what percent- age of children and adolescents would be overweight, obese, and ExHi obese in 2030?

Trends in Body Mass Index and Prevalence of Extreme High Obesity Among Pennsylvania Children and Adolescents, 2007–2011: Promising but Cautionary David Lohrmann, PhD, Ahmed YoussefAgha, PhD, and Wasantha Jayawardene, MD

The economic consequences of obesity in the United States were estimated at $147 billion annually in 2008.1 To better understand these costs, obesity trends to the year 2030 were predicted.2 Obesity prevalence could reach 51% by 2030, but is more likely to stay at more than 40% because of recently emerging posi- tive developments. A subcategory, severe obe- sity, that is, body mass index (BMI; defined as weight in kilograms divided by the square of height in meters) of 40 or greater for adults, has increased faster than overall obesity and is projected to grow from 5% of adults in 2010 to 11% of adults by 2030.2 This growth, with its attendant increased risks of disease, will esca- late costs even if overall obesity prevalence stabilizes.2

Because obesity rates vary across states, the financial burden is not uniform.3 State-specific differences, such as lower cost of less healthy foods, can affect obesity and severe obesity prevalence together with current and projected health care costs.2 Because of the state-specific nature of Medicaid and Medicare expenditures, much of the high cost of obesity-related disease is borne by public sector health plans.

Today’s children and adolescents will be the youngest adults in 2030; therefore, obesity prevention for the future requires monitoring of obesity prevalence rates among this popu- lation over time. Prevalence and trends in obesity among US children from 1999 to 2010 were determined based on National Health and Nutrition Examination Survey data.4 Preva- lence of high BMI in US children and adoles- cents has also been studied.5 By 2010, fewer than 12% of those aged 2 to 19 years nation- wide were at or above the 97th percentile (extreme high obese [ExHi obese]); 17% were above the 95th percentile (obese), and 32% were above the 85th percentile (overweight). A statistically significant increase among 6- to

19-year-old males with a BMI at or above

the 97th percentile was found between 1999

and 2008.4

To inform prevention efforts, state govern- ments have a vested interest in monitoring

obesity prevalence among all age groups, and

especially among children and adolescents.

Pennsylvania, for example, mandates annual

height and weight screening with BMI calcula-

tion for all public school students statewide.6

One recent study assessed child and adolescent

BMI trends in Pennsylvania, excluding Phila-

delphia and surrounding counties, for 2005 to

20097 and found combined overweight and

obese rates decreased from 28.5% to 23.1% at

the middle school level and from 24.6% to

20.9% at high school levels, but increased from

10.9% to 20% at the elementary level. The

largest shift in BMI over the subset of years

from 2007 to 2009 was among overweight

elementary students; 58% of those who were

overweight in 2007 were obese in 2009.

Overweight and obese increased for the study

population as a whole because of this sharp

increase among elementary students. In a sec-

ond, separate study,8 trends in obese (BMI

‡ 95th percentile) and ExHi obese (defined8

as BMI ‡ 35 kg/m2) among 5- to 18-year-old students attending Philadelphia schools in 2006

to 2010 were determined; obesity across all

ages decreased from 21.5% to 20.5% and ExHi

obese from 8.5% to 7.9%. Obese and ExHi

obese were most prevalent among middle

school students, Hispanic boys, and Black girls.8

The purpose of our study was to determine prevalence, trends, and patterns in overweight,

obese, and ExHi obese among Pennsylvania

school children. Specific research questions were:

Objectives.We determined current trends and patterns in overweight, obesity,

and extreme high obesity among Pennsylvania pre-kindergarten (pre-K) to 12th

grade students and simulated future trends.

Methods.We analyzed body mass index (BMI) of pre-K to 12th grade students

from 43 of 67 Pennsylvania counties in 2007 to 2011 to determine trends and to

discern transition patterns among BMI status categories for 2009 to 2011.

Vinsem simulation, confirmed by Markov chain modeling, generated future

prevalence trends.

Results. Combined rates of overweight, obesity, and extreme high obesity

decreased among secondary school students across the 5 years, and among

elementary students, first increased and then markedly decreased. BMI status

remained constant for approximately 80% of normal and extreme high obese

students, but both decreased and increased among students who initially were

overweight and obese; the increase in BMI remained significant.

