Abstract
Abstract
Background:
The aim of this study was to examine the perceptions of the assessment and management of children with obesity of primary, secondary, and tertiary care clinicians across two health districts in western Sydney and a specialty children's health network.
Methods:
Participants were 304 clinicians (medical, nursing, and allied health workers) in primary, secondary, and tertiary pediatric-level services. A questionnaire captured the training, assessment, and management approaches and perceived barriers to managing pediatric patients with obesity. Chi-squared tests and logistic regressions examined the differences in responses between clinicians.
Results:
Clinicians across all levels of health care had only moderate rates of training in obesity (48%), did not routinely measure tandem heights and weights (80%), and infrequently referred children to other services. Only 25% of clinicians frequently referred children to a weight management service (most frequently the dietitian). When comparing across health care settings, those in secondary-level services had higher rates of training (70%) and more frequently initiated treatment for obesity.
Conclusion:
Frequencies of routine identification and initiation of treatment for children with obesity are low among health professionals across health care settings, with some exceptions for secondary care clinicians. Greater and more intensive health professional training on the assessment and management of children with obesity is needed in Australia and may be a key factor in increasing health care for this common chronic condition.
Introduction
Childhood obesity is a major health problem worldwide with approximately one-quarter of school-aged children being overweight or having obesity. 1 Children with obesity present more frequently to medical services 2 and have higher medical costs than their peers with a healthy weight status. 3 Early identification and treatment of children with overweight and obesity are important, 4 with health care professionals playing a pivotal role. 5 However, pediatric health professionals across all levels of health services (i.e., primary-, secondary-, and tertiary-level care) have difficulty recognizing overweight and obesity in patients.
Primary care providers underestimate weight status approximately half of the time, 6 and <1% of children with overweight or obesity who present to their general practitioner (GP) in Australia will be offered specific treatment. 2 Similarly, secondary-level health services have inadequate rates of diagnosis and treatment of children with obesity,7,8 and despite 20% of hospitalized children in Australian tertiary care settings presenting with overweight or obesity, 9 weight problems are rarely addressed when children attend tertiary clinical services. 2
It is important for clinicians and health care professionals to assess the weight status, whether underweight or overweight, of all children and to discuss the outcomes with families. For those children with obesity, perceived barriers to weight management by health professionals include difficulty raising the issue of weight status, uncertainty about advice to offer and appropriate referral pathways, and a lack of local service capacity.8,10 In addition, many primary practitioners do not believe that they play a part in the treatment of children with obesity 11 and question their effectiveness in treating this condition. 12 Secondary-level health care service providers such as pediatricians also feel they have inadequate skills, competencies, and training to effectively manage children above a healthy weight. 8 Furthermore, there are few tertiary pediatric weight management services available within Australia. 13
The management of childhood obesity is a whole-of-health-care priority. 14 The linking of secondary and tertiary services in shared-care has been suggested as one model of care. 8 Research has assessed primary 12 and secondary care physicians' 5 views of managing children with obesity, but there has been no research assessing primary, secondary, and tertiary health care systems within Australia, all of which play a role in the management of childhood obesity. Given the large number of children affected by obesity, it is important that there is effective assessment and recognition of such children, to allow timely referrals to appropriate weight management services for treatment. To achieve this, we need to better understand the associated factors to support each level of care to change their practice. It is also important to know whether factors such as greater training, experience, and knowledge across the different levels of health care may be associated with the appropriate assessment and referral of children with overweight and obesity. The aim of this study was to assess and to compare the perceptions of the assessment and management of children with obesity across all sectors of pediatric health care within a representative sample of clinicians.
