5.08 Writing Assignment 5: Explanatory Synthesis Paper

Anonymous
timer Asked: Dec 19th, 2018
account_balance_wallet $15

Question description

For Writing Assignment 5, you will revise, edit, and polish the draft of your explanatory synthesis paper into a 750-1000 word final paper. Your final paper must contain four sources from the Topics and Sources page.

Make sure your final paper contains the following elements:

  • A fully developed and polished introduction paragraph that includes a thesis statement that's appropriate for an explanatory synthesis paper's purpose
  • Fully developed and polished body paragraphs that present topic sentences, integrate source evidence, consider opposing viewpoints between sources (if they exist), and contain effective transitions both within and between paragraphs
  • A fully developed and polished conclusion paragraph that reiterates the main point of the paper
  • In-text citations in APA format, which indicate where you will include summarized, paraphrased, and quoted material in your paper, and a corresponding reference list, using the APA citation wizard
  • Your final paper must be free from first person reference. Your paper must be written from a third person perspective.

Writing Assignment 5 Work Area

As you work on your Writing Assignment 5, remember to refer to the rubric (click here) to make sure you're fulfilling each aspect of the assignment. You can also download the rubric by clicking here.

You are welcome to use the APA citation wizard below (click on the tab to open up the wizard) if you need to generate new reference citations for the sources you are using.

Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. c h ap t e r 1 How Big Is Normal? Quantifying Children’s Body Size While we do not know much about obesity in babyhood and childhood, we do know that it not a matter to be bragged about. W. A. Evans, physician, 1932 How did childhood obesity come to be a condition requiring medical attention? Fat children existed before the twentieth century. But great fatness in a child was rare enough to be considered a titillating deviation from the much slimmer norm. Indeed, some very large children became famous for their unusual size. In 1896, nine-​year-​old Charley Bilcher was a local celebrity in his hometown of Catasauqua, Pennsylvania, having overtaken Anton Mochty of Austria and brought home to the United States the title of “fattest boy in the world.” Or there was “fat-​boy prodigy,” young Irwin O. Schell (aged eleven) of Reading, Pennsylvania, whose huge hands measured eleven inches over the knuckles. Or Blanche Gray of Detroit, who exhibited herself in traveling shows as “the fat girl.” When Blanche died at age seventeen, her young husband sold her “mammoth cadaver” to a curious surgeon for twenty-​five dollars. Extremely fat “prodigies” of the nineteenth century like Charley, Irwin, and Blanche were regarded as astonishing, even newsworthy—​but not especially medically worrying in spite of Blanche’s death at an early age. Speaking to the newspapers, Charley’s parents were rather proud of their massive son’s size: “Step forward, Charley, and have your picture taken,” said his father. As was popularly believed at the time, they felt that Charley’s fatness reflected favorably on his health and how well they provided for him. Moreover, they didn’t think Charley’s rotundity signified he was sick. In fact, they thought quite the opposite: Charley’s bigness was considered a sign of abundant—​nay, overflowing—​good health. More moderate (and more common) degrees of fatness in childhood than Charley’s impressive heft were seen as healthy and attractive and were popularly believed to confer greater resistance to infectious disease, which was 21 EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 22   Measurement and Diagnosis then such a prominent childhood experience. Underweight children—​those spindly, rachitic, unattractive specimens—​they were the ones doctors and parents of the late nineteenth century worried about.1 Although newspapers of the 1880s and 1890s reported famous fat children as local sensations, just thirty years later such children would be taken to the doctor rather than feted. In that thirty-​year span, fatness in a child had become medically worrying—​no longer a cause for triumph for the fat child, nor something “to be bragged about” by his parents either. Why the change? Why did “prodigies” become patients? This chapter is concerned with the process by which youthful corpulence was medicalized as childhood obesity and with the influences and currents that saw fatness in childhood enter the list of pediatric ills. Two of these influences, and the subject of this chapter, were, firstly, the introduction of the idea that height and weight had medical significance and, secondly, the development of standards or norms for child body size and weight against which children could be compared. Measuring a child’s height and weight and interpreting those measures as both socially and medically significant were critical steps in creating childhood obesity as a clinical condition. Pity the pediatrician. The physician who specializes in treating sick children—​a medical discipline that emerged in about the 1870s—​has a considerable problem. A prominent early American “pediatrist” (an older term) described the dilemma thus: We usually begin by asking our adult patient how he feels or where he has pain if any be present, but our little patients may be too young to speak or if they do speak the pains and discomforts may be referred to in a misleading manner . . . ​A ll the information we are in the habit of getting from the patient’s description of his discomfort may thus utterly fail us. The distress may be as great or even greater, but the infant crying in the night, however definite, however obscure, however complex, or however varied the nature of his misery may be, has nought but a cry.2 In other words, pediatricians are not able to rely on their child patients reporting symptoms correctly or accurately. Consequently, from its inception this specialty has used other means of diagnosis. Birth weight—​routinely recorded for births in lying-​in hospitals since about 1860—​was, and is, used EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   23 as a diagnostic sign of how well the infant was at birth and as a prognostic sign for its early growth. A child’s “failure to thrive” has long been recognized by physicians, midwives, and parents as a “danger signal” that all was not well. To track the infant’s and older child’s general health, nineteenth-​ century pediatric textbooks like William Dewees’s Treatise on the Physical and Medical Treatment of Children (1825) and Job Lewis Smith’s Treatise of the Diseases of Infancy and Childhood (1841) (both popular guides that ran to multiple editions) counseled using appetite as the gauge. “Good appetite” was a “favorable prognostic sign.” “Appetite” was a subjective quantity that relied on a parent’s assessment of what “normal” eating was.3 But as relatively cheap, small scales became available late in the nineteenth century for home use, pediatric text books from about 1890 onward began to cast weight in the role that appetite had formerly fulfilled as an indicator of a child’s general health. “A most important point in the estimation of the development and actual condition of children consists in weighing them,” wrote pediatrist James Finlayson in 1889. Weight had benefits over appetite as a measure. It was objective; it could be recorded and tracked over time (handy, given changes in medical practice to seeing more patients, less frequently); it gave information on how a child was growing as well as the child’s health at any one moment; and it could be assessed in patients too young to describe their own symptoms, and so at least partly answered the pediatrician’s essential dilemma. Early pediatric textbooks like Finlayson’s encouraged physicians to use the tape measure and the scale as part of their standard practice in examining child patients. By the 1920s, taking height and weight was a sufficiently standard part of the pediatric diagnostic work-​ up to make the scale and the stadiometer (the instrument used for measuring height with a vertical ruler and a sliding crossbar lowered to the crown of the child’s head) stereotypical features of the doctor’s office—​ recognized badges of office like the stethoscope and white coat. That height and weight became central to pediatric diagnostics was one of the quietest innovations in twentieth century medicine.4 Slews of height-​weight studies of children have been carried out—​the first, in 1871, is the intellectual progenitor of the now ongoing National Center for Health Statistics and the Center for Disease Control’s rolling child health surveys. Reams of height-​weight tables and graphs, each claiming EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 24   Measurement and Diagnosis Figure 1.1 ​What to expect at a typical pediatric consultation, post-​1920. The l­ocation (a pediatrician’s office) is indicated by the examination table and the ­balance scale. Source: Children’s Bureau. The Child from One to Six, Children’s Bureau Publication. Washington, DC: United States Department of Labor, Government Printing Office, 1937, 12. Courtesy of Ebling Library for the Health Sciences, University of Wisconsin–Madison. status as a standard, have been produced from these huge number-​gathering exercises. Why such a repeated effort to measure children? For one, the effort speaks to ongoing anxieties about children’s health, although exactly what parents, researchers, physicians, teachers, and public servants commissioning surveys or using height-​weight tables were anxious about has changed over time from the prevalence of underweight to overweight in children. For another, the repeated effort was an attempt to get the best data possible to address perceived problems with prior surveys or to use new methods and techniques in data gathering and analysis. And researchers also understood that the population of the United States was changing over time. Children were, in general, getting bigger, so new studies were needed EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   25 to keep the results applicable to “modern” children. And lastly, the continual updating of studies reflects hard-​fought battles over the philosophical question that is fundamental to developing any standard: who ought to be measured? That is, who ought to be the reference population? Beyond questions of childhood obesity and its diagnosis, the different answers to this apparently simple question trace researchers’ changing thinking about how medicine relates to society, what its role ought to be, and where childhood health fits in with these aspirations.5 In the United States, the foundations for investigating children’s height and weight were established in the years between 1870 and 1895 by three notable height-​weight surveys. These studies provided the first indications of the range of American children’s sizes and established auxological (body measurement) investigation as a research discipline in the country. Such studies produced quantitative descriptions of children’s heights and weights—​and started a trend toward establishing the dimensions of the normal child body and pathologizing deviations from this norm. Henry Pickering Bowditch’s study of Boston schoolchildren in 1871 was the first large-​scale study of children’s weight and height carried out in the United States and shows what the anxieties about child health were in the late nineteenth century. Bowditch was a physician, a member of Boston’s Brahmin elite, and Harvard University’s first professor of physiology. His comfortable career spanned private practice, university teaching and administration, and what was then referred to as “state medicine” or what we would now call public health. Bowditch’s research interests were therefore substantial, covering both the scholarly question of human growth and development and the practical problems of community health.6 Bowditch’s first motivation for measuring Boston children was nationalistic. He had read about similar investigations carried out earlier in the century in Belgium by statistician and sociologist Alphonse Quetelet, who had inventively applied statistical analysis techniques to various social phenomena, and in England by Charles Roberts, a fellow physician. Bowditch wanted to know how American children compared with their European peers—​whether the American way of life with its opportunities, its egalitarianism, and its spacious freedom was written on children’s bodies. And, being chair of the Massachusetts State Board of Health, Bowditch wanted to know how the health of Boston’s immigrant children (mostly from Ireland, England, and Germany) compared with that of children from EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 26   Measurement and Diagnosis established American families. Quetelet’s and Roberts’s studies, in line with prevailing belief, showed that people’s bodies, even if they were from the same racial stock, were shaped by the country of their upbringing. Native-​ born American children tended to be larger than Europeans. What then, asked Bowditch, would be the “effect of transplantation into new climatic conditions” on a child’s growth? Would an immigrant child “attain to a stature resembling that of the natives of the States to which they emigrate” and become American in blood and bone and body, as well as citizenship? Or would the children of immigrants always, worrisomely, remain a group apart, an undersized nation within a nation? Reflecting his upper-​crust anxieties about the influx of immigrants, Bowditch believed studying the heights and weights of Boston children could offer answers to this worry about assimilation.7 Beyond national inquiry—​or, rather, national one-​upmanship—​there was also an intellectual aspect to Bowditch’s interest in children’s sizes. Quetelet’s and Roberts’s studies had produced contradictory evidence about how children grew. Quetelet’s data seemed to show that Belgian boys were, on average, always bigger than girls of the same age. But Roberts’s English data quite clearly showed that for a few years during puberty—​the brief “period of female superiority”—​girls were larger. Could both Roberts and Quetelet be right—​did children of different countries grow differently? Or was one of them wrong and girls and boys did in fact grow differently? wondered Bowditch. Did American children grow in the same pattern as Belgian children or as English children?8 Bowditch also envisaged his results having a practical application. In the 1870s, physicians, educationalists, and feminists were waging a heated argument in Massachusetts social club halls over a theory concerning children’s physical and mental growth. The contentious theory was that a child had only a fixed amount of energy available for both physical and mental growth, with the implication that “at those periods when the forces of the organism are engaged in producing rapid growth and development of the physique, the requirements in the way of mental effort should be reduced.” Heavy emphasis on the “Three Rs” should therefore be scheduled during a child’s slower growth times. Feminists were against this theory because of the implications it might have for girls’ education. Harvard medical school professor Edward Clarke had given a talk entitled “Sex in Education or, A Fair Chance for Girls” to the New England Women’s Club, in which he EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   27 argued that girls “being educated and worked just like boys” risked the “healthy development of the ovaries and other accessory organs.” The health of the nation, he argued, required that girls not be educated as wearyingly as boys. If physical and mental growth were antithetical to one another (and Bowditch said he would leave that to others to decide), data on how girls and boys grew could be useful for setting each sex’s educational pace.9 National fortitude, clinical diagnosis, educational theory, and public health—​a ll these issues could be helped, argued Bowditch, by surveying children’s heights and weights. Testing the waters, Bowditch first conducted a small experiment measuring his own family members. From measurements on “thirteen individuals of the female, and twelve of the male sex” (his was a large family), Bowditch felt that the Roberts study in England was more likely to be right than Quetelet’s Belgian study: between