Dietary Interventions in Multiple Sclerosis: Development and Pilot-Testing of an Evidence Based Patient Education Program

Background Dietary factors have been discussed to influence risk or disease course of multiple sclerosis (MS). Specific diets are widely used among patients with MS. Objective To design and pilot-test an evidence based patient education program on dietary factors in MS. Methods We performed a systematic literature search on the effectiveness of dietary interventions in MS. A web-based survey among 337 patients with MS and 136 healthy controls assessed knowledge, dietary habits and information needs. An interactive group education program was developed and pilot-tested. Results Fifteen randomised-controlled trials (RCTs) were included in the systematic review. Quality of evidence was low and no clear benefit could be seen. Patients with MS significantly more often adhered to a `Mediterranean Diet`(29.7% versus 14.0%, p<0.001) compared to controls. 143 (42%) of the patients with MS had tried special MS diets. Important information needs addressed effectiveness of MS diets (44%) and relation between nutrition and MS (43%). A pilot test of our newly developed patient education program with 13 participants showed excellent comprehensibility and the MS-specific content was judged as very important. However, the poor evidence base for dietary approaches in MS was perceived disappointing. Conclusions Development and pilot-testing of an evidence-based patient education program on nutrition and MS is feasible. Patient satisfaction with the program suffers from the lack of evidence. Further research should focus on generating evidence for the potential influence of lifestyle habits (diet, physical activity) on MS disease course thus meeting the needs of patients with MS.


Results
Fifteen randomised-controlled trials (RCTs) were included in the systematic review.Quality of evidence was low and no clear benefit could be seen.Patients with MS significantly more often adhered to a 'Mediterranean Diet'(29.7%versus 14.0%, p<0.001) compared to controls.143 (42%) of the patients with MS had tried special MS diets.Important information needs addressed effectiveness of MS diets (44%) and relation between nutrition and MS (43%).A pilot test of our newly developed patient education program with 13 participants showed excellent comprehensibility and the MS-specific content was judged as very

Introduction
Multiple sclerosis (MS) is a demyelinating disease and the most common cause of neurological disability in young adults worldwide.[1,2] The etiology of multiple sclerosis-especially the pathophysiological mechanisms of neurodegeneration which cause irreversible disability-are still largely unknown.[1,3] Although by now more than a hundred genes are known to increase the risk of MS, all of them only contribute marginally.[4] This as well as the results of twin studies support the premise that, apart from genetic factors, environmental influences modify disease risk and progression, possibly through epigenetic changes which could up-or downregulate immune response and influence neural development.[1,5] Environmental factors that have been proposed to influence disease risk or progression are infection with Epstein-Barr virus, smoking, low Vitamin D status, and a variety of dietary factors such as high sodium and unsaturated fatty acid intake.[1,5,6] Recently, increased attention among researchers has been paid to salt restriction [7], ketogenic diet [8] and the relevance of gut microbiota.[9] However, high level evidence from randomized controlled trials on the influence of dietary factors on disease progression is scarce.Jagannath et al. (2010) [10] only reported on one small trial with 49 participants in their Cochrane Review on Vitamin D treatment in MS.Another Cochrane Review from 2012 included all randomised or controlled clinical trials on specific dietary interventions, diet plans or dietary supplementation except for vitamin D supplementation.[6] They concluded that polyunsaturated fatty acids do not seem to have major clinical effects on disease progression, but may be associated with a tendency in reduction of frequency of relapses over two years.However, data available were insufficient to assess real benefit or harm, because of poor trial quality.[6] Although evidence on the efficacy of dietary interventions is poor, patients with MS (PwMS) have a strong interest in dietary recommendations and many adhere to special diets.There is a large number of books and webpages which support multiple, sometimes conflicting approaches.[11] An internet search using the terms 'diet' and 'multiple sclerosis´carried out by Farinotti et al. in 2012 produced over 27 million links.[6] A German survey among 1573 PwMS showed that the lifetime use of dietary modification was 41%.[12] An Australian survey with 416 PwMS found that a high percentage of participants took supplements (63.2%) and/or followed special dietary recommendations.[13] As the nutritional knowledge of PwMS as well as their interest in understanding diet and nutrition are largely unknown, we aimed to better understand these factors.Being aware of the large gap between scientific evidence on the effectiveness of dietary interventions in MS and dietary behaviour of PwMS, we hypothesized that there is a need for an evidence based patient education program.Accordingly our aim was to design and pilot-test an evidence based patient education program on the influence of diet on MS.

