Communication of children’s weight status: what is effective and what are the children’s and parents’ experiences and preferences? A mixed methods systematic review

The inclusion criteria are listed in Table 1.

We developed one comprehensive literature search strategy, covering both research objectives. It was peer-reviewed by a second search specialist and executed in October 2018. We searched nine databases (MEDLINE, PsycINFO, EMBASE, CINAHL, Web of Science, Cochrane Database of Systematic Reviews, DARE, CENTRAL, HTA). The search strategies are available in Additional file 1. The search strategy was developed using guidelines from the Cochrane Qualitative Research Methods Group for searching for qualitative evidence [33] and those for effect review searches [34]. We also searched the reference lists of all the included studies and relevant reviews.

Two researchers independently assessed the publications according to the inclusion criteria, first the title and abstracts, then relevant publications in full text. Disagreements were resolved through discussion or, if required, by seeking a third researcher’s opinion. Where necessary, we contacted the study authors for further information. We note that although language was an exclusion criterion for objective two, we found only publications in a language mastered by members of the review team, thus no records were excluded based on language.

All methodological quality assessments were done by two researchers, independently of each other. Any disagreements between the two assessors were resolved by discussion or consensus with a third researcher. For randomised control trials (RCTs), we assessed the risk of bias of each included study using the Cochrane ‘Risk of bias’ tool [34]. For the other study designs, we used study appropriate risk of bias domains as developed by the EPOC group [35]. To assess the methodological quality of included qualitative studies, we applied an adaptation of the Critical Appraisal Skills Programme (CASP) quality assessment tool for qualitative studies. Other reviews of qualitative evidence have also used this tool [36‐38].

We used a data extraction form designed specifically for this review, which included; author, year of publication, geographic setting, description of context, data collection methods (sampling, collection, analysis), description of participants, if ethics approval was given for the study, and results. One researcher extracted data and another checked the completeness and accuracy of the data.

We sorted the included effect studies according to categories of interventions and control conditions, and assessed results separately for each comparison. We based judgments about whether meta-analyses were appropriate on recommendations in the Cochrane Handbook for Systematic Reviews of Interventions [39]. None of the included effect studies were sufficiently similar to permit statistical pooling of outcome data. With regard to the study by Prina and colleagues [40], we had to transform the numbers for one outcome (attended parent’s information meeting). A statistician imputed the confidence intervals from the reported effect estimates and their associated standard errors using z-statistics. Where possible, two-sided p-values were calculated in the same way and compared to the reported p-values.

With regard to the qualitative studies, we conducted a best-fit framework synthesis [41]. Four researchers discussed various frameworks that fit the initial themes identified during data extraction. Through consensus, we decided to use the overarching framework developed in Ames and colleagues [36] about vaccination communication. This framework includes six sections: timing of information; availability of information; amount of information; source of information; content of information; and influence of the relationship between information, the way it is communicated and decisions. In addition to this overarching framework [36], we also decided to use the health belief model [42] to analyse the data about behaviour change related to the influence of the relationship between information, the way it is communicated and decisions. We conducted a thematic analysis [43] within each of the framework areas. During the analysis process, we looked to see if different themes emerged from different participant groups or settings, for example, children, adolescents and parents.

As a final analysis step, we brought together the findings of effect and the qualitative findings. We placed all of the findings into the framework identified for the best-fit framework synthesis to explore differences between the topics investigate by the effect and qualitative studies.

We assessed our certainty in the findings using GRADE (controlled studies) [44] and GRADE CERQual (qualitative studies) [45].

The database searches retrieved 7237 references and the manual searches an additional five unique references. We only identified studies reporting on communication and information to children identified as being overweight or obese. None of the included studies reported results related to children identified as underweight. Fig. 1 illustrates the handling of the references. Additional file 2 shows publications read in full but excluded.