about the ages of twelve and a half to fourteen and a half, girls did indeed, on average, seem to be taller and heavier than boys. And, gratifyingly, American children did seem to be impressively large in comparison to their European cousins. Addressing the Boston Society of Medical Sciences, Bowditch proposed conducting “more extended observations” to find out more about the effect of “stock” or “ethnical group” and “climatic conditions” on children’s growth.10 Bowditch got permission to do his extended study from the Massachusetts School Board in 1875. He, however, didn’t lift a tape measure. He had class teachers from Boston schools measure their students’ weight and height without shoes, but with the children remaining clothed out of consideration for logistics and contemporary mores about the appropriateness of children undressing. Bowditch supplied the schools with specially printed forms on which to record the information. Teachers were also asked to note whether each child showed any “deformity,” and (under the same heading) to note “the fact of color . . . ​in order that negro and mulatto children might be distinguished from white children of American parents.” This would allow Bowditch to separate the data along racial lines, with the data from poorer, smaller, African American children separated from wealthier, larger, white children. Although Bowditch did not say as much, if he compared only white Americans with Europeans he was more likely to find that “American” children were bigger than if he included black children in the “American” measurements. Teachers also had to record the nationality and occupa­­ tion of each of the child’s parents. If the child or teacher didn’t know this EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 28   Measurement and Diagnosis information, Bowditch obtained it from the police. For a well-​connected researcher like he was, state resources were made available. Privacy was not a concern in the experimental ethics of 1870. Nearly 11,000 girls and 13,700 boys were measured.11 Once the data cards were returned to him, Bowditch sent the figures to accountants across Boston for analysis. Dividing the labor of calculations in this way was necessary to handle such a large data set, since the “computing” power of the 1870s was limited to the slide rule. The accountants calculated averages (means) for the different measurements for each age, nationality, and sex and the ratio of weight to height for each child—​an index, in Bowditch’s opinion, of the “stoutness” of the child. When he came to analyze the results, Bowditch’s primary assumption was that bigger was better. He referred to tallness and heaviness as being “superior” to shortness and lightness. (Boys were therefore mostly “superior” to girls; Americans—​at least white ones—​were “superior” to immigrants.) Bowditch did not, however, mention the possibility of there being children who were too big: overweight or excessive nutrition was not a concern for him. The underweight child was what he saw as the potential problem.12 With his large sample, Bowditch found, as both his and Roberts’s English study had earlier, that on average girls started to grow rapidly about four years before boys did and then stopped growing earlier as well, so that boys eventually surpassed them in height and weight. “The fact that these periods [of rapid growth] occur at different ages in the two sexes,” said Bowditch, “may therefore be regarded as an argument against the co-​education of boys and girls, except during the earlier years of life in which rates of growth are practically the same.” However, he cautioned, “how much importance is attached to this argument” (and therefore to the resolution of the Sex in Education controversy) would need to be answered with data on mental development.13 From Bowditch’s account of his analysis, it is not clear what he did with the data on African American children. One thing is apparent, however: he did not analyze that data as a group. There are no tables or graphs for the heights and weights of African American children in Bowditch’s work. Given the categories he did in fact use for his analysis—​“German,” “Irish,” “Russian,” “English,” and “American” parentage—​and the fact that teachers were asked to record whether a child was African American in the same section as “deformities,” it may be that Bowditch simply discarded the data from African American children. Alternatively, he also mentions that EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   29 Figure 1.2 ​Girls’ and boys’ heights and weights from Bowditch’s study. Note the “period of female superiority” from about age twelve to fifteen. Source: Bowditch, H. The Growth of Children. Boston: Albert J. Wright, State Printer, 1877, foldout chart. “observations on children of parents of any other single nationality [other than Irish or American] were not found to be numerous” and were therefore sometimes “thrown together into a single group of unclassified nationalities” for analysis. This data, Bowditch said, did not “throw any additional light on the question[s] under consideration,” and so he did not publish it. So it is therefore also possible that, because of the racial mix in Boston in the 1870s, there was too little data on African American children for Bowditch to analyze it as a separate category.14 EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 30   Measurement and Diagnosis On the issue of “stock,” Bowditch found that children whose parents were born in America were taller and heavier than the children of immigrants, although the difference seemed to narrow after just one generation. This, Bowditch suggested, was because the United States enjoyed “a more abundant distribution of comforts” than other countries, and that these comforts were reflected in the larger bodies of even the first generation born in the United States. Moreover, the pronounced class correlations that Roberts had found in England—​children of professional-​class parents (architects, clergymen, students, musicians, teachers) being taller and heavier than the children of unskilled laborers (firemen, pavers, sailors, and watchmen, in Bowditch’s classification)—​were not so apparent in the American data. This was a source of nationalistic pride for the American physiologist—​“the American community” was less divided by class than “English society,” and this showed in children’s bodies. Most interestingly, Bowditch found that the social and environmental influences in the United States, which give a growing boy in this community greater height and weight than are attained by an English boy of the same age, affect the weight more powerfully than the height, and that the Boston boy is therefore by no means to be described as tall and thin in comparison with his English cousin.15 In other words, Boston boys were stockier than their English c­ ounterparts—​their “more abundant comforts” went more to their poundage than to adding inches to their height. (Girls, too, showed this tendency.) Although Bowditch could not have anticipated this, his discovery is notable in light of secular changes in children’s heights and weights over the course of the twentieth century, with children getting both taller and heavier, but more heavy than tall. Weight appears to be more plastic in regard to nutritional change than height. We do not, therefore, currently have an epidemic of tallness in children. Bowditch’s tables and graphs commenced the ensuing century’s worth of work on establishing how tall and how heavy American children were. His work was followed by child measurement studies carried out in 1890–1892 in Worcester, Massachusetts, by Franz Boas (who, later in his career, became famous as “the father of American anthropology”). Boas’s work, mainly EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   31 intended to uncover more information about how children grew, illustrated levels of sophistication in the analysis and interpretation of data on children’s body size beyond Bowditch’s. Boas had intended to collect data for a number of years, but opposition from local newspapers whose editors could see no point to the then-​novel idea of measuring children—​other than for salacious reasons—​made the young researcher leave town after just two years. Boas, however, continued to work with the data he had collected, sometimes combining it with other researchers’ findings, reanalyzing it and investigating different aspects of child growth for the next forty years. The Bowditch and Boas studies were joined in 1892 by a study conducted in St. Louis, Missouri, by physician William Porter. Porter was particularly interested in the ramifications of growth patterns for educational pacing and intended his results to add a Midwest element to the otherwise New England–derived data.16 Bowditch, Boas, and Porter’s work would, however, have remained merely an interesting intellectual inquiry into children’s growth had it not been for the fact that their results were published in the most influential pediatric guide of the century, with height and weight measurements given clinical significance. Luther Emmett Holt, Professor of the Diseases of Children at the College of Physicians and Surgeons, was the preeminent American pediatrician of the early twentieth century. Advisor to New York’s Department of Health and the Child Health Organization (later the American Child Health Organization), founding member and sometime president of the American Pediatric Society (est. 1881), longtime editor of The Archives of Pediatrics (America’s first journal devoted to pediatrics), and head physician at New York’s Babies Hospital, Holt enjoyed a substantial professional and public reputation and had contributed much to nurturing pediatrics in the United States. He was well regarded, but not warmly liked, by his colleagues, who found him stuffy and distant—​he apparently never said good morning—​but he got on well enough with his child patients. Holt’s professional reputation was based not on his personal charms but on two books that made him one of the first celebrity doctors in the United States, and, at the same time, helped make comparing a child’s height and weight measurements against a standard a crucial diagnostic act.17 Diseases of Infancy and Childhood was a textbook Holt wrote for fellow pediatricians. It was first published in 1898 and ran to an impressive number EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 32   Measurement and Diagnosis of editions (eight during Holt’s lifetime) and reprintings. Diseases surpassed all other pediatric texts in sales two years after its release and maintained a selling lead over all rival textbooks throughout Holt’s lifetime. In this book, Holt advocated measurement and record keeping in pediatric diagnosis: tables, charts, and statistics were the essence of his systematic teaching style. Holt himself measured all his patients, even those coming to see him for warts, and he suggested that this practice was critically important for all pediatricians.18 In Diseases, Holt advised measuring children weekly for the first six months, then fortnightly up to one year, and then monthly. But not only that: Holt also gave a table of measurements with which he recommended physicians compare their patients. The table, which showed averages for boys’ and girls’ weight, height, chest, and head circumference for birth up to sixteen years, used data from Holt’s personal observations on infants and Bowditch’s figures for older children, and noted the differences compared with Boas’s and Porter’s figures. (Over the course of subsequent editions, Holt revised the height-​weight estimates upward, saying that it had been his experience in private practice that children were in fact even heavier than Bowditch had found and much heavier than Porter’s findings. Normalcy was getting bigger.)19 The second of Holt’s two famous books told the public about the importance of measurement and comparison with standards. Long before Dr. Benjamin Spock’s Baby and Child Care raised so many baby boomers, Holt’s 1894 parent advice manual, The Care and Feeding of Children: A Catechism for the Use of Mothers and Children’s Nurses, was given the extravagant moniker of “infant bible of the nation.” The New York Times’ book review of 1946 listed it (and its later editions) as one of most influential books on American life and culture, putting it in what today seems far more glamorous and prestigious company with Harriet Beecher Stowe’s Uncle’s Tom’s Cabin, Nathaniel Hawthorne’s The Scarlet Letter, and Benjamin Franklin’s Almanac. When one of the Board of Lady Managers at the hospital where he worked suggested the hospital could raise money by running a training school for nursery maids, Holt devised the teaching material for the training program. He was asked by friends to make the material more widely available: Care and Feeding was the result.20 Care and Feeding spread the new gospel of weighing and measuring to mothers and to children’s nurses. In successive editions of the book, weighing EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   33 and measuring became increasingly central and increasingly defined by reference to quantified norms—​the figures Bowditch, Boas, and Porter had found in their studies. Although in the first edition mothers were told the ages by which a healthy baby should have doubled and then trebled its birth weight, by the third edition in 1903, Holt had greatly expanded the section on measuring and changed from a self-​referential approach to recommending comparison with his table of average measurements. (The third edition gave measures only for birth to ten years for boys; the fourth edition and thereafter gave weights for both sexes, up to fourteen years.) Holt’s guide was adamant about how important weight, in particular, was as a clinical signifier. Of what importance is the weight of a child? asked the guide. And answered it: Nothing else tells so accurately how well it is thriving.21 Parenting guides like Holt’s and its ilk of the early twentieth century were so earnest about this new practice of weighing and measuring that they counseled mothers-​to-​be who were shopping to equip the nursery that, along with the layettes, and bibs, and cribs, they should also buy a scale. They even gave helpful pictures and recommended brands. An exceptionally well-​fitted-​out nursery might also include a special measuring board to make taking baby’s length easier than measuring against the wall or the edge of a table. Mothers were being educated that their child’s size was regarded as medically important.22 The process that Bowditch, Boas, Porter, and Holt had created and promoted had imbued children’s heights and weights with clinical meaning. Holt’s work in particular had suggested that pediatricians and parents should use measures of body size in their assessment of a child’s health. But between the measurement studies