• Open Access

Parents’ views on how health professionals should work with them now to get the best for their child in the future

Authors

  • Joyce L. Marshall PhD MPH BSc (Hons) RGN RM,

    1. Senior Lecturer, Department of Human and Health Sciences, University of Huddersfield, Huddersfield, West Yorkshire
    Search for more papers by this author
  • Josephine M. Green BA PhD CPsychol,

    1. Professor of Psychosocial Reproductive Health, Mother and Infant Research Unit, Department of Health Sciences, University of York, Heslington, York
    Search for more papers by this author
  • Helen Spiby MPhil RGN RM

    1. Professor of Midwifery, School of Nursing, Midwifery and Physiotherapy, University of Nottingham, Queen's Medical Centre, Nottingham, UK
    Search for more papers by this author

Joyce L. Marshall
Division of Maternal and Child Health
The University of Huddersfield
Queensgate
Huddersfield
West Yorkshire
HD1 3DH
UK
E-mail: joyce.marshall@hud.ac.uk

Abstract

Background  Pregnancy and the first years of life are important times for future child well-being. Early identification of families and children who might be likely to experience poorer outcomes could enable health professionals and parents to work together to promote each child’s well-being. Little is known about the acceptability and feasibility of such an approach to parents.

Objective  To investigate parents’ views about how health professionals should identify and work with families who may benefit from additional input to maximize their children’s future health and well-being.

Design  A qualitative study using focus groups.

Setting and participants  Eleven focus groups were conducted with a total of 54 parents; 42 mothers and 12 fathers living in the north of England.

Results  Parents welcomed the idea of preventive services. They strongly believed that everyone should have access to services to enhance child well-being whilst recognizing that some families need additional support. Making judgements about who should receive additional services based on specific criteria evoked powerful emotions because of the implication of failure. Parents projected a belief in themselves as ‘good parents’ even in adverse circumstances.

Conclusions  Targeted additional preventive services can be acceptable and welcome if health professionals introduce them sensitively, in the context of an existing relationship, providing parents are active participants.

Background

There is accumulating evidence that conditions in pregnancy and the early years of life have the potential to enhance or be detrimental to a child’s physical and emotional well-being,1,2 educational attainment3 and future life chances including economic status.4,5 Recognition of the importance of this has led to a shift in emphasis towards future orientated services and significant investment in early years care and provision in the United Kingdom (UK), including increased support for parents through (Sure Start) Children’s Centres. These are sited in disadvantaged neighbourhoods but are universally available to all parents living in that area.

Policy in the UK has for some years advocated the provision of a universal core programme of health promotion (including screening and immunization) with extra support for vulnerable families to enhance each child’s well-being.6–9 This has been considered an effective way of improving children’s health in the context of increasingly scarce professional resources.1,10,11 This approach was also advocated in the Marmot Review to reduce health inequalities:

‘Focusing solely on the most disadvantaged will not reduce health inequalities sufficiently. To reduce the steepness of the social gradient in health, actions must be universal, but with a scale and intensity that is proportionate to the level of disadvantage.’5

(p15)

To operationalize such an approach, it is necessary to be able to identify those families to whom additional services should be offered. When the goal is the child’s future health and well-being, rather than a response to immediate needs, this may not be obvious.

A range of risk factors have been identified from longitudinal epidemiological studies. One of the most consistent findings is that children from poorer households tend to have worse cognitive and behavioural outcomes and lower educational achievement than children from wealthier households.12,13 Hobcraft and Kiernan14 used data from the UK Millennium Cohort Study to identify factors during pregnancy and shortly after birth that were associated with child outcomes aged 5 years. Outcomes assessed included learning and development using Foundation Stage Profile; behaviour using the Strength and Difficulties Questionnaire and health status based on mother’s reports. Factors which were independently significantly related to child development included delayed childbearing, positive parenting, good maternal health, maternal education and employment status.14 Those that were predictive of positive child behaviour were as follows: family context (in particular, lack of conflict), lack of maternal depression, some maternal qualifications and higher incomes.14 However, the predictive value of these models is not strong and they do not replace professional judgement.