Methods
Study Population and Design
This was a cross-sectional quantitative assessment of clinicians, comprising medical, nursing, and allied health workers in primary, secondary, and tertiary pediatric-level services within Sydney, Australia. Participants included clinicians from the following: Western Sydney Local Health District (an urban area with a population of 946,000 residents), Nepean Blue Mountains Local Health District (an urban and semirural area with a population of 350,000), and Sydney Children's Hospitals Network (SCHN, a specialty pediatric network of pediatric tertiary hospitals in Sydney). To account for international differences in the definitions of health care settings, here we refer to primary care clinicians as those who have first contact with patients, and primary care services as those that are easily accessible to the community. 15 Secondary care consists of clinicians who do not have first contact with a patient (e.g., pediatricians), and secondary services can be delivered in the hospital or nonhospital setting. Tertiary care refers to those services that are considered highly specialized and require referral from a secondary care clinician. 16
Recruitment
Recruitment occurred between April 2016 and April 2017. An electronic invitation containing a link to an online survey was sent to clinicians to encourage their participation. Due to the varying nature of the clinicians' areas of work, several strategies were used to recruit participants for this study. First, a list of present primary, secondary, and tertiary public and private health care providers and their contact information was developed from publicly available databases such as the National Health Service Directory, Way Ahead, and connecting with other interagency networks. Contact information was cleaned, and duplicates removed. Second, administrators of relevant e-mail distribution lists were asked to send the survey invitation e-mail on behalf of the authors to (i) clinicians working within the two children's hospitals in Sydney and (ii) clinician members of the two primary health networks within the health districts' geographical regions. Due to some of the surveys being sent via confidential electronic mailing lists managed by clinicians outside of the research team, a response rate was unable to be calculated for the full sample. A sample size could be calculated for those surveys not sent via electronic mailing lists. A total of 1164 invitations to participate in the survey were sent out and 238 survey responses were received, representing a response rate of 20%, which is within an acceptable range for the health care sector.
The study survey aimed to have a wide distribution across present pediatric providers and services within western Sydney and the Nepean Blue Mountains region, to determine the extent of existing weight management services available. Considerable effort was made by researchers to build sound contact lists across primary, secondary and tertiary pediatric, GPs, and allied health services. While every effort was made to survey a wide range of pediatric services, it is important to recognize potential sample skews, due largely to the contact lists available of health professionals. Study ethics was approved by the SCHN Human Research Ethics Committee and participants provided informed consent. The procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008.
Measures
A questionnaire (Supplementary Data) was developed collaboratively by a team of specialists working in childhood obesity, including two medical specialists, a medical advisor from the New South Wales (NSW) Ministry of Health, a clinical nurse consultant, a dietitian, and a project officer. The questionnaire was adapted from The Health Professional Survey used in the Weight4Kids pilot study. 16 The short survey, designed to take participants no longer than 10 minutes to complete, was developed in line with the aims as highlighted in the introduction. The survey was designed to capture the following: (i) geographical location; (ii) primary role; (iii) number of pediatric patients; (iv) pediatric obesity training; (v) assessment approach; (vi) treatment approaches and management options; (vii) referral pathways; and (viii) barriers to managing pediatric patients with overweight or obesity.
The barriers provided in the survey were developed from a review of the literature and from previous unpublished qualitative research conducted within the health districts and network. Information on the clinician's primary role, number of pediatric patients seen per week, and prior training in managing children with obesity was assessed using a 5-point Likert scale (1 = “never” to 5 = “always”). Clinicians were asked how often in the past week they had assessed both the height and weight of a pediatric patient and how often they had plotted BMI on a pediatric growth chart.
The treatment and management of children with obesity were assessed using a list of 17 questions. Participants were asked how often they performed each treatment and management option in the previous week. Responses were collected on a 5-point Likert scale (1 = “never” to 5 = “always”) with the questions based on a previously published survey. 17 Perceived barriers were assessed from a list of 12 potential barriers and respondents reported their perceived significance of the barrier for each one on a 5-point Likert scale (1 = “not a barrier at all” to 5 = “significant barrier”). The barriers were based on data from focus groups of clinicians working within NSW Health. Respondents were grouped into primary, secondary, and tertiary areas of work.