and the way Holt had suggested their results be used in clinical settings was an interpretive gap that pediatricians and parents were encouraged to jump over. Bowditch, Boas, and Porter presented their work as describing what was normal—​quantifying in inches and pounds what size and heft supposedly healthy children’s bodies were. (There was, in fact, little effort given to ensuring the children being measured were actually healthy other than simply ruling out those with an obvious physical deformity. The implication was that the large numbers involved in the survey EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 34   Measurement and Diagnosis would balance any other health variations.) His own work, Bowditch said, was a first “rough approximation” for “establishing a normal standard of development.” When it came to using the data, however, Holt (and later others) interpreted it not as descriptive but as prescriptive. While the Bowditch/Boas/Porter data gave a picture of the group of children measured, it would subsequently be used and promoted to physicians, parents, and school health officers as a standard against which a single child could be compared. Many children to one child; descriptive to prescriptive; research to clinical investigation; auxologists to parents, physicians, and class teachers. In shifting applications, interpretations, settings, and users, height-​ weight measurements took on new meanings and significance. Not only was normalcy quantified, not only did height and weight carry clinical meaning, but a child’s body could be tested and judged as to how well it matched the descriptive criteria.23 One phenomenon more than any other of the early twentieth century illustrates these developments in quantifying normalcy, vesting height and weight with clinical significance, and judging children’s bodies against the Bowditch/ Boas/Porter size standards. “Better Babies” contests were held across America from the early teens to the 1920s. These contests were one example of a wider social anxiety about child health in the early twentieth century. Sparked by concerning levels of infant mortality, government and philanthropic efforts sought to address the matter through a range of means such as improved sanitation, child nutrition, and parent education. Better Babies contests were backed by women’s committees and clubs, such as the Congress of Mothers and the American Medical Association’s Committee for Public Health Education among Women, and supported by the popular magazine the Woman’s Home Companion. The contests aimed at encouraging and rewarding child health and educating mothers about how to achieve it. Community groups like milk committees, state fair organizers, church clubs, agricultural organizations, and individuals could write to the Woman’s Home Companion for advice and help in running a contest—​the magazine would even help with prize money and a handsome medal picturing two “perfectly formed babies” for the winner. “Save the Babies” was the catchcry. 24 Organizers hoped that the combination of free child-​raising information and competitive spirit would drive mothers to raise “better” babies. And not just “babies”—​the competitions covered children up to school age. Organizers EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   35 also hoped that, in contrast to children’s beauty contests, which were well known to be catty affairs, the scientific ideals of the Better Babies contests would encourage mothers of winners to share their recipes for raising pedigree kids. Measuring the young contestants’ height and weight was an important (and the only impartial) aspect of the judging process. Better Babies was hugely popular. In just one year (1914), the Woman’s Home Companion sponsored contests in twenty-​three states. Several organizers noted the wide demographic range of participants, particularly in the large metropolises like New York and Washington, D.C. Along with the clubby bastions of white middle-​and upper-​class women, contests were also organized by organizations with a particular ethnic bent, such as the Hebrew Institute in Chicago, the Kickapoo Indian Produce Fair in Horton, Kansas, and by African American community and church groups. The concept of the “better baby” clearly had broad appeal.25 In some towns, Better Babies contests arose out of baby beauty contests, but Iowa’s 1911 baby contest (which lays claim to being one of the earliest in America) was modeled on its state agricultural fair. In the same way a fat, glossy pig could win the blue ribbon, a physically perfect (meaning both attractive and healthy) baby scored high. Judges—​in this case physicians, rather than farmers—​would mark children on a special scorecard. Scores would then be totted up to find the healthy winner.26 Iowa has long excelled in prize crops of hogs and corn. She is not to be outdone when it comes to the higher order of creation. She is starting to show the world how to raise a prize crop of better babies. 27 Similarly, Denver held its Better Babies competitions at the stockyards. Run by the National Western Live Stock Association, at the 1913 contest, Buffalo Bill rode around the ring with the two winners (one boy, one girl) on his saddle. It was a day out for the whole family. Contest organizers would advertise through local newspapers and their own newsletters. Over the sometimes several days of the contest, mothers would bring their children to the contest location—​often a pavilion within the state fair—​for judging. Moving through different stations, and wearing only a cloth diaper or underwear, contestants would be measured, weighed, and scored on other points of physical and mental development. Better Babies contests, parents were advised, were not meant to be like baby beauty contests. They were “scientifically based,” with prizes for the healthiest specimens of babyhood. In practice, however, cultural standards EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 36   Measurement and Diagnosis have long conjoined child attractiveness with child health: a healthy child is considered attractive; an attractive child is likely to be healthy. And, to help decide on winners among large numbers of contestants based only on brief examination, Better Babies physician judges did incorporate what we would consider today as simply aesthetic standards. To one mother’s “pathetic grief,” her child was marked down for the “defect” of sticking out ears. Marks could be struck off for other unattractive features that had no pathological significance and also for supposedly troubling features of mental development, such as crying. Not just healthy, Better Babies winners also had to be pretty and charming.28 And the other common features of Better Babies winners were that they were tall for their ages—​and chubby. As early measurement studies had shown, medicine of the early twentieth century was worried about the underweight child who failed to thrive and must therefore be sick; the corollary was that the chubby child with his rosy cheeks and dimpled knees epitomized the picture of health. And that was what Better Babies selected for: scoring systems selected winners who were large and heavy. That was achieved by comparing contestants’ height and weight measures with tables of standard heights and weights.29 The two most widely used baby contest scorecards were those supplied by the Woman’s Home Companion and by the American Baby Health Contest Association. Both scorecards used Holt’s table of measurements against which children’s heights and weights were compared. Under the scoring scheme, “physical development” (which height and weight reflected) counted for as much as 80 percent of the total marks. However, after a number of contests, organizers found that many entrants were scoring very highly against the Holt measures, making it difficult to clearly pick winners. The figures, based on regular babies and schoolchildren, were proving to be too low for the plump, pedigree children being entered in Better Babies. The Woman’s Home Companion therefore issued new scorecards with tables of measurements that were arbitrarily larger than the Holt values. The magazine editor explained why they had settled on the new, larger standard: We believed that the highest possible standard should be set for babies entering these contests, even if these standards were above the measurements of the average child. We therefore set as our standard a height which was an inch to an inch and a half greater . . . ​and a weight which was one EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   37 Figure 1.3 ​Plump and perfect. Better Babies Prize winners. Source: North Carolina State Board of Health. “Colorado Prize Winners, 1912.” The Health Bulletin 28, no. 6 (1913): 57. to one and a half pounds greater than [the Holt] standard. The other measurements were correspondingly increased, except in the case of the circumference of the head, a large head often being an indication of rickets or other diseased condition.30 So the standards promoted by the Better Babies competitions, although based on the Bowditch-​cum-​Boas work, were an intellectual step away from those progenitors. The Holt tables were devised as a description of normal children and were intended to be used as a standard against which to judge whether a particular child was normal. Instead, the Better Babies tables would be an indication of what was ideal. Better Babies marked a subtle shift toward using height-​weight tables as aspirational standards, based not on regular, normal children but on ideally healthy, pedigree children. The EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 38   Measurement and Diagnosis Woman’s Home Companion and the American Medical Association collected copies of the scorecards from the competitions and had a statistician, Frederick S. Crum of the Prudential Insurance Company, analyze them. The tables Crum produced were based on over ten thousand Better Babies contestants, ranging from six months to four years of age, making them the first height and weight standards developed in the United States that drew from a national—​a lthough not representative—​sample. The Crum tables were adopted by the American Medical Association (AMA) and by the federal Children’s Bureau and given out to parents as an indication of how big children should be.31 As medical practice, parenting advice, and the cultural phenomena of Better Babies contests in the opening decades of the twentieth century began to make American society sensitive to children’s heights and weights, other areas of life also encouraged parents to take a good hard look at their children’s bodily measurements. Going shopping also brought home to parents the growing significance of children’s body size in American culture. Since the late nineteenth century, shop-​bought or “ready-​to-​wear” clothing for children had been increasing in popularity and availability, first boys’ clothing and then later girls’. Shopping for ready-​to-​wear clothes increasingly made parents aware of their child’s size. Clothing assistants at Macy’s were equipped with a tape measure to help the young customer into properly fitting clothes; Bergdorf Goodman gave customers a special card where a child’s waist, height, chest, leg, and sleeve measurements were recorded.32 But it was the big catalogue sellers, like Sears Roebuck, whose business practices really drove home to parents the need to know how big their child was and also how their child’s body compared with that of his or her peers. In early editions of the Sears Big Book catalogue, parents ordering clothes for their children were instructed to “be particular to always give the height, weight and age in young men’s, boy’s and children’s orders, along with the rest of the information [chest, waist, and inner seam measures], as these are very important.” The catalogue even came with a tape measure to make sizing up the child easy.33 From 1910, Sears introduced an innovation in clothes ordering: standard sizes. Sears saw standard sizes as an important way to increase buyers’ confidence in purchasing clothes and, they hoped, to reduce the costly number EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. How Big Is Normal?   39 of returns for poor fit. (Sears famously offered a money-​back guarantee: “Satisfaction guaranteed or your money back.”) With standard sizes, parents not only had to know their child’s bodily dimensions but then also had to compare the measurements with a table of standards for what was considered—​by Sears—​to be “normal” size for the child’s age. Sears’s sizing designations were based on children’s ages: a size 8 was meant to be the size for an average eight-​year-​old, so parents could get a clear sense of how their child compared with these newly devised norms. (This type of sizing system is sometimes called “age sizing.”) Sears, being America’s biggest catalogue seller, led the industry in developing sizing standards in the 1910s, and the practice became commonplace across all ready-​to-​wear children’s clothing manufacturers by the 1930s. Each store devised its own sizing “standards” depending on its own particular tailoring practices. 34 Although age sizes were meant to fit children of particular ages, the sizes were not based on, say, a formal survey of children’s body measurements. Rather, clothing sizes were developed from in-​house pattern makers’ rules of thumb, trial-​and-​error, passed down “like recipes handed down through families.” Only after a 1939 study by the Bureau of Home Economics was there a move toward developing sizing standards for children’s clothes based on survey measurements. World War II prevented the widespread adoption of new sizing standards, however. It would be 1949 before the Mail Order Association of America—​the trade organization for catalogue sellers such as Sears—​adopted sizing standards based on the bureau’s measurement survey. Parents suffering through the increasingly common experience of buying readymade children’s clothes were having to assess whether their child was too skinny, too fat, or just right according to the particular store’s sizing standards. Clothing sizing therefore contributed to making children’s body size noticeable and important in American life (but didn’t make clothes shopping any easier).35 The Bowditch/Boas/Porter/Holt and Crum tables started an ongoing process of surveying large groups of American children’s heights and weights and presenting them as tabulated or graphed standards. These were sometimes smaller studies, sometimes considerably larger studies, like the massive federal survey of children for Children’s Year in 1918 in which 7.5 million scorecards were used, 2 million were analyzed, and new standards EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102 Copyright © 2014. Harvard University Press. All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law. 40   Measurement and Diagnosis drawn up based on nearly 200,000 children. Ethnic-​group-​specific studies were carried out from the 1920s to investigate physiological questions—​did different ethnicities follow different patterns of growth?—​and social questions—​did impoverishment show up in the bodies of ethnic groups? These studies raised the still on-​going question of whether there ought to be different standards for different ethnic groups.36 Parenting guides and pediatric textbooks routinely included tables and graphs, using one of the available studies—​Holt, Crum, and the Children’s Year data were all popular choices—​as the recommended standard against which to compare a child’s height and weight. The federal Children’s Bureau published its own series of mailing pamphlets on child and infant health called Infant Care and Your Child from One to Six, which gave the Children’s Year data as its standards. Parents were therefore federally advised to assess their child’s height and weight. The normal child had been quantified and described in tabular form.37 By 1920, the idea that parents and physicians wouldn’t know the significance of a child’s height and weight measurements—​a problem Boas contended with when trying to persuade people to support his study in 1892—​was unthinkable. Physical dimensions were firmly established as the preeminent markers of a child’s well-​being—​and of a parent’s ability and skill in caring for them. Both pride and anxiety about children’s health could be quantified by these measures. American society had been sensitized to a child’s height and weight. A child’s body measurements meant something—​something both medical and social—​and could, and should, be judged. This new propensity to measure and judge child bodies was a critical step in framing childhood obesity as a medical condition. The practices of measuring and recording children’s bodily dimensions would shape the way childhood obesity came to be diagnosed, conceptualized, and experienced. EBSCO Publishing : eBook Academic Collection (EBSCOhost) - printed on 5/31/2016 4:02 PM via ARGOSY EDUCATION GROUP AN: 802454 ; Dawes, Laura.; Childhood Obesity in America : Biography of an Epidemic Account: s5351102
