Secondary Logo

Journal Logo


Trends in Attitudes and Practice Patterns of Physical Therapists in Addressing Childhood Obesity in Schools

Kendall, Eydie PT, PhD, PCS

Author Information
doi: 10.1097/PEP.0000000000000466


Seventeen percent of children in the United States are considered obese.1 The childhood obesity rate tripled between 1980 and 20082 before leveling off.1 Obesity in childhood is predictive of obesity as an adult.3 More than 68% of adults are overweight and obesity now effects nearly 35% of the adult population.4 Obesity contributes to illnesses such as heart disease, stroke, diabetes, certain cancers, nonalcoholic fatty liver disease, and other causes of illness and death.4 Obesity also has an effect on the health of our national economy, costing $147 billion annually.4 Obesity is a problem that deserves the attention of multiple disciplines and recognition that the problem starts in childhood.

Children who are overweight and obese experience biomechanical changes during activities such as walking,5 jumping,6 posture,7 and balance.8 They may experience genu valgum,9,10 knee hyperextension,10 and flat feet.9 Changes in biomechanics can put children at risk for injury, including spinal complications, slipped capital femoral epiphysis, Blount disease, and fractures.11 The role of physical therapy (PT) in childhood obesity intervention remains unclear.

The lack of clarity for the role of PT in childhood obesity intervention may be especially true for the school environment. The surgeon general of the United States suggested that school-based programs would be a valuable tool in the fight against childhood obesity.2 Preliminary data suggest the effectiveness of programs.12 Both parents and school personnel recognize the problem but feel helpless to effect change.13 There are divided opinions as to whether school responsibility should include use of special exercise classes.14 One challenge might be the lack of understanding as to the resources that are available. Physical education teachers and school nurses are included in conversations about who should address the problem14 but often not the school-based PT. Some have suggested that there is a specific role for the PT within school-based programs.15 One opinion is that body mass index affects gross motor educational outcomes and should be part of the early health screening by PTs for all students entering the education system.16 Other studies tie gross motor deficits to obesity in older children.17 There is no evidence that the majority of school-based PTs would agree with this view or welcome the additional role within their practice. There have been no studies investigating how school-based PTs perceive their role in addressing childhood obesity.


In 1990s, no states had an adult obesity rate greater than 15%, and most were below 10%.18 The incidence of overweight in elementary school children increased from 4% in 1971 to 19% in 2004.19 In 2007, the terminology used to describe children in the 95th percentile changed and this population thereafter was described as “obese.”20 In 2003, the NIH Obesity Research Task Force was established to define a common vision and promote collaboration among stakeholders.21 The American Academy of Pediatrics (2003) had taken the position that “regular physical activity should be consciously promoted, prioritized, and protected within families, schools and communities” and there is a recommendation to “routinely promote physical activity, including unstructured play at home, in school, in child care settings, and throughout the community.”21 As of 2004, there were more than 1800 active programs to combat childhood obesity registered in the national “Shaping America's Youth” database.22 Of these, 56% were in schools and community programs, 22% were in the health care sector, and 5% were in the area of communication/public advocacy and government affairs.22 Despite these efforts, childhood obesity was not being diagnosed by family physicians in 200523 and parents failed to recognize that their children were overweight.24

In 2006, the Council on Sports Medicine and Fitness and Council on School Health25 encouraged health care professionals to promote physical activity with the realization that overweight children would not outgrow their extra weight.26 Despite added attention, the Centers for Disease Control and Prevention's main antiobesity initiative had inadequate funding in 2006 to start childhood nutrition and exercise programs and only 28 states had programs.27 Part of the issue was that obesity and overweight were not recognized as a disability. Litigation in the form of Equal Opportunity Employment Commission versus Watkins Motor Lines defined the legal definition of disability and excluded obesity as a cause.28–30

Since the mid-2000s, there have been major changes in how obesity is addressed by health care practitioners and in communities. The Centers for Disease Control and Prevention published guidelines on healthy activity.31 Obesity was recognized as a disability by the American Medical Association.32 Between 2012 and 2013, more than 500 state bills that addressed childhood obesity were introduced.33 In 2001, the APTA stated that PT should “restore, maintain and promote not only optimal physical function but optimal wellness and fitness and optimal quality of life as it relates to movement and health.”34 Research focused on attitudes toward the inclusion of childhood obesity prevention in practice.35,36 Even if PTs viewed themselves as qualified, knowledgeable and best equipped to educate parents and teachers about childhood obesity, other stakeholders did not necessarily recognize the value of PT in this area.37

The study explores the attitudes of school-based PTs in addressing obesity and attempts to draw connections between personal/professional demographic descriptors and attitudes. The study also investigates how attitudes affect treatment frequency. A comparison is made between results from 2 separate studies, in 2008 and 2015, in an attempt to investigate emerging trends.