Conclusions. Overall trends in child and adolescent BMI status seemed

positive. BMI transition patterns indicated that although overweight and obesity

prevalence leveled off, extreme high obesity, especially among elementary

students, is projected to increase substantially over time. If current transition

patterns continue, the prevalence of overweight, obesity, and extreme high

obesity among Pennsylvania students in 2031 is projected to be 16.0%, 6.6%,

and 23.2%, respectively. (Am J Public Health. 2014;104:e62–e68. doi:10.2105/



e62 | Research and Practice | Peer Reviewed | Lohrmann et al. American Journal of Public Health | April 2014, Vol 104, No. 4

1. What were the prevalence and trends in overweight, obese, and ExHi obese from 2007 to 2011 among elementary, middle, and high school students?

2. What movement patterns, if any, occurred in normal weight, overweight, obese, and ExHi obese among Pennsylvania elemen- tary, middle, and high school students from 2009 to 2011?

3. If current patterns continue, what percent- age of children and adolescents would be overweight, obese, and ExHi obese in 2030?


Nurses in more than 1157 pre-kindergarten (pre-K) to 12th grade public and private schools located in 43 of 67 Pennsylvania counties, excluding Philadelphia and surrounding counties, used an electronic health record in- cluded in a web-based school health portal called “Health eTools for Schools” to record and report student medical data,7,9 including the annual height and weight for all enrolled students measured by established protocols.10

Along with unique identifiers, gender, and date of birth, medical data were compiled in a data repository maintained by InnerLInk (Lancaster, PA), the company that provides Health eTools at no cost to schools through funding from the Highmark Foundation.9 All appli- cable federal and state safeguards of family and student rights, both medical and educa- tional, were followed in the compilation of these data. Access was provided to de- identified data on the InnerLink server via a password-protected Internet link.

Between 2007 and 2011, a total of 685 531 viable student health records were collected. The number varied, with 71 487 for 2007, 186 585 for 2008, 107 705 for 2009, 107 699 for 2010, and 212 055 for 2011. Files were configured into a relational database by using data processing techniques, which were then summarized and aggregated into 3 categories: age, gender, and school level (i.e., elementary, middle, and high school). Because race/ethnicity was not recorded in student health records, this variable could not be addressed. The total number of data strings was sufficiently robust for analyses.

A SAS program11 for children and adoles- cents developed by the US Centers for Disease Control and Prevention (CDC), with 2000 as the growth reference year for calculation of percentiles and z-score, was used to calculate individual BMI. Because of a number of factors that influence height and weight in children, growth chart percentiles were used to deter- mine high BMI in children and adolescents5; the 97th percentile was adequate for seg- menting ExHi BMI-for-age in children.12

Therefore, overweight was defined for this study as at or above the 85th percentile but less than 95th percentile, obese as at or above the 95th percentile but less than 97th percen- tile, and ExHi obese as at or above the 97th percentile. We validated data to eliminate in- consistencies and unrealistic outliers for BMI, with values of BMI greater than 56.3 (56.3 = 40 + 3 · SD, i.e., 2.25% over the upper normal mass limit 56.3/25 = 2.25) and less than 7 eliminated. Outliers constituted 263 of 685 531 cases (0.04%).

We analyzed BMI trends using the least- squares method, a simple linear regression formula, BMImean = a0 + a1 · Year, which was used to ascertain trends in annual BMI percentage for overweight, obese, and ExHi obese over 5 years for all students. This yielded 3 separate equations (the 1.95% of under- weight students was not a focus of this study). Correlations between the dependent variable (percentage of normal weight, percentage of overweight, percentage of obese, or percentage of ExHi obese) and the independent variable (year) were checked before constructing re- gression models. We used the Pearson v2 test to determine significant differences based on gender, distributed over the 5 years (2007— 2011), controlling for BMI category and school level.

