We took a mixed-methods, three-phase approach to development of the guidance. Phase 1 involved the collection and collation of research evidence around the impact of different types of weight and health communication between parents and children on children’s wellbeing. Where evidence was not available, we addressed these gaps through primary research, stakeholder engagement activities, or drawing on evidence in related settings (23). Phase 2 involved translating the key points from the evidence review into specific content of a guidance document. Phase 3 involved a modified Delphi process to refine and agree a final version of a guidance document for implementation. This final phase facilitated the inevitable compromises needed to navigate between academic evidence, clinical practice expertise, acceptability to stakeholders, and the practicalities of producing a readable and usable resource.
Phase 1: Collection and collation of evidence
1.1 Systematic literature review
A systematic review was initially conducted to analyse the association between parent-child weight-talk and child wellbeing (18). Well-being was defined broadly to include positive and negative mood states, mental health and ill-health (including dysfunctional eating), and body dissatisfaction. Full methods and findings have been previously published (18), however in brief, 38 studies reported on associative outcomes that provided useful insights, but only four were interventions. Of these, only one isolated parent-child communication from other aspects of weight management approaches. A second, broader review was subsequently conducted to synthesise research on the causal links between communication between parents and children in this and other health domains where stigma or concern for wellbeing may deter parents from initiating a discussion (23). This latter review drew on data published up to April 2020, searching five databases and would have identified any more recent studies meeting the inclusion criteria of the earlier review.
1.2 Review of existing guidance
To ensure that new guidance was necessary and would build on current good practice, we searched for any existing guidance through a grey literature review of documents available between October 2019 and April 2020, and updated in September 2020, using the Google search engine. The inclusion criteria were for guidance available in English, aimed explicitly at parents about communication with children about body weight or size. Broad key terms (e.g., ‘child weight talk’) and questions (e.g., ‘how do I talk to my child about weight?’) were used to run searches, and retrievals from the first five pages of results were screened, including informal sources such as blog posts and newspaper or magazine articles. Professionals working in public health and child psychology identified from the authors’ professional networks were also contacted to advise of any known resources. Then a priori criteria for determining the need for new guidance are set out in Table 1.
Table 1
Criteria for determining the need for new guidance
Requirement | Rationale |
---|
Absence of guidance developed or adapted and tested in the UK | - to maximise acceptability and relevance to parents living in the UK |
Absence of guidance with a clear research evidence base | - to be in line with requirements for evidence-based practice in public health - to maximise acceptability to HCPs, and trustworthiness to all stakeholders |
Absence of comprehensive content, including guidance for parents on (i) deciding whether or not to talk to children about weight (ii) raising the topic of weight (iii) continuing conversations about weight | - to address expressed needs of parents, identified through past research and PPIE - to address expressed needs of HCPs in terms of the questions parents ask of them |
Absence of guidance reflecting a biopsychosocial model of obesity (i.e., acknowledging environmental and social determinants of obesity). | - to align with current public health focus on the importance of a systems-wide approach to obesity prevention (24) |
Notes: HCP – Health Care Professional, PPIE – Patient and Public Involvement and Engagement |
Fourteen guidance documents were identified (Additional File 1) which had been developed/updated from 2010 to 2019 (where noted). Not all were comprehensive (e.g., providing guidance solely on whether or not to talk to children about weight, rather than how to go about it). Seven were produced in the USA, three in the UK, two in Australia, one in Ireland and one provided for Europe in general. The two most comprehensive resources appeared to have a similar aim to the current research: ‘Weigh In’ from the US (25), and ‘Confident Body, Confident Child’ (CBCC) developed in Australia (26). Both sets of guidance had been developed with experts from academia and clinical practice, are available to parents on interactive websites, and make reference to sources of evidence (although all references cited for both resources were over a decade old when this review was conducted). CBCC has also been tested as part of a workshop intervention (19, 27). In the UK, brief guidance developed with experts from academia and clinical practice in the mid-2000s was available through the Weight Concern charity website at the time the review was conducted, but the charity is now no longer in operation. As none of the guidance documents met all of our a priori criteria, we progressed with the development of new evidence-based guidance.