Nutrition; Infants The Impact of Breastfeeding on Early Childhood Obesity: Evidence From the National Survey of Children’s Health American Journal of Health Promotion 2016, Vol. 30(4) 250-258 ª The Author(s) 2016 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0890117116639564 ajhp.sagepub.com Francesca V. Hansstein, PhD1 Abstract Purpose: To investigate how breastfeeding initiation and duration affect the likelihood of being overweight and obese in children aged 2 to 5. Design: Cross-sectional data from the 2003 National Survey of Children’s Health. Setting: Rural and urban areas of the United States. Subjects: Households where at least one member was between the ages of 2 and 5 (sample size 8207). Measures: Parent-reported body mass index, breastfeeding initiation and duration, covariates (gender, family income and education, ethnicity, child care attendance, maternal health and physical activity, residential area). Analysis: Partial proportional odds models. Results: In early childhood, breastfed children had 5.3% higher probability of being normal weight (p ¼ .002) and 8.9% (p < .001) lower probability of being obese compared to children who had never been breastfed. Children who had been breastfed for less than 3 months had 3.1% lower probability of being normal weight (p ¼ .013) and 4.7% higher probability of being obese (p ¼ .013) with respect to children who had been breastfed for 3 months and above. Conclusion: Study findings suggest that length of breastfeeding, whether exclusive or not, may be associated with lower risk of obesity in early childhood. However, caution is needed in generalizing results because of the limitations of the analysis. Based on findings from this study and others, breastfeeding promotion policies can cite the potential protective effect that breastfeeding has on weight in early childhood. Keywords prevention research, body mass index, weight control, childhood obesity, obesity prevention, breastfeeding Purpose Childhood obesity has been a widely discussed public health topic for more than the last decade. Statistics indicate that in 2009–2010, 16.9% of children and adolescents aged 2 to 19 years old were obese and 31.8% were overweight or obese. Among infants and toddlers, 9.7% had a high weight for recumbent length, and among children aged 2 through 5 years, 12.1% were at or above the 95th percentile and 26.7% were above the 85th percentile, indicating the presence of an excess in body fat already in early stages of life.1 The growth of childhood obesity requires attention because it is associated with health complications, for example elevated blood pressure, high cholesterol, and type II diabetes.2 In addition to higher pediatric morbidities, children who are overweight and obese are also more likely to be negatively stigmatized by their peers and family members, and to suffer from related psychological disorders.3,4 A longitudinal relationship between childhood body mass index (BMI) and adult adiposity was found in the Bogalusa Heart Study, suggesting that overweight and obese children have higher odds of becoming obese adults.5 Pediatric obesity is also associated with adult morbidity, and several studies have shown a strong association between metabolic abnormalities in youth and cardiovascular disease in adulthood.6 Unlike many other diseases, obesity is preventable, and this evidence calls for the need to explore all possible prevention solutions. In recent years, particular attention has been given to the impact that breast-feeding has on childhood BMI at different ages. The topic has been largely explored in the epidemiological and 1 School of Public Economics and Administration, Shanghai University of Finance and Economics (SHUFE), Shanghai, China Corresponding Author: Francesca V. Hansstein, PhD, School of Public Economics and Administration, Shanghai University of Finance and Economics (SHUFE), 111 Wuchuan Road, Shanghai, 20433, China. Email: f.v.hansstein@mail.shufe.edu.cn Hansstein physiological literature, leading to a total of two systematic reviews, a meta-analysis, and a study that includes both a review and a metaanalysis.7-10 Although there is sound statistical evidence that breastfeeding is protective against obesity, some uncertainty remains because most of the analyses fail to adjust for all possible confounders. Mothers who decide to breastfeed may be different in many respects from mothers who opt to feed their infants with formula.11 These differences are related to socioeconomic status (SES) and to other sociodemographic characteristics. Previous research indicates that the choice to breastfeed is affected by maternal age, education, and income. For example, a study by Chezem12 found that breastfeeding rate increases with higher parental age and education level. Another study13 showed how differences in SES are likely to influence breastfeeding duration rather than initiation. Working women might have an incentive to stop breastfeeding, though breast milk can also be stored and frozen for deferred consumption. Furthermore, less-educated women are more likely to engage in risky behaviors, such as smoking during pregnancy, or avoid doctor visits.14 Residential area is also important, and nonmetropolitan residents and children living in households with a lower SES are more likely to be overweight or obese and, in general, have a worse health status.15 Lefebvre and John 7 have underlined the importance attached to adjusting for as many confounding factors as possible when examining the relationship between obesity and breastfeeding. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analysis, the authors examined a total of 21 academic articles published between 2005 and March 2012. Results show that 10 out of 21 studies did not find any significant relationship between breastfeeding and obesity, whereas 11 found a statistically significant negative relationship, suggesting mixed results about the strength of the association between breastfeeding and obesity. In another work conducted by Arenz et al.8 a total of 28 studies were eligible for review and 9 were included in the subsequent meta-analysis. It was found that breastfeeding had a small, but consistently positive, protective effect against obesity, but this relationship is less pronounced when the analyses are adjusted for at least seven confounders. Harder et al.9 and Owen et al.10 concluded, from respectively a meta-analysis of 17 studies and a review of 28 studies, that prolonged breastfeeding is associated with a decreased risk in overweight and obesity. None of the studies neglected a relationship between obesity and breastfeeding, but all of them called for further evidence and controls. The epidemiological evidence relating breastfeeding to BMI should also be integrated with physiological literature that explains why breastfeeding is protective against the accumulation of adipose tissue. In the last 20 years, six breast milk hormones have been identified, and all of them are involved in food intake regulation. Although there is a need for further research on the precise effects that these hormones have on significant weight adjustments, it has been acknowledged that leptin and ghrelin induce food satiety and increase energy 251 expenditure.16 Previous studies have shown that breastfeeding has a positive effect on BMI even for children exposed to diabetes in utero.17 The present study investigates how breastfeeding initiation and duration affect the likelihood for children aged 2 to 5 of being overweight and obese, controlling for child gender, household socioeconomic confounders, ethnicity, maternal health and physical activity, and residential area (Metropolitan Statistical Area [MSA] or not). According to a detailed review of the published academic literature, no study so far has investigated the effect of breastfeeding duration on BMI in early childhood using a sample representative of the U.S. child population. Furthermore, if breastfeeding is protective against obesity later in life, we hypothesize that the association is likely to be observed at early stages of life when the main environment a child is exposed to is the family, which serves as a filter for the influence of external factors. Methods Design and Sample This research uses the early childhood subsample of the crosssectional data of the 2003 National Survey of Children’s Health (NSCH), a survey that is taken every 4 years and targets U.S. children and adolescents aged 0 to 17. The sampling of the 2003 NSCH was based on the National Immunization Survey. A random-digit-dial sample of households with at least one child below 18 years of age was selected from each of the 50 states and the District of Columbia. Interviews were conducted using the Computer-Assisted Telephone Interviewing system, and the respondent was the adult most knowledgeable about the sampled child’s health and health care. In 2003, a total of 102 353 interviews were performed. The number of households where at least one member was between the ages of 2 and 5 was 8207.18 The 2003 survey was the only wave providing information about BMI in early childhood. Data are acknowledged to be outdated, but the goal of this paper is to test a relationship rather than to observe a phenomenon at a specific moment of time. Measures The BMI is an anthropometric measure largely employed to determine if an individual has a healthy weight or not. It is calculated as the ratio between weight in kilograms and height in squared meters. The same formula is used for children and adolescents aged 2 to 20, but values are then adjusted for age and sex using specific growth charts. During the 2003 NSCH interview round, the child’s height and weight were parent reported; however, in the final version of the dataset, BMI was made available also as a categorical variable. In particular, child and adolescent BMI was categorized as underweight (BMI for age at the 5th percentile or lower), normal weight (BMI for age above the 5th and below the 85th percentile), at risk of overweight (BMI for age at or above the 85th and below 252 the 95th percentile), and overweight (BMI at or above the 95th percentile). These labels are based on the nomenclature used until 2007; after that time an expert committee proposed the use of ‘‘overweight’’ for at risk of overweight and ‘‘obese’’ for overweight.19 This paper uses this most recent nomenclature. The 2003 NSCH provides information about breastfeeding initiation and duration. Breastfeeding initiation was measured by a dummy variable asking whether the child had ever been breastfed. Breastfeeding duration was measured through a continuous variable indicating for how many days the child had been breastfed. The variable was then split into four time periods (less than 3 months, between 3 and 6 months, from 6 to 12 months, and over a year), consistent with official recommendations for the optimal breastfeeding length. The World Health Organization recommends exclusive breastfeeding for at least 6 months20 and the Breastfeeding Report Card of the Centers for Disease Control and Prevention uses the same cutoff points for measuring breastfeeding prevalence in the United States.21 Data transformation was consistent with the categorical form of other key variables included in the analysis and with the choice of using an ordered categorical model to estimate the relationship between breastfeeding and obesity. In the 2003 NSCH interview round, no information about exclusive breastfeeding was available. The analysis controls for these confounders: child gender, household income and education, maternal health status and physical activity, child care attendance or participation in Early Start Programs, ethnicity, and residential area (MSA or not). Income was measured by the poverty indicator, a categorical variable that expresses the total household salary as percentage thresholds of poverty line according to the 2002 Federal Poverty Guidelines. The 2003 NSCH dataset made income level available as a categorical variable, and details on how this indicator was calculated are reported elsewhere.18 Child care attendance and participation in Early Start programs were also included in the analysis. Specifically, respondents were asked if the child regularly attended a nursery school, preschool, or kindergarten, or if he or she attended a Head Start or Early Start program. At the time of the survey, Head Start and Early Start were two federally funded community-based programs for lowincome families providing assistance to pregnant women and their children. Self-reported maternal health status and engagement in physical activity were also included as covariates. In the 2003 NSCH questionnaire only information about white Americans, African-Americans, and Hispanics