Data Analysis
Data analysis was performed using SPSS statistical software (Version 24. 2012; IBM Corp., Armonk, NY). Frequency distributions were tabulated for all variables overall and separately for each of the primary, secondary, and tertiary services and for medical, nursing, and allied health clinicians. Due to the small sample sizes in some response categories, the Likert data were transformed into binary variables where “often” and “always” were combined (hereafter collectively referred to as “frequently”) and “never,” “rarely,” and “sometimes” were combined (hereafter collectively referred to as “infrequently”) for the assessment and management questions. For the barrier questions, “significant barrier” and “very significant barrier” were combined (hereafter referred to as a “major barrier”) and “not a barrier,” “minor barrier,” and “moderate barrier” were combined (hereafter collectively referred to as a “minor barrier”).
Chi-squared tests were used to compare the binary responses between primary, secondary, and tertiary services for assessment, treatment, management options and barriers. Logistic regressions were used to examine the difference between secondary and tertiary care practitioners compared with primary care practitioners among assessment, treatment, and management and barriers for each professional group. The primary care practitioner group was set as the reference group. Odds ratios (ORs) and 95% confidence intervals (CIs) for each outcome were calculated for secondary and tertiary care practitioners relative to primary care practitioners. There was no adjustment made for multiple statistical comparisons.
Results
Demographics
Thirty percent of respondents were medical personnel from the tertiary setting (Table 1). The next largest group was allied health clinicians in the tertiary care setting (22%) and the third largest group was medical personnel from the primary care settings (14%). The allied health clinicians included physiotherapists (n = 45), dietitians (n = 30), psychologists (n = 6), social workers (n = 3), occupational therapists (n = 2), exercise physiologists (n = 1), music therapist (n = 1), and speech pathologist (n = 1). Overall rates of training were low. A higher number of clinicians from the secondary care settings reported training in the management of children with overweight/obesity (70%) compared with those in primary (45%) and tertiary (50%) health care settings (p = 0.04). The type of training received varied and included online modules provided by NSW Health, attendance at conferences and workshops, or as part of their university training in either dietetics or general medicine.
Clinician Characteristics
Anthropometry
Twenty-one percent of all clinicians reported infrequently doing a tandem height and weight assessment when seeing any pediatric patient and there were no significant differences between primary, secondary, and tertiary care clinicians. A similar pattern emerged for BMI calculations when reviewing pediatric patients: 29% of all clinicians reported infrequently calculating BMI, and the proportions for primary (10%), secondary (30%), and tertiary (33%) did not differ significantly.
When the data were compared across medical, nursing, and allied health professionals, only 36% of allied health clinicians reported regularly assessing heights and weights and 31% reported frequently assessing a patient's BMI. In comparison 69% of medical staff and 52% of nursing staff reported regularly assessing heights and weights and 51% of medical staff and 49% of nursing staff reported frequently calculating BMI.
Treatment and Management Options
Table 2 shows the frequency of clinicians' responses for treatment and management options for children with obesity. Across all health care settings, the overall percentages of clinicians using general management and treatment strategies, such as discussing healthy lifestyles, exercise, and positive parenting strategies, for children with obesity were low to moderate (≤53%). When comparing across health care settings, secondary care clinicians were more likely to discuss with the family the health implications of having a child with obesity, compared with primary care clinicians [OR = 2.50 (95% CI 1.04–6.02) p = 0.04].
Clinician Treatment, Management Options, and Barriers for Children with Obesity
When the data were compared across medical, nursing, and allied health clinicians (Fig. 1), there was a higher percentage of medical clinicians (62%) who reported discussing the results of a BMI assessment with families compared with nursing staff (43%) and allied health clinicians (45%). Fewer allied health clinicians reported discussing healthy lifestyles, health issues associated with childhood obesity, or providing dietary advice to families than nursing or medical clinicians. A lower number of nurses reported discussing the health implications of obesity with families compared with medical and allied health clinicians.

