Complementary and alternative medicine (CAM), particularly herbal therapy, is widely used by patients with inflammatory bowel disease (IBD) but controlled data are limited.
Complementary and alternative medicine (CAM), particularly herbal therapy, is widely used by patients with inflammatory bowel disease (IBD) but controlled data are limited.
To systematically review the literature on the efficacy of herbal therapy in the treatment of ulcerative colitis (UC) and Crohn's disease (CD).
Publications in English and non-English literatures (MEDLINE, EMBASE, EBM Reviews, AMED, Global Health) were searched from 1947 to 2013 for controlled clinical studies of herbal therapy in IBD. Outcome measures included response and remission rates.
Twenty-one randomised controlled trials (14 UC; 7 CD) including a total of 1484 subjects (mean age 41, 50% female) were analysed. In UC, aloe vera gel, Triticum aestivum (wheat grass juice), Andrographis paniculata extract (HMPL-004) and topical Xilei-san were superior to placebo in inducing remission or response, and curcumin was superior to placebo in maintaining remission; Boswellia serrata gum resin and Plantago ovata seeds were as effective as mesalazine, whereas Oenothera biennis (evening primrose oil) had similar relapse rates as omega-3 fatty acids in the treatment of UC. In CD, Artemisia absinthium (wormwood) and Tripterygium wilfordii were superior to placebo in inducing remission, and preventing clinical recurrence of post-operative CD respectively.
Randomised controlled trials of herbal therapy for the treatment of IBD show encouraging results but studies remain limited and heterogenous. Larger controlled studies with stricter endpoints and better-defined patient groups are required to obtain more conclusive results on the use of CAM therapies in IBD.
Crohn's disease (CD) and ulcerative colitis (UC) are chronic idiopathic inflammatory bowel disorders (IBD) in which patients often require lifelong medication. Poor adherence to medication has been an important barrier to successful management. Understanding patients' beliefs and concerns and addressing physical and perceptual barriers to adherence is crucial in improving adherence. The conventional treatment for IBD involves the use of corticosteroids, immunosuppressants and antitumour necrosis factor (TNF) antibodies. Some of these agents have been associated with the risks of infection and malignancy.[2, 3]
In recent years, complementary and alternative medicine (CAM) is increasingly being used by patients with IBD because of its perceived natural and healthy properties. Population-based and cohort studies have shown that the use of CAM is common among adult and paediatric IBD patients.[4-11] The prevalence of current or past CAM use in adult IBD populations from North America and Europe ranges from 21% to 60%. In a comparative study of Chinese and Caucasian patients, the overall use of CAM was similar in both groups and similar for CD and UC. Younger age, female gender, a higher education level, adverse drug reactions from IBD medication,[7, 8] extra-intestinal manifestations, perceived stress and prolonged and intensive courses of steroids have been associated with the use of CAM in IBD.
CAM products that have been evaluated in clinical studies for the treatment of IBD include herbal medicine, dietary supplementation (probiotics, prebiotics or fish oil), and mind body medicines such as acupuncture, moxibustion or hypnotherapy. Although research has explored many of these products, scientific evidence regarding their efficacy or safety has not been adequate, and the majority of studies have produced inconsistent results.
The aim of this systematic review is to evaluate the efficacy of herbal therapy in the treatment of IBD. We investigated the use of herbal therapy for both the induction and maintenance of disease remission in UC and CD.
Our search was divided into a few steps. First, two investigators independently identified eligible studies from 1947 to March 2013 in MEDLINE, EMBASE and EBM Reviews using structured headings and keywords of complementary, alternative, herbal, herbs, Chinese medicine, IBD, colitis, Crohn's and proctitis. As some herbs may have slightly different names under different studies we also searched for the alternative names of each herb in Wikipedia identified in the initial search. Subsequently, a more comprehensive search was performed in six search engines (MEDLINE, EMBASE, EBM Reviews, AMED, Global Health and International Pharmaceutical Abstracts) using the name(s) of each specific herb including Tripterygium wilfordii, Wormwood, Lei gong teng, Thunder god vine, Pistacialentiscus, Aloe, Angelica sinensis, Dong quai, Female ginseng, Triticumaestivum, Common wheat, Bread wheat, Boswellia serrata, Frankincense, Plantago, Oenotherabiennis, Oenothera, Common evening primrose, Evening star, Curcumin, Turmeric and Xilei-san in combination with IBD/colitis/Crohn's/proctitis. Such a search strategy was utilised to avoid missing relevant studies. In addition, reference lists of all obtained papers were searched. Manual searches were performed for abstracts published in major international conferences, including Digestive Disease Week, United European Gastroenterology Week and Asia Pacific Digestive Week in the past 10 years, and on the bibliographies of original articles or systematic reviews with the relevant information.