was available. The analysis also included a dummy variable that measured whether the family lived in an MSA or not. Besides household income, all variables were dummy coded. Statistical Analysis Partial proportional odds models were used to estimate the relationships between breastfeeding and obesity. Given the ordered categorical nature of the BMI variable, using a partial proportional odds model is appropriate in this context. An ordered logit model could not be used because these data American Journal of Health Promotion 30(4) violate the parallel slope assumption (beta coefficients of each variable are expected to be the same across all the outcomes of the dependent variable). A generalized ordered logit model also could have been employed, but the partial proportional odds model relaxes the parallel line constraint only for those variables where the assumption does not hold, thus reducing collinearity and improving the quality of the estimates. In the partial proportional odds model, Wald tests were used to examine the parallel slope assumption for each explanatory variable.22 The first part of the analysis investigated the effect that breastfeeding initiation had on the BMI classes, and the second part analyzed the effect of different breastfeeding duration lengths. In particular, breastfeeding duration, which originally was a continuous variable, was split into four time periods: breastfeeding for less than 3 months (BF1; 0 < days breastfed < 90), between 3 and 6 months (BF2; 90  days breastfed < 180), between 6 months and 1 year (BF3; 180  days breastfed < 365), and over 1 year (BF4; days breastfed  365). For each time frame, a dummy variable was created (indicating whether the child was breastfed for that length of time or not) and a partial proportional odds model with all the dummies was estimated. Marginal effects on the BMI outcomes were then calculated, providing insights on the change in the probability of falling into one of the BMI categories for each explanatory variable. Standard errors were estimated using the Taylor linearized standard error procedure. To verify the strength of the results, an additional test and a robustness analysis were performed. The first test repeated the statistical analysis on different age ranges (2–3, 3–4, and 4–5 years old). The second verifies if the relationship between breastfeeding duration and obesity persists after the introduction of a bias in the sample by excluding underweight children. Both the additional analysis and the robustness test confirmed the inverse association between obesity and breastfeeding. These data are not reported here but are available upon request. The NSCH sample was designed following a complex procedure. Sample units were obtained by clustering children within households and stratification of households within states. Observations were also weighted accordingly as indicated in the survey design. The statistical analysis was performed using STATA 12 (StataCorp LP, College Station, Texas). This software provides the capability of variance estimation for complex sample design. Results Table 1 reports descriptive statistics of the study population. BMI classes were evenly distributed across gender, with a slightly higher prevalence of normal weight among females (normal-weight females were 40.28% and males were 38.42%) and obese males (36.22% and 38.88% respectively). At baseline, 29.07% of children were never breastfed. The average age of the sample was 3.5 years old, and of the overall sample, 16.23% were Hispanic and 9.66% were African-Americans. A total of 59.87% of the children attended child care in the form of nursery, preschool, or kindergarten, Hansstein 253 Table 1. Demographic Characteristics of the 2003 National Survey of Children’s Health Early Childhood Subsample (n ¼ 8207)* Variable BMI category Underweight Normal weight Overweight Obese Gender Female Male Household highest education level High school or lower More than high school Household income Below 133% of poverty level 133% to below 300% of poverty level Above 300% of poverty level Age, y Hispanic African-American Ever breastfed Breastfeeding duration BF1 (0 < d breastfed < 90) BF2 (90  d breastfed < 180) BF3 (180  d breastfed < 365) BF4 (d breastfed  365) Maternal health status Good or very good Very good or excellent Fair or poor Maternal physical activity (yes) Attendance of preschool Attendance of HS/ES program Metropolitan Statistical Area residence Yes No % Distribution or Mean (SD) 10.94 36.65 11.00 41.41 48.84 51.16 26.14 73.86 21.55 33.99 44.46 3.5 (1.09) 16.23 9.66 70.86 22.96 19.13 34.02 23.89 93.53 6.47 58.46 59.87 13.15 74.16 25.84 *Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, State and Local Area Integrated Telephone Survey, 2003 National Survey of Children’s Health. (Data Resource Center on Child and Adolescent Health. Available from: http://www.childhealthdata.org. Accessed September 20, 2011.) BMI indicates body mass index; HS, Head Start; and ES, Early Start. and 13.15% attended an Early Start or Head Start program. At the moment of data collection, 74.16% of respondents lived in an MSA. Table 2 presents the results of the partial proportional odds regression on breastfeeding initiation. Results show that breastfed children had a lower probability of being obese and overweight than of being normal weight (.235, p ¼ .002). They were also highly less likely to be obese than to appear in a lower category (.395, p ¼ .001). The odds of becoming obese were higher for Hispanic and African-American populations compared with non-Hispanic and non-African-Americans (.305, p ¼ .010, and .616, p < .010, respectively). Higher levels of income and education were negatively associated with BMI, but the latter relationship was not statistically significant. There was a negative association between BMI categories and child care attendance, but a positive association with the attendance of a Head Start or an Early Start program. Mothers who sent their child to a child care center were likely to have a higher income and to be employed, whereas children enrolled in Head Start or Early Start programs were from low-income families. The signs of these associations can be related to income disparities, and specifically to the inverse pattern between obesity and income observed in the United States.23 Results also showed that women who reported their health status as ‘‘good’’ or ‘‘excellent’’ had children with a lower probability of being overweight and obese, but this relationship was not statistically significant. Maternal engagement in physical activity did not have a protective effect on the children’s BMI and the association was not statistically significant. The lack of a relationship may be attributed to bias due to self-reporting. Children living in a MSA were less likely to be overweight and obese than normal weight and also less likely to be obese than in a lower BMI category. For this variable the parallel slope assumption holds and thus the estimated coefficient is always the same (.200, p ¼ .005). This could be because urban residents have more access to healthy food options, have more information about healthy eating habits, and live in a generally more favorably built environment.15 After analysis of the association between breastfeeding initiation and BMI, the next step was to detect the minimum length of breastfeeding that is protective against obesity. Four cutoff points were calculated, each corresponding to a dummy variable (BF1 ¼ less than 3 months but more than zero days, BF2 ¼ between 3 and 6 months, BF3 ¼ between 6 and 12 months, and BF4 ¼ over a year). Results from the partial proportional odds regression indicated that breastfeeding for less than 3 months was associated with higher BMI classes, as reported in Table 3. Specifically, children who were breastfed for less than 3 months were found to have a higher probability of being overweight or obese than of being normal weight (.213, p ¼ .014). Table 4 reports results from the regression where different breastfeeding lengths BF2, BF3, and BF4 were included in the regression and BF1 was used as a baseline. The estimates of the other cutoff points suggest that a prolonged duration of breastfeeding is protective against overweight and obesity in early childhood, and the longer the duration, the stronger the protective effect. Table 5 reports the marginal effects on three outcomes of the BMI (normal weight, overweight, and obese) for both breastfeeding initiation (model 1) and duration (model 2, BF1). Underweight estimates were not included in the table. Results showed that breastfeeding initiation increased the likelihood of being normal weight by 5.2% (p < .001) and decreased the likelihood of being obese by 8.6% (p < .001), with respect to children who had not been breastfed. Infants who were breastfed for less than 3 months had a 3.1% lower probability of being normal weight (p < .001) and a 4.8% higher probability of being obese (p < .001) compared to children who were breastfed for longer. Compared to non-Hispanic children, Hispanic children had a 9% 254 American Journal of Health Promotion 30(4) Table 2. Partial Proportional Odds Regression, Breastfeeding Initiation Effect on BMI Categories in Children Aged 2 to 5 years, 2003 National Survey of Children’s Health (n ¼ 8207)* BMI Class Coefficients (p)y Variables Breastfed Female Parent education (>HS) Household income Hispanic African-American Maternal health status (good and above ¼ 1) Maternal physical activity (exercise: yes ¼ 1) Attendance of nursery school, preschool, or kindergarten Attendance of Head Start or Early Start program SA residence Underweight 0.042 0.128 0.002 0.043 0.183 0.215 0.091 0.019 0.165 0.099 0.200 (0.714) (0.231) (0.973) (0.072) (0.287) (0.185) (0.497) (0.769) (0.019) (0.520) (0.005) Normal Weight Overweight 0.235 (0.002) 0.255 (0.001) 0.002 (0.973) 0.041 (0.020) 0.297 (0.013) 0.439 (0.203) 0.091 (0.497) 0.019 (0.769) 0.165 (0.019) 0.272 (0.009) 0.200 (0.005) 0.395 (0.001) 0.378 (0.001) 0.002 (0.973) 0.056 (0.002) 0.305 (0.010) 0.616 (0.010) 0.091 (0.497) 0.019 (0.769) 0.165 (0.019) 0.230 (0.026) 0.200 (0.005) *Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, State and Local Area Integrated Telephone Survey, 2003 National Survey of Children’s Health. BMI indicates body mass index; HS, high school; and MSA, Metropolitan Statistical Area. y Underweight indicates a BMI for age at the 5th percentile or lower; normal weight, a BMI for age above the 5th and below the 85th percentile; and overweight, a BMI for age at or above the 85th and below the 95th percentile. Obese corresponds to a BMI at or above the 95th percentile. Table 3. Proportional Odds Regression, Breastfeeding Duration, and Factors Associated With BMI Categories in Children Aged 2 to 5 years, 2003 National Survey of Children’s Health (n ¼ 5664)* BMI Class Coefficients (p) Variables BF1 (0 < d breastfed  90) Female Parent education (.HS) Household income Hispanic African-American Maternal health status (good and above ¼ 1) Maternal physical activity (exercise: yes ¼ 1) Attendance of nursery school, preschool, or kindergarten Attendance of Head Start or Early Start Program MSA residence Underweight 0.213 0.127 0.063 0.045 0.240 0.199 0.175 0.144 0.191 0.164 0.299 (0.014) (0.332) (0.581) (0.133) (0.245) (0.386) (0.313) (0.283) (0.028) (0.412) (0.001) Normal Weight Overweight 0.213 (0.014) 0.262 (0.002) 0.063 (0.581) 0.017 (0.422) 0.371 (0.012) 0.298 (0.065) 0.175 (0.313) 0.038 (0.664) 0.191 (0.028) 0.258 (0.053) 0.299 (0.001) 0.213 (0.014) 0.333 (0.001) 0.063 (0.581) 0.060 (0.008) 0.369 (0.011) 0.585 (0.001) 0.175 (0.313) 0.126 (0.165) 0.191 (0.028) 0.175 (0.195) 0.299 (0.001) *Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, State and Local Area Integrated Telephone Survey, 2003 National Survey of Children’s Health. BMI indicates body mass index; HS, high school; and MSA, Metropolitan Statistical Area. lower probability of being normal weight in the first model and an 11.4% lower probability in the second. They were also more likely to be obese (6.9% and 8.1% respectively). Similar findings emerged for African-American children, who were less likely to have a lower healthy weight and more likely to be obese. Results from the breastfeeding initiation columns show that children from families with higher income levels had a 1.4% higher probability of being normal weight and a 1.2% lower probability of being obese. Similar results apply for breastfeeding duration. Discussion Breastfeeding and Obesity This study adds further empirical evidence to the academic research about the impact that breastfeeding has on obesity rates in early childhood. Both initiation and duration were found to affect the BMI, and the analysis of marginal effects shows that a shorter breastfeeding duration diminishes the effect of variables that usually are negatively associated to obesity. Previous studies have investigated the association between breastfeeding duration and weight adjustments, most of them confirming the latent protective effect of breastfeeding later in life. The assumption of this research is that the protective effect of breastfeeding, if present, is more likely to be observed in early childhood than later. Concerning socioeconomic and sociodemographic factors, in this paper income is negatively related to breastfeeding, suggesting that children from families with higher income are more likely to be breastfed. Higher education levels were not significantly associated with a lower probability of being overweight or obese. In the 2003 NSCH, the variable ‘‘education’’ lacks specificity; it indicates the highest educational Hansstein 255 Table 4. Partial Proportional Odds Regression, Breastfeeding Length, and Factors Associated With BMI Categories in Children Aged 2 to 5 Years, 2003 National Survey of Children’s Health (n ¼ 5664)* BMI Class Coefficients (p) Underweight Variables BF2 (90 < d breastfed  180) BF3 (180 < d breastfed  365) BF4 (d breastfed > 365) Female Parent education (>HS) Household income Hispanic African-American Maternal health status (good and above ¼ 1) Maternal physical activity (exercise: yes ¼ 1) Attendance of nursery school, preschool, or kindergarten Attendance of Head Start or Early Start program MSA residence 0.161 (0.183) 0.214 (0.029) 0.257 (0.018) 0.131 (0.320) 0.067 (0.558) 0.045 (0.135) 0.243 (0.239) 0.195 (0.395) 0.170 (0.324) 0.145 (0.279) 0.191 (0.028) 0.167 (0.404) 0.297 (0.001) Normal Weight 0.161 0.214 0.257 0.259 0.067 0.018 0.367 0.294 0.170 0.035 0.191 0.255 0.297 (0.183) (0.029) (0.018) (0.002) (0.558) (0.417) (0.013) (0.069) (0.324) (0.689 (0.028) (0.057) (0.001) Overweight 0.161 (0.183) 0.214 (0.029) 0.257 (0.018) 0.331 (0.001) 0.067 (0.558) 0.060 (0.008) 0.365 (0.012) 0.580 (0.001) 0.170 (0.324) 0.130 (0.155) 0.191 (0.028) 0.173 (0.202) 0.297 (0.001) *Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, State and Local Area Integrated Telephone Survey, 2003 National