The Spread of Obesity in a Large Social Network over 32 Years
New England Journal of Medicine, 2007 · paper · Sociology
Analyzes Framingham Heart Study data on 12,067 people from 1971 to 2003 to see whether weight gain spreads through social ties. A person's chance of becoming obese rose with obese friends, siblings and spouses, but not neighbors.
Checked: PDF from dash.harvard.edu, fetched 5 Oct 2026. Page numbers are those of that PDF.
2 findings
Reference or label The Discussion cites reference 42 ("health improvements in one person might spread to others.42"), but the reference list ends at entry 41 ("41. Granovetter MS. The strength of weak ties. Am J Sociol 1973;78:1360-80."); there is no entry 42.
- p. 10 Discussion, last paragraph, right column
son might spread to others.42
- p. 11 Reference list, last entry
41. Granovetter MS. The strength of weak
- p. 10 Discussion, penultimate paragraph
health improvements in one per-
- p. 10 Discussion, same sentence
might spread to others.42
Judge's note: Viewed the page 10 image: the superscript after "others." is clearly 42, and the preceding sentence cites 40,41 separately, so it is not a misread. Viewed the page 11 image: the third column of the reference list ends with entry 41 and the rest of the page is blank; page 11 is the last page of the PDF. The text layer agrees (highest entry 41). Nothing in the article explains the missing entry. Which source was meant can only be supplied by the authors or the journal.
Second check: another model agreed
Report a problem with this finding
Wording Methods define obesity as a body-mass index "of 30 or more", but the legends of Figures 1 and 2 both say "yellow denotes an obese person (body-mass index, >30)".
- p. 4 Methods, Statistical Analysis
height in meters) of 30 or more
- p. 5 Figure 1 legend
yellow denotes an obese person (body-mass index, >30)
- p. 6 Figure 2 legend
yellow denotes an obese person (body-mass index, >30)
- p. 5 Figure 1 legend
(body-mass index, >30)
Judge's note: Read the Methods sentence on PDF page 4 and both legends on PDF pages 5 and 6 from the page images; both legends print a plain ">" sign, not "≥". The paper gives only one definition of obesity and nothing says the figures used a different cut-off. The two differ only at a body-mass index of exactly 30, so the legends are most likely loosely worded.
Second check: another model agreed
Report a problem with this finding
10 minor (typos and the like; most checked by the judge only)
Wording "Obesity in a sibling of the opposite sex did not affect the chance that the other sibling would become obese." comes two sentences after the paragraph reports an increase "among siblings of the opposite sex (27%; 95% CI, 3 to 54)".
- p. 8 Results, siblings paragraph
opposite sex (27%; 95% CI, 3 to 54)
- p. 8 Results, siblings paragraph, last sentence
a sibling of the opposite sex did not affect the
- p. 8 Figure 4, row 'Opposite-sex sibling'
Opposite-sex sibling
- p. 8 Results, siblings paragraph, last sentence
Obesity in a sibling of the opposite sex did not affect the chance that the other sibling would become obese.
Only the authors can say what is right here.
Judge's note: Read the siblings paragraph and Figure 4 on the page image for PDF page 8. The Opposite-sex sibling row in Figure 4 is plotted to the right of zero, in line with the printed 27% (3 to 54). The closing sentence follows the separate brother and sister estimates, so it may report further models (brother to sister, sister to brother) for which no numbers are given; the paper does not say, and the Supplementary Appendix is not among the files. This is a question about wording against the printed estimate: which model does "did not affect" refer to? No P value is inferred from the interval.
Second check: another model agreed
Report a problem with this finding
Typo The same word is spelled two ways: "Discernible clusters of obese persons" (abstract) and "in a quantifiable and discernable pattern" (discussion).
- p. 2 Abstract, Results
Discernible clusters of obese persons
- p. 9 Discussion, first paragraph
social networks in a quantifiable and discernable
Judge's note: Read both spellings on the page images for PDF pages 2 and 9. Both are accepted spellings, so this is a consistency point only.