Referral Pathways
A total of 24% of clinicians reported “frequently” providing referral options to families of children with obesity. Figure 2 depicts the frequencies of each of the referral pathways used across all health care settings. There was no difference between primary, secondary, and tertiary care clinicians (25%, 30%, and 23%, respectively). Only 7% of clinicians reported referring to Go4Fun, a free, evidence-based, government-funded 10-week family healthy lifestyle program for children between the ages of 7 and 13 who are above a healthy weight, provided in the Australian state of NSW.

Barriers to Management
As shown in Table 2, while barriers related to engaging the patient/family were reported by a high number of clinicians across all health settings examined (approximately half to two-thirds of clinicians), there were differences in responses between primary and secondary and primary and tertiary care clinicians [families with complex issues (p = 0.03) and a lack of parent/child motivation (p = 0.001)]. There was some evidence that, compared with primary care clinicians, tertiary care clinicians [OR = 2.4 (95% CI 1.1–5.2) p = 0.03] and secondary care clinicians [OR = 2.6 (95% CI 0.95–7.23) p = 0.06] were more likely to report issues with the family being a barrier to the management of the child's weight.
Discussion
This article reports on the frequency of health professionals' engagement with families for the identification and management of children with obesity. It is the first study to assess the different rates of these activities in the primary, secondary, and tertiary health care settings. The key findings across all health care sectors were as follows: (i) relatively low percentages of clinicians reporting having received training in, routinely providing assessment of, or routinely managing childhood weight issues; and (ii) between one-half to two-thirds of clinicians reporting that barriers to providing adequate care were due to a lack of support services and factors relating to challenges with the patient's family.
Given that childhood obesity is a serious and common problem, it is concerning that clinicians across primary, secondary, and tertiary care settings are not routinely assessing their patients' BMI. Allied health clinicians had lower self-reported rates of assessment of BMI than medical or nursing clinicians. A frequently used objective measure of weight status is important for two reasons. First, obesity treatment is more successful if weight issues are recognized and treated early. 19 Second, overweight and even moderate obesity are often not recognized by parents 20 and clinicians 6 without a formal assessment. Frequent and objective measurement of children's weight status will normalize this practice as a typical pediatric measure of growth and development. Primary care physicians are well placed to provide routine measurement of weight status to a large proportion of the pediatric population, but they showed a very low rate of measuring weight status (4% in this study). The rates of measurement of a child's weight status by primary care clinicians in this study are lower than the 45%–90% rates reported by primary care clinicians in the United States.15,21 At the other end of the health care spectrum, clinicians in pediatric tertiary settings, where childhood growth and development are at the core of their service, also reported low rates of measurement of weight status (23%).
The survey respondents also reported low rates of treatment and/or management of patients with obesity. Only half of the clinicians reported discussing the results of the weight status assessment, and <53% provided at least a discussion of some strategies for managing weight. However, secondary care clinicians were more likely than primary care clinicians to report at least discussing with families the health implications of having a child with obesity. This may be due to the secondary care clinicians reporting greater training in childhood obesity, rather than it being a factor related to the secondary setting per se, or a lack of time with patients. Alternatively, it may be related to a lack of services being frequently reported as a major barrier. Previous work assessing tertiary-level pediatric obesity services in Australia found a lack of specialty services available. 13 The perception of a lack of such services may be a reason that clinicians in the secondary services are more likely to treat the children and families themselves. Previous studies have shown that many primary care clinicians do not perceive they have a role in treating children with obesity 11 and are likely to refer these onto secondary services.