Randomised controlled studies in which human adult IBD subjects were treated with herbal therapy were identified. For this systematic review, studies on prebiotics, probiotics[15-17] fish oils/omega 3-fatty acids,[18, 19] and moxibustion/acupuncture  were excluded as these therapies have been extensively reviewed recently. Studies were included if they were placebo-controlled or controlled against a conventional treatment. Only abstracts with adequate information were included. Two independent investigators reviewed studies for eligibility, extracted the data and assessed study quality. Both English and non-English papers were included. Studies were divided into those that have assessed the use of herbal therapy for the induction of remission and maintenance of remission in UC, and the induction of remission and prevention of operative recurrence for CD.
Data were independently extracted into a standardised data extraction form by two independent reviewers (YTL and KKT). When discrepancies were found, a third investigator made the definitive decision for study eligibility or data extraction (SCN). All data were double-keyed into a predesigned database. Data collection included the author's name, year of publication, language of publication, study location, number of subjects, mean age of subjects, proportion of male subjects, types of herbal therapy, treatment duration, follow-up period, remission and response rates and adverse effects.
Our searches identified 1094 abstracts of which 21 clinical studies met inclusion criteria (Figure 1). There were 14 randomised controlled studies of herbal therapy in UC and seven studies in CD between 1947 and March 2013, involving a total of 1484 subjects. The mean age of subjects was 41 years old. The ratio of female to male was 1:1. Eighteen studies were published in English language, three in Chinese language and one in German.
Eleven randomised studies assessed the use of herbal therapy in the induction of remission for UC,[21-28] and three studies assessed their use in the maintenance of remission in UC.[29-32] Six studies have compared herbal therapy with placebo, five studies have compared herbal therapy with mesalazine/sulphasalazine/steroid enema, and three studies have evaluated the efficacy of herbal therapy in addition to conventional therapy. The number of subjects included in the studies ranged from 14 to 224. In the inductions studies, treatment duration ranged from 4 to 12 weeks, and in the maintenance studies treatment duration ranged from 6 to 12 months. Most herbal therapies were not associated with any major adverse effects. Minor adverse effects reported were nausea, constipation and flatulence.
A randomised double-blind controlled trial from the United Kingdom showed that aloe vera gel, administered for 4 weeks, to patients with moderately active UC, was superior to placebo. Clinical remission, improvement and response were reported in 9, 11 and 14 patients on aloe vera, respectively, compared with one, one and two patients on placebo. In a randomised, double-blind, placebo-controlled study from Israel, 23 patients with active distal UC were randomised to wheat grass juice (Triticum aestivum), or placebo, daily for 4 weeks. Ten of 11 patients (91%) on active treatment responded compared with 5 of 12 on placebo (42%). Seven of nine (78%) patients on active treatment had sigmoidoscopic improvement compared with three of ten (30%) patients on placebo. Rectal bleeding, abdominal pain and physician global assessment score significantly reduced in the active treatment arm. The most common adverse effect in patients on wheat grass juice was nausea. A recent large randomised, double-blind, placebo-controlled trial compared Andrographis paniculata extract (HMPL-004) to placebo in 224 adults with mild-to-moderate UC. Treatment with HMPL-004 at a higher dose (1800 mg daily) was associated with greater clinical response than placebo (60% vs. 40%; P = 0.018), but remission rates at 8 weeks were not different between both groups (38% vs. 34%; P = 0.101).
Two studies had compared the efficacy of herbal therapy to mesalazine. In the first study, 30 patients with chronic active UC were randomised to gum resin of Boswellia serrata (900 mg daily in three doses; n = 20) or sulphasalazine (3 g daily in three doses; n = 10) for 6 weeks. Fourteen of 20 patients treated with Boswellia gum resin and 4 of 10 treated with sulphasalazine achieved remission. Eighteen of 20 patients treated with Boswellia gum resin and 6 of 10 patients on sulphasalazine showed an improvement in one of more of the parameters including stool properties, histopathology and scanning electron microscopy. The second study was a randomised double-blind multicentre 8-week parallel-group study which showed that HMPL-004 was as effective as mesalazine (response 76% vs. 82%; remission 21% vs. 16%) in the treatment of mild-to-moderate UC but there was no difference in endoscopic remission rates at 8 weeks between the two groups (28% vs. 24%).
