The aims of this study were to compare the incidence of hip fracture i
n different regions of Turkey and to investigate causes for potential
differences. Hip fracture cases from two cities (Istanbul and Ankara)
and three rural regions (Samsun, Erzurum and Diyarbakir) were compared
with non-fractured controls from the same area with similar age and o
f the same sex. The risk of hip fracture was higher among persons livi
ng in rural areas than among persons living in urban areas, RR = 3.2 (
p < 0.001) and 2.3 (p = 0.009) in men and women, respectively. Adjustm
ent for differences in age and BMI between cases and controls did not
substantially change these findings, RR(adj) = 2.7 (p < 0.001) and 2.1
(p = 0.036), neither did adjustment for exercise or differences in go
nadal status in women. Education was the only adjustment factor that s
eemed to reduce differences between urban and rural areas, RR(adj) = 2
.0 (p = 0.109) and 1.2 (p = 0.816). No difference in the risk of hip f
racture could be detected between persons who migrated from rural to u
rban areas and persons born in urban areas. When restricting the analy
ses of differences between rural and urban areas to low-energy fractur
es, no difference in risk could be detected. When adjusting for differ
ences in age and body mass index (BMI) the relative risks were RR(adj)
= 0.8 (p = 0.873) and 1.2 (p = 0.852). The conclusion of these result
s is that the observed higher total risk of hip fracture in the rural
areas of Turkey primarily can be explained by a larger proportion of h
igh-energy fractures in the rural areas, whereas the risk of low-energ
y fractures seems to be similar.