In the field of pediatric gastroenterology, general knowledge of nutrition, including nutritional assessment and nutrition support, is an essential component of every patient's care. Often, nutrition support takes on a primary role in the management of diseases common to the pediatric gastroenterologist such as pancreatitis, pancreatic insufficiency, celiac disease, short bowel syndrome, chronic liver disease, and liver and small bowel transplantation (1).
The North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) has recognized the importance of nutrition education. In 1999, NASPGHAN published broad guidelines for training in pediatric gastroenterology (2). These guidelines provide a core curriculum that defines the minimum knowledge and technical skills required of fellows upon graduation. Within these guidelines, the nutrition curriculum distinguishes between level 1 of training, which is the basic training required for all trainees (Table 1), and level 2, which defines an advanced curriculum for the fellow who intends to become an expert in nutrition. For completion of level 2, the guidelines suggest a minimum of 1 year of advanced training at an academic center under the supervision of a full-time faculty in nutrition and participation in basic or clinical research in nutrition. Level 2 is intended for trainees who plan to conduct research in the area of nutrition or direct a nutrition support service.
To assess the status of nutrition education during fellowship training, the NASPGHAN Nutrition Committee surveyed the affiliated fellowship training programs in 2008. The goal of the survey was to assess the methodology and content of nutrition education and the variability of training among the different programs. Ultimately, it is hoped this knowledge will help NASPGHAN enhance the consistency among programs and refine the nutrition curriculum to meet the needs of the membership, the trainees, and our patients.
A survey was created using an Excel (Microsoft, Redmond, WA) spreadsheet. Questions and style for the survey were developed through discussions with members of the NASPGHAN Nutrition Committee as well as outside advisors with expertise in surveys who assisted in determining the appropriate content and structure of the questionnaire. Earlier iterations were piloted with members of the Nutrition Committee and the advisors. The types of questions and their wording were refined until consensus was achieved on the optimal questionnaire version, which then was used in the survey. The questionnaire was composed of 42 questions that were answered by drop-down menu, and it also included a section in which program directors were specifically asked to make comments regarding the obstacles faced in teaching nutrition and suggestions for NASPGHAN to help improve nutrition education. The questions were based on the NASPGHAN guidelines, specifically level 1 requirements in nutrition education (2). The questionnaire was distributed via e-mail to the fellowship program training directors of all of the affiliated programs in 2008 and 2009. Follow-up e-mails and telephone calls were then made to those program directors who had not responded to the initial request. Data are reported as mean ± standard deviation and frequency (%) when applicable. Correlations are computed with the Pearson method, and their 95% confidence intervals are also reported. All of the statistical analyses are conducted with R 2.13.0 (R Development, Vienna, Austria).
Of the 65 programs listed, 3 program directors responded that they did not have any fellows at the time and thus were considered inactive. Of the 62 active programs, 43 completed and returned the survey and 19 did not respond. Thus, there was a 69% overall response, which included 36 of 50 US programs, 4 of 7 Canadian programs, and 3 of 5 Mexican programs (Table 2).
The programs’ demographics are shown in Table 2. Programs from the United States, Canada, and Mexico showed a similar number of fellows per program. There was a wider distribution in the number of faculty among the programs. Approximately half of the US MD and dietary faculty had some certification in nutrition (eg, Clinical Nutrition Certification Board, American Clinical Board of Nutrition), and among 33 US programs providing information, 24 (67%) had at least 1 certified faculty.
Survey questions relating to nutrition education are summarized in Table 3. MD degree faculty provided the majority (61% ± 27%) of teaching, but a substantial minority of teaching was provided by registered dietitians. Teaching through clinical care provided the largest percentage of teaching opportunities, with problem-based learning providing the smallest. Competency was assessed through multiple means, but written tests (35%) were less likely to be used than oral questions to individual fellows (77%) or to groups of fellows (65%). Although not clarified further on the survey questionnaire, oral questions in the context of the options provided are meant as less formal questions such as those asked during clinical rounds or conferences. The majority of the programs (74%) reported that fellows had the opportunity to participate in nutrition-related research projects. Data regarding the number of fellows who pursued these opportunities were not collected nor did we obtain the number of fellows who pursued level 2 training because we were concerned with the time burden these additional questions would pose to the fellowship training directors and the resultant adverse effect on response rate.
Questions concerning specific topics covered as part of nutrition education training are summarized in Figure 1. For the purpose of this summary, the topics are grouped into 4 broad categories. The topics covered by ≥80% of the programs are considered to be consistently covered, those covered by 50% to 79% of programs are considered somewhat consistently covered, and those covered by <50% are considered inconsistently covered. In the category of nutrition and development, only the topic of recommended foods related to disease was covered consistently, and the 2 topics that were inconsistently covered were support of the breast-feeding mother and physiology of lactation. For the category assessment of nutritional status and estimation of nutrient requirements, 2 topics were consistently covered, clinical manifestations of nutrient excess and deficiency and classifications of overweight and underweight. Within this category, 5 topics were inconsistently covered, namely anthropometry, body composition measurements, nitrogen balance measurements, triceps skinfold, and midarm circumference measurements. Under the category nutrition support, the topics covered the least include the role of diet in the development of adult diseases and management of feeding disorders. Finally, within the category nutrition in disease, the topics inconsistently covered were nutrition issues and management in renal disease, acquired immune deficiency syndrome, and malignancy. The total number of topics covered at each program did not correlate with the number of faculty (correlation 0.045, 95% confidence interval −0.27 to 0.35; P = 0.78) or the number of fellows (correlation −0.085, 95% confidence interval −0.38 to 0.22; P = 0.59) per program.
Trainees often are able to rotate through specialized clinics or programs outside their own gastrointestinal division; thus, we evaluated the frequency with which this occurred. Of a total of 43 responding programs, 29 (67%) reported the use of specialized clinics as part of the fellowship training curriculum. The home total parenteral nutrition clinic (35%) and feeding team clinic (30%) were the most common rotations. Other less common rotations included cystic fibrosis clinic (21%), weight management program (14%), feeding disorders clinic (11%), lipid disorders clinic (7%), and bone clinic (2%).
To evaluate the training in clinical nutrition support, a series of questions relating to enteral and parenteral nutrition (PN) were provided (data not shown). Of significance, only 65% of programs reported that general ward PN or feeding orders were written by their fellows. Whereas only 32% of programs allowed for fellows to write pediatric intensive care unit (ICU) PN orders and 7% allowed for neonatal ICU PN orders.
Finally, 26 of 42 programs (62%) completed the comments and suggestions section of the survey. The main obstacle reported was the lack of interested faculty and the high workload (Table 4).
Nutrition education is one of the pillars in the training of pediatric gastroenterologists. The NASPGHAN guidelines provide a nutrition curriculum that sets out broad requirements for both a basic and an advanced level of nutrition training (2). To understand how well these guidelines are being implemented and understand the challenges to teaching nutrition, the fellowship program directors of NASPGHAN-affiliated programs were surveyed. The survey was designed to address the demographics, the methodology, and the content of teaching, particularly focusing on the published minimum requirements (level 1).
Nutrition education should be and is handled mainly by medical doctors. Although dietitians account for approximately 30% of the teaching in the United States, they probably are underused in that at least half of the programs in the United States have a nutrition support–certified dietitian.
It should be noted that nutrition is not recognized as a medical subspecialty by national regulatory agencies such as the American Board of Medical Specialties. Certifications in nutrition and nutrition support are awarded to health care providers of different academic backgrounds by a number of organizations including the American Clinical Board of Nutrition, the National Board of Nutrition Support Certification, the American Board of Physician Nutrition Specialists, and the Clinical Nutrition Certification Board after successful completion of examinations that assess nutritional support competencies specifically.
The size of the program did not correlate with the overall coverage of nutrition topics. In fact, neither the number of faculty nor the number of fellows per program was significantly associated with the percentage of topics covered during training.