To reveal possible BMI transitions from 2009 to 2011, we calculated BMI categories via conditional probabilities, based on Bayesian statistics. We applied the v2 test to determine significance levels. Only students with matched identification numbers for 2009, 2010, and 2011, and only those who remained exclu- sively within a school level (i.e., elementary, pre-K–5; middle, 6—8; and high school, 9—12) over the 3 measurement years (2009—2011) were included in the analysis. This approach helped avoid cross-contamination for school

level type, yet still yielded viable data from more than 80 000 students.

Using Vinsem13 software, we created a sim- ulation covering 20 years that calculated future rates of overweight, obese, and ExHi obese based on (1) the number of students within each BMI category in 2009, (2) current con- ditional BMI movement patterns, and (3) as- sumed continuation of the current BMI move- ment patterns. Vinsem software was previously used to simulate epidemics of both infectious14

and chronic disease.15 We confirmed the sim- ulation results by Markov chain modeling.16


Regarding BMI trends, yearly percentages for overweight increased somewhat from 2007 to 2009, but the linear slope lines for all 3 categories declined from 2007 to 2011 (Figure 1).

Because food services, availability of food in school, and opportunities to be physically active, along with prevention and intervention initiatives might have varied, BMI status data were segmented by school level. School level also separated students by developmental cat- egories—childhood (elementary school), young adolescence (middle school), and middle ado- lescence (high school). Therefore, percentages of students in the overweight, obese, and ExHi obese categories were provided by school level (Figure 2) for 2007 to 2011. Combined rates of overweight, obese, and ExHI obese de- creased steadily from 2009 to 2011 for all school levels. The combined rates for middle and high school steadily declined across all years. After increasing from 2007 to 2009, the combined rate at elementary school peaked in 2009 and receded thereafter. With the excep- tion of elementary students in 2008 and 2009, the combined percentage of obese and ExHi obese students was greater than the percentage overweight at all school levels for all years. Likewise, the highest percentages of combined obese and ExHi obese students were found in middle schools. For all school levels and across all years, the percentage of ExHi obese students was more than double the percentage of obese students. Based on the Pearson v2

test, elementary school boys were more likely than girls to be overweight (P = .01) or obese (P = .04). Both middle and high school


April 2014, Vol 104, No. 4 | American Journal of Public Health Lohrmann et al. | Peer Reviewed | Research and Practice | e63

boys were more likely to be ExHi obese (P = .01).

Figure 3 depicts transitions in student BMI status from 2009 to 2011, with results pro- vided as percentages for students in grades pre- K to 12, as well as separately for elementary, middle, and high school students as provided, respectively, in parentheses. The subset of 80 770 students included in these percentages had their BMIs calculated in 2009, 2010, and 2011, and were linked by unique member identifiers for this analysis. (Data from the very low percentage of underweight students in the sample were excluded to assure more accurate v2 results.) Although overweight, obese, and ExHi obese prevalence rates for this subset were somewhat lower than that for the overall study population, these differences were not statistically significant (P= .723).

Between 2009 and 2011, more than 80% of students who were normal or ExHi obese did not change category, whereas almost half of the students initially in the overweight category and approximately three quarters of those in the obese category decreased or increased their BMIs; rates at which students remained within their initial BMI category were relatively con- sistent by school level. For example, the per- centage of obese students in each school level

(elementary: 25.04%; middle school: 23.66%; high school: 23.06%) all clustered around the overall rate of 24.38%.

Several BMI transition patterns were evident (Figure 3). Loop 1 presents BMI patterns for normal and overweight, and loop 4 shows BMI patterns for normal and obese students. For all students, movement from overweight to nor- mal was 19% higher than for movement from normal to overweight (loop 1), a pattern that was somewhat more pronounced for middle and high school students than for elementary students. Additionally, 7 times more students moved from obese to normal (loop 4) than moved from normal to obese; this ranged from 5.92% of elementary to 10.03% of high school students who were obese in 2009 and normal in 2011.