Participants were school-based physical therapists, mostly female, and mostly employed by the school district. The mean age was around 45 years, with about 15 years of experience as a school-based PT, on average. See Table 1 for further details on the demographics of the respondents.

TABLE 1 - Difference in Respondent Characteristics From 2008 to 2015
Demographic Characteristics of School-Based PT Respondents 2008 2015 P Value
Employment status District employee 61% 51% .258a
Private contractor 35% 39%
Contract (medical center, service agency) 4% 10%
Gender Female 82% 87% .478b
Male 18% 13%
Entry-level degree Bachelor's or certificate 72% 51% .009a
Master's 28% 36%
Doctorate 0% 13%
Highest degree Bachelor's or certificate 53% 35% .000a
Master's 44% 36%
Doctorate 3% 29%
2008 Mean (SD) 2015 Mean (SD) Effect Size Power P Value
Age of therapist, y 45.4 (9.3) 45.2 (10.4) 0.01 0.06 .940c
Experience in school-based PT, y 15.7 (9.2) 15.9 (9.1) 0.02 0.07 .892c
Adult obesity rate in state of practice 20.6 (5.7) 28.3 (3.0) 1.69 1.00 .000c
Caseload that is overweight, % 13.4 (10.4) 17.9 (1.9) 0.40 0.77 .020c
2008 2015 P Value
Consults for obesity only? 18% 7% .036b
Provides parent education about obesity? 30% 23% .234b
Teaches within a PT program? 2% 1% .620b
Member of APTA? 32% 38% .267b
Member of the pediatric section? 19% 28% .150b
Member of the SIG? 4% 11% .091b
Member of the sports and fitness SIG? 2% 2% .440b
Abbreviations: PT, physical therapy; SD, standard deviation; SIG, School Improvement Grant.
aThe threshold for statistical significance was set at .05 and determined via Pearson χ2.
bThe threshold for statistical significance was set at .05 and determined via the Fisher exact test.
cThe threshold for statistical significance was set at .05 and determined via an independent-samples t test. For 2008, n = 57, and for 2015, n = 92 (with up to one case missing for some variable).

For the original pilot, participants were selected randomly from a list resulting from a search of the Academy of Pediatric Physical Therapy, APTA, membership directory.38 One hundred twenty-two therapists were contacted by phone to gain access to an e-mail address. Participants were invited until a sample of 50 therapists responded. An e-mail invitation with a SurveyMonkey link was sent to therapists agreeing to participate. It was determined that contacting participants through the school districts would most likely better target the desired population. For the studies in 2008 and 2015, participants were contacted via the school districts from a national data base of public school districts.39,40 The 2 studies involved different groups of participants.

For the 2008 survey, possible respondents were chosen via a stratified cluster sampling technique, based on obesity prevalence of their state.18 Five states were randomly chosen from each category and 5 school districts were randomly chosen from each of those states. Districts were contacted directly via telephone, and e-mail addresses were requested. If the district did not have a PT, as was typically the case in smaller districts, another was chosen to replace it. In the 2008 round, responses were received from 57% of the 100 districts to which the invitations were sent.

For the 2015 study, the state obesity categorizations had changed and the clusters were no longer representative of the states chosen for the 2008 study. Instead of the stratified approach, 400 participant districts were randomly chosen from a list of all school districts in the United States40 with relative distribution according to population of each state. Districts with fewer than 3 schools were excluded, and potential participant districts were limited to those with public schools. District personnel were contacted either via telephone or e-mail obtained from the database or the school district Web sites. Invitations were sent either directly to the PT or to the special education director or other administrator, with requests to pass on the survey invitations to the PT who served that district. Four hundred invitations were sent. The survey completion rate for the 2015 study was 23%.


