Food allergy is an increasing health care concern. Food allergy is defined as an adverse health effect secondary to a specific immune response that occurs reproducibly following exposure to a given food. The immune reaction may be IgE mediated, non-IgE mediated, or mixed. Cow's-milk protein is the leading cause of food allergy in infants and young children.1 Without an appropriate diagnostic workup, there is a high risk of both over- and underdiagnosis and therefore, over and undertreatment. This could affect dramatically to children's quality of life. Moreover, milk avoidance, when it is not necessary, may disturb child's nutritional status or growth development.2,3

Although cow's milk protein allergy (CMPA) is a prevalent condition, its management is quite different depending on the attending physician. Several guidelines4–7 from different organizations have been recently published in highly read journals, trying to unify the management of this condition among pediatric practitioners involved, including primary care pediatricians, pediatric gastroenterologists and pediatric allergologists.

The aim of our study was to analyze the variability in CMPA management by different pediatric gastroenterologist divisions in Spain.

A fifty-item questionnaire (Appendix A) was composed and sent through the Spanish Society for Pediatric Gastroenterology, Hepatology and Nutrition (SEGHNP) e-mail list, where the vast majority of pediatric gastroenterologists in Spain are included. This Society has a total of 321 members, among which 121 of them are full members. The questionnaire comprised open and closed items in a Likert's scale from 0 to 5, including questions about CMPA diagnosis, treatment and prognosis. The items score were grouped as follows: 0–1 “disagree”; 2–3 “undefined”; 4–5 “agree”. The questionnaire was sent January 15th 2016 and the study was closed on January 31st. Data were anonymously included in a database and then analyzed with SPSS 20.0.

Seventy-three questionnaires were received back. 56.9% of the respondents were SEGHNP full members. The median time of professional experience of the participants, was 16 years, ranging from 5 to 43 years.

Only 3 of the items reached more than 90% of concordance among the survey respondents. In other words, for those 3 items more than 90% of the survey respondents agreed with the same answer. In 11 items, the median score fell into the undefined category (2–3 points).

More than 90% of the survey respondents disagreed with the statement that a negative specific IgE test, ruled out CMPA. On the contrary, a 90% agreement was observed with the two following items: after a positive oral challenge, it is mandatory to avoid CMP during, at least, 6 months and by 2 years of age, 80% of CMPA infants will develop tolerance to CMP.

A huge variability was observed regarding answers to the open label questions so those data are not shown in the results.

Although CMPA is a very prevalent condition which pediatric gastroenterologists have been treating for decades, there is still huge variability on its management. Without an appropriate diagnostic workup, including food challenge procedures, there is a high risk of both over- and under-diagnosis. In consequence, we are under risk of over- and under-treatment.

We have observed a huge variability among pediatric gastroenterologists in Spain regarding CMPA diagnosis and management. Not surprisingly, the vast majority of the pediatric gastroenterologists in Spain, were aware of the existence of clinical practice guidelines on the problem, and most of them acknowledged to follow the ESPGHAN guidelines, published in 2012.1 Nevertheless, significant differences in clinical practice were found compared to the ESPGHAN recommendations both in diagnosis and treatment.

ESPGHAN guidelines indicate clearly when and when not to perform an OC in patients suspected to have CMPA. All cases of non IgE mediated CMPA, with the exception of Food Protein Induced Enterocolitis Syndrome (FPIES) should undergo an OC to be adequately diagnosed. In IgE mediated cases, with a clear reaction close to the moment of milk ingestion and with a positive specific IgE testing to CMP, the OC could be avoided. In our survey, 36% participants considered OC to be mandatory always. One in each four pediatric gastroenterologists considered symptom resolution on CMP exclusion diet was enough to diagnose CMPA.

A high percentage of respondents considered that exposition to milk could be performed at home in non severe, non IgE mediated CMPA patients, a scenario considered in the ESPGHAN guidelines. When asked to describe the protocol used for milk reintroduction at home, there were almost no coincidences, being nearly one different protocol per unit in the whole country.

Nearly half of the participants reported regular use of patch tests in their offices for the diagnosis of CMPA, a slightly higher percentage than reported in the literature.8 However, 76% of the respondents thought that non IgE mediated CMPA diagnosis should not be made exclusively based on its results. In current ESPGHAN guidelines, these tests are not recommended. The patch test is difficult to read and subjective. Moreover, there are difficulties on its standardization and application of antigen. In any case, recent studies indicate that they are reliable and safe but its use cannot substitute oral challenge in most of the cases.9

Only 11% of the respondents considered soy formula to be an adequate option for infants below six months of age. ESPGHAN discourages soy formula consumption in this age group because of its phytates and isoflavones content. Not surprisingly, only 5.5% considered milk from other mammals as a suitable option in the treatment of CMPA infants. This option should always be avoided8 because cow, goat and sheep milks evidence 87–98% identity of amino acids and immunological cross reactivity is a real issue.

