How to read medical evidence
Trials, reviews, guidelines and headlines: a plain-English guide to judging what a Crohn's study really shows
In short
- Randomised controlled trials, where people are put into groups by chance, are the fairest way to test whether a treatment works.
- Systematic reviews, such as Cochrane reviews, pool all the good trials on a question and say how certain the evidence is.
- Studies of health records show how common things are and what goes together, but on their own they cannot prove cause and effect.
- 'Four times the risk' can still be a small risk. Look for the actual numbers in each group.
- Check who was studied, for how long, and who paid for the study.
- Guidelines turn evidence into advice, but your own IBD team decides with you what fits your situation.
On this page
Why do sources sometimes say different things?
You may notice that two trustworthy sources give different figures. For example, NHS inform and the US National Institutes of Health say smokers are about twice as likely to get Crohn's, while NICE's summary for GPs quotes a pooled study with odds about three-quarters higher. Each is quoting a different set of studies. They agree on the direction of the effect, but not the exact size.123
Figures can also differ because studies measure things differently. A UK study of how common IBD is found higher numbers than earlier research, and its authors said the answer depends heavily on how a 'case' is defined.45
Guidelines from different countries or different years can also differ, because they weigh new evidence at different times and for different health systems. The British Society of Gastroenterology's 2025 guideline, for example, replaced its 2019 version and was written mainly for the UK.6
What kinds of study are there?
Different questions need different kinds of study. These are the main types you will see quoted about Crohn's:7
- Randomised controlled trial (RCT): people are put into groups by chance, usually by a computer. One group gets the treatment being tested; the 'control' group gets a standard treatment or a dummy treatment (placebo). The NHS explains that researchers try to keep the groups as alike as possible in everything except the treatment being tested.
- Blinded trial: nobody, sometimes including the doctors, knows who is getting which treatment. The NHS says this helps reduce bias. The SONIC trial of infliximab and azathioprine, for example, was randomised and double-blind.
- Head-to-head trial: two active treatments are compared directly, rather than against a placebo. SEAVUE compared ustekinumab with adalimumab, and found no meaningful difference after a year.
- Systematic review and meta-analysis: researchers find every suitable study on a question, judge how reliable each one is, and combine the results. Cochrane reviews are a well-known example.
- Observational or records study: researchers look at what happens to large groups of people, often using GP or hospital records, without giving any treatment. These are how we know how common Crohn's is.
- Case report: a description of one patient, often with something rare. These alert doctors to unusual problems but cannot tell you how common they are.
- Expert consensus: a panel of specialists agrees on advice where trials are lacking. The STRIDE-II treatment targets were agreed when at least three-quarters of the experts rated each one highly.78910111213
New medicines are tested in phases. The NHS explains that early phases check safety and dose in small numbers of people, and later phases test the medicine in larger groups of patients against a placebo or existing treatment. Safety monitoring continues after a medicine is licensed, because rare side effects may only show up once many people use it.7
Does 'linked to' mean 'causes'?
Not necessarily. Records studies can show that two things often go together, but something else may explain the link. NICE's GP summary gives a good example: Crohn's is diagnosed more often in the first years after having the appendix out, but later studies disagree, and part of the link may be because early Crohn's is hard to tell apart from appendicitis.3
Gut bacteria are another example. A large study of children found clear differences in gut bacteria at diagnosis, and the US National Institutes of Health says researchers are still working out how the microbiome and IBD are connected.214
What do the numbers actually mean?
Relative risk and actual risk are different. NICE's GP summary says the risk of Crohn's is about four times higher in the year after a stomach bug, but adds that the actual risk is still low. Four times a very small number is still a small number.3
Look at the result in each group, not just the difference. In SONIC, about 57 in 100 people on infliximab plus azathioprine were in remission without steroids at 26 weeks, compared with 44 in 100 on infliximab alone and 30 in 100 on azathioprine alone. That tells you both that the combination did best and that many people in every group did not reach remission.8
Some reviews give a 'number needed to treat'. A Cochrane review of vedolizumab found that about 13 people needed to be treated for one extra person to reach remission, compared with placebo. The smaller this number, the bigger the effect.10
Check what 'remission' means in the study. Many Crohn's trials define it as a score under 150 on a disease activity index called the CDAI, where higher scores (up to about 600) mean more active Crohn's. Others measure 'endoscopic remission', meaning the bowel lining looks healed on a camera test, which is a harder target. The SEQUENCE trial measured both.9151617
Watch for uncertainty. In SEAVUE, 65 in 100 people on ustekinumab and 61 in 100 on adalimumab were in remission at one year, but the researchers' range of likely true differences ran from adalimumab doing 6 points better to ustekinumab doing 14 points better. In other words, the trial could not separate them. Cochrane reviews say this openly, grading their evidence from high to very low certainty.918
Who was studied, for how long, and who paid?