Systematic literature search
We performed a systematic literature search in PubMed in September 2010 with an update in June 2015 on the topics MS and dietary interventions (full search terms see S1 Text).We defined inclusion criteria listed in Table 1.

Web-based survey
We developed a questionnaire and performed a web-based survey among PwMS and healthy controls.Visitors of the website of the German Multiple Sclerosis Society (DMSG) were invited to the survey in spring 2011.The link on the DMSG website directed participants to the webbased questionnaire.Informed consent was obtained via the first page of the questionnaire.In order to recruit participants for a control group, an additional call for participation was put on the DMSG website addressing friends and relatives of PwMS.
Survey questionnaire.The questionnaire consisted of four parts with a total of 36 items, 31 of which were newly developed by the research team.Most of the questions were directed to PwMS and control group participants but some could only be answered by PwMS.The questionnaire covered the following topics.
Part 1 (12 items): Knowledge and Attitudes towards Nutrition/Diet.Questions addressed beliefs about the influence of different lifestyle factors (e.g.stress, sports) on MS.Patients could attribute 0 to 100 points to the different factors, 0 meaning no influence, 100 meaning maximum influence.Additionally we assessed knowledge about MS diets, motives for adherence to a diet, self-perceived effects of dietary behaviours and we asked about the use of different sources of information.
Part 2 (17 items): Dietary Habits.Questions addressed intake of supplements and adherence to MS-specific diets.Eating habits in general were assessed based on a questionnaire from the `Genes and environment in MS (GEMS)`project.[15] Participants were asked to select a diet that best fit their usual dietary habits (see Table 2).
As the absolute number of responses was low, we combined the three types of vegetarian diets into the category `Vegetarian Diet´for data analysis.`Glycemic Index Diet´and Other Diets´were put together into the category `Other Forms of Diet´.We assessed the risk of bias of the included studies using the Cochrane risk of bias assessment tool.[14] doi:10.1371/journal.pone.0165246.t001 Part 3 (5 items): Physical Activity Assessment with the "Godin Leisure Time Exercise Questionnaire" [16,17] and one additional item.Data not shown.
Part 4 (2 items): Patients´Interests and Preferences regarding different topics for a patient education program on diet and multiple sclerosis.We presented six possible major themes and participants were asked to rank them from 1 = most important to 6 = less important.An open question gave patients the possibility to name additional topics of interest.
Additionally, we asked for socio-demographic data, type and course of disease, weight and smoking status.Mental health status was assessed with 2 questions about depression from Mohr et al., 2007 [18] and self-efficacy with 5 questions from Schwarzer et al., 1999 [19] (data not shown).

Development, pilot-testing and evaluation of the education program
Based on the results of the systematic literature search and the results of the survey, a groupeducation program was developed in accordance with the concept of evidence-based patient information [20,21] (see also S2 Text) and the Medical Research Council (MRC) framework on the development and evaluation of complex interventions.[22] Patients from the MS Day Hospital were invited to participate in pilot training sessions.The first session was conducted in May 2011 with 4 participants.After revision of the program according to the feedback of the first group, the second training session was conducted with 9 participants.
Directly after the education program participants were asked about the extent and quality of the information presented.Using visual analogue scales with a scale from 0 to 10 the following domains were assessed: novelty of information (0 = new, 10 = already known), comprehensibility (0 = comprehensible, 10 = incomprehensible), importance (0 = important, 10 = not important), extent of information (0 = too extensive, 10 = not sufficient) and impact of the information (0 = disappointing, 10 = encouraging).Two open questions asked for the personally most relevant information received and changes that should be made to the training.Additionally, participants were encouraged to express their opinion in an open discussion at the end of the training program.

