Diseases associated with overeating, undereating, and dietary or nutritional imbalance rank among the leading causes of illness and death in the United States. The relationships between specific dietary elements and specific health outcomes have been widely researched and are reasonably well understood; similarly, the role of primary care providers in providing or arranging for dietary counseling has been extensively investigated, but controversy exists about the magnitude of change than can be achieved and the effectiveness of different counseling strategies.
To update the chapter from the 1996 examining the effectiveness of counseling to promote a healthy diet and to assist the US Preventive Services Task Force in making recommendations on this topic.
To produce this systematic evidence review, we developed an analytic framework and 7 key questions that represent the logical chain between dietary counseling (especially about intake of total and saturated fat, fruits and vegetables, and fiber) and health practices and outcomes, together with linkages between diet and nutritional constituents and health outcomes for a wide array of disorders (e.g., cardiovascular disease, cancer). To supplement citations from the 1996 , we sought studies examining the effectiveness of dietary assessment and counseling using searches of MEDLINE for publications appearing from 1966 to 2000, by combining Medical Subject Headings related to diet and nutrition, primary care settings and practices, and counseling. We supplemented these searches with searches of the Cochrane Collaboration database and various bibliographies for recent systematic reviews and meta-analyses on the link between dietary patterns and health outcomes or between counseling and dietary behaviors.
To examine the relationship with diet and health outcomes, we selected systematic reviews, observational studies, and randomized trials relating specific dietary patterns and health outcomes. For studies of dietary assessment, we selected studies that examined test accuracy compared with a criterion standard. For studies linking counseling interventions with dietary change, we selected randomized controlled trials with pre- and post-test measures.
Trained reviewers and the authors abstracted data from the eligible articles onto evidence tables; the first authors checked all abstractions.
The relationships between dietary patterns and health outcomes have been examined in a wide range of observational studies. Few randomized trials have examined the effect of dietary interventions on health outcomes. The majority of studies show that persons consuming diets high in fruits, vegetables, fish, and whole grains or fiber and low in saturated and trans-unsaturated fats have lower rates of coronary heart disease and some forms of cancer. Similarly strong evidence supports the relationship between dietary intake of calcium and the risk of low bone mineral density. High intake of dietary sodium and low intake of dietary potassium are associated with higher blood pressure levels and increased incidence of hypertension. Efforts to reduce sodium intake and increase potassium have shown moderate effects on blood pressure, with greater effects seen in African-Americans and persons with hypertension. Several brief, valid dietary assessment instruments are feasible for the primary care setting. Although these instruments have not been evaluated as to their impact on health outcomes, they serve an important role of identifying dietary counseling needs and monitoring change over time. Many of these instruments are designed for specific patient populations or nutrients. We identified 33 articles examining the effect of nutritional counseling in primary care patients. Among primary care patients, nutrition counseling can produce modest improvements in saturated and total fat consumption, as well as fruit and vegetable consumption. The evidence is insufficient to determine the effectiveness of counseling in changing consumption of whole grains or fiber, calcium, sodium, or fish. Intensive interventions are more likely to produce large changes, but typical strategies pursued in primary care settings tend to be of lower intensity and produce smaller changes. Interventions using mailed or computer-generated materials appeared moderately effective, particularly in increasing fruit and vegetable consumption. Isolating the effect of a single counseling approach as more or less effective is made difficult by the tendency for counseling interventions to test multiple approaches simultaneously. Studies employing 3 or more well-proven counseling elements were more effective than those employing fewer elements.
Diets low in saturated and trans-unsaturated fat and high in fruits, vegetables, fish, and whole grains are associated with better health outcomes. Counseling patients can improve dietary behaviors, including reduction in dietary total and saturated fat and increases in fruit and vegetable intake. More intensive counseling and counseling directed to higher-risk patients have generally produced larger changes than less intensive interventions delivered to low-risk populations.