In children older than 3 years of age, a diet with no dairy content is considered to be a safe option for treatment if CMPA persists beyond that age, given that calcium and vitamin D are adequately supplied in the child's diet. In our survey, close to 70% of the respondents did not consider this as a safe option. In the Spanish National Health System, dietitians are scarce. Those pediatric gastroenterologists, with no help from experienced pediatric dietitians, could feel insecure to recommend this diet to children under these circumstances.10

Although most of the CMPA infants will outgrow their allergy before the age of three, a significant proportion of children will continue to be allergic to cow's milk, being this percentage higher for those with an IgE mediated allergy.11 Four out of 10 of the respondents were not sure if this statement was true. In the last years, oral desensitization protocols (OTI) have been developed for these IgE patients and are widely used, even when several concerns have been raised regarding the possibility of developing eosinophilic esophagitis during the treatment. In our study, almost 90% of the respondents have access to these protocols in their practices.12

There are several reports indicating that accidental exposure to cow's milk has been responsible of deaths, and all CMPA children with an anaphylactic reaction or at risk of developing one and their parents should be instructed in the use of epinephrine injectors for that cause. Close to 10 percent of the responders to our survey, did not recognize accidental exposure to milk as a hazardous risk at any age.13,14

Finally, one out of four pediatric gastroenterologists in Spain, considered that their knowledge on CMPA was outdated and scarce, even when almost all of them were aware of the existence of practice guidelines. Continued medical education systems are not always capable of maintaining practitioners updated. Several initiatives have been conducted in order to enhance medical knowledge on food allergy management. Yu et al., carried out a study that developed and piloted a food allergy educational resource with very good results. Self-paced online training modules were selected by their survey respondent as one of the most preferred learning methods. Based upon this, the educational material comprised a slide set and written material as well as a self-injectable epinephrine demonstration. Significant improvement in knowledge and confident was observed after completed pre-and post-knowledge test.15

Another brief educational tool, is the Food Allergy Comprehension Tool (FACT) that is a rapid way to address known knowledge gaps among pediatricians and to identify areas in need of further intervention. It was designed to integrate with the current food allergy guidelines. It includes a pre- and post-evaluation, allergy case studies and several case-related questions with detailed explanations. Significant improvements in knowledge and confidence were observed.16

We have found surveys regarding CMPA management focused on primary care physicians. The goal of a survey carried by Gupta and coworkers,17 was to characterize food allergy knowledge, attitudes, and beliefs among primary care physicians and family physicians in the United States. Participants answered 61% of knowledge based items correctly. The overall score was significantly higher for pediatricians (62%) than for family physicians (54%).

A Spanish study18 found low awareness of existing guidelines for CMA among general pediatricians, pediatric nutritionists and allergologists. 39% of their survey respondent from 52 countries were unaware of CMA management guidelines. They also found a potential over-diagnosis among general pediatricians in Europe due to variable diagnostic practices.

To the best of our knowledge, this is the first survey conducted among pediatric gastroenterologists on this topic.

The strengths of our study were that in Spain, nearly all non-IgE mediated and many IgE mediated CMPA patients are managed by pediatric gastroenterologists, and the vast majority of them were reached through the mailing list. Twenty-three percent of the Spanish pediatric gastroenterologists answered the survey with a high percentage of full members, and data were collected and analyzed in a short period of time, avoiding communication among participants. The median time of professional experience was over 15 years, and more than half of the respondents were full members, indicating that participants in the study are experienced professionals with a high degree of specialization.

The main limitation of the study is a possible selection bias in the participants. Data were anonymously included in a database, but we can assume that respondents were those more confident in their answers or with a greater interest in CMPA, so we could reasonably expect to get better results in this group. It may well be that the final results may underestimate the real magnitude of the problem.

In conclusion, although CMPA is a prevalent condition that pediatric gastroenterologists have been treating for decades, we have found huge variability on its management. There are multiple opportunities for improvement in CMPA diagnostic workup and management among pediatric gastroenterologist in the future.

The authors declare that they have no conflict of interest.