Edited by: Kristin M. von Ranson, University of Calgary, Canada
Reviewed by: David Hunter Gleaves, University of South Australia, Australia; Andrea Sabrina Hartmann, University of Osnabrück, Germany
This article was submitted to Eating Behavior, a section of the journal Frontiers in Psychology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Eating disorders are serious and potentially fatal health problems that constitute a considerable burden of mental health problems (
Despite an increasing number of studies about ON, the body of scientific evidence is still highly fragmented. At present, ON does not have its own classification in the Diagnostic and Statistical Manual of Mental Disorders (DSM)-5, which is partially due to an ongoing discussion about how to properly classify ON (
The controversy regarding the underlying pathology of the disorder and the lack of research on the subject have resulted in an absence of validated diagnostic criteria.
While attempts to determine the prevalence of ON have been made, the lack of consensus on diagnostic criteria for the condition means there is not yet a reliable estimate available. The available estimates of prevalence of ON range from less than 1 to 88.7%, with studies conducted in different population groups (
Additionally the knowledge on practicing health professionals’ opinions regarding ON is lacking. This mixed methods study was conducted in the Netherlands (Spring-Summer 2018) in an attempt to fill this knowledge gap and move one step closer to understanding the phenomenon of ON. To the best of our knowledge, this study is the first to focus on the opinions and ideas of health professionals with regards to Orthorexia Nervosa.
The objective of the study was to assess the level of recognition of ON as a distinct disorder amongst health providers in the Netherlands and to uncover their opinions on the pathology, classification, and diagnostic criteria for the condition. Given that different health professionals have different experience and knowledge, their opinions and ideas on a specific topic may vary. For this reason, a distinction has been made between mental health professionals (MHP) (including psychiatrists and psychologists) and physical health professionals (PHP) (dietitians and physiotherapists).
A mixed-methods design with concurrent triangulation was adopted, unifying the results from a self-administered online questionnaire and semi-structured interviews. This design was believed to provide the most insight on different health professionals’ opinions, as well as a deeper understanding of these in a smaller group of MHPs.
We designed our questionnaire in order to collect initial data on the opinions of health professionals about ON in the Netherlands.
As no validated measures were available for the data collection required for this opinion study and no similar studies had been previously conducted, the co-authors created a questionnaire based on a framework that visualizes factors possibly influencing a health professional’s opinions regarding ON (Appendix
The questionnaire was digitally distributed among psychologists, psychiatrists, dietitians and physiotherapists (total
The providers were then contacted via email and through social media and were presented with an anonymous link to the questionnaire. They were also asked to forward the questionnaire link to their colleagues.
The questionnaire consisted of three sections: recognition, diagnosis and classification of ON; potential influence of modern Western culture on ON; and demographic characteristics of the respondents. Questions within the section on recognition, diagnosis and classification of ON included items such as whether participants recognized the described pattern of symptoms, what their subjective estimation of the prevalence of the disorder was, whether they found current diagnostic criteria sufficient and opinions about ON classification. Results regarding the potential influence of the modern Western culture can be found in another article published in this journal (
Semi-structured interviews were conducted to triangulate and enrich the quantitative data. An interview guide (Appendix
Clinically active psychologists and psychiatrists in the Netherlands were eligible to participate in the interviews if they had handled at least one eating disorder case in the last year. The list of health professionals used for recruitment of questionnaire participants was also used for recruitment of interview participants. Due to privacy protection we could not check for any overlap, i.e., whether interview participants were also the ones filling in the questionnaire. The recruitment occurred via email and telephone calls.
The participants signed an informed consent prior to the interviews. The interviews were then recorded and transcribed verbatim.