Loops 2, 3, and 5 present the BMI patterns for overweight, obese, and ExHi obese. The combined percentages of all students who moved from obese (loop 2) or ExHi obese (loop 5) to overweight (36.59%) were substantially higher than the combined percentages of stu- dents who moved from overweight to obese and ExHi obese (23.55%). Conversely, 4.5 times more students moved from obese to ExHi obese than moved from ExHi obese to obese (loop 3). This pattern was similar for students

from all 3 school levels, with a slightly higher percentage of elementary students moving from obese to ExHi obese. In addition, a greater combined percentage of elementary (52.16%) than middle (44.01%) or high school (42.12%) students moved from obese to ExHi obese and overweight to ExHi obese, and fewer elemen- tary (14.51%) than middle (15.57%) or high school (17.52%) students moved in the op- posite direction from ExHi obese to obese and ExHi obese to overweight. Based on the simulation of BMI category transitions (Figure 4), the prevalence of overweight, obese, and ExHi obese among Pennsylvania students in 2031 was projected to be 16.0%, 6.6%, and 23.2%, respectively, with the highest prevalence of ExHi obese among elementary students (31%; middle school, 17%; high school, 13%).


The year 2009 appeared to have been a watershed for child and adolescent obesity in Pennsylvania. The rapidly escalating over- weight and obesity prevalence among elemen- tary students peaked in that year, and then decreased in 2010 and 2011 to approximately 2007 levels. Although, in retrospect, the 5-year trend began declining for all 3 conditions in 2007, this decline was not detectable before 2009. By 2010, a similar trend was identified for obese and ExHi obese students in the Philadelphia, Pennsylvania area.8 Based on overall percentages, Pennsylvania made nota- ble progress toward achieving the Healthy People 2020 obesity prevalence objectives for children (aged 6—11 years; 15.7%) and adolescents (aged 12—19 years; 16.1%).17

Despite these promising findings, the preva- lence of overweight, obese, and ExHi obese among Pennsylvania children and adolescents was still more than 2% points higher in 2011 than for the United States in 2010.4 Consistent with national findings,5 middle school- and high school-aged boys were more likely than their female counterparts to be ExHi obese. If all individuals with BMIs at or above the 95th percentile were considered, approxi- mately one third were classified as obese and two thirds as ExHi obese; the percentage of children and adolescents who were ExHi obese in 2011 already exceeded the 2030

y = –0.6833x + 1390.9 R² = 0.51

y = –0.47x + 950.1 R² = 0.93

y = –0.1647x + 344.91 R² = 0.60






Pe rc

en ta

ge o

f S tu

de nt



Overweight Obese ExHi obese

Linear (overweight) Linear (obese) Linear (ExHi obese)

2007 2008 2009 2010 2011

Note. ExHi = extreme high. The sample size was n = 685 531. The city of Philadelphia and its surrounding counties were

excluded from this analysis.

FIGURE 1—Trend in overweight, obese, and extreme high obese prevalence by percentage:

Pennsylvania schools, 43 of 67 counties, 2007–2011.


e64 | Research and Practice | Peer Reviewed | Lohrmann et al. American Journal of Public Health | April 2014, Vol 104, No. 4

severe obese projections for US adults by more than 2 percentage points (13.7% vs the projected 11%).

Uniquely, our study and 1 previous study of the Pennsylvania school population7 employed mathematical modeling to determine whether

student BMI status remained static or changed over time. Our study confirmed the previous finding7 that child and adolescent BMI status moved substantially in both desirable and un- desirable directions, especially among over- weight and obese categories, within relatively short periods of time. Results indicated that the movement of students from overweight toward obese and ExHi obese and from obese to ExHi obese, especially among elementary students, tended to overpower movement in the oppo- site direction. Therefore, the 20-year simula- tion determined that the prevalence of ExHi obese among Pennsylvania pre-K to 12th grade students could almost double by 2031, pri- marily driven by current transition patterns among elementary school children. The prev- alence of obesity and ExHi obesity among today’s children when they are adults 15 to 20 years hence cannot be predicted; however, previous research showed that children with higher levels of obesity18 and who were obese as adolescents19 were likely to be obese as adults.18,19 Obesity prevalence was shown to double twice from adolescents to adults in their early 30s, with obese adolescents most likely to remain obese as adults.20

