Navigation bar--use text links at bottom of page.

(Effects of a Nutrition Education Program on Physicians/Residents--continued)


The purpose of this study was to evaluate the effectiveness of a 9-month, 11-session nutrition education lecture series on the related knowledge and professional behaviors for a group of first, second, and third-year family practice residents. The objectives of the study were:

  1. To demonstrate that residents who received nutrition education would increase their nutrition knowledge as measured by scores on a 55-question comprehensive nutrition test;

  2. To demonstrate that increased nutrition knowledge of the residents would result in an increase in the frequency and number of nutrition handouts, nutrition discussions, and nutrition recommendations, and;

  3. To demonstrate that increased nutrition knowledge would result in an increase in the adequacy of the residents’ responses to patient requests for nutrition information.

Nutrition knowledge of residents/physicians is lacking, and there is no consensus on how to resolve this issue. The analysis of the literature relative to the purpose of the study showed that the nutrition knowledge of medical students, residents and physicians was lacking (Krause and Fox, 1977; Mlodinow and Barrett-Connor, 1989; Morgan et al., 1988; Lasswell et al., 1995), that nutrition education in medical schools and residencies was inadequate (Committee on Nutrition in Medical Education, 1985; Young, 1992; Nuhlicek et al., 1989; Wincik, 1993) and no definitive system on how best to incorporate nutrition education into medical training had been devised (Walsh et al., 1987; Murphy, 1990; Weinsier et al., 1991). While nutrition was identified as important, few physicians scored high on nutrition tests, and/or rated themselves as competent in their nutrition knowledge. Researchers also found a significant negative correlation between years in practice and nutrition knowledge (Krause and Fox, 1977).

Formal nutrition education is lacking in medical schools. Several studies of medical schools and residency programs showed that few had formal nutrition education programs in place or offered any formal educational curriculum in nutrition. Surveys of practicing physicians, residents, and graduating medical students showed that most felt their nutrition training was inadequate.

The literature review also showed that several studies have reported on programs that have been implemented in an attempt to provide nutrition training to residents and physicians. Some of the problems identified included inadequate time available, unqualified educators, a shortage of nutrition oriented physician role models, and poor lecture attendance by the physicians. Components of the successful programs included having a qualified nutrition educator who built rapport with faculty, residents and staff, sharing a common conference area with residents and faculty, providing mandatory nutrition presentations at regularly scheduled times, and providing practical learning materials.

Features of effective nutrition education programs. It was also shown that family practice residency is an effective time for nutrition education, that noontime lecture conference series are an effective educational tool for family practice residents, that the co-counseling interactive learning approach can be an effective model, and that registered dietitians can be effective in educating residents (Gray, 1988).

The specific null-hypotheses tested at the 0.05 level of significance in this study were as follows:

  1. H0(1): program does not change resident nutrition knowledge. That is, no differences exist between the residents’ scores on a standardized nutrition test before and after the 11-session nutrition education program.

  2. H0(2): program does not change resident nutrition behaviors. No differences exist between the frequency of the residents’ distribution of nutrition handouts, discussions of nutrition, and nutrition-related patient recommendations, during patient visits before and after the 11-session nutrition education program.

  3. H0(3): program does not change resident responses to patients. No differences exist in the adequacy of the residents’ responses to patient requests for nutrition information before and after the 11-session nutrition education program.

The 15 residents used in the study underwent a nutrition education program that consisted of 11 nutrition education sessions conducted 1-2 times monthly, over a period of 9 months. The residents were tested twice using a comprehensive, 55-question nutrition test. A pre-test was administered to the residents immediately prior to the beginning of the educational program. A post-test was administered immediately after the completion of the program. The test consisted of five questions from each of the 11 lecture topic areas.

A patient questionnaire was developed to evaluate the patients’ perception of the nutrition-related behaviors of the residents and the amount, type and effectiveness of nutrition information presented during visits with their physician. The questionnaire monitored the frequency and number of nutrition-related handouts distributed, nutrition-related discussions, and nutrition recommendations made. This distribution occurred both immediately prior to the beginning of the program, and immediately after its completion. A total of 610 patients were surveyed, 304 pre-test and 306 post-test.

The first null-hypothesis dealt with changes in nutrition knowledge of residents undergoing the 11-session nutrition education program. The t-test analysis indicated significant difference at the 0.01 confidence level for the residents when all years were grouped together. The t-test analysis indicated significant difference at the 0.05 confidence level for the second-year residents, when residents were grouped by year. Therefore, the first null-hypothesis was rejected. In other words, the data indicated a significant increase in the nutrition knowledge of the residents undergoing the 11 session nutrition education program, as was predicted in the objectives.

Mlodinow and Barrett-Connor (1989) reported physicians answered only 69% of test scores correctly with individual scores ranging from 46.8% to 83.9%. The residents in this study scored similarly with a post-test average of 69.3% and individual scores ranging from 58.2% to 78.2%.

The second null-hypothesis examined the differences in frequency of nutrition-related behaviors of the residents during their patient interactions as perceived by the patient. For the difference in frequency of discussion of nutrition in relation to health maintenance, the change in score was from 17.1% pre-test, to 23.1% post-test. The difference between pre- and post-test was found to be statistically significant at the 0.05 confidence level. The change in frequency of nutrition recommendations was from 7.9% pre-test, to 15.0% post-test. The difference between pre- and post-test was found to be statistically significant at the 0.01 confidence level. The change in frequency of dietitian referrals was from 1.3% pre-test, to 5.9% post-test. The difference between pre- and post-test was also found to be statistically significant at the 0.01 confidence level. No significant differences occurred in the frequency of nutrition-related handouts or overall nutrition discussions. Based on the analysis of these data, the second null-hypothesis was partially rejected. The findings indicated support for the interpretation that the nutrition education program was partially effective in increasing the nutrition-related behaviors of the family practice residents.

The third null-hypothesis examined the level of patient requests for nutrition information and the adequacy of the residents’ response to patient requests for nutrition information. The analysis of the data indicated a significant increase (p<0.05) between the pre-test and post-test frequency in the number of patient requests for nutrition information. While not significantly different using chi-square statistics, there was a significant increase (p < 0.05) in the resident responses evaluated by patients as "Very complete" when the data were analyzed using the normal approximation test for equality of proportions. Based on the analysis of these data, the third null-hypothesis was rejected. These findings indicate support for the interpretation that the nutrition education program was effective in increasing the adequacy of the residents’ responses to patient requests for nutrition information.


The analysis of the data collected relative to the principal objectives of the study indicated significant differences between the pre- and post-test scores of the residents after an 11-session nutrition education program. While only the overall score for all residents and the scores for the second year were statistically significant, a tendency for scores to increase was also observed for the first and third-year residents. No differences were noted for the reasons and duration that the patients were visiting their physician, as expected. The data also indicated improvement in several nutrition-related education behaviors of the residents. Significant increases were noted for the discussion of nutrition related to health maintenance, for overall nutrition recommendations, and for recommendations to see a dietitian. Additionally, the data indicate that the nutrition education program was effective in significantly increasing the adequacy of the residents’ responses to patient requests for nutrition information.

Given the opportunity, physicians who have a better understanding of nutrition seem more likely to discuss nutrition with their patients. These physicians are also more likely to understand the important role nutrition plays in health maintenance and to try to communicate this importance to their patients. Therefore, successful nutrition education programs like this one can help increase the proportion of primary-care providers who provide nutrition counseling and/or referral to qualified nutrition specialists to over 50% as recommended in the publication, Promoting Health/Preventing Disease: Year 2000 Objectives for the Nation (U.S. Department of Health and Human Services, 1989).

This study showed that an 11 session nutrition education program during residency can not only impact the knowledge of residents, but also their interactions with their patients. The study also showed that regularly scheduled noon time conferences are an effective time for the program, and that dietitians can be effective educators. Co-counseling, or involving the residents in the preparation and presentation of the programming, along with providing practical handouts, are also beneficial. Therefore, educating residents about nutrition may not only help contribute to patient nutrition education, but also improve their health, as will be discussed later.

Externalities may have significant impact on program. Some of the key issues in the success of this program may have happened outside of the formal educational infrastructure. Having the dietitian available on site to the residents was important. The program director, a dietitian, was available to the residents during the day and during their noontime lunches. This allowed the residents and the dietitian informal time to build rapport, discuss nutrition, and (discuss) their individual patients. Other programs have also found such informal interaction with a dietitian to have a positive impact (Adams and Jackson, 1984).

The findings here are supported by the literature. The 1983 Young study showed that those physicians who obtained their information about clinical nutrition from nutrition-specific resources, such as nutritionists and dietitians, had significantly higher favorable attitude scores then those physicians who obtained their information from non-nutrition-specific resources. These physicians also tended to determine their patients’ nutritional status, to advise and teach desirable health habits, and to identify patients at risk for malnutrition more frequently then their peers. These physicians also held significantly more favorable attitudes on the topic of how diet can affect their patients’ health, and a greater confidence in their ability to provide nutrition counseling, than those who had not studied nutrition.

The 1988 study by Gray et al. also found that residents benefited by participating in co-counseling sessions with a dietitian. In their study, the residents were evaluated through chart reviews and video observations. These residents showed significant increases in their nutrition counseling scores for hypertension and trends for improvement in pregnancy and diabetes.

In 1989, Murphy investigated the effects of a nutrition curriculum on the counseling skills and behaviors of family physicians. A trend towards higher scores, in terms of frequency and duration of nutrition counseling was seen among the physicians who received nutrition education.

In 1993, Lazarus et al. studied the effects of a nutrition program provided by a physician nutrition specialist in a family practice residency. Results showed significant increases in the nutrition knowledge of the physicians, the nutrition knowledge of the patients, the patients’ belief in the importance of complying with a diet order, and an increase in diet recommendations. The study also found that residency is an appropriate time for nutrition education programs.

1995: similar study with positive effects. At the same time that the current study was underway, Kirby et al. (1995) were also testing the impact of a nutrition education program in a family practice residency program. Their nutrition education program included four one-hour teaching sessions with interactive demonstrations and case studies, resident participation in a three-day dietary analysis, and knowledge based pre- and post-tests. The intervention group of residents showed a significant increase in their test scores, from 50% on pre-tests before the four teaching sessions to 70% on post-tests, providing evidence of the effectiveness of the nutrition education program. The intervention group of residents also rated their own knowledge of nutrition higher than the control group did.

A number of problems and limitations were encountered in the process of implementing the current research study. These should be considered when interpreting these results and should be dealt with when continuing this type of research. One problem encountered was that there was no way to ensure 100% attendance at all the lectures. In order to encourage better attendance at each session, gourmet healthy vegetarian meals were catered in and served. Residents were sometimes called away on emergencies and occasionally had conflicts with other requirements of their residency. To compensate for this, materials from the sessions were provided to any resident who was unable to attend. It would be important to take this into consideration in future studies, and try to find a way to present all the information to every resident. Additionally, in the future, statistical analysis should be set up to be able to compare the scores of those residents who attended sessions, versus those who didn’t.

Another problem encountered was proper training of educators. While an effort was made to use only dietitians with special training or education in specific areas, some dietitians were less adequately trained then others to educate physicians. Some of these dietitians worked in the community with low income and minority populations. They were not prepared or qualified to answer and discuss some of the questions and issues at the level needed by the residents and occasionally did not cover the topic material included in the nutrition knowledge pre- and post-test. After the first two sessions, the decision was made to use more university professors and experienced educators. In replications of this study, it is recommended that educators be carefully selected, and highly trained and familiar with educating physicians.

While each one of the residents was involved in the preparation and presentation of one of the education sessions that they chose beforehand, there was no testing to see if this had an effect on knowledge for that specific area. In other words, did involving a resident in a specific area, such as diabetes, improve his or her test score more on the diabetes section of the test?

A number of findings can be derived from the study which may have practical importance for individuals directly involved in educating residents and/or physicians. The findings from this study confirmed that regularly scheduled noontime conferences during family practice residency are an effective time for nutrition education sessions. Finding time to add a new program to the existing resident schedule is difficult, and noontime conferences can be an effective time for this. Mandatory attendance would also help in increasing attendance figures. Winick, in a 1993 special article on nutrition education in medical schools, also concluded that nutrition should be an integral part of primary care residency, and identified family practice residencies as the ideal place and setting to have the greatest impact on educating physicians about nutrition knowledge and skills.