The survey consisted of 3 parts. Part 1 was demographic information (Table 1), part 2 was an attitude survey (Appendix 1, Supplemental Digital Content 1, available at:, and part 3 was a case series of 2 fictional, yet plausible, patient scenarios designed to investigate practice with respect to recommended treatment frequencies (Appendix 2, Supplemental Digital Content 2, available at:

The attitudes survey was developed to address attitudes of school-based PTs toward intervention for childhood obesity. A focus group of 5 local practicing school-based PTs developed/reviewed a list of 18 statements that supported the role of school-based physical therapists in addressing childhood obesity and additional constructs to support this role. The statements were equally distributed among 3 constructs: seriousness of the problem, role of the PT, and role of the school. A 7-point Likert-type scale was attached to the statements. The scale was pilot tested with a sample of 50 PTs randomly selected from a list of Academy of Pediatric Physical Therapy members.

The cases were written by the author based on experience as a school-based PT. Both cases were reviewed by the task force. Respondents were asked whether they would opt to treat each child, and if so, at what frequency. The frequency options were scripted to be consistent with Individual Education Plan (IEP) software choices available at the time.

Procedure and Analysis

For the 2008 and 2015 studies, the initial request was sent and a follow-up request was sent 2 to 4 weeks later. The response rate for the 2008 round was 57% and the response rate for the 2015 round was 23%. Statistical analyses were performed with SPSS version 23. Effect size and power were determined with G*Power3.0.10. Nonparametric tests were for variables that violated the rules of normality, which was defined as skewness or kurtosis of more than twice the value of the standard error for the respective statistic. An α level of 0.05 was predetermined to be the threshold for a significant difference between groups and for correlations between variables. Internal reliability for the attitude survey was assessed via Cronbach α, and was 0.887.

To determine the difference between the demographic variables from 2008 to 2015, parametric and nonparametric analyses were used. A Pearson χ2 was used to determine the difference in variables with more than 2 options, with nonnormal distributions. The Fisher exact test was used to determine differences between 2-option variables with nonnormal distributions. This included gender and all of the treatment and membership descriptive variables. The differences between normally distributed groups that involved interval variables, such as age, years of experience, adult obesity rate, and percentage of caseload overweight were determined using an independent-samples t test. This test and the Levene test for equality of variances were used to compare overall attitude survey scores between the 2 dates.

Treatment frequency changes from 2008 to 2015 for the 2 cases were analyzed separately with Pearson χ2. The influence of attitude survey scores on treatment frequency recommendations was analyzed with analysis of variance and Tukey post hoc analysis.

Investigation into the veracity of the subconstructs was completed by principle component analysis with varimax rotation method and Kaiser normalization. The number of components was determined by a minimum of 1 eigenvalue. Items were assigned to a component if their value in the component was at least 0.5 and at least double the value found in any other components. The 3-factor components aligned with the 3 original subconstructs with which the items were designed. The first component aligned with the PT subconstruct, containing all 5 items from the subconstruct and picking up 1 that was originally created to fit within the seriousness subconstruct. The second factor contained 3 of the school subconstruct items. The third factor contained only 1 item, but was the quintessential representation of the seriousness subcontract, so this factor was included in the analysis.

For the investigation into the component factors influence on treatment decisions, the frequency categories were reduced to 2: treat (which contained the “monthly,” “2 times per month,” “weekly,” and “2-3 times per week” responses) and no treat (which contained the “none” and “consult” responses). Statistical analysis was used to determine whether there was a difference in the component scores with respect to the 2 groups. A 1-way analysis of variance was used to determine whether there was a difference in the component scores with a normal distribution, PT factor, and total score. The Mann-Whitney U test was used to determine significance for the nonnormally distributed component scores, school factor and seriousness factor.


The demographic descriptors of physical therapists remained consistent between the 2 surveys (Table 1). There was a significant change in entry-level highest degree, increases in the obesity rate for the adult rate of the respondent's state and the rate of obesity for the children on caseload. There was a statistically significant decrease in the number of therapists who consulted for obesity alone.

The attitude scores are in Table 2. Higher scores indicate a more positive attitude about PT having a role in childhood obesity prevention within the school setting. The range of scores was 37 to 88 for 2008 and 33 to 81 for 2015, with a mean of 64.0 for 2008 and 60.6 for 2015. The standard deviation changed from 14.4 (2008) to 11.3 (2015), indicating reduced variability in 2015, that was statistically significant (P = .012). Only 1 of 14 individual items demonstrated a difference between the 2 years. Statement 14 (Appendix 1, Supplemental Digital Content 1, available at: was originally part of the seriousness subconstruct but better aligned with the PT component after factor analysis. The statement by itself demonstrated an increased value in 2015—mean (standard deviation) = 5.117 (1.753) in 2015 versus 4.228 (1.937) in 2008 (P = .004).

TABLE 2 - Attitude Scores and Relationships of Attitude Scores With Demographic Variablesa
Attitude Survey Scores 2008 Mean (SD) 2015 Mean (SD) Effect Size Power P Value
64.0 (14.4)b 60.6 (11.3)b 0.26 0.45 .140c
Correlations of Attitude Scores With 2008 2015
Age of therapist −0.286b −0.186
Years of experience in school-based PT −0.332b −0.231b
Caseload that is overweight, % 0.300b 0.141
Adult obesity rate in the state of respondent's practice (prevalence) 0.094 0.082
2008 2015
Mean (SD) Effect Size Power P Value Mean (SD) Effect Size Power P Value
Attitude scores: effect of associated membership
APTA member Yes (n = 18) 66.6 (13.3) 0.27 0.24 .358c Yes (n = 35) 63.7 (9.7) 0.49 0.73 .028c
No (n = 39) 62.7 (14.9) No (n = 57) 58.4 (11.9)
Pediatric section member Yes (n = 11) 65.5 (13.5) 0.13 0.10 .703c Yes (n = 26) 64.5 (8.4) 0.55 0.76 .029c
No (n = 46) 63.6 (14.7) No (n = 66) 58.8 (12.0)
Attitude scores: practice effects
Consults for obesity alone Yes (n = 10) 75.9 (9.5) 1.21 0.96 .003c Yes (n = 6) 70.2 (8.7) 1.03 0.78 .030c
No (n = 47) 61.4 (14.1) No (n = 85) 59.9 (11.1)
Provides parent education about obesity Yes (n = 17) 71.2 (14.3) 0.74 0.81 .012c Yes (n = 21) 63.7 (11.8) 0.35 0.40 .156c
No (n = 40) 60.9 (13.5) No (n = 70) 59.7 (11.0)
Attitude scores: effect of degree
Entry-level PT degree Bachelor (n = 41) 64.5 (14.3) 0.42 0.41 .036c Bachelor (n = 46) 59.3 (11.1) 0.32 0.75 .037d
Master (n = 16) 70.3 (13.0) Master (n = 33) 58.8 (11.4)
Doctorate (n = 0) N/A Doctorate (n = 12) 69.2 (9.2)
Highest degree attained (other than PT education) Bachelor (n = 30) 62.2 (15.2) 0.18 0.20 .593d Bachelor (n = 32) 59.5 (11.2) 0.27 0.60 .043d
Master (n = 25) 65.5 (14.0) Master (n = 33) 57.7 (11.3)
Doctorate (n = 2) 70.0 (8.5) Doctorate (n = 27) 64.8 (10.7)
Abbreviations: PT, physical therapy; SD, standard deviation.
aOverall attitude score comparisons made with an independent-samples t test.
bThe Levene test for equality of variances demonstrated a significant difference (.012). Correlations of attitude scores with continuous variables determined by Pearson bivariate 2-tailed correlation coefficients. It indicates statistical significance at .05.
cFor discrete demographic variables, statistical significance was determined by an independent-samples t test.
dOne-way analysis of variance.

There was a weak negative correlation between both age and years of experience with attitude scores (Table 2). Younger therapists with less experience tended to better accept their role in childhood obesity intervention in the schools. The negative correlation persisted in 2015 but was reduced and it remained statistically significant for years of experience but not for age. Percentage of caseload that was overweight has a minimal effect with attitudes. Professional association membership, both APTA and Academy of Pediatric Physical Therapy, related to attitude scores in 2015 but not in 2008. The practice strategies, consulting or providing parent education, diminished as evidenced by the effect size, but were significant. There was a significant effect of the highest degree (other than PT) on attitude scores in 2015 (Table 2).

For the 2008 study, there was no difference in the mean attitude score for the 4 cluster groups of adult obesity percentage for the state (P = .594). The mean (standard deviation) for each of the 4 groups was as follows: the 15% to 19% obesity rate cluster = 64.05 (13.67), the 20% to 24% cluster = 61.00 (14.77), the 25% to 29% cluster = 62.60 (14.03), and the more than 30% cluster = 69.10 (16.72). Cluster comparisons were not possible for the 2015 survey, as a different method of participant recruitment was used. However, there was no statistically significant correlation between state obesity rate and the attitude scores for either 2008 or 2015 (Table 2).

