Data from a prospective study were used to investigate risk factors fo
r hip fracture among a representative population of middle-aged adults
. During the years 1974-1978, all women (n = 25,298) and men (n = 27,0
15) aged 35-49 years in three Norwegian counties were invited to atten
d a cardiovascular screening (attendance rate = 91.5%). This cohort wa
s followed throughout 1990 with respect to hip fracture, for a total o
f 572,006 person-years. A total of 281 new fractures were identified,
of which 71 were excluded from the analysis due to high-energy trauma
or fracture in metastatic bone. Age-adjusted relative risks (RR) and 9
5% confidence intervals (CI) for hip fracture increased with body heig
ht in women (RR = 3.62, 95% CI 1.46-8.97, greater-than-or-equal-to 1.7
0 m vs. < 1.55 m) and men (RR = 2.92, 95% CI 0.94-9.05, greater-than-o
r-equal-to 1.85 m vs. < 1.70 m). A history of diabetes mellitus also g
ave elevated risk of fracture (RR = 5.81, 95% CI 2.15-15.71 in women a
nd RR = 7.67, 95% CI 2.40-24.53 in men). In addition, hip fracture was
related to body mass index (inverse), disability pension, and marital
status. An increased risk for smokers appeared only among those consu
ming greater-than-or-equal-to 15 cigarettes per day. In multivariate a
nalysis, all these risks remained elevated. This study suggests that,
as in older populations, known risk factors for low bone mass are rela
ted to hip fracture among middle-aged adults. In addition, body height
seems to have an independent influence on hip fracture incidence in t
his age group.