Statistical analysis
This survey was conducted using the online software Enterprise Feedback Suite (EFS) Survey 8.0 (Globalpark) and the online platform www.unipark.info.Data were analysed using SPSS 22.0 for Windows.Based on own previous surveys and work of others on lifestyle factors in MS a sample size of at least 100 participants in PwMS and control group was considered sufficient to detect major differences.For dichotomous data we calculated absolute numbers and percentages and performed Pearson's chi-squared test to test for significance.For continuous data we calculated mean and standard deviation and used the t-test for independent samples to test for significance.Pearson's chi-squared test was calculated to assess the difference in usual dietary habits and supplement intake between PwMS and the control group.

Ethical considerations
Ethical approval was provided by the Ethics Committee of the University of Lübeck, Germany (document number .

Results of the systematic literature search
Screening process.Our search in June 2015 yielded 1048 hits after removal of duplicates.One investigator (KRL) reviewed citations on the basis of title and abstract information and excluded those clearly not meeting inclusion criteria.We retrieved the full text of the remaining studies which were subsequently reviewed independently by two investigators (KRL, SK) to assess whether inclusion criteria were met and to rate quality.All differences were settled by discussion.A flow chart of the screening process is shown in Fig 1 .For additional information see S1 Checklist.We included 9 trials modifying fatty acid intake and 6 trials on Vitamin D. The main characteristics and results of the trials are displayed in Tables 3 and 4.

Risk of bias of included studies. The risk of bias of included studies generally was substantial (Fig 2) with only two recent Vitamin D trials being judged as of low of risk of bias (Fig 3).
Most trials could not show significant effects of supplementation considering patient relevant outcomes like disease progression or relapse rate (see Tables 3 and 4).This is in line with the findings from previous systematic reviews.[6,10] Results of the web-based survey Socio-demographics.Between May and July 2011, 583 persons read the information sheet for the survey and 550 (94.3%) actually started the survey.473 (86%) participants completed the survey, 337 PwMS and 136 healthy controls.Socio-demographic characteristics were comparable between patients and controls except for the age, which was significantly lower in the control group (Table 5).
Adherence to MS diets or special dietary recommendations.42% of the patients in our survey indicated that they had already tried to follow an MS diet or had adhered to special dietary recommendations for PwMS.Of these 143 PwMS who followed dietary recommendations, 36 (25%) tried to modify fatty acid intake (less saturated fat, more unsaturated fatty acids), 30 (21%) ate no meat, 31 (22%) adhered to a vegetarian diet (some of them including fish) and 24 (17%) indicated that they tried to eat less or no meat and more fish.Several other diets that were occasionally mentioned included diets according to the principles of Swank [39], Jelinek [40], Adam [41], Evers [42], Kluge [43], a low-carb-diet, paleo diet, and diets without sugar or food additives.Patients who adhered to dietary recommendations (n = 143) reported on their goals from a list of possible objectives (Fig 5).
Intake of dietary supplements.Dietary supplements were very commonly used in our sample.About three quarters (76%) of PwMS as well as control group participants indicated, that they had used dietary supplements.In both groups a wide range of vitamins (multivitamins, C, D, E, folate, other vitamins of the Vitamin-B-group), minerals (calcium, magnesium, copper, iron) and other nutritional supplements (e.g.gingko, green tee extract, Indian psyllium seed) were used.However, there were relevant differences in the frequency of use of different supplements between PwMS and controls.PwMS used Vitamin D (PwMS 62%, controls 40%) and omega-3-fatty acids (PwMS 67%, controls 31%) significantly more often than controls, whereas controls used Vitamin C (PwMS 60%, controls 82%) significantly more often (p<0.001).Selenium was used more often by PwMS and controls tended to use zinc more often, but these differences were not statistically significant.
As adhering to a diet might be a challenge, we asked PwMS how they felt following dietary recommendations and if they had ever interrupted their diet.Interestingly, only about 25% of PwMS reported that they had stopped their diet and about 67% reported that they felt better (49%) or even clearly better (18%) when being on a diet.30.1% of PwMS felt unchanged.Those who stopped their diet (n = 36) mainly reported the following reasons: too restrictive, no effect, too much effort, too expensive (for details see S1 Table ).
Perceived influence of diet and other lifestyle factors on MS.Diet is clearly a factor many patients in our survey perceived to impact the development and disease course of MS.About one third believed that dietary factors influence the development of multiple sclerosis and about 60% believed that the course of disease can be influenced by dietary factors.Asked retrospectively, the importance diet/nutrition had for PwMS increased after they were diagnosed with MS.Before diagnosis about 54% of respondents considered diet/nutrition as important (13.1%) or rather important (40.7%),whereas after diagnosis this number increased to about 82% (important: 46.3%, rather important: 35.3%).
However, diet is only one and not the most important influencing factor on disease course in the perception of PwMS.Respondents perceived the influence of stress/emotions, physical activity and Disease Modifying Drugs (DMD) to be larger than the influence of diet (Fig 6).
Information seeking behaviour.Only 9.5% of PwMS in our sample never sought information about dietary recommendations for MS.In contrast 43% thoroughly looked for such advice and 47.5% sought some information.Major source of information was the internet (83%), followed by books (64.3%) and physicians (36.7%).Other patients (23.6%), hospitals/ rehabilitation centres (23%) and self-help organisations (12.1%) played a minor role in getting information on dietary advice.When asked about their knowledge about special MS diets, most patients had heard about the Evers diet (48.4%) [42], followed by the Fratzer diet (28.5%) [44], gluten-free diet (26.1%) and the Swank diet (20.8%).[39] About one third of PwMS (37.1%) had never heard of any of the special MS diets mentioned in the survey.
What information on diet and MS do patients perceive as most relevant.Participants were asked which topics would be of greatest interest to them for an education program on diet and MS.Unfortunately, the request to assign a hierarchy of importance to the presented themes was not understood by a large proportion of respondents (66.5%) leading to wrong answering patterns.Therefore results of this question can only be displayed for 113 PwMS as shown in Fig 7.  The most important topic to PwMS in our survey was to get information about"Current scientific evidence for the benefits of MS diets"(44.2% of PwMS assigned rank 1 or 2 to this topic) closely followed by the question if "there is a connection between MS and nutrition" (42.5%).However, the other topics were also of interest for at least a quarter of PwMS in our survey.