A previous Swiss-based study assessing nursing and medical practitioners found a lack of knowledge and confidence to assess and mange people with obesity and highlighted the need for education to health care providers. 22 When comparing self-reported assessment and management of children with obesity between medical, nursing, and allied health clinicians, there appeared to be a difference between these groups of clinicians. Clinicians from a medical background reported more frequently assessing heights and weights and providing some discussion to families about their child's weight compared with nursing and allied health clinicians. It has been widely recognized that further training in the assessment of children with overweight and obesity is needed across all levels of health professionals, especially medical practitioners. 23 Our study continues to highlight the need for more education to health care professionals about the assessment and management of children with obesity. There does appear to be a difference in knowledge and confidence between health care professionals from the medical, nursing, and allied health backgrounds, as well as those from primary, secondary, and tertiary care settings, and thus, education may need to be tailored to specific groups of clinicians.
While there is indeed a need for more clinical pediatric weight management services, underutilization of available services is a concern. This is shown by low rates of referrals to Go4Fun, an evidenced-based, family-centered group program, free for children between the ages of 7 and 13 years and their families, run by the NSW Government in Australia. Despite Go4Fun being readily available and able to successfully target children with overweight and obesity, 24 only 7% of clinicians across primary, secondary, and tertiary care settings referred to this service. General education in obesity does not appear to affect this behavior, since secondary care clinicians did not differ from other health care professionals in self-reported referral rates to Go4Fun despite higher training in obesity. A potential reason for this underutilization may be the perception of their effectiveness in this population, where parent-related factors such as motivation were perceived barriers, especially in secondary and tertiary care settings. Similar barriers to management were found in a separate survey of dietitians, nurse practitioners, and pediatricians. 5 The research literature shows that a high percentage of families drop out of treatment programs 25 and family-related factors likely underpin this large attrition rate. 26 Other reasons for the lack of referral to Go4fun may be that clinicians are unaware of the existence of the program or are unsure of the referral criteria to the program.
There did not appear to be a lack of interest or a lack of confidence in treating children with obesity in our survey respondents. Despite this, clinicians were not always raising the issue of a child's weight status with the families, providing treatment, or referring to appropriate services. Secondary care clinicians had higher self-reported rates of training in obesity, provision of weight management advice, and treatment than clinicians in the primary and tertiary care areas. We were unable to assess clinician's level of training. Thus, those with comprehensive training in pediatric obesity may be more likely to initiate weight management care than those clinicians will little or no training. Indeed, high rates of training, specifically on behavior change therapies, improve the confidence of clinicians in managing families. 23 Improving clinician training in the management of pediatric obesity improves the use of obesity-related discussions with families. 27 Adequate training for clinicians may also increase confidence in, and/or highlight the importance of, performing anthropometric assessment of children, to increase rates of identification. The medical management of children with obesity by a clinician during medical appointments is an important foundation for treating this condition. 28 The literature suggests that the assessment and provision of basic education to families and patients can be the responsibility of all medical practitioners 28 and this should be across all health care settings. 29 The fact that a group of clinicians with higher rates of any training (secondary health care professionals) also provided higher rates of treatment and referral options highlights the importance of training and begs the question of how thorough training needs to be. Perhaps very short training is sufficient. If this is the case, training rates may increase if clinicians know that only a small commitment to training makes a difference. This is an important question for future studies to explore.
Another way to increase clinicians' recognition and/or treatment of children with obesity is by utilizing automatic alerts in electronic medical records (eMRs) when a child has had a tandem weight and height assessed, to highlight the child's weight status. 7 In Australia, some eMR systems have the capacity to calculate a child's BMI percentile and z-score, when the child's height and weight are entered into the eMR system in tandem. In most health services in Australia, the measurement of heights and weights of children is not routine, and when heights and weights are assessed they may not be entered into the eMR system, with the result that assessment of a child's BMI does not occur routinely. When tandem heights and weights are entered into eMR, clinicians are still required to interpret the BMI percentile/z-score calculation to determine if the patient has obesity. The integration of eMR systems with clinical practice guidelines and tools can increase the identification of obesity in both primary care30,31 and tertiary care settings. 32 Given that many clinicians, especially those in the primary and tertiary care setting, report inadequate training in the management of children with overweight or obesity, the implementation of automatic alerts in eMRs may improve the recognition of children with overweight and obesity. Automatic alerts along with increased training to improve clinician confidence in assessing and educating families will further facilitate improved recognition and treatment of children with obesity. Initiatives such as a health pathways program has commenced in Australia for primary and secondary care clinicians,33,34 which could also enhance the improvement in education and appropriate referral of children with obesity.