We included four effect studies [28, 40, 46‐52]. The studies were published between 2014 and 2017, all but one was an RCT, and they were conducted in Mexico, New Zealand, the UK and the US, with a total of 2649 participants (Table 2). All examined different ways of providing weight-screening feedback to parents: face-to-face, various written formats, with or without additional information. Two studies had the same comparison, thus we could group the studies into three comparison groups. We describe the results of these below. Evidence profiles for all the findings can be found in Additional file 4.

A two-phase RCT, conducted in New Zealand, compared the effect of two different formats of face-to-face feedback on a child’s weight-screening results [28, 46‐49]. (We do not present results of the second phase, because it describes a treatment program for overweight or obese children.) The study conducted anthropometric measurements of 1093 children, of which the 271 children with BMI > 85th percentile and their families were further assessed. One group received weight feedback using a “traffic light” approach, considered best practice care, and another group received weight feedback using a “traffic light” approach combined with motivational interviewing (MI). Table 3 presents the findings from this comparison and the summary of findings table is available in Additional file 3. The results show that for parents, there is probably little or no difference between the two conditions, with regard to their: attendance of further treatment sessions; recognition of their child’s overweight or obesity; reaction (being upset) about the way information is given; motivation for lifestyle change. However, the parents in the MI condition probably have somewhat greater satisfaction with the way the healthcare provider supports them.

Two studies examined the effect of parents receiving written weight-screening feedback plus additional resources or information in comparison to only receiving written feedback [40, 50‐52]. The study designs were RCT and control before and after study (CBA) (embedded in a cohort study), and they were conducted in the USA and UK, respectively. In the RCT, the additional resources were access to web-based information, personal screening, and educational tools. In the CBA study, the additional resources were a call from a school nurse and, in a subsample, a face-to-face appointment. Table 4 gives the findings from this comparison and the summary of findings table is available in Additional file 3. For parents, the results show that there is probably little or no difference between the two conditions, with regard to whether they perceive they get information/resources that help them understand their child’s weight status or help to reduce the risk of overweight, and whether they contact a healthcare provider or perceiving they get useful weight status information. There is insufficient evidence to conclude whether feedback letters plus additional resources, compared to standard feedback letters, improve parents’ ability to classify their child’s weight status or recognise the risks of obesity.

Lastly, we included an RCT from Mexico that examined the effect of three different formats of written feedback letters to parents after school-based weight screening (results of a fourth group receiving no information is not included in this review) [40]. The letters differed with regard to whether BMI and health information was presented i) without comments, ii) with messages about the health risks, or iii) with information about other children’s weight status. The parents of 824 children identified as obese and overweight receiving any of the written weight-screening feedback letters are included in the analyses. Table 5 presents the findings from this comparison and the summary of findings table is available in Additional file 3 The results show that for parents, there is probably little or no difference between the three feedback conditions, with regard to whether they attend parents’ information meetings and in taking any action to address their child’s BMI. Similarly, there may be little or no difference between the three feedback conditions with respect to the children’s subsequent BMI, but parents may have somewhat lower ability to classify their child’s weight status correctly when they only receive simple written feedback.

We included 23 qualitative studies, presented in 25 publications [53‐77]. As summarized in Table 6, 15 studies were conducted in the USA [53, 54, 56, 57, 59‐63, 65, 68‐70, 72, 74‐76], five in the United Kingdom [55, 58, 66, 67, 77], and one each in Australia [71], Canada [64] and Norway [73]. Twelve of the studies concerned information sent from elementary/middle schools or preschools [53, 55, 58, 60, 61, 66‐70, 72, 76, 77], 11 regarded face-to-face communication with healthcare providers in primary healthcare centres [54, 56, 57, 59, 62, 64, 65, 71, 73‐75], and one study explored parental preferences regarding communication about their child’s weight [63]. Parents were participants in 21 studies [53, 54, 56‐65, 67‐77], 10-year-old children the participants in two studies [55, 66] and children/adolescents in three studies [57, 64, 71].

In the following section we present a summary of the qualitative findings identified during the best fit framework synthesis. The categories from the framework are used to group the summaries. For the individual findings within each framework category and our confidence assessments see Table 7.