Patient education program-pilot study
Structure and content.The education program lasted 2 hours and was conducted by a medical student with an interest in nutrition sciences.It consisted of power point presentations on the content shown in Table 6.Within the program there were two short group discussions, one to work on understanding the quality of studies and one opportunity for sharing their experiences with MS diets, as well as a final group discussion.Pilot study results.The cohort consisted of 13 participants (including two accompanying persons) in two sessions (four and nine participants respectively) with a mean age of 38.5 years (±SD 12.3).Ten participants were female, three were male.The results of the program evaluation for PwMS are displayed in Table 7.
Median novelty of the information was 7.8 for the first part of the program and 5.5 for the second part.However, the wide range of scorings suggests great differences in patients´knowledge before training.Comprehensibility of the displayed information was rated as very good.
Considering the importance of the information delivered, information from the second part was judged more important compared to the information of the first part.Extent of the information presented was judged to be adequate.The first part of the program was neither disappointing nor encouraging for the participants.However, the second part about MS diets and evidence based information on the results of MS-studies was thought to be disappointing (impact of information, mean score 2.4).
At the end of each part of the education program participants were asked about the most relevant information they had received and possible changes that should be made.Information mentioned as valuable was related to understanding the quality of studies ("How to identify good studies"; " . . . to critically evaluate study results"), comprehending the influence of fatty acid intake on MS ("No effect of Omega3 /6 fatty acids"), and understanding the lack of knowledge of diet and MS ("Demystification of diets"; "No reliable results on diet and MS").In addition, participants expressed their disappointment that so little is known about the influence of diet on MS ("that I know nothing"; "all I know is not proven").Participants made several proposals for future education programs, which can be categorised as follows: Fewer information on studies ("less on studies"; "less theory"), more information on food and nutrition ("more information on food"; "fewer studies, more nutrition, the pros and cons of different foods"), more practical information ("more practical ideas and feasible proposals"), shorter duration ("maybe a little shorter"; "theoretical part a little shorter").Direct quotes of all comments and proposals are given in S2 Table.