Treatment frequencies for the 2 cases are shown in Figure 1. For case 1, there was no change in the frequency of therapy sessions based on the year (P = .234). For case 2, there was a significant increase in the frequency of PT intervention sessions recommended (P = .007). The largest noticeable difference is a decrease in the PTs who chose no treatment (changed from 45% to 20%) and an increase in those who chose to treat weekly (changed from 5% to 24%). The same trend occurs for case 1 (the “none” category changed from 16% to 5% and the “weekly” category changed from 38% to 47%) but was not statistically significant.

Fig. 1.
Fig. 1.:
Treatment frequencies for the 2 cases. Statistical analysis was performed via Pearson χ2. For case 1, there was no demonstration of statistically significant difference (.234) in treatment frequencies between years. For case 2, there was a difference, statistically significant at .007.

Figure 2 graphs the attitude score and the frequency of treatment for cases. In 2008, there was a difference in scores between the “no treat” group and some of the treatment options for both of the cases. The attitude survey score tended to trend upward in relationship to the frequency recommendations for treatment. In 2015, this difference was not present, and the attitude score was similar regardless of recommendations for treatment frequency.

Fig. 2.
Fig. 2.:
Attitude survey score (mean, 95% confidence interval) based on frequency of treatment choices for the 2 cases for both years. Analysis was performed via 1-way analysis of variance. Statistically significant differences (.013 for case 1 and .006 for case 2) with Tukey post hoc analysis revealing differences between * and ** in 2008.

Table 3 includes the factor scores for each of the 2 cases. For 2008, the overall score and the PT and school factors had an influence on whether or not to treat, in one or both of the cases, as did the total score. This relationship is not present in 2015, but influence from the seriousness factor in treatment determination emerged for both cases.

TABLE 3 - Relationship of Attitude Score Factor Components With Treat/No Treat Decisiona
Attitude Survey Component Mean (SD) Effect Size Power P Value
2008 Case 1 Direct Treat (n = 47) No Treat (n = 9)
PT factor 26.51 (6.71) 15.78 (4.68) 1.85 1.00 .000b
School factor 11.74 (2.22) 9.56 (2.88) 0.85 0.75 .020c
Seriousness factor 4.60 (1.49) 4.00 (2.20) 0.30 0.22 .269c
Total attitude score 66.43 (13.30) 48.33 (7.37) 1.68 0.99 .000b
Case 2 Direct Treat (n = 31) No Treat (n = 25)
PT factor 27.84 (7.21) 21.00 (6.16) 1.02 0.98 .000b
School factor 11.68 (2.61) 11.04 (2.23) 0.26 0.25 .337c
Seriousness factor 4.65 (1.82) 4.44 (1.39) 0.12 0.12 .644c
Total attitude score 68.68 (14.10) 57.12 (11.61) 0.90 0.95 .002b
2015 Case 1 Direct Treat (n = 87) No Treat (n = 5)
PT factor 4.43 (1.93) 2.60 (2.07) 0.91 0.62 .099b
School factor 11.29 (2.40) 13.20 (0.84) 1.07 0.74 .062c
Seriousness factor 4.43 (1.93) 2.60 (2.07) 0.91 0.63 .043c
Total attitude score 60.91 (10.93) 51.80 (15.85) 0.67 0.42 .080b
Case 2 Direct Treat (n = 74) No Treat (n = 18)
PT factor 26.05 (6.22) 23.17 (9.76) 0.35 0.38 .121b
School factor 11.49 (2.19) 11.00 (3.07) 0.22 0.21 .776c
Seriousness factor 4.57 (1.91) 3.33 (1.91) 0.65 0.79 .016c
Total attitude score 61.22 (10.59) 57.11 (13.82) 0.33 0.35 .169c
Abbreviations: PT, physical therapy; SD, standard deviation.
aNo treat (no and consult) versus direct treat (monthly, every other week, weekly, 2-3 times per week).
bSignificance determined by 1-way analysis of variance.
cSignificance determined by the Mann-Whitney U test. The physical therapy factor contained 6 survey items, the school factor contained 3 survey items, and the seriousness factor included 1 survey item.