Some parents felt that there was a general lack of communication about the routine weight screening- and notification process [53, 70] and that the notification process prior to weighing was weak [53, 61, 70, 76]. Others wanted to be notified about when to expect the weight screening results in the mail so that they could prepare [53, 61] and that the information should be sent out quickly so that it is up to date with recent measurements [54].

Although schools provided a letter at the beginning of the school year to opt out of the weight screening, many parents did not remember receiving or seeing this letter [61]. Some parents felt that the screening had taken place without their knowledge, “behind their back”, when the referral letter arrived home without warning [61, 68]. This issue also applied to the follow up information received by parents that often accompanied the weight notification letter. Many parents confessed that the supporting information they received with the letter was not seen, disregarded or placed in the bin often due to the emotional reaction to the letter itself. Some suggested that it would be better to send this supporting information later once the parent had absorbed the results from the notification letter [67].

Many parents were aware of growth charts and BMI weight reports and felt that they were useful tools [73]. However, they were unsure of how to read and interpret them [65, 74], and needed and wanted a better explanation to understand them [53, 73, 76, 77]. When there were no explanations, parents often misunderstood the growth charts and BMI weight reports [70, 75].

Many children found the weighing process to be secretive [55]. They did not know what to expect [55] and this could cause fear and anxiety [55, 66]. Children who were familiar with being weighed at home did not experience the same fear or worry [66]. To make children feel more comfortable authors recommended an introduction session before weighing and a drop in session after to discuss questions and concerns the children had [55].

Most parents agreed that healthcare providers played an important role in addressing their child’s weight [54, 56, 57, 59, 71, 74] and reported high trust in providers [57, 64, 74]. This trust could lead to greater comfort with the provider and feelings of better quality of care [57, 64]. Some parents felt that it was not the role of the school system to comment on their child’s weight [53, 65, 70]. They felt more comfortable and preferred to have their healthcare provider address weight issues [53, 59, 60].

Parents had clear expectations of the healthcare providers. They should be forthright, direct, address and initiate conversations about weight [54, 56, 57, 59, 73], thus taking the pressure off parents to initiate a difficult discussion [54]. They should use a sensitive approach [54, 57, 71], be positive [56, 57, 65], show interest [56], intervene early [56, 73], and talk directly to the child in a caring positive manner [54, 57, 65], sending a clear message [56]. Parents and children also had a preference for the use of open ended questions in a respectful tone [64] and motivated by concern for the child [65]. They wanted them to present and discuss the health risks associated with being overweight [56, 62, 65]. They also wanted support from their healthcare providers in developing a step-by-step specific, practical and individualised plan and accessing local information to support behaviour change [54, 57, 61] and to explain these concepts in a way that both parents and children could understand [65]. Some parents and children also felt that healthcare providers should tailor the conversation to the child’s age and be flexible about when children should be involved in the conversation and how often the conversation should take place [64].

Adolescents (aged 14–16) and parents felt that the adolescents took information coming from providers more seriously and responded better to them [57]. Adolescents also reflected on the providers’ intentions, reporting them as being supportive and interested and that this motivated them to change health behaviours [57].

Confidentiality and privacy were important during the screening process. Parents and children felt strongly that weight screening should be performed in a private setting [55, 60, 61, 63, 69, 70, 76] in order to avoid embarrassment, teasing and stigmatisation [60, 61, 63, 65, 76].

Some parents experienced that the weight screening results from their healthcare provider were different from those received from the school, or the healthcare provider’s reaction to the school screening led them to question or totally disregard the school results [53, 70].

The majority of parents who talked about their experiences with and preferences for information approved of the information being sent home by letter, but had concerns about how the letter would be sent [53, 61]. Confidentiality and privacy were important with regard to the delivery of the weight screening results [61, 68‐70]. They preferred delivery by standard mail directly to them [53, 60, 65, 76]. Some parents did not want the letter sent home with the child [60, 61, 63, 76], as they were concerned that if the letter was given to the child, the child could open, forget or discard it [53, 61, 63, 76]. Parents did not like that the child might see the letter first, as they worried that this could have a negative impact on the child [65]. Parents who did not approved of the weight screening information being sent home by letter wanted a more personal form of communication, for example, a call from the school nurse [53, 61], having teachers deliver the information at parent-teacher conferences [63] or email [53, 54].