Discussion
Our aim was to design and pilot-test an evidence based patient education program on dietary factors in MS based on the information gathered from a systematic literature review and the data gathered in a web-based survey on dietary habits, perceived effectiveness of dietary interventions and patients`information needs.
About one third of PwMS in our survey believed dietary factors to have an influence on the development of multiple sclerosis and about 60% believed that the course of disease can be influenced by dietary factors.When asked about other factors, the influence of stress/emotions, sports and disease modifying drugs on the course of disease was perceived to be bigger than the influence of diet.Motivation to adhere to special dietary recommendations seemed to follow a realistic assessment of the evidence.78.3% of PwMS adhering to a diet aimed at improving overall wellbeing and only 8.4% hoped to cure their disease.However, 73% of PwMS adhering to dietary recommendations in our survey aimed at slowing down disease progression.This is in line with findings from an Australian survey (n = 428) that reported that 70% of survey participants aimed at improving overall health and well-being, 60% to tackle fatigue and about 50% to alleviate general MS symptoms when taking supplements or adhering to diets.[13] Previous studies have shown that diet as well as complementary and alternative medicines (CAM)-which often consist of dietary supplements-is considered highly relevant by PwMS and is widely used for a variety of reasons.[11][12][13] Although evidence for specific dietary intervention is limited, 42% of PwMS in our survey indicated that they had tried to follow a MS diet or had adhered to special dietary recommendations, which is in line with the findings of an earlier German survey.[12] Attempts to modify fat intake, eat less or no meat and more fish were reported more frequently by PwMS in our survey than specialized MS diets.Accordingly, compared with healthy controls PwMS in our survey were significantly less likely to follow a "Mixed Diet" and significantly more often indicated, that they adhered to a `Mediterranean Diet`.
Intake of dietary supplements was common.In our survey, 76% of both PwMS and control group participants indicated that they had ever used dietary supplements.However, PwMS  and controls preferred different types of supplements.Vitamin D, omega-3-fatty-acids and selenium were used more frequently by PwMS whereas controls tended to use Vitamin C and zinc more often.This finding indicates that PwMS tend to take supplements which have been discussed as potentially beneficial for PwMS and are currently under investigation in MS trials, despite limited evidence for a substantial benefit of supplementing Vitamin D or PUFAs.In the Australian survey slightly lower numbers (63.2%) of overall use of supplements have been reported.[13] Data from the German National Nutrition Survey showed that in the general population 27.6% were current users of supplements.[45] Unfortunately, representative data about "ever use" of supplements in the German general population are not available.
The internet was the main source of information on diet and MS used by our survey participants followed by books.Interestingly, although recruitment was performed using the German MS self-help organisation's website, self-help organisations only played a minor role as information source.These findings differ from those of Leong et al. (2009) who stated that in South Australia conventional health care practitioners and friends and family were the most common sources of information on CAM products and dietary interventions.[13] Most of the trials included in the systematic review did not report relevant benefits of supplementation on relapse rate or disease progression which is in line with previous reviews by Farinotti et al. (2012) and Jagannath et al. (2010).[6,10] We found no controlled studies on comprehensive change of dietary patterns as a life style intervention.However, 143 (42%) of the 337 PwMS in our survey, who adhered to a diet or special dietary recommendations, made substantial changes to their eating pattern (e.g.modifying fatty acid intake, eating no meat, adhering to a plant-based, vegetarian diet or a Mediterranean Diet) and 67% of those reported that they felt better or even clearly better when being on a diet.Besides a possible therapeutic effect, the psychological effects of increasing patients' sense of control by adhering to dietary concepts need to be considered.
The absence of controlled studies for broad dietary approaches leaves patients and caregivers largely clueless and might also have added to the lack of satisfaction with the education program.Our results show that adequately performed studies examining the influence of nutrition on MS are urgently needed in order to support patients in their ability to make informed decisions.Ideally, adequately powered RCTs with comprehensive lifestyle changes as a multimodal intervention [22,46] and carefully selected outcome measures should be conducted.Taking into account the unpredictability and great variability of disease progression in MS, the conduct of inception cohort studies closely monitoring lifestyle habits (type of diet, level of physical activity) of PwMS over several years could also be considered to improve the amount of available evidence.
In addition to scientific evidence on the influence of diet on MS, patients also requested information on a variety of other topics in a nutritional education program.While our focus was on the methodological problems of dietary studies in MS, future attempts to develop a patient education program on diet and MS should take into account more general aspects of healthy diets.The survey had a number of limitations.As recruitment was web-based and not population-based, people with a specific interest in the topic might have been more likely to access the web-platform.Additional research should attempt to acquire population-based information.Also representativeness of the control group can be questioned as an unexpectedly high proportion of control group participants indicated to follow a vegetarian diet.Additionally, our questionnaire had some limitations.Most questions were self-developed and not rigorously tested before applying them.This caused misinterpretations of some questions by the participants.For an in depth assessment of dietary habits of PwMS standardized tools for nutritional assessment should be applied in the future.Regarding our evidence based patient education program, a group of 13 consumers is a very limited number to draw inferences.On the other hand this study was meant to be a pilot program to test feasibility and get first, direct feedback from patients.The results could be the starting point for developing a follow-up program which then should be evaluated in different settings and with a substantially higher number of participants.

