Dietary modifications, nutritional supplements and alternative medicine in paediatric patients with inflammatory bowel disease


Prof. C. A. Edwards, Human Nutrition Section, Division of Developmental Medicine, University of Glasgow, Glasgow G3 8SJ, UK.


Background  Data on use of complementary and alternative medicine in children with inflammatory bowel disease are scarce.

Aim To assess prevalence, predictors and parental attitude to the use of complementary and alternative medicine in a paediatric population with inflammatory bowel disease in the west of Scotland.

Methods  Questionnaire survey encompassing alternative therapists, nutritional supplements and dietary modifications used in the management of inflammatory bowel disease was used. Demographics, disease and treatment data were also collected.

Results  86 guardians completed the survey. 61% had used at least one type and 37% were using complementary and alternative medicine recently. The most common types were probiotics (44%) and dairy-free diet (28%). Higher parental education and young parental age were predictors of complementary and alternative medicine use. An increased number of oral steroid courses and prior use of complementary and alternative medicine for other health reasons were associated with use of alternative therapists. Personal recommendation and to complement conventional medicine were the main reasons for using complementary and alternative medicine. 48% judged complementary and alternative medicine at least partially effective. 86% agreed that doctors should be supportive of use. 89% would give complementary and alternative medicine to their children.

Conclusions  Use of complementary and alternative medicine was higher in children with IBD than in adults with the same disease. The gastrointestinal nature of the disease could explain the high use of nutritional supplements and dietary modifications in this survey.


Inflammatory bowel disease (IBD) is a chronic debilitating disease with a continuous remitting and relapsing course. The disease itself presents at any age and 15–20% of patients are diagnosed during childhood.1 The aetiology of the disease remains unknown and the most effective medications commonly have side effects that limit patients’ acceptance and long-term compliance. In children, the additional issues of growth, maturity and image make disease management more complicated and troublesome for health professionals, the patients themselves and their guardians.2 Although most patients rely on conventional medicine, some may combine or replace this with unconventional-alternative remedies for the management of their condition.3, 4

Complementary and alternative medicine (CAM) is defined by the National Centre of Complementary and Alternative Medicine, as a group of diverse medical and healthcare systems, practices and products not presently considered part of conventional medicine.

Complementary and alternative medicine use by the general population is widespread around the world5 and increasing over time.6–8 Likewise, CAM use is reported as high in chronic illness and gastrointestinal patients.3, 6–15 In particular, the prevalence of CAM use in IBD patients has been reported to be higher than in healthy people11 and although early studies estimated that 6% of patients use CAM,14 recent surveys reported a sharp increase to 72%.9 However, no definite figures can be drawn. The use of CAM varies widely between studies and this could reflect differences in study design, responding populations,4, 16 cultural specificities or preferences,17 but most of all different definitions of CAM.5 There is no clear definition of which specific methods and remedies comprise CAM and which do not. Some studies included therapies-like prayer17, 18 and exercise in the definition and others excluded use of multivitamins purchased over the counter.19 Therapies with good evidence of clinical efficacy, but which are not being prescribed as conventional medicine, like probiotics, were included in only a few studies.10, 20 In fact, with more scientific evidence accumulating around the efficacy of such products21, 22 and their commercialization, it is anticipated that more patients will use them in the future. New branches of established CAM disciplines are continually being developed, taught in medical schools, used in hospital or in primary health care,8, 23, 24 reimbursed by insurance companies, and integrated into conventional medicine.25

Although there have been some studies on the use of CAM in adult IBD patients4, 17, 19, 23 the prevalence of CAM in paediatric populations is not well studied. Differences in disease management regimes between adult and paediatric patients,2 as well as the issues of growth, maturity and image make the extrapolation of the results from adult studies difficult. Indeed, the use of CAM by paediatric patients reflects the choice of their guardians and not the patients’, and rather how the former perceive and cope with their child’s disease.

Most of paediatric surveys have been conducted in the US20 and Australia10 with only one multinational study which included only a small sample of UK patients from a single hospital in London, UK.26 Disease management protocols differ between countries27 and as the use of CAM in adults was associated with the type and use of specific medications,4, 28 different patterns of CAM use should be expected worldwide.