A trial result applies most clearly to people like those in the trial. SEAVUE only included adults who had never had a biologic. The LIR!C trial of surgery versus infliximab only included people with a short stretch of disease, under 40 cm, at the end of the small bowel. The REACH trial was in children. Ask whether the people studied were like you.91920
Many trials last a year or less. A 2026 Cochrane review comparing biologic medicines for Crohn's said long-term safety data were limited, and called for more head-to-head trials and more focus on bowel healing.18
Papers list who funded them, so readers can take this into account. Many drug trials are paid for by the company that makes the medicine: SEAVUE was funded by Janssen, and CALM and SEQUENCE by AbbVie. The 2026 Cochrane review was funded by the UK's National Institute for Health and Care Research, and the BSG's 2025 guideline was written without commercial funding.69161821
Good trials are registered before they start, on registries such as ISRCTN. ISRCTN says registering studies supports openness and helps reduce selective reporting, where only some results get published. The NHS adds that researchers should publish their results and share them with participants.722
Where do guidelines fit in?
Guidelines bring the evidence together and turn it into advice for doctors. NICE publishes the evidence documents behind its Crohn's guideline, showing which studies it looked at. The BSG says its 2025 guideline reflects the best current evidence but is not meant to be followed rigidly, because individual situations may need a different approach.623
NICE says decisions about treatment are best made together: your team should explain your options, including the pros and cons, and listen to what matters to you. Bringing a study you have read to your appointment is a good way to start that conversation.24
Where can I read research myself?
The NHS says trial results are usually published in medical journals and libraries of evidence such as the Cochrane Library. Short summaries (abstracts) are usually free to read, but full articles often need a subscription, and papers are written in technical language. Some papers are free in full, such as the UK PROFILE trial.725
Sources
This article is written in our own words from these sources. Each number in the text points to one of them.
- NHS inform (Scotland) Crohn's disease Read on NHS inform: Crohn's disease
- NIDDK (US National Institutes of Health) Symptoms & Causes of Crohn's Disease Read on the NIDDK (US National Institutes of Health) website: Symptoms & Causes of Crohn's Disease
- NICE CKS Crohn's disease: What are the causes and risk factors? Guide, May 2024 Read on the NICE website: Crohn's disease: What are the causes and risk factors?
- BMC Gastroenterology The incidence and prevalence of inflammatory bowel disease in UK primary care: a retrospective cohort study of the IQVIA Medical Research Database Data, Mar 2021 Read the paper: The incidence and prevalence of inflammatory bowel disease in UK primary care: a retrospective cohort study of the IQVIA Medical Research Database
- NICE Clinical Knowledge Summaries Crohn's disease: How common is it? Guide, May 2024 Read on the NICE website: Crohn's disease: How common is it?
- British Society of Gastroenterology (Gut) British Society of Gastroenterology guidelines on inflammatory bowel disease in adults: 2025 Guideline, Jun 2025 Read on the British Society of Gastroenterology (Gut) website: British Society of Gastroenterology guidelines on inflammatory bowel disease in adults: 2025
- NHS Clinical trials Web page, Dec 2022 Read on the NHS website: Clinical trials
- New England Journal of Medicine Infliximab, azathioprine, or combination therapy for Crohn's disease (SONIC) Study, Apr 2010 Read the paper: Infliximab, azathioprine, or combination therapy for Crohn's disease (SONIC)
- The Lancet Ustekinumab versus adalimumab for induction and maintenance therapy in biologic-naive patients with moderately to severely active Crohn's disease (SEAVUE) Study, Jun 2022 Read the paper: Ustekinumab versus adalimumab for induction and maintenance therapy in biologic-naive patients with moderately to severely active Crohn's disease (SEAVUE)
- Cochrane Vedolizumab for induction and maintenance of remission in Crohn's disease Systematic review, