|Skip left side navigation and go to content||
Predictors of Dietary Behavior
Karen Glanz, Ph.D., M.P.H.
|Clearly defined?||Measures available?||Associations found?||Changeable?||Which dietary?|
|Category of Influence|
|Biological & Demographic||Yes, mostly||Yes||Yes||
Use to target
|Energy, fat, fruit/veg, etc.|
|Psychological||Yes, for many||Yes||Yes, but modest||Yes, in the short-term||Fat, fruit/veg, eating|
|Social/Cultural||Wide variability||Yes, for some||Yes, mostly for kids||Some variables||
|Organizational||No, few definitions||Some - schools, worksites||Some; limited||Yes||Fat, fruit/veg|
|Physical Environment||Limited||Few||Promising||Yes||Fruit/veg, fat, grains|
|Policies/Incentives||Limited||Some, often proprietary||Yes, ecological level||Yes||Total food, and types|
Provisional answers to these questions are summarized in the table. From this preliminary review, a few key conclusions are possible:
There is a need to clarify definitions and terminology within categories, across disciplines. For example, the emphasis on “biology” varies widely and is evolving over time – does it include race? Is it a focus on metabolism? Are their genetic/biologic factors that drive behavior? Also, the label “environment” is often used by epidemiologists to refer to lifestyle, i.e., behaviors, whereas social scientists more often use it to refer to influences outside the person. Further confusion might derive from the public health focus on “environment” in reference to physical environments and related health threats (e.g., air pollution, water quality, chemical exposures).
The further we go up the social ladder (more upstream), the less clear are the definitions, less well defined are the measures, and the more limited is the body of research. While many questions can be answered by studying mainly demographic/biological, and psychological factors, substantial unexplained variance remains. At the social/cultural level, some variables have been defined, measured, and studied; but others remain scientifically vague. There is a need to work toward better definitions and measures, and to study associations in multi-level studies (not just ecologic associations and casual observations). There has been recent progress in measuring home, school and worksite food environments, and in identifying access to foods at a general level (i.e., are there grocery stores in a neighborhoods?). However, more work is needed to see whether available data can be useful, and/or to undertake new research.
In addition to addressing the above questions, it will be important to consider applications in which predictors of both diet and physical activity are examined. Key issues include: 1) How are predictors of dietary behavior related to predictors of physical activity? and 2) How do their respective predictors operate together? Also, different research foci for different stages of the life cycle (children, adolescents, adults, older adults) are needed to assure relevant measures and research designs, and appropriate interpretation.