During the years between the 2 studies, there was no change in the average attitude score but there was a decrease in variability of scores. At the time of the 2005 study, the concept of treating for obesity was relatively unfamiliar to PT practitioners and it is plausible that members of the profession had not yet come to a consensus on scope of practice. For 2015, there is a better centralization of responses. This may be due to increased exposure of therapists to practice in obesity-related work, which may now be a part of the weekly routine41 or increased efforts on the part of the professional components to provide resources to members. The gain in scores based on membership to the APTA and the Academy supports to the latter. Younger therapists scored higher on the attitude survey than did older therapists for studies. While this might indicate the potential for change in attitude as younger therapists replace the older ones, a significant change had not yet occurred during this study.

While there was not an increase in the attitude of physical therapists toward intervention for obesity, there was a change in practice. In 2015, while fewer therapists consulted for obesity alone, more chose to treat the child in case 2 whose issues were solely obesity and obesity-related conditions. Perhaps therapists are gaining the ability to look past obesity being the cause of the issues and focus on the functional and health deficits that can be addressed by PT. In 2008, the notion of PT responsibility for the problem had the most effect on whether or not to treat, while in 2015 the seriousness of the problem had a greater effect. The change in referral, and the response to consulting, may indicate that therapists are beginning to accept the role and look further into the cause of disability for this case, to see past the obesity to see the functional problems and physical limitations. This might also be the reason for the significant change in response to the 1 item “If physical therapists concentrate on obesity, children with real disabilities might be neglected (opposite).” If therapists begin to see the effects of obesity as being a disability, then they would disagree more with this statement and score higher on the item.

While the treatment frequency for case 2 did increase, it was not as high as for case 1, despite his health and education being more greatly impacted by his weight. The 2 major differences in the cases were age and special education participation. Participants might have felt less effective with an older child or may have cast more blame on him because he had greater ability to make choices. Also, they (or their employers) may not have been as willing to actively treat a child who is not eligible for an IEP and/or have more policy restrictions on the use of Section 504 (Rehabilitation Act of 1973) for related services. The disparity of service allocation seen in these cases could potentially bolster the call by some to make obesity an IDEA classification.42

Another possible reason for a positive increase in the referrals for case 2 and not case 1 might be a recognition of the need for PT intervention dependent on the extremeness of the obesity. PTs are just now beginning to acknowledge different categories of obesity and how they can impact the need for more specialized services. Between 6%1 and 12%43 of children are in the extreme obesity category, and this subset of the population is still on the rise.43 Greater health issues are associated with severe obesity, which may indicate the need for more medically oriented intervention.43 Extreme cases have different underlying pathologies and genetic predispositions.44 PTs, as the medically oriented exercise experts, have the appropriate science background, and are better prepared to address these concerns,45 monitor soundness of systems, and devise more appropriate outcome measures.46 As the focus turns toward extreme cases of obesity, the role of PT is better recognized by the medical community as part of the multidisciplinary program required for success in intervention.43


Difficulty locating and recruiting participants was a challenge, which increased in the 2015. The districts were less willing or able to forward the invitation to the PTs. Changes in employment structure or district structure, emergence of charter schools, and shifts in PT staffing may have impacted the response rate. PTs were easier to contact in general, as many had e-mail addresses on the district Web sites, which were more accessible in 2015. However, that easy access may have resulted in more survey requests and survey fatigue. Of note, for the 2015 round, some districts had enacted application processes for requesting research participation for the district personnel. These processes were time-consuming and resulted in exclusion of the PTs from those districts. For many of the comparisons, if there was no statistically significant difference, power was not adequate to be able to conclude equality of the groups.

Obesity is a sensitive and controversial topic and self-selection might have exacerbated the problem of response rate. The 2008 survey contained a question asking how many pounds the PT would like to lose in order to be the ideal weight. Many respondents chose not to answer that question, and of the ones who did, nearly all responded with zero, indicating a possible bias. There were several respondents who started the survey and then did not complete it after coming to that question. There were also some negative comments about that question within the qualitative section of the 2008 survey, which is not included in this report. The question was not included in the 2015 survey and is not reported here. However, it highlights the potential sensitivity of the topic for the respondents themselves.

Face validity was the primary means used to assess the tool, as it was designed around a novel concept and could not be compared with other existing instruments. While some efforts were made to triangulate the information by including clinical cases, the relationship with the attitude score was not consistent. The number of items was limited, especially in the third factor (seriousness) in which conclusions were based on responses to 1 item alone. The sample was small, especially for 2008, and there was a difference in sample size from 2008 (n = 57) to 2015 (n = 96).