The interviews were semi-structured in nature and consisted of four parts: opinions on diagnosis and categorization, a hypothetical case based on current proposed criteria for ON, potential link to the modern Western culture and demographic information. The potential link to the modern Western culture has been discussed elsewhere (
When discussing the hypothetical case, participants were presented with an infographic containing a figure surrounded by text boxes with symptoms of ON (Appendix
A sample size calculation revealed that a total number of 88 participants was required to yield a power of 0.8 at alpha-level 0.05 with a Pearson’s
Questionnaire participant characteristics.
Male | 20 | 12.5 |
Female | 138 | 86.3 |
Not listed | 1 | 0.6 |
25 and younger | 10 | 6.3 |
26 to 35 | 43 | 26.9 |
36 to 45 | 28 | 17.5 |
46 to 55 | 39 | 24.4 |
56 to 65 | 33 | 20.6 |
66 and older | 2 | 1.3 |
Higher Professional Education | 75 | 46.9 |
Bachelor | 10 | 6.3 |
Master | 61 | 38.1 |
Other a | 12 | 7.5 |
Psychologist | 41 | 25.6 |
Psychiatrist | 3 | 1.9 |
Dietitian | 71 | 44.4 |
Physiotherapist | 34 | 21.3 |
Otherb | 9 | 5.6 |
0 to 7 | 56 | 35 |
8 to 15 | 30 | 18.8 |
16 to 23 | 21 | 13.1 |
24 to 31 | 25 | 15.6 |
32 to 39 | 21 | 13.1 |
40 or more | 1 | 0.6 |
Yes | 138 | 86.3 |
No | 20 | 12.5 |
Yes, within the last year | 76 | 47.5 |
Yes, more than one year ago | 25 | 15.6 |
No | 59 | 36.9 |
Total |
160 | 100.0 |
Fifteen individuals participated in the interviews: thirteen psychologists and two psychiatrists. Four participants were expatriates (origins were Norway, Romania, Turkey, and Greece), with the remaining eleven being native Dutch. The sample consisted of fourteen women and one man. Eleven had undertaken a master’s program, ten of whom had already or were in the process of further specializing. Work experience ranged from 3 to 32 years. Twelve of the individuals regularly met patients with eating disorders, while three participants only rarely encountered this patient group (see
Interview participant characteristics.
Male | 1 | 6.7 |
Female | 14 | 90.3 |
25 and younger | 0 | 0.0 |
26 to 35 | 2 | 13.3 |
36 to 45 | 5 | 33.3 |
46 to 55 | 4 | 26.7 |
56 to 65 | 4 | 26.7 |
66 and older | 0 | 0.0 |
Psychologist | 13 | 86.7 |
Psychiatrist | 2 | 13.3 |
Dutch | 11 | 73.3 |
Non-Dutch | 4 | 26.7 |
Master | 11 | 73.3 |
Doctoral | 6 | 40.0 |
Specialization | 10 | 66.7 |
The Netherlands | 9 | 60.0 |
Other | 3 | 20.0 |
0 to 7 | 2 | 13.3 |
8 to 15 | 7 | 46.7 |
16 to 23 | 3 | 20.0 |
24 to 31 | 2 | 13.3 |
32 to 39 | 1 | 6.7 |
40 or more | 0 | 0.0 |
Regular contact with patient(s) with an eating disorder | 12 | 80.0 |
Total |
15 | 100.0 |
Statistical analysis was carried out in IBM SPSS Statistics Version 24. Binomial tests were conducted to test whether a significant proportion of the entire sample thought that ON should have its own diagnosis or whether they thought that exercise related symptoms should be part of the diagnostic criteria. The null-hypothesis in these tests was that the yes/no-groups would be of the same size. Between-group differences were analyzed using Chi-squared tests, with Fisher’s exact test being applied where less than 80% of the expected frequencies exceeded 5. The professional category “other” was excluded in the comparative analyses. All tests were two-tailed, and significance level was set at an alpha of 0.05. To calculate the achieved power, a
A
Qualitative analysis was carried out in ATLAS.ti 7. Thematic analysis with open and closed coding was applied. A code book (Appendix
According to Dutch legislation, no ethical approval was required for this study (
Following the triangulated, mixed methods nature of the study, the results of qualitative and quantitative analysis are presented together and divided in four themes. First, the beliefs about relative prevalence of ON in Dutch population and level of recognition of ON in the daily practice will be presented. Next, opinions about how ON can be categorized within the framework of DMS will be discussed, followed by the views on a potential need for a separate diagnosis for ON. Finally, the impressions of the current proposed diagnostic criteria for ON will be presented.