This study confirmed that properly trained dietitians can be effective educators, if they are skilled and knowledgeable in the topic area and experienced in speaking and lecturing to an audience such as physicians. The 1994 American Dietetic Association position paper (White et. al., 1994) regarding nutrition and medical education encouraged resident programs to use registered dietitians with advanced degrees to provide physician education and patient care.

Another finding was that providing practical handouts on each topic was important. This gave the residents additional information on each topic that they could review on their own time and refer to when necessary. The information was also provided to the residents as a file on their laptop computers The 1991 Weinsier et al. study showed that one of the key elements in creating and implementing a strong nutrition program was practical learning materials.

Although there was some increase in knowledge scores, perhaps the most important outcome of the study was an increased understanding by the residents of the need to include nutrition in patient care and the value of the registered dietitian to the health care team. This is evidenced by the increase in nutrition-related professional behaviors, specifically referrals to registered dietitians. This is the ultimate goal of a nutrition education program for physicians--not to make physicians become dietitians.

Recent studies on nutrition education for physicians (published after this study was done) report similar findings. An exhaustive search of the literature was completed to see if there were any changes in the status of nutrition education for medical schools and residents since the completion of this study. Several published studies looked at nutrition in medical education in Europe and Asia. The results of those studies were consistent with the results indicated in the studies in the earlier literature in regard to:

Additionally, several other studies done in the United States in this topic area were published in the period. A 1997 study by Lazarus reported that a six-month nutrition education program improved physician knowledge scores and the frequency with which they discussed nutrition and recommended diets to their patients. The study also concluded that residency is an ideal time for nutrition education. A 1995 study (Ockene, Ockene, Quirk, Hebert, Saperia, Luippold, Merriam, and Ellis, 1995) examined the effect of a three-hour education program on physician knowledge, attitudes and skills. No change in physician knowledge occurred; however, physicians reported positive changes in 4 out of 6 measures of their attitudes regarding nutrition counseling, and their counseling skills significantly improved. This suggests that a longer program over a more extended period of time is more effective in improving knowledge.

A 1996 study (Ockene, Herbert, Ockene, Merriam, Hurley, and Saperia, 1996) examined the effect of a nutrition education program on the counseling skills of primary care internists. The study concluded that primary care residents, when provided with training in nutritional counseling techniques and a supportive office environment, would counsel patients effectively. A 1995 study (Johnson, Murphy, and Michener, 1995) reported on the preliminary results of the authors’ efforts to improve medical students’ nutrition knowledge, to encourage application of this knowledge, and to help maintain a positive attitude towards nutrition over their training. The initial preliminary results indicate success in improving initial nutrition knowledge and attitudes.

The first steps in educating physicians are finding the adequate time during their education and a medical program that is interested. Another step is to find competent educators who can spend additional time interacting with the residents. An additional step would be to utilize effective training methods, such as co-counseling, case studies, and practical material to get the residents involved in the education process. And finally, an ongoing nutrition education program during medical training, along with materials and easily available referral nutritionists to support the practical aspects of nutrition education for their patients is needed.

Research Recommendations

One suggestion for further research might deal with the effect of the nutrition education of the physician on the health of their patients. The results of this study focused on the knowledge of the physicians and their nutrition-related professional behavior with their patients. It would be important to know if this information is affecting the health of their patients. While this study focused mainly on the impact on the physician, the ultimate goal of this effort is to help reduce the incidence of nutrition-related diseases in the public. A 1996 study (Evans, Rogers, Peden, Seeling, Layne, Levine, Levin, Grossman, Darden, Jackson, Ammerman, Settle, Stritter, and Fletcher) evaluated whether a nutrition education intervention program for physicians improved the cholesterol levels of their patients. Despite successful changes in physicians’ attitudes and behaviors, there was no significant change in the patients’ cholesterol levels. However, the type of intervention used consisted of only two one-hour education sessions. A longer program, such as the one used in the current study, might have more impact. This type of research would provide a better understanding of the effectiveness of trying to improve the public health by educating physicians/residents in nutrition.



Return to beginning of article


Back to Special Topics: Nutrition

   Beyond Veg home   |   Feedback   |   Links