In a randomised controlled trial, 153 patients with UC were randomly assigned to three groups. Group I: Jian Pi Ling (JPL) tablet with retention-enema of Radix Sophorae Flavescentis and Flos Sophora decoction; group II: sulphasalazine and retention-enema of dexamethasone; group III: placebo and retention-enema of decoction as that in group I. Remission rates at 3 months in group 1 were significantly higher (53%) than those in the other two groups (28% and 19% respectively). However, the very low efficacy rate of conventional therapy makes this study hard to interpret.
In another randomised controlled trial from China, 118 patients with active UC were treated with a traditional Chinese remedy, Kui jie qing (KJQ) enemas four times daily (n = 95) whereas 11 patients treated with sulphasalazine, oral prednisolone and prednisolone enemas served as controls. Effective ‘cure’ was shown in 72% of KJQ-treated patients but only in 9% of controls although the definition of ‘cure’ or ‘improvement’ in this study was not clear.
One study has assessed the use of herbal therapy in addition to conventional treatment. 118 patients with active UC were treated with oral Yukui tang tablets (‘decoction for ulcer healing’) and herbal decoction enemas, in addition to oral prednisolone 15 mg daily, neomycin and vitamin B for 40 days. Eighty-six control patients who received only low dose prednisolone, neomycin and vitamin B were used for comparison. The remission rates and response rates were 33% and 51%, respectively, in the active group, compared with 17% and 43%, in the control group.
Three studies assessed the use of herbal enemas in the treatment of ulcerative proctitis. Fourteen patients with mild-to-moderate active UC were treated with bovine colostrum enemas or placebo containing albumin solution twice daily for 4 weeks in addition to mesalazine. The colostrum, but not the placebo group, showed a mean reduction in symptom score, and an improvement in histological score. In an 8-week double-blind randomised study, Xilei-san, a traditional Chinese herbal medicine with anti-inflammatory properties, was compared with dexamethasone enemas in 35 subjects with mild-to-moderate active ulcerative proctitis. Subjects were followed up for 12 weeks. Both treatments showed significant improvement in clinical, endoscopic and histological score compared with baseline. In a separate randomised controlled trial, 30 subjects with intractable ulcerative proctitis despite 4 weeks of topical mesalazine or corticosteroid were randomised to xilei-san suppositories or placebo suppositories for 2 weeks. The number of patients who achieved remission (clinical activity index ≤4) was significantly higher in the Xilei-san group compared with placebo (P < 0.04). Endoscopic and histological improvements were also significantly higher and relapse rate at 180 days was significantly lower in the active treatment arm compared with placebo. Rectal Xilei-san appeared to be well tolerated in both studies. Table 1 shows studies of herbal therapy used for the induction of remission in UC.
|Author, reference||Country||Year||CAM||Number of subjects||Comparator||Duration||Remission/Response on CAM (%)||Remission/Response on comparator (%)|
|Langmead||UK||2004||Aloe vera||44||Placebo||4 weeks||30||7|
|Ben-Arye||Israel||2002||Triticum aestivum||23||Placebo||4 weeks||91||42|
|Khan||UK||2002||Bovine colostrum enema||14||Placebo||4 weeks||–||–|
|Sandborn||5 countries in the USA and Europe||2013||HMPL-004||224||Placebo||8 weeks|| |
|Fukunaga||Japan||2013||Xilei-san suppository||30||Placebo suppository||2 weeks||46||0|
|Zhang||China||2013||Xilei-san enema||35||Dexamethasone enema||8 weeks||–||–|
|Gupta||India||2001||Boswellia serrata||30||Sulphasalazine||6 weeks||70||40|
|Chen (article in Chinese)||China||1994||Jian Pi Ling (JPL) tablet||153|| |
|90 days||53|| |
|Wang||China||1997||Kui Jie Qing enemas||106||Sulphasalazine, oral prednisolone, prednisolone enema||20 days||72||9|
|Chen||China||1999||Yukui tang tablets||118||Oral prednisolone, neomycin and vitamin B||40 days||33||17|
Curcumin, a biologically active phytochemical substance found in turmeric, has been shown to have anti-inflammatory and antioxidant properties in cell culture and animal studies. A randomised, double-blind, multicentre Japanese study assessed the efficacy of curcumin as maintenance therapy in patients with quiescent UC. Eighty-nine patients were randomised to curcumin (1 g twice daily), or placebo, in addition to sulphasalazine or mesalazine, for 6 months. Relapse rate was significantly lower in the curcumin than placebo group (4.7% vs. 20.5%; P = 0.04).