If families are to be selected to receive personalized additional services based on increased risk of poor outcomes, then it is important that such an approach is acceptable to these families. A recent government White Paper advocates increased local involvement and puts individuals in the ‘driving seat for all aspects of their and their family’s health well-being and care’.15 (p24).There is however a paucity of literature relating to what parents themselves think about the idea of receiving targeted additional preventive services now to improve children’s health and well-being in the future. Therefore, the aim of this study was to investigate parents’ views about how health professionals should identify and work with families who may benefit from additional input to maximize their children’s future health and well-being.

Methods

Study design

Focus groups were chosen as the most appropriate method to explore participants’ attitudes, beliefs and reactions to the provision of additional services to improve future child health as richer data can be generated through conversations between group members as well as with the group facilitator.16 Ethics committee approval was granted for the study, and research governance procedures were followed.

Recruitment, sampling and participants

English-speaking parents were recruited predominantly from existing groups in Children’s Centres and National Health Service (NHS) antenatal classes, the majority of these in less affluent areas of three different towns in the North of England. Non-English-speaking Pakistani parents were recruited through a specialist maternity co-ordinator, a Pakistani community centre and English for speakers of other languages (ESOL) classes in a fourth town. It was considered appropriate to recruit in more than one locality to guard against the undue impact of local factors. Mothers and mothers-to-be were seen as the principal informants, but fathers’ views were also sought, both in groups of couples and in a fathers-only group. Purposive sampling to include some parents during pregnancy and some who already had young children was considered important to elicit a range of perspectives. For example, parents who have pre-school children are likely to be able to draw on personal experiences to look back on the input they would have liked.

Existing groups were identified via key people in maternity services and Children’s Centres. A researcher then attended each group, with the permission of the group facilitator, to provide full information about the research and to recruit potential volunteers who were contacted later to confirm attendance. A £10 supermarket voucher was offered to each participant. There was generally 1 week between the introductory session and the focus group.

Data collection

Focus groups were carried out between October and December 2009. The aim was to represent a range of parental groups and geographical locations. This resulted in the identification of 11 focus groups. English-speaking groups were facilitated by an experienced researcher, supported by a second researcher. The non-English-speaking groups were recruited and facilitated by an experienced bilingual researcher. Groups were generally conducted in the locations where participants were used to meeting and, where possible, at a time when they would already be on the premises, for example before or after a regular group session. They were usually held around the middle of the day and lunch was provided. Immediately before the formal start of the discussion, participants were given a list of statements to read to encourage them to think about what they might want for their child in the future and the kinds of things in their lives that helped or hindered this (Fig. 1). This was used to facilitate initial discussion and to encourage parents to focus on the idea of future orientated conversations – to think about how they might want health professionals to work with them now to improve their child’s outcomes in the future. A topic guide in the form of a poster visible to all participants (Fig. 2) showed the broad areas for discussion around this central topic. Sessions were arranged for one and a half hours, with most discussions lasting approximately 1 hour.

Figure 1.

 Example of questions to stimulate future orientated conversations.

Figure 2.

 Poster used to structure and facilitate focus groups.

Analysis

All focus groups were audio recorded, fully transcribed and checked for accuracy by comparing the transcript with the sound file. The groups with non-English-speaking Pakistani parents were translated and transcribed by the bilingual researcher. All transcripts were reread several times to ensure familiarity with the data. Conceptually, child health and well-being were considered broadly to encompass psychological, physical and social factors and therefore when services are referred to these span health and social aspects of care. One researcher produced an initial hierarchical coding framework that was based on concepts relating to the aims of the research17 and issues prioritized by participants. This was further refined and developed through discussion between all three researchers. Codes included the following: child health and well-being (parenting, social environment, finance, lifestyle, genetic, education), health and social care (providing information, able to choose, service provision, interactions with health professionals, judging/labelling and targeting services) and data collection/linking. Parents’ views about data collection and storage were not the issues of concern to parents and are therefore not reported in this paper but can be found in the full study report.18 Data were coded line-by line using the software package NVivo. Codes were added and refined as analysis progressed. Data within each code were initially scrutinized in detail, and this was followed by considering linkages across codes. Comparative analysis was carried out using charts to highlight similarities and differences across key concepts (groups of linked codes) as we sought to make sense of parents’ views and provide explanations.19 Key concepts included targeting services, choice and control, future orientated thinking/services and preferred service provision. Reflexivity was encouraged by discussions between all three researchers, with different views and backgrounds (two clinical and one psychosocial) at all stages of analysis and writing. Trustworthiness was enhanced by initial familiarization and continually interacting with the data throughout the analysis, and this included checking the data carefully for disconfirming cases. To protect participants’ anonymity, all names have been removed from the transcripts. Each participant has been assigned a code consisting of a letter M (mother) or F (father), followed by the number of the focus group and a unique identifier allocated by order of speaking.