1.3 Interviews with children on their experience and expectations of weight monitoring
Although much has been published on parents’ views of weight measurement programmes, such as the NCMP, and of the challenges health practitioners experience when talking to parents about children’s weight, the voices of children are largely absent from published research. Similarly, little is known about what children who are living with overweight and obesity expect and want from parents; our systematic review search terms were designed to identify studies reporting on this if available (18, 23). To ensure children’s perspectives were represented we carried out an engagement activity with children, through 11 interviews with children aged 9–11 years (7 female, 4 male) to explore their expectations and understanding of weight measurements in a hypothetical scenario. Participants were recruited through a convenience sampling technique, following advertisements in local media and online fora. As the guidance is intended to be suitable for all regardless of the child’s weight, and as an explicit aim to avoid parents having to identify their child as overweight in order to take part, families of children of all body weights were eligible, and the interviewed child’s weight status was not requested. In line with ethical approvals, parents were provided with a written information sheet and asked to provide informed consent ahead of participation. Children were provided with an age appropriate information sheet, and the researcher checked through each aspect of participation with them and their parent at the start of the video call, before asking them to formally confirm assent.
Illustrated story cards were used to guide the interview, depicting a scenario in which a child took part in a weight measurement programme at school and their parents received a feedback letter informing them that the child was overweight. The researcher read a story card aloud then asked the child questions to explore their thoughts and feelings on each part of the story, before moving on to the next card. Participants’ parents/caregivers were present with the child but were asked not to speak on behalf of, or to prompt their child. All interviews took place remotely via Microsoft Teams or Zoom and were transcribed verbatim; the study was disrupted by the COVID-19 pandemic as the first interview took place in March 2020, but was then paused until October, ending in November 2020.
The key findings are summarised in Fig. 1, with further detail in supplementary materials (Additional File 2). Similar views were expressed by all children and suggested that children understand that weight is an indicator of health and trust their parents to let them know and take action if their weight is unhealthy.
Phase 2: Guidance development
2.1 Conceptual model
The insights from each source of evidence that could inform the content of the guidance document were identified and extracted by FG and EG (Additional File 3). These were translated into either a point of content (e.g., piece of advice) or frame for content (e.g., the spirit, tone or emphasis). A logic model was developed to set out the predicted direct and indirect effects of the guidance through mapping identified parent barriers and needs to specific techniques or content (Fig. 2). Content was included to; reduce stigma, increase confidence in having constructive conversations and deal with arising scenarios, promote social support seeking, normalise challenges in talking about weight with children, and support parents’ autonomy in deciding if and when to have conversations about weight with their child. Immediate benefits were predicted to be on parents’ emotions (e.g., less anxiety, shame or anger), confidence (e.g., in their ability to frame conversations appropriately) and behaviour (e.g., having conversations at the appropriate time, and/or changing the home environment). Indirect benefits for children were predicted to be the receipt of more parental support for dealing with worries about their weight, more support to change their health behaviours to achieve a healthy weight, and experiencing fewer negative conversations about weight (teasing, criticism, blame).
Figure 2 here
2.2 Framing the guidance
Psychological reactance is an unpleasant arousal state that occurs as a defence mechanism to feeling labelled within an undesirable and stigmatised group (28); it is evident in parents’ reported feelings of anger, anxiety, guilt and annoyance when receiving feedback that their child does not have a healthy weight status (29–31). Reactance results in people rejecting support and acting in the opposite ways to those intended (32). We drew on theories of health communication, and specifically narrative persuasion (33), to frame the guidance in a way that aimed to (a) challenge and reduce stigma and stereotypes of childhood obesity, and (b) reduce barriers to parent engagement stemming from negative psychological reactance. To prevent potential reactance we incorporated narrative messages (34–36) (i.e., case studies and stories) that provide information about health and social issues by describing how a real or fictional character thinks, feels and responds to an unwelcome health message, often reacting negatively at first before ultimately responding positively (37). Narratives have been reported to be effective in conveying health information to people who are expected to be resistant (35), more effective than didactic messages in changing attitudes to increase support for obesity prevention policies in adults (34, 35, 38) and more effective in making complex information more comprehensible to people with lower educational levels (39).
The narrative accounts included were based on interviews with parents who had previously received feedback from the NCMP informing them that their child was considered to be overweight or very overweight (the term obesity is not used) (40). Pilot testing of the narrative messages [16] found them to be acceptable and to introduce a novel element to parents by depicting scenarios from the child’s perspective; novelty in messaging is thought to increase readers’ engagement (41) and so, as well as using the narratives in the guidance, throughout the resource we included pull-out quotes depicting children’s perspectives. One new narrative was constructed during the Delphi process, in response to stakeholder requests for the presentation of a different perspective. This was designed using a similar strategic approach (i.e., specification and inclusion of elements, reference to previous parent interview transcripts), but was not included in the pilot work.