Colonic fermentation of Plantago ovata seeds (dietary fibre) produces butyrate, and butyrate enemas may be effective in the treatment of UC. In an open label, parallel-group, multicentre, randomised clinical trial, 105 patients with UC in remission were randomised into groups to receive Plantago ovata seeds (10 g twice daily), mesalazine (500 mg three times daily), and Plantago ovata seeds plus mesalazine at the same doses. Relapse rate at 12 months were similar in the three groups (40% vs. 35% vs. 30%). Mild side effects included constipation and/or flatulence.
In a placebo-controlled study, 43 patients with stable UC were randomised to receive either MaxEPA (Oenothera biennis) (n = 16), super evening primrose oil (n = 19), or olive oil as placebo (n = 8) for 6 months, in addition to their normal treatment. Compared with MaxEPA and placebo, evening primrose oil significantly improved stool consistency and the difference was maintained even after treatment was discontinued. There was, however, no difference in stool frequency, rectal bleeding, disease relapse, sigmoidoscopic appearance or histology in the three treatment groups. Table 2 summarises studies of herbal therapy used for the maintenance of remission in UC.
|Authors, reference||Country||Year||CAM||Number of subjects||Comparator||Duration||Remission on CAM (%)||Remission on comparator (%)|
|Fernandez-Banares||Spain||1999||Plantago ovata seeds||105||Mesalazine||12 months||60||65|
|Greenfield||UK||1993||Oenothera biennis||43||Evening primrose oil and olive oil as placebo||6 months||–||–|
Four studies have evaluated the use of herbal therapy to induce remission in active CD,[36-38] one study to maintain remission in CD, and two placebo-controlled trials assess the use in the prevention of disease recurrence after surgery for CD.[40, 41]
In a German double-blind study, 40 patients with CD on a stable daily dose of steroids were randomised to receive the herbal blend containing wormwood herb (Artemisia absinthium) or placebo for 10 weeks. Clinical remission was seen in 13 of 20 (65%) patients on woodworm compared to none on placebo at 8 weeks. In a separate controlled trial, 20 patients with active CD were given in addition to their existing CD therapy either dried powdered wormwood or placebo. At 6 weeks, 8 of 10 (80%) on wormwood and two of 10 patients (20%) on placebo achieved clinical remission defined as a Crohn's disease activity index (CDAI) below 170 or a reduction in CDAI by 70 points. Six of ten, and none of ten patients, on woodworm and placebo had a clinical response.
In a randomised, double-blind, non-inferiority, parallel-group study, 102 patients with active CD were randomised to Boswellia serrata extract (H15) or mesalazine. The mean reduction in CDAI was 90 for H15 and 53 for mesalazine. A separate double-blind, placebo-controlled, randomised, parallel study from 22 centres in Germany evaluated the long-term efficacy and safety of Boswellia serrata extract (Boswelan, PS0201Bo) in maintaining remission in 108 patients with CD. At 52 weeks, there was no significant difference in the proportion of patients in clinical remission between the actively treated and the placebo group (59.9% vs. 55.3%). The mean time to relapse was also not different between the two groups (171 vs. 185 days). The traditional Chinese medicine, Tripterygium wilfordii Hook F (TWHF), has both immunomodulatory and anti-inflammatory activities. Twenty adult patients with active CD were treated with T2 pills for 12 weeks. CDAI scores dropped during the first 8 weeks and endoscopic improvements were observed at week 12. Furthermore, a significant decrease in serum levels of C reactive protein and pro-inflammatory cytokines was reported.
