Epidemiologic studies of carotenoids and disease can be divided broadly into two categories: observational studies and intervention trials. Observational studies examine the
association between carotenoid intake and disease inci- dence, or between blood or tissue levels of carotenoids and disease incidence. Although these studies have contributed enormously to the literature regarding effects of carotenoids on disease, the interpretation of these studies
is difficult for many reasons. First, fruits and vegetables
Figure 1. Reported prOtective effects of carotenoids in humans.
are complex mixtures of vitamins, minerals, fibers, and
numerous other phytochemicals. Thus, people who con- sume more carotenoids in their diet also consume more of
these other substances, many of which may have disease- preventive properties (2). Also, individuals who consume more fruits and vegetables may also consume less dietary
fat or may be more health-conscious in other ways than
individuals who consume relatively few fruits and vegeta- bles. Observational epidemiologic studies of carotenoids and human health must therefore be interpreted cautiously, as it is entirely possible that effects observed may
result from dietary or other factors correlated with carote- noid intake rather than from carotenoids themselves. Another complexity in the interpretation of observational epidemiologic studies of carotenoids and disease
arises from the lack of a standard, reproducible approach
for quantifying carotenoid intake. The older literature is
replete with terms such as “vitamin A from plants,” “carotene index,” and “yellow-green vegetables.” Exactly what constitutes these indices is far from clear. This problem
is a consequence of the lack of availability of adequate
carotenoid food composition tables, which until recently
have not contained information regarding the amounts of various carotenoids in foods commonly consumed. The U.S. Department of Agriculture released a carotenoid
food composition database in 1993 (3); the availability of
this database now allows researchers to estimate intake of several major dietary carotenoids including alpha-carotene, beta-carotene, beta-cryptoxanthin, the combination
of lutein plus zeaxanthin, and lycopene. These advances are enhancing observational epidemiologic studies of carotenoids and disease. Advances in analytical technology allow for the relatively rapid measurement of a number of carotenoids in
human blood and/or tissues, facilitating the conduct of biochemical epidemiologic studies of carotenoids and dis- ease. However, carotenoid levels in blood have been
shown to reflect fruit and vegetable intake (4), and thus biochemical assessment of carotenoid status in observational studies does not overcome the problem of carotetioids potentially acting as a marker for other correlated
etiologic factors.
I