Conclusions
The development and pilot-testing of an evidence-based patient education program on nutrition and MS has proven the feasibility of such a program.Patient satisfaction with the program suffered from the lack of evidence and the lack of studies on comprehensive dietary approaches.Further research should focus on generating evidence for the potential influence of lifestyle habits (diet, physical activity) on MS disease course thus meeting the needs of PwMS.Dietary counselling is considered highly relevant by patients.However, with the given evidence

Sharing of experiences with MS diets
Introduction to common MS diets (Evers,Fratzer,Swank [42,44,39] Randomized controlled trials studying diet and MS and their results

Final discussion
doi:10.1371/journal.pone.0165246.t006patient education needs to mainly focus on the many open questions in the area.Hence, an additional aim should be to enhance patients´ability to critically appraise the evidence.

Fig 2 .
Fig 2. Overall risk of bias across all included studies as judged by the reviewers.Note: 'Other sources of bias'are relevant in certain circumstances, relating mainly to particular trial designs (e.g.carry-over in cross-over trials and recruitment bias in cluster-randomized trials).[14].doi:10.1371/journal.pone.0165246.g002

Fig 3 .
Fig 3. Risk of bias of included studies as judged by the reviewers.Note: 'Other sources of bias'are relevant only in certain circumstances, relating mainly to particular trial designs (e.g.carry-over in cross-over trials and recruitment bias in cluster-randomized trials).[14].doi:10.1371/journal.pone.0165246.g003

Fig 6 .
Fig 6.Influence of different factors on disease course in the perception of PwMS.Patients could attribute 0 to 100 points to the different factors, 0 meaning no influence, 100 meaning maximum influence.n = 337 DMD = Disease Modifying Drugs.doi:10.1371/journal.pone.0165246.g006

Table 1 . Eligibility criteria used for literature search and screening.
PwMS = patients with MS, RCT = Randomized controlled trial, CCT = Controlled clinical trial.

Table 3 . Characteristics of included studies-trials modifying fatty acid intake by supplementation. Study Year and Country Design Intervention Outcome parameters Main results (as described by study authors)
Treatment A: 8 capsules of 0.6 ml oil, containing 360 mg linolenic acid and 3.42 g linoleic acid daily Control B: 8 capsules of 0.6 ml of oleic acid Treatment C: 11.5 g of linoleic acid/d as a spreadControl D: 4 g of oleic acid/d as a spread Disability (DSS); relapses (frequency, site, duration, severity)

Table 4 . Characteristics of included studies-trials modifying Vitamin D intake by supplementation. Study Year and Country Design Intervention Outcome parameters Main results (as described by study authors)
ARR = Annualized Relapse Rate; DMT = Disease Modifying Treatment; DSS = Disability Status Scale; EDSS = Expanded Disability Status Scale, IU = International Units; ns = not specified; MRI = Magnetic Resonance Imaging; MSFC = Multiple Sclerosis Functional Composite; PwMS = Patients with MS; RCT = Randomized Controlled Trial, RRMS = Relapsing Remitting Multiple Sclerosis.doi:10.1371/journal.pone.0165246.t004