In total, 95.6% reported believing that ON is, at least to some extent, prevalent in the general population in the Netherlands. None of the respondents thought that ON is highly prevalent in the Netherlands and seven participants (4.4%) stated believing that ON is not prevalent at all. There was no significant association between opinion about prevalence and professional group (
The results of the questionnaire were supported by the interview data. Participants generally reported believing that ON is prevalent in the general population and recognized the condition in their daily practice as well as in their personal lives. They mentioned how healthy eating and caring about what one eats is becoming more and more prominent, with individuals looking not only to nutritional value, but also to often overlooked information about the quality of the food e.g., content of e-numbers (codes for substances that are permitted to be used as food additives) etc., (
Regarding categorization, the majority of questionnaire respondents reported believing that ON falls under the Eating and Feeding Disorders category in the DSM, followed by the category Obsessive-Compulsive and Related Disorders. Approximately a quarter said that they believed it could fit within Anxiety Disorders. Additional categories mentioned were Autism Spectrum Disorders and Personality Disorders (
Questionnaire frequencies.
1 | 7 | 4.4 |
2 | 84 | 52.5 |
3 | 63 | 39.4 |
4 | 6 | 3.8 |
5 | 0 | 0.0 |
124 | 77.5 | |
Anorexia Nervosa | 10 | 29.4 |
Bulimia Nervosa | 2 | 5.9 |
ARFID | 18 | 52.9 |
OCD | 18 | 52.9 |
Generalized Anxiety Disorder | 4 | 11.8 |
Other | 4 | 11.8 |
Eating and Feeding Disorders | 118 | 74.2 |
Obsessive-Compulsive and Related Disorders | 89 | 56 |
Anxiety Disorders | 38 | 23.9 |
Other | 8 | 5.0 |
In favor of including exercise related symptoms in diagnostic criteria | 110 | 68.8 |
Total |
160 | 100.0 |
Although interview participants reported believing that ON fits within several categories of the DSM, their responses were consistent with the questionnaire participants’, in that a clear majority (
As shown in
Statistical testing.
Separate diagnosis for ON | Binomial test | <0.001∗ |
Profession group x Separate diagnosis for ON | χ2 test | = 0.003∗ |
Profession x Separate diagnosis for ON | Fisher’s exact test | = 0.003∗ |
Level of education x Separate diagnosis for ON | Fisher’s exact test | = 0.117 |
Profession group x Diagnosis ON fits within | χ2 test | = 0.112 |
Profession x Diagnosis ON fits within | χ2 test | = 0.432 |
Level of education x Diagnosis ON fits within | χ2 test | = 0.067 |
Profession group x Category ON fits within | χ2 test | = 0.985 |
Profession x Category ON fits within | χ2 test | <0.001∗ |
Level of education x Category ON fits within | χ2 test | = 0.267 |
Exercise related symptoms in diagnostic criteria | Binomial test | <0.001∗ |
Profession group x Exercise related symptoms in diagnostic criteria | χ2 test | = 0.971 |
Profession x Exercise related symptoms in diagnostic criteria | Fisher’s exact test | = 0.373 |
Level of education x Exercise related symptoms in diagnostic criteria | Fisher’s exact test | = 0.740 |
Regarding giving ON its own diagnosis, the interview participants were uncertain. Approximately half of the participants (
Participant quotes.