Two placebo-controlled studies assessed the role of Tripterygium wilfordii (GTW) in preventing post-operative recurrence of CD. Forty-five patients with CD were randomly assigned to receive GTW or mesalazine after their operation. No clinical recurrence occurred in both groups at 3 months. There were no significant differences in clinical relapse at 6 months (18% vs. 22%) or 12 months (32% vs. 39%) between the GTW and mesalazine groups. Endoscopic recurrence at 12 months was also similar in the two groups (46% vs. 61%). The second study randomised 39 CD patients to GTW (n = 21) or sulphasalazine (n = 18) 2 weeks after resection for CD. Clinical recurrence was reported in 6% on GTW and 25% on sulphasalazine, and endoscopic recurrence was reported in 22% on GTW and 56% on sulphasalazine. GTW appeared to be as effective, if not more effective, than mesalazine in preventing recurrence of post-operative CD. Table 3 shows studies of herbal therapy in the treatment and prevention of post-operative recurrence in CD.
|Author, reference||Country||Year||CAM||Number||Comparator||Duration||Remission on CAM (%)||Remission on comparator (%)|
|Omer||USA||2007||Artemisia absinthium||40||Placebo||10 weeks||65%||0%|
|Krebs||Germany||2010||Artemisia absinthium||20||Placebo||6 weeks||80%||20%|
|Gerhardt (article in German)||Germany||2001||Boswellia serrata extract H15||102||Mesalazine||8 weeks||36%||31%|
|Ren||China||2007||Tripterygium wilfordii||20||Placebo||12 weeks||–||–|
|Holtmeier||Germany||2010||Boswellia serrata extract (Boswelan, PS0201Bo)||108||Placebo||52 weeks||60%||55%|
|Tao (article in Chinese)||China||2009||Tripterygium wilfordii (post-op CD)||45||Mesalazine|| |
82% (6 months)
68% (12 months)
78% (6 months)
61% (12 months)
|Liao (article in Chinese)||China||2009||Tripterygium wilfordii (post-op CD)||39||Sulphasalazine||–||94%||75%|
At least 50 percent of patients with IBD have used some form of CAM during their disease course. This appeal is enhanced by the common belief that CAM therapies are safer and less toxic than conventional treatment. In many surveys, vitamin supplements and herbal therapies are the most commonly used form of CAM in IBD patients accounting for up to 65% and 58% of the proportion of CAM users respectively.[14, 45, 46] Other forms of CAM used in IBD include homeopathy, traditional Chinese medicine, acupuncture or probiotics. The frequency of use for individual therapy varies greatly and likely reflects local patterns of CAM use.
We have focused on herbal therapies as others have extensively reviewed the evidence for other forms of CAM. In two systematic reviews, omega-3 fatty acids appear not effective for the maintenance of remission in CD[18, 19] In contrast, a meta-analysis showed that acupuncture and moxibustion are superior to western medicine in the treatment of UC with good safety profile. In this systematic review, several studies have reported superiority of herbal therapy or at least equivalent efficacy, whereas others have shown that herbal therapies are less efficacious than conventional therapies. For example, curcumin a natural remedy used to cook curry was reported to be more effective than placebo for maintaining remission in patients with quiescent UC. Herbs such as Aloe vera and Boswellia serrata have also demonstrated efficacy in UC based on their anti-oxidant properties. Wheat grass juice appeared effective and safe as a single or adjuvant treatment of active distal UC. HMPL-004 showed better clinical response than placebo, and comparable efficacy as mesalazine in patients with active mild-moderate-UC whereas Plantago ovata seeds might be as effective as mesalazine in preventing relapse of UC. In patients who are intolerant to mesalazine, these therapeutic options may serve as attractive alternatives. Most subjects with UC had found greater benefit from herbal therapy than those with CD.
However, not all studies have shown consistent and promising result of herbal therapy in IBD. The reasons for the discrepancies from studies could be related to the different study designs, small sample size, lack of adequate controls, overlapping treatment effects, variety of drug formulations and doses used and poor reporting of results, which have led to results that are, in some instances, difficult to interpret.[48, 49] The major problem in interpreting these results is the generally low quality of the randomised controlled trials. Publication bias is also likely to be a significant problem whereby positive results are being preferentially published. Most herbal medicines should undergo a similar level of rigorous testing as pharmaceutical drugs. With the increasing demand from the public, the manufacture, marketing, and prescription of herbal medicines is expected to continue to grow.
