Associations between head circumference, leg length and dementia in a Korean population Jae-Min Kim 1 , Robert Stewart 2 , Il-Seon Shin 1 , Sung-Wan Kim 1 , Su-Jin Yang 1 and Jin-Sang Yoon 1 * 1 Department of Psychiatry, Chonnam National University Medical School, Kwangju, Republic of Korea 2 Section of Epidemiology, Institute of Psychiatry, London, UK SUMMARY Background Adult leg length is influenced by nutritional intake in childhood. Shorter leg length has been found to be associated with adverse health outcomes in late life, including dementia. Smaller head circumference has also been found to be associated with dementia. The independence of these two potential markers of risk for dementia has not been investigated. Methods Community residents aged 65 or over (n ¼ 916) within a defined geographic area of South Korea were screened clinically for dementia and dementia subtypes. Data on anthropometric measures (head circumference and leg length), demographics (age, gender), apolipoprotein E, and early life environment (birth order, number of siblings, parental occupation, area of residence, education) were gathered. Reproductive history was also ascertained in women. Results Both smaller head circumference and shorter leg length were associated with increased age, rural residence in childhood and lower education. After adjustment for these factors, they were both independently associated with Alzheimer’s disease but only in women. Conclusions Smaller head size and shorter limb length were associated with lower early-life socio-economic status. Both factors were apparently independent markers of risk for dementia which may indicate risk factors occurring in childhood affecting both brain and skeletal development. Associations were principally present in women. Reasons for gender differences in associations require clarification but, for this population and age group, may include preferential treatment of male children. Copyright # 2007 John Wiley & Sons, Ltd. key words — head circumference; leg length; dementia; Alzheimer disease; Korea INTRODUCTION Reduced birth weight and/or reduced growth in infancy have been found to predict many adverse outcomes much later in life (Eriksson et al., 2003; Yliharsila et al., 2003; Forsen et al., 2004; Sayer et al., 2004). The early-life environment is also important in the etiology of cognitive decline and dementia. Educational level (Mortimer and Graves, 1993; Stern et al., 1994), area of residence (Moceri et al., 2000; Yano et al., 2000), parental occupation and household size in childhood (Moceri et al., 2001), have been identified as factors associated with dementia. One possible explanation is that an adverse early life environment affects maturation of the brain, and may increase vulnerability to later neurodegeneration. For disorders such as dementia, skeletal dimensions pro- vide potentially important proxy information on child- hood environment remaining stable into old age. Smaller head size has been associated with prevalence (Schofield et al., 1997), and incidence of dementia (Borenstein Graves et al., 2001), as well as with age at onset (Schofield et al., 1995), severity (Graves et al., 1996), and progression (Mori et al., 1997) of Alzheimer’s disease (AD), and with lower cognitive function in non-demented older populations (Reynolds et al., INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY Int. J. Geriatr. Psychiatry 2008; 23: 41–48. Published online 30 May 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/gps.1833 *Correspondence to: Prof. J.-S. Yoon, Department of Psychiatry, Chonnam National University Medical School, 5 Hak-dong, Dong-ku, Kwangju, 501-757, Republic of Korea. E-mail: jsyoon@chonnam.ac.kr Contract/grant sponsor: Chonnam National University Hospital; contract/grant number: CUHRICM-Y-200505. Contract/grant sponsor: Korea Health 21 R&D Project. Copyright # 2007 John Wiley & Sons, Ltd. Received 23 November 2006 Accepted 5 April 2007