Reference | Quote | Reference | Quote |
---|---|---|---|
Expatriate psychologist: |
Expatriate psychologist: |
||
Expatriate psychologist: |
Expatriate psychologist: [...] |
||
Dutch psychologist: |
Dutch psychologist: |
||
Expatriate psychologist: |
Expatriate psychologist: |
||
Expatriate psychologist: |
Expatriate psychologist: [...] o |
||
Dutch psychologist: |
Dutch psychologist: |
||
Dutch psychologist: |
Dutch psychologist: |
||
Dutch psychiatrist: |
Expatriate psychologist: |
||
Dutch psychologist: |
Expatriate psychologist: |
||
Expatriate psychologist: |
Dutch psychologist: | ||
When discussing a hypothetical case, further opinions about separate diagnosis for ON were expressed. Despite only seven participants being outspokenly in favor of a separate diagnosis for ON, eleven of the fifteen participants reported that they would use the new diagnosis rather than the currently available ones for the hypothetical case, had it been an option. Only four participants would continue with the currently available ones, and two of these reported that they would do so because it is “safer” – because they feel more secure with the current ones due to them being more established (
Regardless of favoring or opposing a separate diagnosis, interview participants reported several advantages and disadvantages of introduction of ON as a diagnosable disorder.
Two large themes in prospective advantages of ON diagnosis arose: improvements for patients, and improvements for health professionals. The participants reported that a separate diagnosis would result in more research on the condition, more programs to learn about it and specialize within it, and more facilities to treat this specific condition; thereby leading to better treatment (
Potential improvements for health professionals included helping them to talk about the condition with their clients and knowing which direction to look in, as well as the creation of guidelines to help practitioners to treat this patient group. Keeping the DSM up-to-date with societal changes was also mentioned (
Regarding disadvantages of a separate diagnosis, no differentiation between disadvantages for patients and for health professionals was clear. The main disadvantage reported in the interviews was that there are already too many diagnoses and potential of overdiagnosing. It was also mentioned that, for the DSM-5, the idea was indeed to slim it down, and reduce the number of diagnoses (
Among different aspects of the diagnostic criteria, exercise related symptoms were specifically covered in the questionnaire. A majority of the participants reported thinking that these exercise related symptoms should be part of the diagnostic criteria for ON (
Although some parts of the diagnostic criteria were considered suitable by the interview participants, most agreed that the criteria needed to be refined. The main refinement mentioned was to make the criteria clearer and more specific. A concrete suggestion to improve this was to have more examples (
Although some previous researchers have deliberated on a potential link between ON and other disorders, no empirical studies have been conducted to investigate the opinions of clinically active health professionals on the matter, making our study the first of its kind. This is an important step to take in the early stages of research on the condition, and it was an attempt by the co-authors to acquire some clarity and direction on the topic. It was clear that participants in this study recognized ON, reported that they believed it to be prevalent in the Netherlands and reported having met clients who fulfilled the proposed diagnostic criteria. Furthermore, the majority of participants agreed that the condition should have its own diagnosis, placed in the DSM category Eating and Feeding Disorders.
Questionnaire results revealed that PHP were more in favor of a separate diagnosis than were MHP. A possible explanation for this may be the different clientele and focus of the profession groups, as PHP work more with the body and MHP more with the mind. Existing literature suggests that individuals working with the body professionally may be predisposed to ON, as prevalence among dietitians has been estimated to 41.9% (
An important question in this discussion is to what extent a diagnosis will be beneficial for the patient. On the one hand, patients will benefit from the recognition of ON as a distinct disorder, as it may lead to improved quality of treatment as well as potentially improving access to treatment should insurance companies decide to view it as a condition for which reimbursement can be provided. On the other hand, there is always a risk of stigmatization of individuals with a distinct condition, calling for caution in regard to the decision. Regardless of the potential advantages or disadvantages of a diagnosis, one may argue that a condition that is pathological and differs from currently available diagnoses should be represented in the DSM. The diagnoses of the DSM should be representative of reality; thus, if it is occurring in real life as a condition, it should have a place in the DSM.