Survey of Children’s Health. BMI indicates body mass index; HS, high school; and MSA, Metropolitan Statistical Area. Table 5. Marginal Effects of Breastfeeding Initiation and Duration on Early Childhood BMI Categories* Breastfeeding Initiation (n ¼ 8207) Normal Weight (p) Breastfeeding initiation (yes/no) BF1, length (< 3 mo/>3 mo) Female Parent education (>HS) Income (low/middle and high) Hispanic (yes/no) African-American (yes/no) Maternal health status (good and very good/poor) Maternal physical activity (yes/no) Attendance of nursery school, preschool, or kindergarten (yes/no) Attendance of Head Start or Early Start program (yes/no) MSA residence (yes/no) Overweight (p) Obese (p) Breastfeeding Duration (n ¼ 5664) Normal Weight (p) Overweight (p) Obese (p) 0.053 (0.002) 0.032 (0.005) 0.089 (0.001) — — — 0.074 (0.001) 0.023 (0.029) 0.085 (0.001) 0.000 (0.73) 0.000 (0.973) 0.000 (0.973) 0.014 (0.001) 0.002 (0.229) 0.012 (0.002) 0.090 (0.001) 0.002 (0.870) 0.069 (0.009) 0.085 (0.001) 0.033 (0.022) 0.139 (0.001) 0.013 (0.497) 0.001 (0.499) 0.020 (0.251) — — 0.031 (0.013) 0.005 (0.018) 0.047 (0.013) 0.076 (0.001) 0.009 (0.484) 0.073 (0.001) 0.009 (0.581) 0.001 (0.581) 0.013 (0.582) 0.009 (0.001) 0.008 (0.001) 0.013 (0.001) 0.114 (0.001) 0.009 (0.684) 0.081 (0.011) 0.053 (0.159) 0.056 (0.002) 0.129 (0.008) 0.025 (0.312) 0.004 (0.317) 0.038 (0.013) 0.002 (0.769) 0.000 (0.769) 0.004 (0.622) 0.024 (0.019) 0.002 (0.023) 0.037 (0.019) 0.023 (0.265) 0.037 (0.015) 0.028 (0.164) 0.028 (0.027) 0.004 (0.032) 0.042 (0.027) 0.075 (0.001) 0.001 (0.375) 0.053 (0.027) 0.079 (0.005) 0.029 (0.001) 0.003 (0.008) 0.045 (0.005) 0.024 (0.221) 0.038 (0.194) 0.043 (0.001) 0.007 (0.001) 0.066 (0.001) *Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, State and Local Area Integrated Telephone Survey, 2003 National Survey of Children’s Health. BMI indicates body mass index; HS, high school; and MSA, Metropolitan Statistical Area. level in the household, but not specifically the one of the mother. Results from this analysis showed that there is a small but significant negative association between longer duration of breastfeeding and obesity in the Hispanic population. Interestingly, in the United States, the Hispanic population is subject to the so-called ‘‘Hispanic paradox,’’ the phenomenon in which health outcomes of the Hispanic population are better than their income levels would indicate. Rates of breastfeeding among Hispanic women are unrelated to SES and are usually higher than those of other disadvantaged groups. Previous studies have also shown that Hispanic mothers born in the United States have lower breastfeeding rates than non-U.S.-born Hispanic women,14,24 providing evidence that immigrant status and acculturation are also important in affecting health outcomes and behaviors. However, another study found that Hispanic women with a college education were more likely to be exclusively breastfeeding at 6 months than Hispanic women without a college education.25 Examining the effect of all possible confounders would require randomized trials where some women are assigned to the treatment group (breastfeeding) and others to the control group (formula). In this case, randomization is unethical and not easy to implement. A randomized trial was conducted in Belarus during an intervention for promoting breastfeeding practices. Results of this study showed that breastfeeding had 256 no protective effect against obesity and it did not affect insulin growth factor in children who were 11.5 years old.26 According to this literature review, the Belarus study is the only randomized trial on the relationship between obesity and breastfeeding conducted so far. Although the empirical evidence suggests the presence of a beneficial effect of breastfeeding, the difficulty of controlling for all of the confounders has led experts to be cautious about broadcasting the scientific evidence to the general public, and this also applies to the present study. This paper is able to control for important confounding factors but does not overcome the limitations of earlier studies. Strengths of the research presented here are that this study focused on early childhood and used a U.S. national representative sample. However, in general, there is no uniform evidence that breastfeeding protects against a higher BMI in late childhood and early adolescence. As previous studies have suggested, this relationship is difficult to measure because of the many environmental influences that shape eating habits. Early childhood is, however, a stage of life where family is the center of a child’s development and external influences are still limited. If there is a negative effect of breastfeeding on children’s weight, it is more likely to be observed during the first years of life than later on. Policy Implications Understanding what is the optimal breastfeeding duration for a beneficial effect on the BMI is of key importance for developing adequate prevention. This involves exploring both if there is a minimum length of duration after which lower obesity association is detected and if an increase of this duration produces additional benefits. Breastfeeding is a low-cost health prevention policy with many benefits for the child’s future health development. Bartick and Reinhold27 estimated that if 90% of U.S. mothers followed the recommendation of exclusively breastfeeding for 6 months, this would result in a savings of $13 billion and in a decrease of 911 deaths per year. Currently, both the policy statement of the American Academy of Pediatrics and the guidelines of the World Health Organization20,28 acknowledge that breastfeeding protects children from childhood obesity. In order to protect against obesity and other infectious diseases, both organizations recommend exclusive breastfeeding for at least 6 months, but additional information specifying for how long breastfeeding is protective against childhood obesity should also be given. Results of this analysis suggest that breastfeeding for at least 3 months is protective against obesity in early childhood. Furthermore, as shown in Table 1, among women who initiated breastfeeding, 57.9% are still breastfeeding after 6 months, including, however, both exclusive and nonexclusive breastfeeding. In the light of the positive health outcomes of prolonged breastfeeding, health care professionals should be encouraged to promote policies that would minimize and eliminate barriers, and encourage mothers to feed their infants with breast milk. Among the American Journal of Health Promotion 30(4) possible strategies to increase the awareness about the benefits of breastfeeding, educational programs on the importance of prolonged duration could be promoted in maternal and childcare facilities, especially in areas with lower breastfeeding rates. An exchange of best practices with other countries might also be a strategy in order to understand what policies work better than others in favoring promotion for breastfeeding.29 Studying optimal targets is extremely important for an efficient allocation of funds, and policy makers should pay attention both to sociodemographic factors and to parental attitudes towards breastfeeding. New mothers from disadvantaged groups are less likely to breastfeed their infants. Breastfeeding intentions have been found to differ between obese and normalweight women. In particular, overweight and obese mothers are less likely to exclusively breastfeed, compared to nonoverweight mothers, in the hospital and also upon discharge.30 Factors associated with a woman’s feeding choice are not stable and significant changes may occur over time.31 When working regulations and facilities are changed to favor maternal needs, for example introducing flexible hours or nursing rooms, working mothers may adjust their behaviors more easily. Limitations This study has limitations because of the nature of the 2003 NSCH. All of the variables were self-reported and the BMI was parent reported. BMI for age was reported only in years, and children were thus assumed to be at the midpoint of the age year for purposes of calculating BMI for age. Comparing the 2003 NSCH to national estimates, it emerged that, for children under the age of 10, height was generally underreported and weight was overreported. Although maternal health is controlled for, no information about prepregnancy BMI was available. The survey also lacked data about length of exclusive breastfeeding, maternal age and employment, and marital status. Data are from 2003 and this impedes observation of whether the relationship has changed over time. However, data on overweight and obesity prevalence in early childhood were available only for this wave. Because data were cross sectional, no causality can be inferred from the results. Conclusion This study aims at contributing to the academic debate on breastfeeding and obesity by providing further empirical evidence on the role of breastfeeding initiation and duration on BMI levels later in life. Children breastfed for less than 3 months were found to have higher chances of becoming overweight and obese in early childhood. Though these results are statistically significant, caution is needed in generalizing them because of the several limitations discussed earlier. The strength of this study is that it is the first that tests the relationship on a large sample size representative of U.S. children aged 2 to 5 years. The sampled nature of the NSCH allows the generalization of research findings for U.S. children aged 2 Hansstein to 5 and avoidance of the self-selection bias of studies where participation is voluntary. Although important covariates were not included, the analysis includes key confounders such as household income or maternal health. Breastfeeding practices and policies should be encouraged and carefully targeted. Mothers should be encouraged to acknowledge the health benefits of breastfeeding, including the possible positive effect on the child’s BMI. SO WHAT? Implications for Health Promotion Practitioners and Researchers What is already known on this topic? In the academic literature, breastfeeding has been found to be protective against overweight and obesity later in life, but some uncertainty remains because of the many confounders that affect the relationship. The choice to breastfeed depends on many factors, for example income, education, and ethnicity. Previous findings have shown that mothers from higher socioeconomic status, as well as younger mothers, are more likely to breastfed their children. Both the American Academy of Pediatrics and the World Health Organization recommend exclusive breastfeeding for at least 6 months, and they acknowledge the protective role of breastfeeding on obesity later in life. What does this article add? This study aims to provide further empirical evidence on the relationship between breastfeeding and obesity by using data of a representative sample of the U.S. population, namely the 2003 National Survey of Children’s Health. This is the first study to employ an extensive survey in estimating the association and, at the same time, to focus on early childhood. The paper finds that breastfeeding for at least 3 months is protective against overweight and obesity in early childhood, though no information on breastfeeding exclusively was available and there were other important limitations. The paper targets children aged 2 to 5, assuming that if a protective effect is present, this is more likely to be observed at an age when the influences outside the family environment are limited. What are the implications for health promotion practice or research? Results support policy statements from the American Academy of Pediatrics and the World Health Organization and may be of interest to both researchers and practitioners in the field for further investigation. 257 Acknowledgments I express my gratitude for the feedback on an earlier draft of this manuscript to Professors Feng Liu from the School of Economics at the Shanghai University of Finance and Economics and Mario Mazzocchi from the Department of Statistics and Sara Capacci from the Department of Economics of the University of Bologna. I am grateful to the colleagues of the Health Sciences Center of West Virginia University for their useful comments on the initial work of the paper. I also thank the staff at the Child and Adolescent Health Measurement Initiative, Data Resources Center, for their responsiveness and support in helping me to have access to the National Survey of Children’s Health datasets. I thank the participants at the 2013 Child and Teen Consumption Conference and at seminars at the School of Public Economics and Administration and at the School of Economics at SHUFE University for their useful comments. Finally, I would like to that the staff of the library of the SHUFE for their research assistance. Any analyses, interpretations, or conclusions are reached by the author and not by NSCH, which is responsible only for the initial data. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) received no financial support for the research, authorship, and/or publication of this article. References 1. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity and trends in body mass index among US children and adolescents, 1999–2010. JAMA. 2012;307:483-490. 2. Krebs NF, Jacobson MS. Prevention of pediatric overweight and obesity. Pediatrics. 2003;112:424-430. 3. Puhl RM, Brownell KD. Confronting and coping with weight stigma: an investigation of overweight and obese adults. Obesity. 2006;14:1802-1815. 4. Latner JD, Stunkard AJ. Getting worse: the stigmatization of obese children. Obes Res. 2003;11:452-456. 5. Freedman DS, Khan LK, Serdula MK, et al. Inter-relationships among childhood BMI, childhood height, and adult obesity: the Bogalusa Heart Study. Int J Obesity. 2004;28:10-16. 6. Allcock DM, Gardner MJ, Sowers JR. Relation between childhood obesity and adult cardiovascular risk. Int J Pediatr Endocrinol. 2009;2009:108187. 7. Lefebvre CM, John RM. The effect of breastfeeding on childhood overweight and obesity: a systematic review of the literature. J Am Assoc Nurse Pract. 2013;26:386-401. 8. Arenz S, Rückerl R, Koletzko B, von Kries R. Breast-feeding and childhood obesity— a systematic review. Int J Obes. 2004;28: 1247-1256. 9. Harder T, Bergmann R, Kallischnigg G, Plagemann A. Duration of breastfeeding and risk of overweight: a meta-analysis. Am J Epidemiol. 2005;162:397-403. 10. Owen CG, Martin RM, Whincup PH, et al. Effect of infant feeding on the risk of obesity across the life course: a quantitative review of published evidence. Pediatrics. 2005;115:1367-1377. 258 11. Amir LH, Donath S. A systematic review of maternal obesity and breastfeeding intention, initiation and duration. BMC Pregnancy Childbirth. 2007;7:9. 12. Chezem JC. Breastfeeding attitudes among couples planning exclusive breastfeeding or mixed feeding. Breastfeed Med. 2012;7:155-162. 