Strengths and Limitations
Treating children with overweight and obesity is not done in isolation, but services are required across the whole of health care. A strength of this study was that it is the first of its kind to assess and compare the views of primary, secondary, and tertiary care clinicians. A further strength of this study is that it included clinicians from medical, nursing, and allied health, all of whom play a role in the assessment and management of children with obesity.
This study aimed to assess the views of clinicians across primary, secondary, and tertiary care settings. Due to the nature of the recruitment strategy, there were a larger number of clinicians from the tertiary care setting representing a wider range of clinician roles across medical, nursing, and allied health professionals. As with many foundational studies that attempt to reach a large cross section of participants for the first time, the limitations of the study were primarily due to low and unequal participant recruitment. Participants from primary and secondary care settings were from a limited number of professions and there was a small sample size compared with clinicians from a tertiary setting. Having an adequate representation of primary and secondary clinicians is important because clinicians in primary and secondary care should be responsible for identifying and providing treatment for children with overweight and obesity, given their greater contact with the general pediatric population. In addition, due to the nature of the study, recruitment was completed via web-based surveys, which resulted in a low response rate of 20%. While this low response rate is not unusual—it is similar to a study of Canadian pediatric clinicians that found a 30% response rate for a web-based survey 35 it could produce in “no-response” bias. 36 Due to these issues with low and unequal participant recruitment, the views of the clinicians from this study may not be representative and need to be interpreted with caution. However, it is important to also note that the response bias is likely to underestimate the extent of the problem, as many of the clinicians who participated in this study may have had some interest in managing children with obesity.
This highlights the poor management of children with obesity—in a sample that is likely biased to an interest in obesity, rates of treatment and referrals are still low. To build on this first survey of clinicians across multiple health care levels, future studies could include monetary rewards, which has been shown to more effectively increase response rates among clinicians. 35
This study relied on self-reported rates of assessment and education. Without objective measures of the rates of assessment and treatment of children with obesity by clinicians, the rates of assessment and education may not be accurate. It may be that clinicians overestimated their rates of assessment and education provided to children with obesity. Further information on the amount of training clinicians received on assessing, referring, and treating children with overweight and obesity would have been beneficial. The authors asked clinicians whether they have received formal training and a yes/no response was given. No other details were sought in the interests of keeping the survey short and simple for busy clinicians.
The findings that secondary clinicians were more likely to provide treatment and referral options for children with obesity, and also had higher rates of training, compared with primary and tertiary clinicians, suggest that there is something about training that needs to be clarified. This would have been interesting to explore but cannot be answered by this foundational study. Despite this, the low levels of training noted further highlight one of the core findings of the study regarding the lack of training in this area. Details on clinician's level of training should be considered for future studies.
Conclusion and Implications for Health Care
This is the first study to assess primary, secondary, and tertiary health care clinicians' views of managing children with obesity in Australia. Overall, clinicians showed low rates of training in obesity, assessment of weight status, and initiation of treatment/management of obesity, including under-referring to a freely available program. As treatment of childhood obesity is a whole-of-health problem, greater and more intensive and targeted health professional training on the assessment and management of children with obesity is needed in Australia and may be a key factor in increasing health care for this chronic condition. Additional strategies, such as automatic eMR alerts and specific training on the importance of tandem height and weight assessment, may improve outcomes for children with obesity.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