Results

A total of 54 parents participated in 11 focus groups (Table 1). Groups of between six and ten participants were sought as the optimum size to encourage interaction.16 More parents than expected attended one group but typically not all who signed up attended, which resulted in three groups each having only two participants. Another group had to be abandoned when only one person attended. Personal details, such as age and financial situation, were not requested from participants, although many were volunteered during the course of group discussions. A few parents had professional occupations but others alluded to social and economic difficulties in their lives that represented a range of social disadvantage. Three groups included teenage parents. Many of the Pakistani women had been brought up in Pakistan and had recently arrived in the UK.

Table 1. Focus group participants
Group numberTotal n = Description
  1. *Three mothers attended, but one was engaged with child care and did not participate.

English-speaking parents
  1 2*Mothers of toddlers who were both also in early pregnancy
  2126 couples in late pregnancy, all first time mothers
  3 64 mothers + 1 couple, all mothers in late pregnancy with first child
  4 6Mothers of young babies, most with other children
  5 2Mothers of pre-school plus older children
  6 4Mothers of young babies, 3 also had toddlers
  7 2Mothers in mid pregnancy, each with an older child
  8 2Fathers, both had young baby plus older child
Non-English-speaking Pakistani parents
  9 63 couples with pre-school children plus older children
 10 5Pregnant women, 2 with no other children, 1 with a toddler and 2 with older children
 11 7Mothers of pre-school plus older children

Parent’s beliefs about why some children do better than others

When thinking about factors that could affect children’s health and well-being, parents in all focus groups believed in the importance of a child’s social environment, including the immediate family situation, the wider family network and the neighbourhood or community that determined the child’s peer group and schooling. In addition, a supportive family environment was given prominence. Emphasis was placed on activities such as spending time with children and developing a good relationship with them, teaching children right from wrong (either by being a good role model or through discipline) and receiving support from the wider family (e.g. grandparents). Not taking an interest in children’s activities and family conflict, particularly differences of opinions about children’s upbringing, were considered to have a negative impact on children.

Parents felt that other factors such as geographical location, the family’s financial situation and lifestyle interacted with these social influences. The interplay between genetics and environment was also highlighted in that parents thought that some children would do well even when exposed to adverse social situations. Whilst this was usually ascribed to the child’s personality or ability, the parental role was also emphasized:

M6.2:... I think it don’t really matter what the area’s like, as long as you look after your kids and teach them right from wrong. I grew up on a Council estate but I’ve never touched drugs in my life, so, and that’s from my parents. They … tell you the good and bad of it and then it’s your decision what you’re gonna do with it…

Future orientated thinking about child health and well-being

Most respondents believed that thinking about their aspirations for their child to identify what they could do now to improve the future health and well-being of their child was a ‘good idea’ because it encouraged them to think about things they might not otherwise have considered.

M3.4: Yeah, I suppose the, the points on it [future orientated questions] make you think so rather than just bowling on not thinking about anything, it actually makes you sit down and think ‘mm, yeah, what do I want for my child?’

However, this kind of thinking was not generally considered easy, especially for parents having their first baby. Perhaps unsurprisingly, the difficulty encountered in thinking about aspirations for children seemed to be due to the abstract nature of it. Once parents had concrete experiences to which they could relate their ideas, further thinking became easier. Acquiring information and having discussions about child health and well-being to plan for their children’s future were generally considered to be a process rather than a single event – a process during which parents’ ideas and thinking could be modified and changed to incorporate further information, understanding of their child and to accommodate life events as they occurred.

The challenge of future orientated thinking about child well-being was reflected in some of the suggestions made by parents about how and when it should occur. All who commented felt that face-to-face discussion was a very important part of the process and it was generally assumed (perhaps, shaped by their understanding of existing services) that these discussions would be with a midwife or health visitor. A key aspect was considered to be the existence of a relationship with the health professional prior to such important, and potentially threatening, discussions.

Services available to all

When considering services to support parents with the aim of promoting child health and well-being, respondents across all focus groups felt strongly that everyone should have access to such services. However, it was recognized that some people might need more support than others.