Regardless of decisions regarding diagnosis, the level of impairment caused by the condition needs to be central in the treatment of the condition. With this in mind, it is essential to talk to patients and to know what they experience. Patient experiences are central in formulating good and valid diagnostic criteria, however the diagnostic criteria are needed to identify study participants. This discrepancy may pose a problem with identification of study participants.
Besides determining whether ON should have its own diagnosis at all, there is also the question of where this potential new diagnosis should be documented. In this study we focused on the DSM, however, many respondents expressed mixed opinions regarding this manual and mentioned several shortcomings. As diagnoses are fluid and constantly changing, the DSM is regularly updated. Although it is positive that the manual attempts to stay up to date with current situations, the processes regarding the development of the DSM and whether the updates fulfill their purposes have previously been criticized (
The importance of conducting a study like this one lies largely in that it raises awareness in the research community as well as among clinicians. It gives insight into the current state of knowledge and beliefs among practicing health professionals, while also letting them know that their experiences and opinions are valued. The mixed-methods design allowed for identification of a broad range of opinions of some health professionals in the Netherlands on whether, and how, ON should be classified, as well as their reasoning behind the opinions.
A large limitation to this study is the representativeness of the sample and thus generalizability of the results. A total N of 1165 clinicians were directly contacted and invited to participate in the study, however, only 13.7% of these decided to do so. With such a small proportion participating, the representativeness of the sample may be jeopardized, as it may differ systematically from the Dutch clinicians in these professions. It is also quite possible that only individuals with a specific interest in this topic participated, which may have influenced the results of this study further. Additionally, as the inclusion criteria were broad, the participants’ pool was a rather diverse yet small group. Although this means that the sample is not representative of all different kinds of health professionals in the Netherlands, and less so globally, it does protect against “tunnel vision” in these early stages of exploring the condition. As it is not yet known how ON should be categorized, and we do not even know whether it indeed is an eating disorder, only including participants with extensive experience with eating disorders may lead to biased results. For generalizability and to assure representativeness, this study would need to be replicated. After having conducted this study it is clear that this design is feasible, therefore, the study also serves as a feasibility study.
Additional limitations to this study were the use of external assistance with coding of the interviews as well as the validity of the questionnaire. Due to language barriers, external assistance was necessary to code the interviews, possibly resulting in nuances being lost in the analysis. To minimize this, a close and regular contact was held between FR and the coding assistant during this process, ensuring that no questions were left unanswered. Regarding the validity of the questionnaire, no specific tests were conducted for this. Pre-designed testing of construct validity could have enhanced the quality on this matter.
Mental disorders, such as AN and OCD, have been proven to result in tremendous suffering for the individual, as well as an increased risk of suicide. ON seems to have many similarities with these disorders, while also possibly reaching prevalence numbers several times higher than those of AN (
In conclusion, a heterogeneous sample of health professionals in the Netherlands seem to be of the opinion that ON should have its own diagnosis and that it should be placed in the DSM category Eating and Feeding Disorders. The results of this study suggest that, despite some possible disadvantages, giving ON its own diagnosis might be helpful to both patients and health professionals by improving treatment and facilitating the therapy process. However, as the sample cannot be considered to be representative, more research is needed to allow for further generalization.
The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.
FR participated in all parts of the research, conceiving the idea, design and conceptualization of the study, collecting and analyzing the data and writing a first draft of the manuscript. ES participated in conceiving the idea and supervised and guided all parts of the research as well as participated in preparing the manuscript for publication. ZB participated in the design of the study and collecting the data. TC participated in the data analysis and in preparing the manuscript for publication. All authors discussed the results and contributed to the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors gratefully acknowledge Alice Geurtsen for her help in forming the methods of this study and the collection of the data, as well as Isabelle Saberi-Far for her assistance in the coding of the Dutch interview transcripts.
The Supplementary Material for this article can be found online at: