Birth weight and risk of type 2 diabetes a systematic review

Peter H. Whincup, Samantha J. Kaye, Christopher G. Owen, Rachel Huxley, Derek G. Cook, Sonoko Anazawa, Elizabeth Barrett-Connor, Santosh K. Bhargava, Bryndís E. Birgisdottir, Sofia Carlsson, Susanne R. De Rooij, Roland F. Dyck, Johan G. Eriksson, Bonita Falkner, Caroline Fall, Tom Forsén, Valdemar Grill, Vilmundur Gudnason, Sonia Hulman, Elina HyppönenMona Jeffreys, Debbie A. Lawlor, David A. Leon, Junichi Minami, Gita Mishra, Clive Osmond, Chris Power, Janet W. Rich-Edwards, Tessa J. Roseboom, Harshpal Singh Sachdev, Holly Syddall, Inga Thorsdottir, Mauno Vanhala, Michael Wadsworth, Donald E. Yarbrough

    Research output: Contribution to journalReview article

    615 Citations (Scopus)


    Context: Low birth weight is implicated as a risk factor for type 2 diabetes. However, the strength, consistency, independence, and shape of the association have not been systematically examined. Objective: To conduct a quantitative systematic review examining published evidence on the association of birth weight and type 2 diabetes in adults. Data Sources and Study Selection: Relevant studies published by June 2008 were identified through literature searches using EMBASE (from 1980), MEDLINE (from 1950), and Web of Science (from 1980), with a combination of text words and Medical Subject Headings. Studies with either quantitative or qualitative estimates of the association between birth weight and type 2 diabetes were included. Data Extraction: Estimates of association (odds ratio [OR] per kilogram of increase in birth weight) were obtained from authors or from published reports in models that allowed the effects of adjustment (for body mass index and socioeconomic status) and the effects of exclusion (for macrosomia and maternal diabetes) to be examined. Estimates were pooled using random-effects models, allowing for the possibility that true associations differed between populations. Data Synthesis: Of 327 reports identified, 31 were found to be relevant. Data were obtained from 30 of these reports (31 populations; 6090 diabetes cases; 152 084 individuals). Inverse birth weight-type 2 diabetes associations were observed in 23 populations (9 of which were statistically significant) and positive associations were found in 8 (2 of which were statistically significant). Appreciable heterogeneity between populations (I2=66%; 95% confidence interval [CI], 51%-77%) was largely explained by positive associations in 2 native North American populations with high prevalences of maternal diabetes and in 1 other population of young adults. In the remaining 28 populations, the pooled OR of type 2 diabetes, adjusted for age and sex, was 0.75 (95% CI, 0.70-0.81) per kilogram. The shape of the birth weight-type 2 diabetes association was strongly graded, particularly at birth weights of 3 kg or less. Adjustment for current body mass index slightly strengthened the association (OR, 0.76 [95% CI, 0.70-0.82] before adjustment and 0.70 [95% CI, 0.65-0.76] after adjustment). Adjustment for socioeconomic status did not materially affect the association (OR, 0.77 [95% CI, 0.70-0.84] before adjustment and 0.78 [95% CI, 0.72-0.84] after adjustment). There was no strong evidence of publication or small study bias. Conclusion: In most populations studied, birth weight was inversely related to type 2 diabetes risk.

    Original languageEnglish
    Pages (from-to)2886-2897
    Number of pages12
    JournalJAMA - Journal of the American Medical Association
    Issue number24
    Publication statusPublished - 24 Dec 2008

    ASJC Scopus subject areas

    • Medicine(all)

    Cite this

    Whincup, P. H., Kaye, S. J., Owen, C. G., Huxley, R., Cook, D. G., Anazawa, S., ... Yarbrough, D. E. (2008). Birth weight and risk of type 2 diabetes a systematic review. JAMA - Journal of the American Medical Association, 300(24), 2886-2897.