Skip to main content

Table 5 Comparative features of previous and current study about effect of socioeconomic status on BMI change.

From: Mid-term Body Mass Index increase among obese and non-obese individuals in middle life and deprivation status: A cohort study

Ref. Population (% coverage) No of individuals included in final analysis (% women) Min-Max age at baseline Min-Max follow-up (mean follow- up) No of BMI measurements/method SES measurement Whether baseline BMI adjustment (or otherwise taken into account) in analysis Main findings in relation to SES effect on BMI change Comment
11 Subset of "Whitehall II" civil servant cohort study (73%, actual coverage higher as ~4% of invited persons had moved) 2,466 W
5,507 M
25 y – 25 y ~25 y – ~25 y (~25 y) 2 / 1st recalled, 2nd measured Employment grade (I-III) Yes Significant SES effect, particularly among those with largest BMI increase (i.e. > 6 kg/m2) Individuals who lost weight / BMI during follow-up were excluded.
12 Subset of Malmo Diet and Cancer Study, excluding those with history of cancer, heart attack and stroke inter alia.
Initial "invited" sample random. (NR)
5,464 W (100%) 20 y – 20 y 25 y – 53 y (36.6 y) 2 / 1st recalled, 2nd measured Employment status
Own occupational group Paternal (bread-winner) occupational group Educational attainment
Yes Significant SES effect, for all different SES measures  
13 Subset of the Medical Research Council National Survey of Health and Development Cohort Study (socially stratified cohort of 1946 newborns) 2,659 M + W (% W not explicitly reported in this study, originally cohort 47.5%) 20 y – 20 y 6 y (f-up 1)-23 y (f-up 4) (NR) 4 / First 2 recalled (some indication of underestimate), last 2 measured Paternal Social Class at age 14 Also Educational attainment Yes Significant childhood SES effect, even adjusting for educational attainment  
14 Finnish Twin Cohort Study (89% to 1st questionnaire, follow-up q'rres coverage of 84% and 77%) 2,482 monozygotic and 5,113 dizygotic twin pairs (56% of participants were W) 18 y – 60 y 6 y (f-up 1)
15 y (f-up 2) (6 y and 15 y for f-up 1–2 respectively
3 / All self-reported (validation study proves good validity) Educational attainment Yes Significant SES effect for BMI change between 1975–1981, but no effect between 1981 and 1990  
PS Borough residents (53.7% W 47.5% M, actual coverage higher as 10.8% "excluded" cases also included in denominator) 11,158 W
9,831 M (53.1% W)
35 y – 55 y 1 y – 10 y (4.8 y) 2 / Both measured Ecological (based on small area deprivation measurements) Yes Null SES effect for non-obese individuals, significant SES effect for obese individuals models adjusting for baseline BMI Stratification of analysis by baseline obesity status
  1. Ref: Reference, No: Number, BMI: Body Mass Index, SES: Socioeconomic Status, PS: Present study, W: Women, M: Men, F-up: Follow-up, NR: Not reported, y: years.