On the positive side, substantial percentages of students moved from ExHi obese back to either obese or overweight and from obese to overweight or normal weight in 2009 to 2011. The previous study7 found that 56% of over- weight elementary students moved to obese status between 2007 and 2009, but based on findings of our present study, that percentage then dropped by more than half (24.7%) in 2009 to 2011. These developments, if sus- tained, could help reduce the prevalence of ExHi obesity20 because they clearly demonstrated that movement in the desirable direction by a considerable percentage of individuals is possible. Additionally, this type of information, when known, could be used to target interven- tion programming for the greatest impact.7

Determining the exact reasons for emerging BMI trends and movement patterns was not possible in this case because the kinds, amount, and intensity of healthy eating and physical activity programs in participating schools were not monitored and might have varied. None- theless, some circumstantial information about improvements in school health policy, envi- ronment, and programs nationally, and for





























































Percentage of Students

G ra

de a

nd Y

ea r

Overweight Obese ExHi obese

Note. BMI = body mass index; ExHi = extreme high; G6–8 = middle school; G9–12 = high school; Pre-K–5G = elementary

school. The sample size was n = 685 531; elementary school: n = 328 687; middle school: n = 182 851; high school: n =

173 993. Female: n = 335 111; mean BMI = 20.773 (95% confidence interval [CI] = 20.714, 20.751). Male: n = 305 420;

mean BMI = 20.647 (95% CI = 20.647, 20.683). The city of Philadelphia and its surrounding counties were excluded from

this analysis.

FIGURE 2—Percentages of overweight, obese, and extreme high obese students by school

level: Pennsylvania schools, 43 of 67 counties, 2007–2011.


April 2014, Vol 104, No. 4 | American Journal of Public Health Lohrmann et al. | Peer Reviewed | Research and Practice | e65

Pennsylvania, were known. As previously in- dicated, personnel in all Pennsylvania public schools were mandated through state policy to accurately measure every student’s height and weight annually, notify all parents or guardians, in writing, of their child’s BMI status, and encourage them to bring this to their child’s physician’s attention if the BMI was in the overweight or obese ranges.6 Other school policies and practices were more supportive of healthy eating and increased physical activ- ity.21 Through its reauthorization of the school breakfast and lunch programs in 2004,22

Congress mandated that, by 2006, all partici- pating US schools adopt a wellness policy aimed at improving nutrition education, op- portunities for physical activity, and the food environments in schools. In May 2006, the Pennsylvania State Board of Education ampli- fied this broad federal mandate by adopting specified physical activity and nutritional stan- dards for public schools intended to incorpo- rate opportunities for students to be physically active, including recess and physical education, promote Safe Routes to School, and assure that all students participated in 30 minutes of daily physical activity.6 Nutrition standards for competitive foods in schools were also man- dated.23 Again in 2006, the Pennsylvania De- partments of Education and Health partnered with Highmark Foundation’s Healthy High 5 program, a 5-year, $100 million initiative that

supported a variety of strategies in schools designed to address physical activity, nutrition, and other critical health issues.24 Pennsylvania is the fourth largest recipient of US Department of Agriculture Supplemental Nutrition Assis- tance Program Education funding nationally, and in 2010, it devoted $21 million to serving 221227 school-aged children.25 Previous re- search found that student fat, sugar, and calorie intake was reduced26 and BMI was positively affected27 in states with laws regulating foods sold in schools outside of the federal school meal program (i.e., competitive foods).

At the national level, the Clinton Foundation negotiated an agreement with the soft drink industry that subsequently resulted in a 90% reduction in calories distributed to schools.28

Related positive changes were documented in Pennsylvania at the school level.29,30 Data collected biannually from school administra- tors by the Pennsylvania Department of Edu- cation, and reported by the CDC, indicated that the presence of at least 1 vending machine decreased to 68% of schools in 2010, down from 77% in 2006, with content changes as well. The presence of soda pop and fruit drinks that were not 100% juice decreased from 51% of schools to 24%, and sports drinks decreased from 62% to 49%.30 Programmatically, 77% of Pennsylvania schools instituted some type of wellness advisory board by 2010, and nearly all established expected outcomes for physical

education.29 Also by 2010, 77% of schools required students to complete 2 or more health courses, up from 65% in 2006.