Some parents wanted additional materials for addressing a child’s above normal BMI and for the family in general, such as websites, phone numbers, information letters or pamphlets [53, 54, 57, 63, 70, 72]. These parents felt that a letter with an explanation of the weight result was not enough to support them with further action and decision-making [53, 67, 70, 72]. Some mentioned that they also lacked support, such as a support hotline to phone, after receiving the weight notifications [53] and wanted links to local programs or resources [54]. When parents experienced that follow up and guidance were lacking, some experienced frustration and confusion [54, 57, 67]. This was also seen as a barrier to addressing the child’s weight issue [53, 54, 57].

Some parents expressed concerns about the content of the weight screening notification letter [53]. One of these concerns was the verbiage [53] and the format of the letter [53]. Parents wanted a simple, easy to understand, visual explanation of BMI and how to interpret the results [63, 72, 77]. Some felt that the letter was too general [53, 70], impersonal [53], and many parents felt that the content of the letter lacked necessary information [53, 63, 76, 77] (See Table 8).

Some parents felt that the tone of the weight screening notification letter was judgemental and negative [53]; judging their parenting abilities [53, 67] or insulting their child when words like overweight were put in bold [67]. Parents wanted the letter to be written using sensitive language and a supportive tone [54]. Parents suggested that the content of the letter should focus on what the whole family can do, rather than just the target child [63, 72].

Parents stressed the importance of appropriate terminology in order to communicate respect and promote engagement [64]. They believed that healthcare providers should explore a family’s preferred terminology when communicating about a child’s weight [64]. In many cases, parents preferred the terms overweight and obese [56, 64] and suggested to use these in reference to national norms to aid in understanding [56] or to discuss health and growth rather than weight and size [64]. Still, some parents avoided using the term obese altogether as they found it to be an untrue description of their child, hurtful, insulting and judgemental [61, 62, 65, 72]. They considered colloquial terms such as fat, chubby or plump offensive and inappropriate for use in official letters or interactions with healthcare providers [56, 64].

Some parents said that hearing the word overweight would be motivating and convey a strong message [56]. They felt the same way about the term obese as it would be an “eye opener” [56, 75]. However, terms like “at-risk for overweight” and “unhealthy weight” and “normal weight” were vague and confusing and would not motivate them to take action [56, 62, 72].

The next section presents findings related to the influence between the relationship of information, the way it is communicated and the action taken by participants. The headings in this section represent the sub areas of the health belief model which was the framework used to analyse the data in this section.

A few parents accepted the results of the weight notification letter [53, 61, 65, 67, 70]. These parents mentioned finding the letter helpful and welcomed it compared to a lack of notification in the past [65]. Others said it confirmed what they already knew [53, 61, 67, 70]. Parents who had overweight children and who accepted the letter, viewed it as an opportunity to make some needed changes [70] and were happy that their child’s weight issue had been brought to their attention [70].

Many parents questioned the results they received from the BMI testing [53, 60, 77]. They described receiving feedback about a child’s overweight in negative terms [58, 61, 65, 67, 69, 70], using words such as “cross”, “angry”, “annoyed”, “upset”, “insulted”, “distressing” and “perturbed” to describe how they felt [58, 61, 67, 70]. The letter caused a great deal of panic and worry among parents of overweight children, as they felt they had been caught unawares [67, 70]. Other parents felt as if they were being judged [58, 77], responding with anger and defensiveness in some cases [77]. Some parents felt disappointment in not being able to live up to their own expectations for themselves in regards to managing their child’s weight [77].

Some parents disregarded, ignored, disagreed with or did not believe the results they received from the weight notification letter [53, 58, 61, 70, 73, 77]. Several study authors believed that this disagreement highlighted the misconceptions that parents have regarding their child’s weight classification [53, 58, 60, 61, 67, 70, 73, 77]. Some parents had received contradictory remarks from the child’s healthcare provider [53, 61, 70]. Parents who did not believe the school’s findings or disregarded the letter had a lower perception of the severity of being overweight and/or a decreased level of concern about the problem [53].