There was a slightly different selection process between 2008 and 2015. In 2008, a cluster sample was used based on the obesity categories for the states for that time frame. These clusters did not have an effect on the attitude survey score nor did the participant's state. The method changed for 2015 because the obesity categorizations for the states changed and the clusters were no longer intact.


The author thanks Rick Tivis for consultative support in statistical analyses, Dr Jerry McMurtry and Dr Chris Williams for assistance with the original instrument design and analysis, Diva Lamotte for assistance with respondent recruitment, and my clinical pediatric physical therapy colleagues for input on the initial survey items.


1. Ogden CL, Carroll MD, Lawman HG, et al. Trends in obesity prevalence among children and adolescents in the United States, 1988-1994 through 2013-2014. J Amer Med Assoc. 2016;315(21):2292–2299.
2. Committee on Health, Education, Labor, and Pensions. Childhood Obesity: Beginning the Dialogue on Reversing the Epidemic: Hearing of the Committee on Health, Education, Labor, and Pensions, United States Senate, One Hundred Eleventh Congress, Second Session, on Examining Childhood Obesity, Focusing on Reversing the Epidemic, March 4, 2010 [e-book]. Washington, DC: US Government Printing Office; 2012.
3. Simmonds M, Llewellyn A, Owen C, Woolacott N. Predicting adult obesity from childhood obesity: a systematic review and meta-analysis. Obes Rev. 2016;17(2):95–107.
4. Centers for Disease Control and Prevention. Overweight and obesity. Published September 2015. Accessed July 4, 2016.
5. Gushue D, Houck J, Lerner A. Effects of childhood obesity on three-dimensional knee joint biomechanics during walking. J Pediatr Orthoped. 2005;25(6):763–768.
6. McMillan A, Auman N, Williams D, Collier D. Frontal plane lower extremity kinematics and kinetics in boys who are overweight versus healthy weight during walking. Pediatr Phys Ther. 2008;20(1):116–117.
7. Dunn P, Kelley K, Mund T, Riechers L, Gajdosik C, Fehrer S. Obesity and posture in adolescents. Pediatr Phys Ther. 2008;20(1):106–106.
8. Pathare N, Haskvitz E, Selleck M. Comparison of measures of physical performance among young children who are healthy weight, overweight, or obese. Pediatr Phys Ther. 2013;25(3):291–296.
9. Jankowicz-Szymanska A, Mikolajczyk E. Genu valgum and flat feet in children with healthy and excessive body weight. Pediatr Phys Ther. 2016;28(2):200.
10. Omalley G, Hussey J, Roche E. A pilot study to profile the lower limb musculoskeletal health in children with obesity. Pediatr Phys Ther. 2012;24(3):292–298.
11. Wills M. Orthopedic complications of childhood obesity. Pediatr Phys Ther. 2004;16(4):230–235.
12. Wagner B, Grant-Beuttler M, Borja M, et al. Exercise and nutrition school program for adolescents classified as at risk for or overweight. Pediatr Phys Ther. 2006;18(1):109–110.
13. Patino-Fernandez A, Hernandez J, Villa M, Delamater A. School-based health promotion intervention: parent and school staff perspectives. J School Health. 2013;83(11):763–770.
14. Clarke J, Fletcher B, Lancashire E, Pallan M, Adab P. The views of stakeholders on the role of the primary school in preventing childhood obesity: a qualitative systematic review. Obes Rev. 2013;14(12):975–988.
15. Bucher H, Hoban S, Cojocaru C, Nichols S, Lovelace-Chandler V. Overweight children in preschool classrooms: prevalence, functional characteristics and possible roles for physical therapists. Pediatr Phys Ther. 2006;18(1):85–85.
16. Nervik D, Martin K, Rundquist P, Cleland J. The relationship between body mass index and gross motor development in children aged 3 to 5 years. Pediatr Phys Ther. 2011;23(2):144–148.
17. Nunez-Gaunaurd A, Moore J, Roach K, Miller T, Kirk-Sanchez N. Motor proficiency, strength, endurance, and physical activity among middle school children who are healthy, overweight, and obese. Pediatr Phys Ther. 2013;25(2):130–138.
18. Centers for Disease Control and Prevention. Overweight and Obesity: U.S. Obesity Trends 1985-2006. Published December 2006. Accessed December 14, 2007.
19. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA. 2006;295(13):1549–1555.