The main purpose of the present study was the evaluation of current and past CAM use by a paediatric IBD population in the west of Scotland, including types of CAM with good records of clinical efficacy, the reasons of their choice and parental attitudes on CAM use. A secondary outcome was to report any factors associated with CAM use including new predictors-like growth and anthropometry and serial postdiagnostic laboratory markers of disease activity.

Materials and methods

All paediatric (<18 years old) IBD patients, along with their guardians, who attended their follow-up clinical appointments at Yorkhill Royal Hospital of Sick Children, Glasgow were eligible to participate. Yorkhill Royal Hospital for Sick Children is the biggest tertiary paediatric IBD referral centre in Scotland, serving approximately half of the Scottish paediatric population and is predominately in charge of the management of almost all children with IBD in the greater region of the west of Scotland. At the time of the study, there was no private paediatric clinic. This allowed the study to include nearly all paediatric patients diagnosed with IBD and residing in the west of Scotland.

Given that each patient visits the IBD out-patients clinic at least once a year for medical review (independent of disease activity), it was anticipated that all IBD patients, and their guardians, could be approached within a year. This survey asked carers to document CAM use 3 months postdiagnosis and we excluded new patients (<3 months). Although the time from symptoms onset to diagnosis is quite variable and patients may have had the opportunity to use CAM before diagnosis, we wanted to make sure that all patients had significant time to try CAM after diagnosis of IBD. Patients with other severe concomitant chronic illness were also excluded.

The study protocol was approved by the Local Research Ethics Committee and the Research and Development Office at Yorkhill Hospitals.


An independent clinical researcher explained the research to the patients and their guardians. An information leaflet was given to the guardians and an age-adapted version to the patients themselves. If both agreed to participate, a blank questionnaire was provided with a postpaid, preaddressed envelope and the method of completion was demonstrated on a template questionnaire by the researcher. Participants could ask questions on clarity. The guardian filled in the questionnaire at home and returned it by post. A second questionnaire was posted to all participants that had not responded within 15 days.


The instrument consisted of 40 questions and took approximately 15 min to complete. A preliminary questionnaire was compiled using two existing questionnaires.10, 20 The questionnaires were then modified with the inclusion and exclusion of questions to reflect the use, and types of CAM commonly used in the UK. CAM was defined as those unconventional remedies and treatments that are not normally taught in the British medical schools, as established approaches to IBD management, are not reimbursed by the National Health Service nor recommended by our medical staff. Exercise and prayers were not included.

The face validity of the questionnaire was checked by several health professionals. Moreover, the first 10 participants were asked anonymously to comment on questionnaire clarity and ease of use. Appropriate amendments were made to the questions and the structure of the instrument.

The first section of the questionnaire described the patient’s socio-demographic and disease characteristics. The following section focused on the use of CAM and listed four groups of therapies, namely Dietary modifications, Herbals, Nutritional Supplements and Alternative Therapists. The guardians had to answer whether the child used, or had recently used any of these especially for the management of IBD.

In total, 35 different types of CAM were described and there was an open question at the end of each group for any CAM not on the list. Although multivitamins and dietary modifications were included, it was made clear to the participants to report only those that were purchased over-the-counter or not recommended by their hospital medical staff.

The next section, applied only to those patients who had used CAM. Participants were asked to report sources of information on CAM, reasons they had chosen these, their effectiveness, cost and doctor’s awareness of their CAM use. The last section, directed to all the respondents, consisted of five questions addressing parental attitudes about CAM, and the use of CAM by other members of their family or the patients themselves for other health reason.

To ensure anonymity, confidentiality and undistorted responses the questionnaires were not numbered and parents asked not to add any identifiers. An invisible number was located on each questionnaire to identify nonrespondents for subsequent mailing.

On a supplementary form, detached from the main questionnaire but with the same invisible number, the researchers recorded information on disease characteristics, postdiagnostic laboratory data, conventional treatments, growth and anthropometry. This form was reattached to the main questionnaire when the latter was returned to the researchers by the participants.