M4.1: Mm. It’s like someone with …say depression or something, not everyone has it, do they? So it’s …‘one size fits all’, it doesn’t fit all cos it, it doesn’t. If you haven’t got depression you’re not gonna go to some services …

Parenting was not perceived to be easy and was seen as a dynamic process with high and low points; as several parents pointed out, anyone could potentially feel the need for extra support or services and it was not always possible to predict this. Thus, there was a general feeling that a range of services should be available for parents to access if they chose to do so.

There was a recognition that targeted services could result in missing out on some services if individuals were not seen as meeting certain criteria:

M5.2: I feel a lot of Children’s Centres … they’ve been put out to be for everybody, but then when you actually get down to the nitty-gritty of everything … well we have to be seen to be doing stuff for the under privileged. But, again, who’s to say everybody’s not under privileged? We are, everybody is in some, some way, do you know what I mean, some people show it more than others. So how can you specify, yes, they’re under privileged, no, they’re, they’re not?

The mothers in this group could be considered to be ‘under privileged’ in that they were living in low-income households with low educational attainment but they were married and were not receiving benefits. They clearly articulated what they felt they had gained from universal access to free services:

M5.1: But when you’ve got two [children] and when you … haven’t got a lot of money if you’re stuck at home, you know, you just get in a rut, don’t you?

M5.2: That’s it, you’re living for the next pay day, aren’t you? …

M5.1: So if, if these facilities are there you’re getting out of the house, you know, you’re, you’re mixing with other mothers, you’re mixing with people…

‘Judging you’: views about targeted services

In addition to the issues of missing out on services, parents were asked how they would feel if they were to be offered certain services because they belonged to a group with an increased likelihood of poor child outcomes. Although every effort was made to phrase this question neutrally, the reaction was very negative, using powerful words like ‘gutted ’, ‘patronized ’ and ‘insulted ’. Parents in many groups were concerned that this would make them feel they were being assessed, stereotyped and judged and that their abilities as parents were being questioned. The following extract was spoken in tones of heavy sarcasm and incredulity:

M7.1: So you’re gonna actually have to say to these individuals, there’s a probability owing to your circumstances that your child is not gonna do well and could (a) have this, (b) have that, so therefore now we’re going to give you this extra support to try and prevent that happening. I mean it could upset (…) (laughs) couldn’t it?

M4.4: From like the stereotypical, oh you’re a kind of young mum or you’re a single parent. It’s, it’s hard to actually, cos you do constantly get judged …

These reactions appeared to be underpinned by the need to feel that they were good parents who were doing the best for their children – they did not want to be perceived as failing.

M1.1: I’d have been a little bit gutted actually (laughs) cos I’d think why, am I not doing a very good job?

However, when this question was asked of Pakistani parents, framed in relation to inability to speak English, the responses were much more positive.

Targeted services appeared to be most acceptable where a current problem exists (such as depression or inability to speak English) about which something could be addressed. Extra support in such situations is perhaps less likely to threaten parents’ view of themselves as ‘good parents’. However, this is a somewhat different scenario from the future orientated philosophy, which is based on epidemiological predictions of adverse future outcomes. Generally, criteria such as being young, not financially well off, living in particular geographical areas or being less well educated were not considered acceptable criteria upon which to make judgments. Parents recognized the potential for considerable variation within each of these categories and many provided examples from personal experience to illustrate this. For example, M3.4 made the point that where she currently lives does not reflect her upbringing and her sense of self, but rather her current financial circumstances:

M3.4: I was brought up in a really nice, big house in the middle of nowhere but now I’m out there on my own, I’m twenty-three … and I’ve … to get myself on the property ladder I’ve moved in, into town,... But obviously then will your perception of me be different because I live in a cheap little terraced house … that doesn’t change who I am … it could possibly offend people that you think well just because of where I live…

M3.2: Doesn’t mean to say that I’m … not capable of looking after my child…

Discussions about child health and well-being were considered to be an on-going process in which health professionals could play an important role. Whilst on the one hand, parents felt that health professionals should have a role in deciding which families required more support services, parents across all groups felt that such decisions should not be made on the basis of a 10-minute visit, but should be made once they had developed a relationship or rapport.

