguistically diverse backgrounds (i.e.migrants) [30].There are some limited examples that have focused on developing culturally sensitive content for maternal health [18] emphasising the value of considering intergenerational knowledge, unconventional or community-led approaches and more holistic practices like mindfulness [44].But overall, there is a lack of representation and inclusivity in existing mHealth and a striking example of this is the predominant use of white-skinned women and 'pink-looking babies' in visual designs [72].Research in Digital Mental Health and Self-care also highlight the tendency towards medicalised approach for wellbeing overlooking individuals' diverse needs and sociocultural backgrounds [7,65].Future work needs to challenge the Western-centric biomedical approaches, norms and cultures relating to maternity therefore calling for a more Participatory Design (PD) agenda.Indeed, PD researchers have recently turned towards decolonising discourses [63] to shift worldviews and develop transcultural approaches to technology design in which 'narratives and counter-narratives are freely colliding to form joint reflected narratives' [79].Moving forward in this space, we see the development of PD processes and co-creative methods promising and complementary to phenomenological approaches described by Oguamanam et al. [44] in order to better integrate the experiences of marginalised groups whose needs tend to be overlooked and generalised.We contribute to this in the following section by highlighting our methodological insights.

Research limitations and strength
We acknowledge the limitations of this study.Firstly, we focused on a small participant sample with a majority of older participants and only four women who had children within the past five years Disrupt monolithic thinking by avoiding ethnic categorisation and stereotyping.Design for the complex identity and representation needs of women from migrant communities instead.
Consider a family-centred approach to design for different sharing preferences that enable women to retain agency and control over shared data about themselves and their family members; and to control what may be accessed between family members.Explore design of community-centred technologies for Roma women facilitating peer-to-peer sharing on a small scale and linked to physical places, to foster support and connections locally.
including one who gave birth recently (Ruzena).We build on previous HCI work that reported barriers to engage women from minority groups in the design process because of unpredictable working hours, caring responsibilities and living conditions [13].Despite working closely with the community worker, we experienced similar challenges for recruiting young mothers.However, our work aligns with best practice such as research co-production with leading community members whilst developing cultural sensitivity through multiple and longitudinal engagements [7,67].We adapted our research plan accordingly to fit our participants' needs (i.e.schedule) and used design-based methods to sustain engagement, which in turn strengthened participation.For example, we found the character-driven scenario method (i.e.Matylda, Figure 1) helpful for synthesising previous contributions and focusing conversation on particular topics at the workshops.Indeed, it provided us with an anchor point for staying close to our research population (young women from a Roma background going through pregnancy), and we observed how over time, the women became acquainted with Matylda.This method also worked well with our older participants who were able to empathise with younger women whilst prompting them to remember their own pregnancy and sharing their memories with our younger participants.This encouraged intergenerational sharing and the women were keen to share with others what they learned from the workshops: "I learned lots of information, not for me, because I won't get pregnant again, but for my kids, I will be support them to have more information, so I can help them [. ..] pass it the information that I get in here" (Gabriela).
Unpacking our methodological contributions is beyond the scope of this paper but we highlight key insights based on our work in Table 4 for other researchers working in similar spaces.We want to highlight our contribution to the growing HCI portfolio of creative methods generated from research with people from minority ethnic backgrounds [8] including migrants [11,15] and refugees [69].

CONCLUSION
In this paper, we have reported on a series of Co-creation Workshops with 11 Roma women to investigate their experiences of pregnancy and perceptions towards antenatal mHealth technologies.The Roma community experiences significant health disparities and discrimination in healthcare settings therefore providing us with a complex socio-medical context for exploring how to better support the antenatal care needs of ethnically marginalised women.By drawing on Critical Digital Health, Intersectional HCI and Women-centred Design discourses in HCI, we have highlighted the limitations of current mHealth technologies pointing to design opportunities for better accommodating the experiences of women from marginalised groups.We discussed the following implications; firstly, the need to design for the emotional and mental wellbeing needs of pregnant women whilst gaining a greater socio-cultural understanding of mental health to address barriers like stigma.Our findings re-emphasised the potential of mHealth technologies to Use creative and participatory methods to encourage reflection, discussion, making, and ownership of participants' creations; Provide a celebratory debriefing session and opportunities for contributing to reflective feedback on the research process and outcomes e.g, we created a short film that was shared with key stakeholders in a webinar context.Provide opportunities for participants to take part in dissemination activities e.g.community-led event.
support information seeking and sharing needs but greater consideration is needed to address women's concerns related to online harms.Finally, as digital technologies are increasingly used and becoming more pervasive with the integration of apps into routine antenatal care [30], we have offered empirical insights that further reinforced the need for a more critical, intersectional and womencentred design approach to mHealth technologies.To conclude, we evidence qualitative understandings of migrant women's health and wellness needs through using co-creation as a vehicle for research engagement, further addressing the research gap in HCI on research with immigrant populations [5,11].

Figure 1 :
Figure 1: Snapshots of the co-creation activities: character-driven scenario, pregnancy journey and Zine making.

Figure 2 :
Figure 2: The co-created Zine printed with key messages and contributions from participants.

4. 1
Part 1: Understanding the health and wellness needs of Roma women 4.1.1Familiarisation and discrimination within a new health system.

Figure 3 :
Figure 3: Snapshot #1 of the Zine: The women chose to represent key steps when having a baby in the UK.Online resources were also co-identified about the planning and timeline of the pregnancy.

Figure 4 :
Figure 4: Snapshot #2 of the Zine:The women highlighted the different emotions they experienced using emojis to express the ups and downs of the pregnancy journey.Online resources were also co-identified about mental health and pregnancy.

Figure 5 :
Figure 5: Snapshot #3 of the Zine: Detail showing the Zine unfolded.The women included images related to self-care and featured compassionate messages related to both physical and mental wellbeing.

Table 1 :
Summary of the workshops and topics explored.

Table 3 :
Summary of key takeaways.

Table 4 :
Summary of our methodological insights.insights Description Engage the community through a gatekeeper Utilise existing trusted networks and use community liaison colleagues (gatekeepers) to recruit participants in the community; Provide gatekeepers greater opportunities for research involvement through peer researcher training and mentoring; Utilise a community venue that is familiar, accessible, and acceptable to participants for the research engagement.Conduct research in a trusted place.Co-Creation in a trusted space Use the first meeting to build rapport, go through the information sheet and consent form together with the gatekeeper, answer any questions using clear language and provide visual aids (i.e.printed workshop guide, table 1); Plan for the space to be familiar to them and child-friendly, and make provision for participants' needs e.g.we provided breakfast with different food options according to participants' preferences each time we met; Agree on ground rules i.e. each person should get a chance to speak, one at a time; personal stories stay in the group; agree to disagree and treat one another with respect.Have a flexible work plan to address women's Build in time to be flexible to accommodate the needs of the group members, e.g., needs (i.e.childcare) childcare, school holidays, work commitments; Provide the option for interpretation and translation but allow participants to speak in English if they are comfortable to do so; Translate all shared materials and use visual displays e.g., projected websites, PowerPoint presentations, charts, to share information and aid discussion.Embed co-creation throughout