The prevalence of CAM use was estimated as the total number of patients reporting CAM use of any type divided by the total number of respondents. Categorical variables were presented as frequencies and mean values with standard deviations and medians with interquartile range for parametric and nonparametric variables, respectively.

Binary and nominal logistic regression analyses were performed to determine potential predictors of CAM use. Candidate predictors of CAM use were suggested a priori. These included age, sex, disease type, phenotype and duration, total number and types of medication since diagnosis, serial postdiagnostic laboratory data on C-reactive protein, haemoglobin, albumin, erythrocyte sedimentation rate, IBD operation, body mass index (BMI), height and height velocity z-score at recruitment, 6 and 12 months before, parental use of CAM, CAM use for other reason, social deprivation score (Scottish Index of Multiple Deprivation;, parental education and age. For purposes of analysis, patients were grouped into ‘Self-prescribed CAM Users’ (including dietary modifications, nutritional and herbal supplement users) and ‘CAM Therapist Users’ (those who consulted an alternative therapist) as proposed by Harris and Rees5 to permit comparison with other studies.18 The strength of association between use of CAM and associated factors was measured by odds ratio and associated 95% CI.

The process of selection and inclusion of variables in the multivariate model was based on univariate logistic regression analysis. Associated predictors with P-values <0.1 were entered in a multivariate model and their independent association was tested using backward stepwise regression. The final model included only those predictors with P-value <0.05 as independent predictors of CAM use for the binary analysis, whereas for the nominal regression analysis the final model considered only predictors with P-value <0.05 for at least one of the comparisons. Statistical analysis was performed using minitab 13 for Windows.



Approximately 480 000 children are serviced by the Yorkhill hospitals, and 159 patients visited the IBD out-patient clinic between June 2005 and July 2006 and this covers the IBD population serviced by the IBD clinic. Twenty-four patients were not eligible to participate. From 135 eligible patients 104 were approached (77% of the eligible population) along with their guardians and all consented to participate. Because of logistic constraints, it was not possible to approach all patients and 31 who visited the clinic at the time of the study were not recruited. Eighty-six including 40 female patients completed and returned the survey (83% response) and all were used in the analysis. All excepting one patient were of Caucasian origin.

Respondents demographic characteristics

The median age of patients at recruitment was 12.7 (4.8–17.5) years. There was almost equal distribution of the family social deprivation score (Scottish Index of Multiple Deprivation; Table 1). The majority of the responding parents were more than 31 years old and 56% had higher education (college and university combined; Table 1).

Table 1.   Demographics, disease characteristics and post diagnostic medication of participants in a Scottish Survey on the use of CAM by children with IBD
Disease locationPer cent of patientsHospitalization (last 2 years)Per cent of patients
  1. * Scottish Index of Multiple Deprivation 2004; based on six individual domains of Current Income, Employment, Housing, Health, Education, Skills and Training and Geographic Access to Services and Telecommunications;; † Responding parent.

  2. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine; GI, gastrointestine.

Lower GI57040
Upper and lower GI38124
Upper GI5217
Total postdiagnostic medication number Total postdiagnostic medication type 
614Exclusive EN60
710Caloric supplements43
Gut resection12Methotrexate18
Social deprivation* (quintiles) Parental education† 
120Standard grade (year 11)33
218Highers (year 12)11
316Further education – college24
Parental age (years)   

Patients’ disease characteristics

Crohn’s disease patients accounted for two-thirds of participants, whereas 26% had ulcerative colitis (UC) and 9% were classified as indeterminate colitis according to standard endoscopy, histology and radiology criteria. The median age at diagnosis was 10.3 years (range: 1.9–14.5) with median disease duration of 2.4 years (0.4–7) at the time of recruitment. Majority of patients had been treated with 5-aminosalicylates, azathioprine, steroids and exclusive enteral nutrition (Table 1). 12% of patients had had a major operation for IBD that included resection of part of their gut.