The authors of one study [66] found that children reacted very differently to their weight notification. Children receiving normal weight feedback often reacted with joy and happiness. However, children who were told they were overweight were often surprised about the result, entering a phase of denial or shock. Many felt that the results must be a mistake and questioned if the measurements were right. The reactions to weight feedback were often very emotional, with those who were overweight reacting with negative emotions or disbelief. This caused a lot of worry, which in turn could influence their mental health and well-being.

One group of parents who had overweight children participated in an othering process (to view or treat someone as intrinsically different from you) where they believed their children were fine and not the target group of the BMI measurement program, but others with overweight children were [58, 61, 67]. Parents of normal weight children also participated in othering [58, 67]. This othering process allowed parents/guardians to identify themselves as part of the group that is doing the right thing, and viewed others, especially those with overweight children, as not doing things correctly [67]. The process of othering contributed to the dismissal of overweight feedback that parents received themselves, legitimising their rejection of the feedback for their child [58, 67]. Parents used distinct language to define themselves from the other group who they perceived did need to be targeted [58, 61]. Parents described themselves as; educated, responsible, middle class, and interested [58]. They described the other parents as; irresponsible, ignoring healthy living advice, and fed their children unhealthy foods [58, 67]. Many also believed that these others who did need to change were not listening [58] and so questioned the impact of the notification letters [58].

Some parents felt that the school was not doing enough to address the results of the weight screening as students spent more of their time in school environments than at home [53]. Parents suggested more time for activities such as recess or gym time [53, 68] or offering healthier food options [53, 61, 68].

Some parents discussed how receiving the weight notification letter had been a cue to action for them [53, 61, 67, 69, 70, 77]. They used the letter as a tool, showing it to their children [53, 70] or spouses, friends or neighbours [67, 70] to start a discussion and create awareness and opportunity [53, 70]. Some implemented changes or planned to implement changes in diet and activities with or without the knowledge of their children [53, 61, 67, 70, 77]. Finally, for some the letter was a cue to action to contact their family physician for follow up [70].

Other parents ignored, downplayed or dismissed the letter, taking no action to address the weight of their child [53, 61, 67, 77]. This was often because they were not fazed by the results so ignored or downplayed the severity of the information they had received [53, 61], stating that their child was already very active and/or was eating a healthy diet and so did not need to implement changes [61, 67]. For others, they just did not believe the results of the letter and were angry at receiving it [61]. Finally, some parents said the letter had no impact as they had already implemented changes in their homes to address weight issues before receiving the letter and so continued with these [53, 67].

Many parents talked about how difficult they found it to control their child’s weight [53], expressing feelings of concern, lack of knowledge, hopelessness and being overwhelmed [53, 70]. Many felt that despite trying to make changes in eating and exercise habits they were unable to significantly reduce their child’s weight [53]. They were unsure of where to go for help and what actions to take [70].

Many parents found it difficult to talk to their children about their weight [70, 73]. They found it stressful as the children would often become emotional and shut down and parents were unsure of how to react [70]. Many parents felt that they lacked the knowledge on how to communicate with their children about the topic leading to fear and frustration [70]. They were unsure of how to respond when their children started commenting on their own weight as well in order not to have a negative impact, for example, on self-esteem [70]. Parents wanted to know more about how to discuss BMI findings with their children, including kid friendly suggestions to use in the family [53].

Some parents felt a lack of knowledge and fear of doing harm when discussing weight issues with their children, parents want guidance in how to talk to their children about their weight notification. Some parents preferred to discuss their child’s weight without the child present either between themselves or with a health care provider as they feared the child would understand the conversation and this could lead to the development of low self-esteem or eating disorders [64, 71, 73]. Some parents felt that it was important to consider the child’s age when deciding if they would be involved in the conversation [57, 64, 71] or if the child was older whether the parent would be involved in the conversation [71]. Other parents supported a phased approach where the child would be increasingly included in the conversations over time [64]. Others chose to not have any conversations with their children about the letter as they did not want their child to think they were overweight or be labelled or believed that talking about it could lead to other problems [67, 70].