M2.4: I think it would also help if someone was to come to me and say, I think you need extra help I’d like to think that it was somebody that knew me, that spent time with me and saw the circumstances that I was in, not somebody just walking in and having one look around and saying well you need help here.

Some parents were unsettled by the handover from midwife to health visitor and had concerns about the health visitor’s child surveillance role. Several respondents mentioned that they would like to meet the health visitor during pregnancy rather than have to cope with the introduction of this new professional at a time when many felt most vulnerable.

Overall, there was a sense that targeted services could be acceptable and welcome if health professionals introduced them sensitively. For example, health professionals had sometimes encouraged women to attend groups that they thought might offer them support and this had been considered to be helpful.

Support from groups and from health professionals

Almost all women (except the pregnant Pakistani women) spoke of various ways in which they felt supported by attending a range of different groups. They gained from the social contact for both themselves and their baby. In order to learn parenting skills, women prioritized listening to real-life experiences, this seemed to provide embodied knowledge akin to the idea of an apprenticeship for parenthood.20

M3.2: … they did things like baby massage and all sorts of stuff so because there are mothers there with babies, you can go along and see what’s going on for when your baby is born. So it kind of helps that transition from, … being pregnant to ‘ah! there’s a baby’ kind of thing, what am I going to do now, nobody’s told me. So I’ve really found that helpful

Such ‘real life’ experiences seemed to be easier for many to understand and remember and could alleviate anxiety. The informal nature of group settings was appreciated. Women gained much information from their peers in these settings, but they also valued the presence of a health professional at the groups as it offered regular opportunistic contact that they would not have otherwise had. This often provided an opportunity to build a rapport that seemed to make it easier to discuss any concerns. Such regular informal contact may contribute to the development of enhanced parenting skills in a way that is not seen as threatening but supportive. This was contrasted with interactions with some health professionals where parents felt that they were simply being told what to do.

It was important that parents perceived input from health professionals as supportive – perhaps as a ‘gentle nudge in the right direction’ (M3.3), leaving parents feeling they were in control and had choices rather than being told what to do in a ‘bolshy’ (M6.2) or ‘bossy’ (M4.6) way. Parents wanted information to help them to prepare for future events, and specific, detailed information about healthy behaviours and services available to them – balanced information that was not ‘wishy washy’ (M7.2) ‘explaining the pros and cons’ (M10.1) to enable them to make their own decisions.

Discussion

This qualitative study has explored parents’ views on how health professionals might work with families in pregnancy and early childhood to identify those who may benefit from additional input to maximize their children’s future health and well-being. Whilst parents thought being encouraged to consider their aspirations for their child was a good idea, they did not find this easy to do because of the abstract nature of such thinking. Access to free universal services for young children was valued highly. Parents recognized that not everyone would want or require the same services and a key point was that parents felt that they rather than health professionals were best placed to judge whether or not to access services. Making judgements about who should receive additional services based on criteria such as being young, not financially well off, living in particular geographical areas or being less well educated were not considered acceptable because of concerns about stereotyping and the need to be seen as good parents. However, there was a general feeling that this could be more acceptable at an individual level if performed sensitively within the context of an existing relationship with a health professional.

There were some limitations of this research. The extent to which participants were able to engage with the topics under discussion appeared to be influenced by previous and current use of services and their parenting experiences. Several of the Pakistani mothers had recently arrived in the UK; therefore, their knowledge of the NHS may have been rather different from that of other respondents. As participants were recruited through existing groups, the views of the most socially disadvantaged families may not be represented as they are least likely to take part in research.21,22 Although only two participants were recruited to three of the groups, this did not seem to be detrimental to the quality of the data.

There is clear evidence of the relationship between social disadvantage and poorer outcomes.5 The difficulty of reaching these families has been well documented over several decades – the ‘inverse care law’ remains relevant.23–25 This considered alongside the findings of this study present a conundrum – how can services be provided to disadvantaged families who are least likely to access services if targeting is not acceptable? This is an underlying aim of Childrens’ Centres. These are purposefully sited in disadvantaged neighbourhoods but are universally available to all parents living in that area and thus provide free universal services to parents. The recently recommended remit of Childrens’ Centres is to:

‘maintain some universal services so that Centres are welcoming, inclusive, socially mixed and non-stigmatising, but aim to target services towards those who can benefit from them most.’9

(p7)