Patients anthropometry and growth data

The mean BMI z-score of the respondents was 0.2 SD with 2% classified underweight (BMI z-score: <2 SD) and 4% obese (BMI z-score: >2 SD). The mean height z-score was −0.4 s.d.s with 7% presenting values <2 SD suggesting possible growth retardation. Their median height velocity at 6 and 12 months was 0.9 and 0.8 SD, respectively.

Use of CAM

Fifty-two of the 86 respondents (61%) reported prior use of CAM for the management of IBD and 37% were using some form of CAM on recruitment. A quarter of the respondents (21 of 86) had consulted at least one alternative therapist, whereas only three patients were using that kind of CAM on recruitment (Table 2).

Table 2.   Past and recent users of CAM in a Scottish Survey of children with IBD grouped into ‘Self-prescribed CAM Users’ and ‘CAM Therapist Users’
Type of CAM userPer cent ever used (n = 86)Per cent recently used (n = 86)
  1. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine.

‘Self-Prescribed CAM Users’3734
‘CAM Therapist Users’243
Overall CAM use6137

Probiotics, dairy-free diet, omega-3/fish oils and aloe were the most common therapies, followed by gluten-free diet, homeopathy, massage and over-the-counter multivitamins and megavitamins (Table 3). The median and interquartile range of total number of CAM therapies ever used was 3 (2–6) CAM and 2 (1–2) in the recent users.

Table 3.   Use of dietary modifications, nutritional supplements, herbals and alternative therapists in a Scottish Survey on the use of CAM by children with IBD
 Per cent ever used (n = 86)Per cent recently used (n = 86)
  1. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine.

Dietary modifications
 Dairy-free diet2814
 Gluten-free diet152
 Low residue diet93
 Low sugar diet50
 Vegetarian diet50
 Other diets61
 Evening primrose80
 Other herbals231
Nutritional supplements
 Fish/omega-3 oils278
 ‘Active’ dairy products92
 Other nutritional supplements133
Alternative therapists
 Other alternative therapists160

Seventy-three per cent (38 of 52) of the CAM users answered the relevant section on CAM use. Of 14 respondents that skipped the section only one was ‘CAM Therapist User’. 22 revealed the use of CAM to their doctor. 50% reported that their doctor reacted positively, 37% neutrally and 13% negatively to the use of CAM. Most of the participants who did not report the use of CAM to their doctors, thought that it was not important for their doctor to know or they forgot to do so.

Parents reported that they received information about CAM from personal recommendations, magazines, newspapers and the Internet (Table 4).

Table 4.   Reasons of use and sources of information on CAM in a Scottish Survey on the use of CAM by children with IBD
  1. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine.

Reasons of CAM use (%)
 Complement conventional treatment45
 Personal experience42
 CAM is natural and harmless32
 Frustration with medication side effects 29
 Dissatisfaction with medication13
 Other reason10
 CAM is more effective3
 Lack of confidence in medication3
Sources of information (%)
 Read about it47
 Recommended by friends47
 Internet search26
 Recommended by health professional18
 Other reason18

Reasons given for using CAM included an attempt to complement conventional treatment, a personal experience with CAM use, frustration with conventional medication side effects and the belief that CAM is natural and harmless (Table 4).

Approximately equal numbers (16 and 17, respectively) of CAM users judged CAM use effective and not effective. One patient reported deterioration of the disease condition and one mentioned side effects. More ‘CAM Therapist Users’ (12 of 16) labelled CAM effective compared to ‘Self-prescribed CAM Users’ (P = 0.022). Fifteen of 38 users found CAM expensive although most of these (10 of 15) had consulted an alternative therapist.

Twenty-seven per cent of all respondents had discussed CAM with their doctors and most were ‘CAM Therapist Users’ (P = 0.003). 16% of the children had used a type of CAM for other reasons, whereas 32% of the patients’ relatives had used CAM.

