IBS Research: Truth, Post Truth and Statistics
To the sceptics among us, there is no such thing as truth. Memory is fallible and remodelled every time it is recalled; so we can never be sure of what really happened. Religious truths are faiths based on a mythology that we may or may not choose to believe in. We cannot even trust our own perceptions, since these are so often distorted by the way we think about them based on our experience.
Scientific Evidence
Scientific truths, on the other hand, are based on evidence. Science works by creating hypotheses and then devising experiments to try to disprove them. Scientists will often create models of illness in, for examples, isolated tissue, cells or experimental animals, against which to test their hypotheses. The conclusions, however, only apply to the specific context of the observation or the stringent conditions of the experiment. Nevertheless, the harder a hypothesis is to disprove, the more likely we are to regard it as true. The reality is that most ‘scientific facts’ are hypotheses based on probabilities. The evidence that helicobacter pylori is the cause of peptic ulcers seemed pretty convincing after Dr Barry Marshall drank his bacterial broth, but it has since become clear that many of us harbour helicobacter, yet most don’t have ulcers, so there must other factors operating.
It is of course so much harder when the evidence for the very existence of a specific disease is lacking, as for The Irritable Bowel Syndrome. Then the whole scientific exercise can be like shooting at shadows; as soon as you think you’ve got a fix on it, it disappears or changes.
The best that scientists can do in such circumstances is remove a factor or administer a treatment and see if symptoms improve in a statistically significant number or patients compared with those who have no active intervention. But they need to keep the conditions of the trial as near consistent as possible and to recruit enough matched patients in order to prove that the active treatment is better than a blank or placebo.
Drug Trials
Using tight conditions and sufficient numbers, drug companies may claim a statistically significant benefit when only 50% of patients responded to their new drug compared with 40% on placebo. Although this result may achieve statistical significance, based on the demonstration that the chance of achieving the same result by chance was less than one in twenty, the biological significance of such a result is low. Moreover, when their drug is released in clinical practice, factors, such as diet, stress, medications and other illnesses cannot be kept constant, and the potential clinical effect could be negligible. So would you take a new treatment with unknown long term side effects based on such data? Or would you try to think of other ways to enhance the placebo effect?
Another major difficulty in such trials is whether the presence of the active treatment can be concealed from both patient and investigator. In many instances it can’t. Patients often know or can guess. In drug trials, clinical investigators can prescribe placebo pills or capsules that look identical to the active treatment, but they rarely check whether their subjects guessed what they were taking based on how they felt. If they did, it would completely invalidate the principle of a double blind, randomised, controlled clinical trial.
Diets and more holistic treatments
Diets, psychological treatments and complementary therapies cannot be blinded. Both the patient and the investigator know what treatment they are getting. This is where we get the spectacular results; the 70% of people who respond to a low FODMAP diet, the 80% that respond to hypnotherapy, the glowing patient testimonies for everything from silicol gel to aloe vera.
Clinical investigators may choose to compare how many people taking the new treatment still report benefit some months later compared with those who either had no treatment or were receiving the best treatment available at the time. That has indicative value and a modest degree of scientific validity providing the new treatment was not endorsed by continued publicity.
Trials of FODMAP exclusion have tried to get round the problem of blinding by conducting a double blind trial of food challenge in patients stabilised on a minimal FODMAP diet. Dr Susan Shepherd and her colleagues from Monash University compared the effect on 25 patients of food challenge with drinks of poorly absorbed fructose or fructans, compared with glucose as an absorbable control1. Their results demonstrated that feeding fructose and fructans (raffinose) revoked the symptoms of IBS.
The study design is as good as it gets for dietary intervention, but Shepherd and her colleagues only tested two components of FODMAPs, monosaccharide fructose and the trisaccharide raffinose (a combination of fructose, galactose and glucose). Raffinose may qualify as a FODMAP, but is not the form in which most FODMAPs are consumed. Furthermore, since these low molecular weight sugars were ingested in liquid form, it is likely they travelled down the intestine much faster than solid FODMAP food, potentially arriving in the colon as a bolus which would be rapidly fermented releasing sufficient gas to cause symptoms. This study has been widely quoted as supporting the efficacy of the low FODMAP diet for IBS, but we need to look at the context of the study and decide how much the results support that conclusion.
Post Truth
The difficulty in conducting trials of treatment in IBS begs the question: how much should we care about scientific rigour? Isn’t it enough to know that the treatment makes scientific sense and just try it for ourselves? After all, if we are convinced by the publicity surrounding a new treatment and believe in the treatment, surely the powerful psychological effect of belief will summate with any scientific efficacy and we will have a win/win situation. Yes, but isn’t that a bit like choosing a particular brand of shampoo and speculating on the stock market.
This brings us to the Oxford Dictionaries’ new word for 2016: post-truth. Post truth is defined as ‘relating to or denoting circumstances in which objective facts are less influential in shaping public opinion than appeals to emotion and personal belief’. It was ever thus. There is an inherent distrust of experts. People only believe what makes sense and works for them.
Specific treatments need specific evidence to support them. Otherwise, there is always the risk of unwanted and potentially damaging side effects. In adopting a laissez-faire attitude, treatments for IBS may come to resemble the quack medicines, sold at country fairs around the time of The Enlightenment. Despite the scientific evidence or perhaps because of it, I am concerned that specific IBS remedies like The Low Fodmap Diet and the broad spectrum antibiotic rifaximin could risk causing a persistence of intestinal hyper-sensitivity by depleting the colonic microbiome.
While holistic therapies that help the patient calm their symptoms of IBS through well being, relaxation, mindfulness, and insight, may also have little statistical evidence of efficacy, they encourage the patients own healing resources and do not carry the same degree of risk.
Reference
- Shepherd SJ, Parker FC, Muir JG, Gibson PR (2008). Clinical Gastroenterology and Hepatology 6: 765-771
For more information,
log in or sign up to The IBS Network and click on the following links
https://www.theibsnetwork.org/medical/medicines/
https://www.theibsnetwork.org/diet/fodmaps/
https://www.theibsnetwork.org/therapies/
The Sensitive Gut 