Our findings indicate that parents are sensitive to any implication that they are ‘failing’ and that being steered towards appropriate universal services, such as those within Childrens’ Centres is likely to be more acceptable. This is in line with a recent government White Paper, ‘Healthy Lives, Healthy People’ that advocates a partnership approach to health throughout life with the underlying principle of ‘proportionate universalism’ (p32) – whereby action increases with the level of disadvantage.15

Offering additional targeted services for those who are likely to gain most from them has the potential to use scarce resources to best effect. However, explicit targeting of services even in the presence of some universal services could still lead to parents feeling ‘stereotyped’ or being concerned they would ‘miss out’ if they were not selected to receive such services. This paradox has been described in other similar studies; for example, Roche et al. (2005)26 explored parents’ views on child health surveillance and health promotion programmes and found that parents often described contact with the health visitor negatively as they felt they were being judged on either ‘appearances’ or markers of socio-economic status, whereas other parents felt excluded from accessing support. It seems important that interventions and/or support offered should be offered sensitively based on the circumstances of each individual family – a ‘one size fits all’ type approach was not seen as acceptable.15

Mothers appeared to benefit from attending groups. They emphasized the importance of social support and learning from ‘real life’ experiences of other mothers.20 This finding should be interpreted with caution as our sampling method will have favoured those who value groups but this has also been highlighted in other studies.27,28 Providing easily accessible local groups with a health professional present may offer at least a partial solution as, in addition to the advantages above, mothers also valued regular informal contact with an known health professional. This facilitated easy access to further services if required and enabled parents to resolve issues at an early stage.

The non-personal nature of offering services on the basis of neighbourhood or parents’ age might be expected to be a more acceptable way of selecting families to receive additional services, but this elicited a strong negative response in this study. The interpretation of what targeting actually is may be important. It may mean some people are excluded from services (not acceptable to participants in this study) or it could mean ‘nudging’ people towards services. Offers of additional services from health professionals are likely to be more acceptable within the context of an existing relationship and if they are presented sensitively as suggestions. If the ‘offer’ of additional services is mishandled, this could potentially lead to upset and non-engagement with services.26

The involvement of parents in local initiatives may also be important. In England, Public Health Support Teams were established in 2006 with the aim of providing support to improve infant, child and maternal health outcomes to disadvantaged groups and areas.29 By visiting areas and carrying out interviews with key local partners, they aimed to build-up a picture of maternal and child services and support to develop action plans based on local needs. One of the key recommendations from this work is that community engagement is necessary to understand the needs of the local population in order to provide flexible, responsive and acceptable services. The findings from our study support this approach in that parents felt they were best placed to decide whether or not they would access services and being involved in the development of services may be the next step to achieving further engagement with services.

It may be that health professionals have a relatively small part to play in the support some parents receive. Perhaps, for those who are happy to attend groups, efforts should be focused towards helping parents to build-up their own social support network. For others, perhaps, more structured individual support such as that provided by the Family Nurse Partnership (FNP) may be required.30,31 Early evaluation of the FNP approach in the UK is promising,32 although results of the on-going randomized controlled trial (RCT) are awaited. Whatever the level or type of support or service, a key factor appears to be parents’ ability to build a positive relationship with the health professional33,34– one that is not perceived as judgemental. This may be easier to achieve within the context of increasing investment in health visitor services in the UK as set out in the Health Visitor Implementation Plan.35

In summary, making judgements about who should receive additional services based on specific criteria from epidemiological studies evoked powerful emotions because of the implication of failure. However, targeted services could be acceptable and welcomed if introduced sensitively by a known health professional providing parents remain active participants within this process. These important findings will need continued consideration within the context of rapidly developing policies that aim to improve child health and well-being.

Acknowledgements

We thank all of the focus group participants and those who enabled access to them. We acknowledge the contribution to the development of the focus group topic guide and arrangement and facilitation of focus groups by Sue Hennessey and Felicia McCormick, Research Fellows, The Mother and Infant Research Unit and Sara Christensen for transcription services. We are particularly grateful to Dr Shenaz Ahmed, Leeds Institute of Health Sciences, University of Leeds, for facilitating and translating the non-English-speaking groups with Pakistani participants.

Sources of funding

Funded by the Department of Health via a grant to the Child and Maternal Health Observatory (CHIMAT).

Conflict of interest

There are no conflicts of interest.

Ancillary