Most of the parents were positive about the use of CAM, but the majority of them (40 of 53) had already used CAM for their child’s IBD (Table 5). 89% of the respondents would give their child CAM, if they felt it to be useful although 50 of 73 were already CAM users. In addition, 52% of the respondents were not wary of using CAM, 57% of them would be happy for their child to use any type of CAM and 86% agreed with the statement that ‘doctors should be supportive of people using CAM’. On the other hand, most of the respondents had no opinion about CAM safety and agreed with the statement that ‘not enough is known about CAM’ (Table 5).

Table 5.   Attitude on CAM use by parents of children with IBD in a Scottish Survey on the use of CAM
 Disagree (%)No opinion (%)Agree (%)
  1. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine.

Not a person that would use CAM741511
Doctors should be supportive of CAM41086
Wary of using CAM52444
CAM is very safe214930
Not enough is known about CAM112069
Would be happy my child to use any CAM251857
Would give CAM only if doctor says so 391249
Would give CAM if I felt to be helpful7489

Predictors of CAM use

Parents under 46 years and high parental educational level were the strongest independent predictors that distinguished CAM users in the multivariate analysis (Table 6). Use of CAM for other health reason, and an increased number of oral steroid courses were additional variables that differentiated ‘CAM Therapist Users’ from non-users although a positive association between disease duration and use of ‘CAM Therapist Users’ revealed in the univariate analysis was lost in the multivariate model.

Table 6.   Predictors of CAM use by Scottish children with IBD based on univariate and multivariate logistic analysis with P-values, 95% CI and odds ratio
 All users‘CAM Therapist Users’‘Self-prescribed Users’
  1. IBD, inflammatory bowel disease; CAM, complementary and alternative medicine.

  2. * Age groups 18–30 and 31–45 combined; † Combined into two groups (standard grade and highers; college and university).

Univariate analysis [P-value; OR (95% CI)]
 Diagnosis age0.127; 0.9 (0.7–1.0)0.051; 0.8 (0.7–1.0)0.388; 0.9 (0.8–1.1)
 Disease duration0.517; 1.1 (0.9–1.4)0.048; 1.3 (1.0–1.8)0.541; 0.9 (0.7–1.2)
 Parental age*0.007; 4.0 (1.5–11.3)0.047; 4.2 (1.0–17.1)0.022; 4.0 (1.2–12.9)
 Parental education†0.001; 5.8 (2.0–16.2)0.001; 12.7 (2.9–55.9)0.017; 3.9 (1.3–11.9)
 CAM for other reason0.155; 2.7 (0.7–10.5)0.028; 5.4 (1.2–24.2)0.659; 1.4 (0.3–7.0)
 CAM use by family0.360; 1.6 (0.6–4.1)0.061; 3.1 (0.9–10.0)0.951; 1.0 (0.3–2.9)
 Steroid use0.065; 2.3 (0.9–5.8)0.144; 2.5 (0.7–8.5)0.116; 2.3 (0.8–6.3)
 Number of steroids courses0.588; 1.1 (0.8–1.5)0.074; 1.4 (1.0–2.0)0.519; 0.9 (0.6–1.3)
Multivariate analysis [P-value; OR (95% CI)]
 Parental age*0.005; 6.6 (1.8–24.3)0.012; 21.4 (2.0–232.6)0.021; 5.5 (1.3–23.8)
 Parental education†0.001; 9.1 (2.4–33.7)0.002; 18.9 (2.8–125.3)0.008; 7.0 (1.7–29.3)
 CAM for other reason0.128; 4.5 (0.6–30.8)0.009; 25.3 (2.3–283.7)0.434; 2.3 (0.3–18.8)
 Number of steroid courses0.436; 1.2 (0.7–2.0)0.021; 2.5 (1.1–5.4)0.908; 1.0 (0.5–1.7)


Knowledge of CAM use by children with IBD is imperative. Some of these treatments have severe adverse effects with reports in the literature of death, anaphylaxis, renal failure and malignancies,29, 30 whereas for those with good safety records in adults, the same doses used in adults, may not be safe in children with a smaller drug distribution volume. Moreover, possible unpredicted interactions with conventional medicine could reduce or delay the efficacy of the prescribed treatment with subsequent detrimental prognostic effects.31

Paediatric data on CAM use by IBD patients are scarce and to the best of our knowledge this study is the biggest paediatric study so far in the UK and Europe. Unlike previous surveys26 only patients younger than 18 years were recruited to ensure that our measured CAM use was not confounded by ‘adult children’.