This study had several limitations. Informa- tion about race/ethnicity was not collected in student health records; therefore, no analyses based on this variable were conducted. How- ever, some applicable demographic informa- tion was available. The racial/ethnic composi- tion of the 43 counties containing study schools was 82.4% White, 7.9% Black, 9.7% other, 8.4% Hispanic, and 91.6% non-Hispanic.31 Of the19 Pennsylvania counties classified in 2010 as urban,3112 (63%) were represented in this study. In these 12 counties, 17.9% of children lived in poverty compared with12.4% in the 31 rural counties (16.0% combined).31,32

Furthermore, the number of student data strings available for analysis varied because the number of schools using Health eTools for Schools changed yearly, with some schools dropping off and others joining. Additionally, no comparisons could be made with students attending schools located in the 24 excluded Pennsylvania counties because health record data, including BMI, were only available from schools that used Health eTools for Schools. Because of the pattern of new children begin- ning school and others graduating from high school each year, some students’ height and

Note. BMI = body mass index; ExHi = extreme high; G6–8 = middle school; G9–12 = high school; pre-kindergarten–5G = elementary school. The sample size was n = 80 770; elementary school: n = 48 309;

middle school: n = 24 384; high school: n = 8077. Conditional probabilities for individually matched BMI, 2009–2011. Normal→OverW = P(OverW11 Normal09) = 9.16%. Obese→Normal = P(Normal11 Obese09) = 7.11%. For percentages enclosed in parentheses, the first percentage pertains to elementary school, the second percentage pertains to middle school, and the

third percentage pertains to high school. The city of Philadelphia and its surrounding counties were excluded from this analysis. P < .001 based on the v2 test compared with the expected values.

FIGURE 3—Pattern of student body mass index migration reported by percentage: Pennsylvania schools, 43 of 67 counties, 2009–2011.


e66 | Research and Practice | Peer Reviewed | Lohrmann et al. American Journal of Public Health | April 2014, Vol 104, No. 4

weight could not be measured for the 3 times required to be included in some analyses. Regardless, the total number of student data strings provided for any 1 year was sufficiently robust, as was the number of data strings avail- able for multiyear comparisons, to generate reli- able results. Because environments, medical technology, and behaviors might change, the simulation of obesity prevalence was not a pre- diction. Rather, simulation results suggested the prevalence rates should the child and adolescent BMI transition patterns of 2009 to 2011 remain unchanged over time. The simulation results also provided information that policymakers could use for generating better-informed decisions about obesity prevention resource allocation.


Overall trends in child and adolescent BMI status seem to bode well for Pennsylvania’s future. BMI transition movement patterns, however, told a somewhat different story. Overweight and obesity prevalence were es- sentially leveling off. However, ExHi obesity, especially among elementary students, is projected to increase over time. The public health challenge most crucial to reversing the obesity epidemic is preventing the overweight

and obese children and adolescents of 2011 from moving into the obese or ExHi obese categories along with accelerating movement from ExHI obese and obese back toward over- weight and normal weight. To this end, evalua- tions should be conducted at the school level to assure compliance with mandated obesity pre- vention policy, environment, and program initia- tives, as well as to determine which, if any, school-based initiatives are clearly associated with improved BMI trends, and therefore, might pro- vide the greatest benefit. Given the fiscal impli- cations, state officials should be motivated to invest the current resources required to substan- tially improve the obesity and severe obesity trends among the adults of tomorrow. j

About the Authors David Lohrmann and Wasantha Jayawardene are with the Department of Applied Health Science, Indiana Uni- versity School of Public Health—Bloomington. Ahmed YoussefAgha is with the Department of Epidemiology and Biostatistics, Indiana University School of Public Health— Bloomington. Correspondence should be sent to Wasantha Jayawardene,

Department of Applied Health Science, SPH Bldg. 116, 1025 E 7th Street, Bloomington, IN 47405 (e-mail: wajayawa@indiana.edu). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints” link. This article was accepted December 14, 2013.

Contributors D. Lohrmann contributed to the interpretation of find- ings and writing of the article. A. YoussefAgha contrib- uted to the data mining and analysis. W. Jayawardene contributed to data validation and review of the article.