When considering the methodology of teaching, we found that nutrition is taught principally through clinical care, with only 20% of programs reporting the use of lectures. This observation suggests that nutrition teaching during fellowship for the most part is unstructured. Indeed, several programs, when asked to comment, specifically identified the lack of a core curriculum as an important obstacle in teaching nutrition. The lack of structure may explain why a number of topics considered part of level 1 training were covered by less than half of all of the reporting programs. These inconsistently covered topics include the support of the breast-feeding mother, the physiology of lactation, anthropometry and measurements of triceps skinfold and midarm circumference, body composition and nitrogen balance, the role of diet in the development of adult diseases and management of feeding disorders, the nutritional management in renal disease, acquired immune deficiency syndrome, and malignancy.
The Accreditation Council for Graduate Medical Education has published a list of tools to assist in competence assessment (3). Perhaps the more commonly used are global ratings of recorded performance, chart stimulated recalls, oral examinations, checklist evaluations, patient surveys, simulations and models, and written examinations. Each method has its strengths and limitations, and overall competence may be best measured through a combination of tools. Our survey showed that competence assessment was accomplished through a variety of methods, but mainly via less formal oral questions to individuals or groups of fellows as opposed to written tests.
The majority of programs report using rotations in specialized clinics or programs as part of the fellows’ training in nutrition. Interestingly, some of these clinics seem to have a more important contribution than others to the teaching of specific nutritional topics. As such, the feeding team clinic is one of the more common rotations, used by 30% of all of the programs, yet the topic of nutritional management of feeding problems is only somewhat consistently covered (63% of all programs), presumably primarily by bedside teaching. On the contrary, the weight management clinic is one of the least common rotations (14% of programs), yet nutritional management of obesity is covered by 77% of all programs. These data suggest that other factors, such as the number of patients with a particular clinical problem the fellow is exposed to in all clinical settings, also may influence the coverage of nutritional topics.
Regarding clinical nutrition support, it is important to note that fellows are fairly well exposed to PN, although not uniformly because 84% of programs reported that fellows wrote PN orders. This percentage may be higher if one considers fellow participation in PN order decisions and not strictly order writing. Nevertheless, it is important to point out that fellows only wrote PN ICU orders in 32% of programs. These numbers may be affected by the rules at a particular institution regarding the role of fellows versus residents in writing PN orders. Although there was no specific question addressing the topic of ICU nutrition, other related questions such as nutritional consequences of stress were only somewhat consistently covered (53% of programs).
Sobering, in our view, is that some respondents identified a lack of interested faculty and/or lack of nutritionists as obstacles to nutrition education. We presume a lack of interest may be more connected to a lack of time to provide formal nutrition education (eg, lectures) rather than actual disinterest given that many of the faculty have some type of nutrition certification. Also noted were low reimbursements for nutrition consultations contrasted with financial incentives for procedure-based skills. These findings are similar to those previously reported in a survey of adult gastroenterology fellows regarding factors that prevented them from pursuing further training or research in nutrition (4). The authors of the present study identified the recruitment of a physician with expertise in nutrition as potentially the most beneficial step in improving nutrition training.
In view of the above findings, we have identified potential areas for improvement. First, it will be important to emphasize the minimum requirements as defined by the nutrition curriculum level 1 training. Second, the creation of a lecture series or computerized learning modules could assist programs in providing a comprehensive and structured education in nutrition. Third, a structured competence assessment is of utmost importance because it benefits both trainees and programs. Fourth, it is important to provide greater nutrition exposure, emphasizing its clinical relevance and application, as part of continuing medical education including at national meetings, including the NASGPHAN annual meeting. Finally, we hope that the present observations will assist in the revision of the Fellowship Training Guidelines, which is an ongoing project.
The authors appreciate the input of the 2007 NASPGHAN Nutrition Committee and Drs Jenifer Lightdale and Anthony Otley in the development of the survey and Inderpreet Jalli for preparing the Excel file.