In our study, the number of nonrespondents was small and any difference between respondents and nonrespondents should not change the overall findings significantly.

Our study design approached a geographically diverse sample, representative of half of the whole Scottish paediatric IBD population in contrast to other surveys with predominantly severely ill patients from a single tertiary hospital.26 This was achieved because there was no other public or private paediatric clinic and the hospital management protocol ensured that each patient visits the hospital at least once a year independent of disease activity. Indeed, our sample’s disease and demographic characteristics mirrored the results of a Scottish national study,32 and a prospective epidemiological survey in the UK.1 This study surveyed only those patients who were diagnosed in a conventional clinic. However, it is unlikely in the area surveyed that any child with IBD will never visit a National Health Service clinic and therefore very few who distrust conventional treatment will have been missed. Demonstration of the questionnaire and its completion at home ensured confidentiality and clarity of the questions. The anonymity of the questionnaire allowed undistorted responses and encouraged reliable answers and may have contributed to the high response rate.

Consistent with previous paediatric studies,10, 20, 26, 33 this study found that a large proportion of IBD children used CAM, to a similar extent as adults or higher extent than adults.19, 28 More than half of the children used some form of CAM and one of four visited alternative therapists. This is much higher than the 10% use of alternative therapists that has been reported for the general population in the UK National Omnibus Survey.34 Our estimated use of CAM is lower than the 72% of the survey population in an Australian study,10 but higher than the 41% found in a multinational study,26 22% and 51% in a Canadian and American paediatric survey, respectively.20, 33 It is possible that the increased use of CAM in our study reflects an increase over time, but this needs to be verified in a longitudinal study in the same population. A direct comparison between paediatric studies is hampered by differences in the study design, responding population and definitions of CAM.5

In the current survey, probiotics, dairy-free diet and fish oils were the most prevalent types of CAM followed by aloe and homeopathy. This is consistent with previous paediatric studies in the USA20 and Australia,10 which used similar questionnaires. In contrast, the multicentre survey by Heuschkel and his colleagues26 found that only 6% of their participants used probiotics and only 2% in their centre in London compared to the 44% found in this Scottish survey. The higher use of probiotics in the current study parallels the increase in the number of studies suggesting a beneficial effect in IBD management21, 22 and increased availability and advertising. The belief that dairy products can exacerbate the disease and that patients restrict their consumption has already been well documented.35, 36 The lower incidence of multivitamin or megavitamin use than in other studies26 could be attributed to our questionnaire being specific to vitamin supplements not prescribed by the medical team. It is noteworthy that three of the most commonly used CAM in this survey (probiotics, omega oils and aloe vera) already have controlled trials for adults published in the literature,21, 37, 38 are suggested through support groups and websites and therefore could have affected the choice of these treatments. Although it would be interesting to know whether the patients used these treatments to maintain remission or to suppress a relapse, the questionnaire was not designed to address this.

This study considered a wide range of possible predictors for CAM use. Demographically no relation to age, gender or disease diagnosis (UC vs. Crohn’s disease) has been found in both adult16, 17, 28, 39, 40 and paediatric surveys and our results are in accordance. Young parental age and higher education level were independently associated with CAM usage, a finding commonly reported in the general paediatric population.41 It may be that young and educated parents, are more open or seek information on innovative methods of disease management, or do not trust conventional medicine. It was reported42 previously that early onset CD is associated with affluence; however, in our study the spread of patients between deprivation scores based on SIMD was roughly equivalent and there was no skewing of patient numbers towards the affluent score. Social deprivation was not found to be a predictor of CAM use.