Acknowledgments We thank the Highmark Foundation and Robert G. Gillio, MD, InnerLink Inc., for their support in prepa- ration of this article. The Journal of School Health (February 2013, Vol. 83, No. 2) published a companion article, which was based on Pennsylvania student data from 2005 to 2009.

Human Participant Protection This study was approved by the Indiana University Bloomington institutional review board.

References 1. Finkelstein EA, Trogdon JG, Cohen JW, Dietz W. Annual medical spending attributable to obesity: payer- and service-specific estimates. Health Aff (Millwood). 2009;28(5):w822—w831.

2. Finkelstein EA, Khavjou OA, Thompson H, et al. Obesity and severe obesity forecasts through 2030. Am J Prev Med. 2012;42(6):563—570.

3. Trogdon JG, Finkelstein EA, Feagan CW, Cohen JW. State- and payer-specific estimates of annual medical expenditures attributable to obesity. Obesity (Silver Spring). 2012;20(1):214—220.

4. Ogden CL, Carroll MD, Kit BK, Flegal KM. Preva- lence of obesity and trends in body mass index among US children and adolescents, 1999—2010. JAMA. 2012;307(5):483—490.

5. Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in US children and adolescents, 2007—2008. JAMA. 2010;303(3): 242—249.

6. Commonwealth of Pennsylvania. Public School Code of 1949. The Pennsylvania Code. Vol § 23.7. Height and weight measurements 1959.

7. YoussefAgha AH. Lohrmann DK, Jayawardene WP. Use of data mining to reveal body mass index (BMI): patterns among Pennsylvania schoolchildren, pre-K to grade 12. J Sch Health. 2013;83(2):85—92.

8. Robbins JM, Mallya G, Polansky M, Schwarz DF. Prevalence, disparities, and trends in obesity and severe obesity among students in the Philadelphia, Pennsylva- nia, school district, 2006-2010. Prev Chronic Dis. 2012;9:E145.

9. Wright PM, Li W, Okunbor E, Mims C. Assessing a novel application of web-based technology to support implementation of school wellness policies and prevent obesity. Educ Inf Technol. 2012;17(1):95—108.

10. Nihiser AJ, Lee SM, Wechsler H, et al. Body mass index measurement in schools. J Sch Health. 2007;77 (10):651—671.

11. A SAS Program for the CDC Growth Charts [com- puter program]. Atlanta, GA: National Center for Chronic Disease Prevention and Health Promotion: Centers for Disease Control and Prevention; 2011.

12. Flegal KM, Wei R, Ogden CL, Freedman DS, Johnson CL, Curtin LR. Characterizing extreme values of body mass index-for-age by using the 2000 Centers for







Pr ev

al en

ce , %

2010 2015 2020 2025 2030


Overweight Obese ExHi obese

Note. ExHi = extreme high; G6–8 = middle school; G9–12 = high school; pre-kindergarten–5G = elementary school. The

sample size was = 80 770; elementary school: n = 48 309; middle school: n = 24 384; high school: n = 8077. The city of

Philadelphia and its surrounding counties were excluded from this analysis.

FIGURE 4—Simulation of student overweight, obese, and extreme high obese prevalence:

Pennsylvania schools, 43 of 67 counties, 2011–2031.


April 2014, Vol 104, No. 4 | American Journal of Public Health Lohrmann et al. | Peer Reviewed | Research and Practice | e67


Disease Control and Prevention growth charts. Am J Clin Nutr. 2009;90(5):1314—1320.

13. Vensim� Software [computer program]. Version 5. Harvard, MA: Ventana Systems; 2011.

14. Repenning N, Sterman J, Rahmandad H. Simulating epidemics using VensimPLE. Massaschusetts Institute of Technology Sloan School of Management. 2010. Avail- able at: http://ocw.mit.edu/courses/sloan-school-of- management/15-872-system-dynamics-ii-fall-2010/ assignments/MIT15_872F10_supp01a.pdf. Accessed July 12, 2013.

15. Zhang J, Osgood N, Dyck R. A system simulation model for type 2 diabetes in the Saskatoon health region. University of Saskatchewan. 2012. Available at: http:// www.systemdynamics.org/conferences/2012/proceed/ papers/P1209.pdf. Accessed July 12, 2013.