Other techniques to reduce stigma included acknowledging the environmental and external factors that contribute to the development of obesity, and explicitly stating that it is not helpful to blame parents of children for a child’s weight status (42–44). We also used person-first language (45) and positive imagery of children (and parents) of varying body sizes, and demographics enjoying physical activity and healthy eating.
2.3 Development of initial draft
An initial draft guidance document was created by FG and EG drawing on all data sources outlined, and integrating them into coherent sections (Fig. 3). This was then refined through an iterative review process with all study authors. The guidance was presented in a PDF format, and to increase the accessibility of the document, text was displayed in short segments or bullet pointed tips and hints, broken up by pull-out quotes and positive images. We intended the guidance to be relevant to all parents concerned about talking about weight with their children, whatever their child’s weight status (e.g., in response to social media, or concern from healthy-weight children during changes at puberty). This was considered important to reduce stigma; having a resource relevant to all could encourage parents to talk about the contents with each other and avoid possession of the guidance indicating that the family included a child living with overweight or obesity.
Phase 3: Modified Delphi Study
Recruitment
In line with guidance on conducting Delphi studies (46) we aimed to recruit 20–30 participants representing different stakeholders and areas of expertise. Six core groups that should be represented were identified, with a target of recruiting a minimum of two people within each group: (1) parents of primary school aged children, (2) school nurses involved in delivery of the NCMP, (3) public health professionals working in children’s healthy weight, (4) other health professionals involved in the general care of children (e.g., GPs, health visitors, dieticians), (5) clinical child and adolescent psychologists, and (6) relevant academic researchers.
Recruitment was initially through the existing networks of the study authors, drawing on knowledge of published researchers specialising in this area and contributors to the PHE, now OHID, NCMP board. An open call for expressions of interest was also posted on Twitter (originating from the lead author’s account but retweeted by other authors to their networks). Twitter volunteers were screened from their Twitter/internet profile on the basis of (i) credibility (e.g., employing organisation, qualification, length or nature of experience), (ii) uniqueness of insight (e.g., international experience, representation of charity etc), (iii) impact on the balance of expertise across the Delphi group.
Procedure
The modified Delphi process consisted of two rounds of review and response; a third was scheduled but consensus had been reached at the end of Round 2. Ahead of each round, participants were emailed documents to review, alongside a short video talking through the materials and outlining the aims of the round.
In Round 1, the documents for review included the draft guidance (Draft 1), a brief overview of the findings from Phase1, and links to more detailed information for those wishing to give greater scrutiny (outline of approach sent to participants is provided in Additional File 4). Participants provided feedback via a Jisc online survey (47) with a primarily open response format seeking comment on each section and page of the guidance. For each page in turn, survey questions asked whether there was anything missing, whether anything should be removed and for any other comments (e.g., changes in tone, adaptations etc). A small number of closed questions were also included (e.g., whether case studies were helpful or not). In Round 2, participants had the option of feeding back using the survey, through a webinar (held before the deadline for completion of the Round 2 survey), or both. A short video was produced by the research team setting out the rationale for what was changed or kept the same and why and circulated with the updated guidance. The webinar was provided in order to allow participants to discuss points of contention, giving an opportunity for people to explain their opposing views and a chance to arrive at consensus. Employees from the organisation it was intended would implement the guidance (PHE, now OHID) were also invited to listen to, and contribute to the Round 2 discussion to involve them in discussions that related to future pragmatic and feasibility decisions such as document length, formatting and use (n = 4).
After each phase, all comments were extracted and discussed by the two lead authors, FG and EG. A response to each comment was made and recorded; a “You said, we did” feedback document reporting the key points of feedback received from Round 1 alongside the changes subsequently made was included with the revised version of the guidance (Draft 2) during Round 2 (Additional File 5). Decisions on whether or not to implement a change were based on (i) frequency of similar comments across participants, (ii) background of respondents (e.g., for matters of opinion/preference, views of parents or those working most closely with parents and children were prioritised) and (iii) feasibility within the aims of the guidance (e.g., it was not considered feasible to add sections beyond the scope of communicating about weight). Where there were areas of disagreement or wide variation in views after Round 1 (e.g., over the extent of the use of the word “weight” throughout the document), these were discussed by all authors to agree an approach for compromise and this justification presented to the Delphi group.