Second check: none (judge only)
Report a problem with this finding
Typo Affiliation note reads "the Department of Sociology, Harvard University, Boston (N.A.C.)"; the city may be meant to be Cambridge, MA.
- p. 2 Affiliation note, left margin
Harvard University, Boston (N.A.C.)
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Harvard University's Department of Sociology is in Cambridge, MA, not Boston. The page image for PDF page 2 shows the text as quoted; the same note gives "Cambridge, MA" for Mt. Auburn Hospital and Boston for Harvard Medical School. The location itself cannot be verified from the files.
Second check: none (judge only)
Report a problem with this finding
Typo Glossary, Connected component: "member of any other commponent." should read "component".
- p. 3 Glossary box, 'Connected component' entry
member of any other commponent.
- p. 3 Glossary, Connected component
any other commponent
Judge's note: The doubled m is printed in the Glossary box on the page image for PDF page 3; it is not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo Network Ascertainment: "87% percent of egos with siblings" repeats the unit; it should read "87% of egos with siblings".
- p. 4 Network Ascertainment, left column
87% percent of egos with siblings
- p. 3 Glossary, Connected component
any other commponent
- p. 4 Network Ascertainment
87% percent of egos
- p. 4
and 87% percent of egos with siblings
Judge's note: The page image for PDF page 4 shows "87% percent"; the same sentence has "83% of the spouses". One reviewer note also mentions "commponent", which is covered by C003.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 1 legend: "purple denotes a friendship or marital tie and orange denotes a familial tie", but in this copy the ties in Figure 1 are drawn in blue and red-orange.
- p. 5 Figure 1 legend
purple denotes a friendship or marital tie and orange denotes a familial tie
Judge's note: On the page image for PDF page 5 the ties in Figure 1 appear blue and red-orange, close to the colours of the node borders. In Figure 2 on PDF page 6 the ties do appear pale purple and orange as its legend says. This may be colour reproduction in this repository copy, not a legend error, so it is a minor observation. I viewed the full page only, not the crop at higher magnification.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 2 legend: "The largest connected subcomponent of the whole network and the change in obesity over the 32-year study period, are shown in an animation" has a comma between the subject and its verb.
- p. 6 Figure 2 legend, last sentence
over the 32-year study period, are shown in an animation
Judge's note: The comma is printed on the page image for PDF page 6. Punctuation only; the meaning is unaffected.
Second check: none (judge only)
Report a problem with this finding
Typo Figure 3 legend: "The examination took place at seven time points." (the figure key lists Examination 1 to 7) and "There is no trend in geographic distance." (meaning no trend in the effect across distance groups).
- p. 7 Figure 3 legend
The examina- tion took place at seven time points.
- p. 7 Figure 3 legend
There is no trend in geographic distance.
Judge's note: Both sentences read on the page image for PDF page 7. The figure key and the Methods describe seven examinations, so the singular is loose. The mileage groups rise by construction, so the second sentence must mean the effect shows no trend. The meaning is recoverable in both cases.
Second check: none (judge only)
Report a problem with this finding
Typo Discussion: "the opposite-sex effects and friendship effects may counteract each another." should read "each other" or "one another".
- p. 9 Discussion, right column, second paragraph
counteract each another.
- p. 9 Discussion, paragraph on spouses
counteract each another
Judge's note: "counteract each another." is printed on the page image for PDF page 9, right column.
Second check: none (judge only)
Report a problem with this finding
Typo Discussion: "Our findings that the weight gain of immediate neighbors did not affect ... and that geographic distance did not modify the effect for other types of alters (e.g., friends or siblings) helps rule out common exposure"; the plural subject "findings" takes "help".
- p. 9 Discussion, second paragraph
friends or siblings) helps rule out common expo
Judge's note: The sentence is printed as quoted on the page image for PDF page 9, left column. The subject is "Our findings" and the verb is "helps".
Second check: none (judge only)