13. Ahluwalia IB, Morrow B, Hsia J, Grummer-Strawn LM. Who is breast-feeding? Recent trends from the pregnancy risk assessment and monitoring system. J Pediatrics. 2004;142:486-491. 14. Gibson-Davis CM, Brooks-Gunn J. Couples’ immigration status and ethnicity as determinants of breastfeeding. Am J Public Health. 2006;96:641-646. 15. Lutfiyya MN, Lipsky MS, Wisdom-Behounek J, InpanbutrMartinkus M. Is rural residency a risk factor for overweight and obesity for US children? Obesity. 2007;15:2348-2356. 16. Savino F, Liguori SA, Fissore MF, Oggero R. Breast milk hormones and their protective effect on obesity. Int J Pediatr Endocrinol. 2009;2009:327505. 17. Crume TL, Ogden L, Maligie M, et al. Longterm impact of neonatal breastfeeding on childhood adiposity and fat distribution among children exposed to diabetes in utero. Diabetes Care. 2011;34:641-645. 18. Blumberg SJ, Olson L, Frankel M, et al. Design and operation of the National Survey of Children’s Health, 2003. Vital Health Stat 1. 2005;(43):1-131. 19. Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of high body mass index in US children and adolescents, 2007–2008. JAMA. 2010;303:242-249 20. World Health Organization. Infant and young child feeding: model chapter for textbooks for medical students and allied health professionals, 2009. Available at: http://www.who.int/nutrition/ publications/infantfeeding/9789241597494/en/. Accessed April 8, 2014. 21. Centers for Disease Control and Prevention. Breastfeeding report card—United States, 2009. Available at: http://www.cdc.gov/ breastfeeding/data/reportcard.htm. Accessed April 8, 2014. American Journal of Health Promotion 30(4) 22. Williams R. Generalized ordered logit/ partial proportional odds models for ordinal dependent variables. Stata J. 2006;6:58-82. 23. Haas JS, Lee LB, Kaplan CP, et al. The association of race, socioeconomic status, and health insurance status with the prevalence of overweight among children and adolescents. Am J Public Health. 2003;93:2105-2110. 24. Singh GK, Kogan MD, Dee DL. Nativity/ immigrant status, race/ ethnicity, and socioeconomic determinants of breastfeeding initiation and duration in the United States, 2003. Pediatrics. 2007;119(suppl 1):S38-S46. 25. Holbrook KE, White MC, Heyman MB, Wojcicki JM. Maternal sociodemographic characteristics and the use of the Iowa Infant Attitude Feeding Scale to describe breastfeeding initiation and duration in a population of urban, Latina mothers: a prospective cohort study. Int Breastfeed J. 2013;8:7. 26. Martin RM, Patel R, Kramer MS, et al. Effects of promoting longer-term and exclusive breastfeeding on adiposity and insulin-like growth factor-I at age 11.5 years: a randomized trial effects of breastfeeding on adiposity and IGF-I. JAMA. 2013;309: 1005-1013. 27. Bartick M, Reinhold A. The burden of suboptimal breastfeeding in the United States: a pediatric cost analysis. Pediatrics. 2010; 125:e1048-e1056. 28. Eidelman A, Feldman-Winter L. From the American Academy of Pediatrics: policy statement: breastfeeding and the use of human milk. Pediatrics. 2005;115:496-506. 29. Cameron B, Javanparast S, Labbok M, et al. Breastfeeding support in child care: an international comparison of findings from Australia and the United States. Breastfeed Med. 2003;7: 163-166. 30. Visram H, Finkelstein SA, Feig D, et al. Breastfeeding intention and early postpartum practices among overweight and obese women in Ontario: a selective population-based cohort study. J Matern Fetal Neonatal Med. 2013;26:611-615. 31. Scott JA, Binns CW, Graham KI, Oddy WH. Temporal changes in the determinants of breastfeeding initiation. Birth. 2006;33:37-45. Copyright of American Journal of Health Promotion is the property of Sage Publications Inc. 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.
Running Head: CHILDHOOD OBESITY Childhood Obesity Reginald Sheftall Jr Argosy University 1 Childhood Obesity 2 Childhood obesity continues to become a widely discussed health issue thanks to the startling statistics in childhood obesity. For instance, according to Hansstein (2016) reports, 16.7 percent of children in 2010 were obese, whereas 31.8 were overweight or obese. Similarly, going children, aged 2-5 had a significant percentage that was high and above the normal weight line for young children. Hansstein (2016) further argues that although breastfeeding is quite protective against obesity research has established a little relationship between the length of breastfeeding and the possibility of a child to become obese in later stages of life. Imperatively, it is important to state that early childhood obesity exposes the child to a higher risk of obesity in later stages of life. This paper, however, seeks to ascertain the impact of reduced physical activity, nutritional lifestyle and rapid urbanization on Childhood Obesity. Lack of Physical Exercise Accumulation of adipose is often blamed as the main cause of obesity in individuals. While this is scientifically proven, lack of adequate physical exercise in school and home are generally blamed for the rise in obesity among individuals who are still considered children. Exercise helps burn up excess calories, body fat that is not required by the body. As such with the present trend in lifestyles with video games seemingly replacing outdoor games and constant, if not consistent snacking on junk, children are more than likely to end up obese and overweight. However, much of the failure to put in effort in physical activity is largely dependent on how the society perceived the obesity issue, where some view it as a genetic family issue, while others see it as a failing in character (Hamblin, 2015). Lack of healthy nutritional lifestyles Junk food is quite popular, with modern families, where parents who are actively lost in their occupations and lack kitchen time, result in feeding their young ones on soda and junk. Childhood Obesity 3 While these substances are not as risky per se, they are high in sugar and fat which accumulate in the body, without a removal plan. Additionally, the seriousness placed on leisure activity which often goes well with junk, as well as the inexpensiveness of the edibles, make the control of junk consumption almost impossible in children. As Tarvernise (2015) and Dawes (2016) opines, parents, healthcare makers, and policy givers are responsible for weight and weight control among the population and more specifically children, and they currently are not putting in enough efforts to lower it. Technology Urbanization technological advancement has resulted in the shift of activities that children engage in. Video games, social media, and movies all remove the time needed to take a walk or join a friend in a football game. This shift from traditional forms of leisure increases the sitting hours which in essence leads to accumulation of fats. On the other hand, such activities constantly go with consumption of junk, which further compounds the problem. This results in obesity which affects their self-esteem and general health in the long run. Conclusion Childhood obesity is an indicator of being overweight and or obese in later life. As such it is important that parents, caregivers, and policymakers recognize factors that lead to obesity in childhood. As discussed, these include a lack of physical exercise, poor nutrition, urbanization, which has caused a shift in lifestyle. On the other hand, it is essential to highlight the importance of breastfeeding during the early stages of life. While some theories indicate a relationship between duration of breastfeeding and obesity, comprehensive research has established that it provides an individual infant with the relevant nutrients that influence the uptake of substances Childhood Obesity and metabolism. Finally, in as much as statistics in childhood obesity continue to fall, they are still alarming which necessitates the need for mitigation. 4 Childhood Obesity References Dawes, L. (2016). Childhood Obesity in America: Biography of an Epidemic. Chicago: Argosy Education Group. Hamblin, J. (2015, December 16). Body Weight, Clash of Ideologies. Retrieved from The Atlantic: https://www.theatlantic.com/health/archive/2015/01/body-weight-clash-ofminds/384514/ Hansstein, F. V. (2016). The Impact of Breastfeeding on Early Childhood Obesity: Evidence From the National Survey of Children’s Health. American Journal of Health Promotion, 30(4), 250-258. Tavernise, S. (2015, November 12). Obesity Rises Despite All Efforts to Fight It, U.S. Health Officials Say. Retrieved from The New York Times: https://www.nytimes.com/2015/11/12/health/obesity-rises-despite-all-efforts-to-fight-itus-health-officials-say.html 5
Running Head: CHILDHOOD OBESITY Childhood Obesity Reginald Sheftall Jr Argosy University 1 Childhood Obesity 2 Childhood obesity in the United States continues to be a widely discussed health issue due in part to the startling statistics. For instance, according to Hansstein (2016), “16.7 percent of children in 2010 were obese, whereas 31.8 were overweight or obese. Similarly, going children, aged 2-5 had a significant percentage that was high and above the normal weight line for young children.” Hansstein (2016) argues that although breastfeeding is quite protective against obesity, research has established little relationship between the length of breastfeeding and the possibility of a child to become obese in later stages of life. It is important to state that early childhood obesity exposes the child to a higher risk of obesity in later stages of life. Therefore, childhood obesity is a major issue in the United States and there is a direct relationship between reduced physical activity, nutritional lifestyle, and rapid urbanization on childhood obesity. Good start on the intro and the thesis is improving. (remove all headers) Begin with the topic sentence for physical activity in your own words and then follow with the research. Accumulation of adipose is often blamed as the main cause of obesity in individuals. While this is scientifically proven, lack of adequate physical exercise in school and home are generally blamed for the rise in obesity among individuals who are still considered children. Exercise helps burn up excess calories, or body fat that is not required by the body. As such, with the present trend in lifestyles with video games seemingly replacing outdoor games and constant, if not consistently snacking on junk, children are more than likely to end up obese and overweight. However, much of the failure to put in effort in physical activity is largely dependent on how society perceives the obesity issue, where some view it as a genetic family issue, while others see it as a failing in character (Hamblin, 2015). Good start. Is this entire paragraph paraphrased? Any other research from the other sources to support this point? I would add more in your own words. Childhood Obesity 3 Begin with the topic sentence for this point in your own words. Junk food is quite popular with modern families where parents who are actively lost in their occupations and lack kitchen time, result in feeding their young ones on soda and junk. While these substances are not as risky per se, they are high in sugar and fat which accumulate in the body, without a removal plan. Additionally, the seriousness placed on leisure activity which often goes well with junk food, as well as the inexpensiveness of the edibles, make the control of junk consumption almost impossible in children. As Tavernise (2015) and Dawes (2016) opine, parents, healthcare makers, and policy givers are responsible for weight and weight control among the population and more specifically children, and they currently are not putting in enough efforts to lower it. Good start. Work on expansion of content and adding additional research from the other sources to support this point. Begin with topic sentence on this point in your own words. Urbanization technological advancement has resulted in the shift of activities that children engage in. Video games, social media, and movies all take away time needed to take a walk or join a friend in a football game. This shift from traditional forms of leisure increases the sitting hours which in essence leads to accumulation of fats. On the other hand, such activities constantly go with consumption of junk, which further compounds the problem. This results in obesity which affects their self-esteem and general health in the long run. Good start. Need research to support this point from your sources. Childhood obesity is an indicator of being overweight and or obese in later life. Good! As such, it is important that parents, caregivers, and policymakers recognize factors that lead to obesity in childhood. These include a lack of physical exercise, poor nutrition, and urbanization, which has caused a shift in lifestyle. In addition, it is essential to highlight the importance of breastfeeding during the early stages of life. While some theories indicate a relationship between duration of Childhood Obesity 4 breastfeeding and obesity, comprehensive research has established that it provides an individual infant with the relevant nutrients that influence the uptake of substances and metabolism. Finally, in as much as statistics in childhood obesity continue to fall, they are still alarming which necessitates the need for mitigation. Good conclusion! References Dawes, L. (2016). Childhood Obesity in America: Biography of an Epidemic. Chicago: Argosy Education Group. Hamblin, J. (2015, December 16). Body Weight, Clash of Ideologies. Retrieved from The Atlantic: https://www.theatlantic.com/health/archive/2015/01/body-weight-clash-ofminds/384514/ Hansstein, F. V. (2016). The Impact of Breastfeeding on Early Childhood Obesity: Evidence From the National Survey of Children’s Health. American Journal of Health Promotion, 30(4), 250-258. Tavernise, S. (2015, November 12). Obesity Rises Despite All Efforts to Fight It, U.S. Health Officials Say. Retrieved from The New York Times: https://www.nytimes.com/2015/11/12/health/obesity-rises-despite-all-efforts-to-fight-itus-health-officials-say.html

Tutor Answer

DrReginaldWoof
School: Duke University

Hi, please see the attached paper. Have...

flag Report DMCA
Review

Anonymous
Thanks, good work

Similar Questions
Hot Questions
Related Tags

Brown University





1271 Tutors

California Institute of Technology




2131 Tutors

Carnegie Mellon University




982 Tutors

Columbia University





1256 Tutors

Dartmouth University





2113 Tutors

Emory University





2279 Tutors

Harvard University





599 Tutors

Massachusetts Institute of Technology



2319 Tutors

New York University





1645 Tutors

Notre Dam University





1911 Tutors

Oklahoma University





2122 Tutors

Pennsylvania State University





932 Tutors

Princeton University





1211 Tutors

Stanford University





983 Tutors

University of California





1282 Tutors

Oxford University





123 Tutors

Yale University





2325 Tutors