16. Ching W-K, Huang X, Ng MK, Siu TK.Markov Chains: Models, Algorithms and Applications. 2 ed. New York, NY: Springer; 2013.

17. US Department of Health and Human Services. Healthy People 2020 Topics and Objectives. Available at: http://www.healthypeople.gov/2020/topicsobjectives2020/ pdfs/HP2020objectives.pdf Accessed July 31, 2013.

18. Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamson DF, Byers T. Do obese children become obese adults? A review of the literature. Prev Med. 1993; 22(2):167—177.

19. The NS, Suchindran C, North KE, Popkin BM, Gordon-Larson P. Association of adolescent obesity with risk of severe obesity in adulthood. JAMA. 2010; 304(18):2042—2047.

20. Gordon-Larsen P, The NS, Adair LS. Longitudinal trends in obesity in the United States from adolescence to the third decade of life. Obesity (Silver Spring). 2010; 18(9):1801—1804.

21. Levi J, Segal LM, Laurent R, Lang A, Rayburn J. F as in Fat: How Obesity Threatens America’s Future 2012. Princeton, NJ: Trust for America’s Health and Robert Wood Johnson Foundation; 2012.

22. US Department of Agriculture Food and Nutrition Service; US Department of Health and Human Services. US Department of Education; Centers for Disease Control and Prevention. Local School Wellness Policies: Overview and Action Steps. Washington, DC: US Department of Agriculture; 2012.

23. Pennsylvania Department of Education Division of Food and Nutrition. Nutrition Standards for Compet- itive Foods in Pennsylvania Schools for the School Nutrition Incentive; 2008. Available at: http://www. pears.ed.state.pa.us/forms/files/PDE181.pdf. Accessed July 23, 2013.

24. Highmark Foundation. Highmark Healthy High Five, A Five-Year Initiative Report. 2012. Available at: http:// hhh5report.cg.com. Accessed July 27, 2013.

25. Food and Nutrition Service. Approved federal funds for Supplemental Nutrition Assistance Program education by fiscal year. Supplemental Nutrition Assis- tance Program Education Connection: United States De- partment of Agriculture. 2011. Available at: http://snap. nal.usda.gov/snap/ApprovedFederalFundsSNAP- Ed01202010.pdf. Accessed July 21, 2013.

26. Taber DR, Chriqui JF, Chaloupka FJ. Differences in nutrient intake associated with state laws regarding fat, sugar, and caloric content of competitive foods. Arch Pediatr Adolesc Med. 2012;166(5):452—458.

27. Taber DR, Chriqui JF, Perna FM, Powell LM, Chaloupka FJ. Weight status among adolescents in states that govern competitive food nutrition content. Pediatrics. 2012;130(3):437—444.

28. Wescott RF, Fitzpatrick BM, Phillips E. Industry self-regulation to improve student health: quantifying changes in beverage shipments to schools. Am J Public Health. 2012;102(10):1928—1935.

29. National Center for Chronic Disease Prevention and Health Promotion. Profiles 2010-Chronic Disease Pre- vention Pennsylvania Secondary Schools. Atlanta, GA: Centers for Disease Control and Prevention; 2011.

30. National Center for Chronic Disease Prevention and Health Promotion. Profiles 2008-Chronic Disease Pre- vention Pennsylvania Secondary Schools. Atlanta, GA: Centers for Disease Control and Prevention; 2009.

31. Pennsylvania State Data Center. Local 2010 Census Data Released for Pennsylvania. 2012. Available at: http://pasdc.hbg.psu.edu/Data/Census2010/tabid/ 1489/Default.aspx. Accessed November 1, 2013.

32. University of Wisconsin Population Health Institute. County Health Rankings. 2012. Available at: http:// www.countyhealthrankings.org. Accessed October 23, 2013.


e68 | Research and Practice | Peer Reviewed | Lohrmann et al. American Journal of Public Health | April 2014, Vol 104, No. 4


Copyright of American Journal of Public Health is the property of American Public Health Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s express written permission. However, users may print, download, or email articles for individual use.

Paper Due? Why Suffer? That's Our Job!

Order New Solution