Most of the published herbal therapies used in IBD have been associated with minimal adverse effects. At least in short-term studies, the overall incidence of adverse events reported in patients treated with herbal therapy has been similar to those receiving placebo or mesalazine. This treatment will therefore be attractive to many patients, particularly those with a preference for ‘natural remedies,’ those who have had surgery, and those who have experienced side effects from conventional drugs. The cost is also likely to be less than that of high doses of mesalazine, the most widely used agents for relapse prevention. Cost-effectiveness analysis of CAM is warranted.
Most herbal therapies have been reported to have anti-inflammatory properties and plausible mechanisms of action in IBD. Bovine colostrum consists of immunoglobulins and growth factors and may be efficacious in UC. Preparations from the gum resin of Boswellia serrata have been used as a traditional remedy in Ayurvedic medicine in India for the treatment of inflammatory diseases. In vitro studies have reported the suppression of tumour necrosis factor-α (TNF-α) and pro-inflammatory interleukins by wormwood (Artemisia absinthium) extracts. Boswellic acids inhibit the leukotriene biosynthesis in neutrophilic granulocytes via noncompetitive inhibition of 5-lipoxygenase, resulting in a significant dropped in the mean serum TNF-alpha level; this effect was not observed in the placebo group. Andrographis paniculata has in vitro inhibitory activity against TNF-α, IL-1β and NF-κB.
This study has several limitations. First, we were not able to pool data for analysis due to differences in outcomes and methodology among the included studies. The studies were clinically heterogeneous. Publication bias is likely to be present, but the assessment of bias cannot be performed with very limited number of included trials in each herbal therapy category. Although many of these herbal therapies seem to be effective in the management of UC based on clinical experiences, their long-term efficacies and safety profile need further investigation. Second, selection bias may be present as we have limited our review to herbal remedies that have been tested in human IBD. Other promising herb, for example Bromelain, that has been shown to have anti-inflammatory activities in vitro but has yet been investigated in patients with the disease, was not included. Third, although herbal therapy such as curcumin has promising benefit in IBD, our review is based on only one study and these results need to be reproduced in other populations and larger longitudinal studies are required to prove the efficacy and safety. If such studies continue to demonstrate a favourable safety profile, then it may be possible to combine these herbal therapies with conventional drugs of known efficacy such as corticosteroids, thiopurine and anti-TNF agents, with the aim of achieving synergistic efficacy without incurring synergistic toxicity. Pharmacokinetic and drug interactions studies will be important. Pending results of these studies, the popularity of CAM among IBD patients challenges the physicians who need to enhance their awareness and knowledge about this common practice to better advice or counsel their patients, especially the young women and those with severe disease and those requiring hospitalisation and those who are at increased risk of using CAM. One study from Germany showed that only 48% of all IBD patients regarded a scientific foundation for CAM treatment as being important and 65% will continue to use CAM even if scientific reports showed that the therapy was not effective. Lastly, whether CAM is applicable to all IBD patients has not been fully elucidated. Individual CAM may serve as an alternative treatment in patients who are failing mesalazine. At present, the majority of studies have been conducted in patients with UC and more studies of CD with different disease phenotypes and behaviour are needed. Nevertheless, taken together, all these studies suggest the effectiveness of these new therapeutic approaches, not only when conventional treatment fails or is not tolerated, but also, in some instances, as potential first-line treatment.
In conclusion, clinical studies of herbal therapy in IBD have shown promising efficacy data and acceptable safety profile but studies remain limited and heterogenous. Conducting high-quality clinical trials with appropriate blinding and large number of patients is necessary to obtain more conclusive results on the risk–benefit ratio of herbal therapy in IBD.
Guarantor of the article: Dr Siew C Ng.
Author contributions: SC Ng: study concept and design; data extraction and analysis, drafting of the manuscript, YT Lam and K Tsoi: literature search; data extraction and entry, F Chan and J Sung: revision of the manuscript, J Wu: study supervisor and critical revision of manuscript. All authors approved the final version of this manuscript.
Declaration of personal interests: SC Ng has served as a speaker and received research funding from Ferring and Janssen. F Chan has served as a speaker for Pfizer, Astra Zeneca and Takeda. J Sung has served as a speaker and advisory board member for AstraZeneca and Nycomed. J Wu has served as a speaker for Takeda, AstraZeneca and Lundbeck.
Declaration of funding interests: None.