In contrast to other studies that assessed medication on recruitment, we considered also the total number and types of medication that each patient was on since diagnosis as a better depiction of disease activity history. No association between these indices and use of CAM was found confirming results from other paediatric10, 26 and adult studies.3, 18 Although an association between CAM use and steroid use was not found, in contrast to findings in adults by Hilsden,28 an independent association between the number of steroid courses and the use of alternative therapists was evident. This was previously reported in adults on high intravenous doses of steroids.4 No such association was found between repeated courses of exclusive enteral nutrition and CAM use. Although both repeated steroids and exclusive enteral nutrition courses reflect disease relapses, our findings suggest that the side effects of steroids and patients’ desire to terminate or avoid their use may be more important than disease activity. This was reported as a reason for using CAM in the current study and also by 63% of adults IBD patients in a German survey.4, 28

As enteral nutrition is the standard treatment for paediatric Crohn’s in the UK, caloric supplementation was not a predictor of CAM use in our study unlike the study by Markowitz in the USA20 where different management protocols were followed.27

Reading about CAM, friends’ recommendation and Internet search were the main sources of information about CAM similar to previous adult40 and paediatric surveys.10 The chronic nature of the disease and the lack of a definite cure are highlighted by the fact that most of the participants chose CAM in an attempt to complement the conventional medicine. Frustration with side effects of conventional medicine has been continually reported as the main reason why adult IBD patients used CAM3, 4, 10, 26, 28, 43 and this coincides with the observation that only 4% of patients with gastro-oesophageal reflux disease (for which effective medication with minimal side effects is available) tried to treat their condition with CAM.44 Moreover, recently, medication side effects have been documented as the most important parental concern for their child with IBD45 and could justify the need for alternative therapies with fewer or no side effects.

Most surveys in both adults and paediatric patients have shown a self-reported benefit of CAM use.10, 23, 26 In this study, half of the users considered CAM use effective, although any possible improvement was subjectively assessed by the patients themselves and is impossible to distinguish whether this was attributed to the concomitant conventional medication, to the CAM use per se or to a spontaneous improvement as a result of a documented placebo effect.46 Only two patients reported an adverse effect in contrast with adult surveys.40 This is not surprising as the majority of therapies used by participants in this study included nutritional remedies and exclusion diets generally characterized with good safety records.

Fewer patients in this study told their doctor they were using CAM compared with previous studies.26 This lack of disclosure is undesirable, as CAM may make an impact on conventional medicine or have side effects. The majority of our respondents wanted their doctor to be supportive of CAM. Adult IBD patients in Germany4 believed that a combination of methods using CAM and conventional medicine should be offered within one integrative clinical institution.

Most of the parents in the present study said ‘they would be happy for their child to use of CAM’ but agreed that ‘not much is known about CAM safety’. The chronic nature of the condition and the desire to ensure the best possible outcome for their child are reflected in the fact that almost 90% of the respondents’ stating that they would choose CAM, if they felt these were useful and some of them would do so even if medical advice was against it.

Use of CAM is high in paediatric IBD patients in Scotland and paradoxically parallels the development of new and more effective medications and modes of disease management. With increasing use of CAM, we urgently need more quality studies and clinical trials to shed light on their efficacy and better inform parents, patients and medical staff. So far, the results of clinical trials are too limited and most too flawed to suggest use of CAM in clinical practice.

Rather than actively discouraging or encouraging use, health professionals should encourage IBD patients to report CAM use and should be aware of all the alternative therapeutic choices that are available to their patients. Several hospitals worldwide have published policies and guidelines on CAM use and the American Academy of Paediatrics published a relevant article on counselling families who choose CAM for their child with chronic illness or disability.47


The authors would like to thank Dr Markowitz and Dr Day for their generosity in sharing the questionnaires that served as the starting point for our study, and Dr David Young for statistical help. Declaration of personal interests: C. A. Edwards has spoken at meetings and published a chapter in a book which was sponsored by Yakult. She has received funding from BBSRC, Foods Standards Agency, UK, EU commission, Broad Medical and acted as consultant to Mead Johnson. Declaration of funding interests: K. Gerasimidis was funded by a full scholarship from the Greek State Scholarship foundation and a grant from the Hellenic Association of Gastroenterology and Nutrition.