20. Krebs NF, Himes JH, Jacobson D, Nicklas TA, Guilday P, Styne D. Assessment of child and adolescent overweight and obesity. Pediatrics. 2007;120:S193–S228.
21. Huang TT-K, Horlick MN. Trends in childhood obesity research: a brief analysis of NIH-supported efforts. J Law Med Ethics. 2007;35(1):148–153.
22. Donato KA. National health education programs to promote healthy eating and physical activity. Nutr Rev. 2006;64(2):s65–s70.
23. Louthan MV, Lafferty-Oza MJ, Smith ER, Hornung CA, Franco S, Theriot JA. Diagnosis and treatment frequency for overweight children and adolescents at well child visits. Clin Pediatr. 2005;44:57–61.
24. Baughcum AE, Chamberlin LA, Deeks CM, Powers SW, Whitaker RC. Maternal perceptions of overweight preschool children. Pediatrics. 2000;106(6):1380–1386.
25. Council on Sports Medicine & Fitness and Council on School Health. Active healthy living: prevention of childhood obesity through increased physical activity. Pediatrics. 2006;117:1834–1842.
26. Reany P. Eleven Seen as the Key Age for Obese Children. Reuters Health Information. Published 2006. Accessed May 18, 2006.
27. Associated Press: Efforts to help obese children floundering. Accessed October 10, 2007.
28. Institute of Management & Administration. Legal insights: morbid obesity not ADA impairment: 6th circuit. HR Focus. 2006;83(12):2.
29. Tolle NL. Non-physiologically caused obesity is not an impairment under ADA, circuit court says. Employee Benefit Plan Rev. 2007;61(7):28–29.
30. Zachary MK. Obesity and the ADA—the reason matters. Labor Law. 2006;67(12):23–26.
31. Centers for Disease Control and Prevention. Physical activity basics. Accessed December 15, 2016.
32. Pollack A. AMA recognizes obesity as a disease. New York Times. June 2013. Accessed July 1, 2015.
33. Bleich S, Jones-Smith J, Jones H, O'Hara M, Rutkow L. The voices for healthy kids campaign and US State Legislation to prevent childhood obesity. Am J Public Health. 2016;106(3):436–439.
34. American Physical Therapy Association. Guide to Physical Therapists Practice. 2nd ed. Alexandria, VA: American Physical Therapy Association; 2001.
35. Kendall EA. Attitudes and practice patterns of school-based physical therapists regarding intervention for childhood obesity. Pediatr Phys Ther. 2014;26(1):164–165.
36. Huber GM, McKenna S, Shah M, Healey WE. Health promotion of pediatric patients in physical therapy practice: a nationwide study. Pediatr Phys Ther. 2011;23(1):99–100.
37. Schlessman A, Martin K, Ritzline P, Petrosino C. The role of physical therapists in pediatric health promotion and obesity prevention: comparison of attitudes. Pediatr Phys Ther. 2011;23(1):79–86.
38. Pediatric Section of the American Physical Therapy Association. Membership Directory. Accessed November 2008.
39. National Center for Education Statistics. Search for public school districts: Institute for Educational Sciences, U.S. Department of Education. Accessed December 2008.
40. US Department of Education: National Center for Education Statistics. Search for Public school Districts. Accessed June 2015.
41. Milne N, Choy N, Leong G, Hughes R, Hing W. Child obesity service provision: a cross-sectional survey of physiotherapy practice trends and professional needs. Aust J Primary Health. 2016;22(2):140–146.
42. French R, Sanborn C, Dimarco N, Stephens T. Childhood obesity: classification as an IDEA disability. Palaestra. 2016;30(2):17–26.
43. Wickham E, DeBoer M. Evaluation and treatment of severe obesity in childhood. Clin Pediatr. 2015;54(10):929–940.
44. Bass R, Eneli I. Severe childhood obesity: an under-recognized and growing health problem. Postgrad Med J. 2015;91(1081):639.
45. Nichols S, Cojocaru C, Hoban S, Bucher H, Lovelace-Chandler V. Implications for physical therapy for severely overweight children in preschool classrooms: a clinical case report. Pediatr Phys Ther. 2006;18(1):101–101.
46. Kendall E. Exploration of common pediatric outcome measures for assessing a child in the class 3 obesity category. Pediatr Phys Ther. 2015;27(4):E8.

attitude; obesity; school

© 2018 Academy of Pediatric Physical Therapy of the American Physical Therapy Association