Some children felt that they had limited information about what they themselves could do about their weight. They had to rely on the adults around them, their parents and guardians, for information about how to tackle their weight issues. This lack of information about what action they could take often caused the children to worry [66].

Above, the results of the studies about effect and the studies about experiences are presented separately. Here, we show the results of placing all of the findings into one framework (Table 9). It shows that the effect studies had a narrower scope of research covering five framework areas; source of information, content of information, perceived susceptibility of being overweight, perceived severity of being overweight, and the parent’s cues to action. The research focus and findings from the qualitative studies were broader, covering all but two areas of the framework; the perceived benefits of being overweight and the perceived severity of being overweight.

The findings once placed in the framework show that future effect studies could also look at the impact of the timing of the information to parents, information availability, the amount of information parents and children would like to receive, as well as issues related to barriers to addressing weight issues in schools and feelings of self-efficacy.

Our systematic review comes with strengths and limitations. A strength of this mixed methods systematic review is the close collaboration between the commissioner and the research team in coming to an agreement on the objectives, protocol, and types of studies to be included. This ensured high relevance. Further, we used systematic and transparent methods throughout the review process, and combined evidence from both experimental studies about effect and qualitative studies about people’s experiences and perspectives. By viewing such studies through the same lens, it is possible to enhance the/our understanding of how the findings interrelate. With regard to limitations, our literature search is more than a year old, and it is possible that new relevant studies have been published after this date. We relied on the information and data presented in the published articles, which in turn are limited by issues such as word restrictions. Children identified as underweight were not covered in any of the included studies.

As part of the qualitative synthesis process, the authors working with objective 2 reflected on how our backgrounds and positions might have influenced our choice of review topic, study selection, data extraction, analysis, and interpretation of data. Our backgrounds are in health systems research, social sciences, and pedagogy and, while working on the synthesis, we were all employed by The Norwegian Institute of Public Health. None of the reviewers have been involved in primary research related to weight assessment programs or communicating to parents or children about their weight. HA has been involved in research related to childhood vaccination programs where she routinely saw children being measured and weighed but weight and weight feedback were not the focus of the research. Before working on the synthesis, we did not have any preconceived ideas regarding weight assessment and weight status notification interventions. However, we believed that the implementation of programs should be informed by robust evidence of effectiveness, acceptability and feasibility.

The following questions, derived from our findings, may be helpful to consider when implementing or planning for routine childhood weight screening communication strategies in order to address issues of importance to their target population. It is important to consider local contextual factors including gender, age, cultural group, and education when implementing new strategies for communicating with parents and children about their weight status. Consider:

While we believe we systematically have addressed the review objectives, we found that there are several relevant ongoing trials. Thus, an updated review of research objective one seems indicated. Related, more research on objective one would be useful because the four studies we identified only represent some types of participants, interventions, and outcomes. For example, none of the studies addressed communication when children are underweight, online feedback, notification tools, or information delivered through digital technologies to portable devices such as smart phones or tablets. Additionally, the follow up times of our included studies were short, and it would be important to learn about long term effects. With regard to the qualitative studies, a larger spread of countries, contexts and participants from a variation of backgrounds would be beneficial as well as insights on the views of children and adolescents, including those who are underweight. We also encourage future studies to provide better reporting of context, sampling, methods, and with regard to qualitative studies, researcher reflexivity.

More research is needed on parents’ and children’s’ preferences around the details of timing, amount, and content of weight notification methods. The findings once placed in the framework show that future effect studies could also look at the impact of the timing of the information to parents, information availability, the amount of information parents and children would like to receive as well as issues related to barriers to addressing weight issues in schools and feelings of self-efficacy. Future effect studies could be linked with process evaluations including qualitative studies on order to explore why the intervention work or not.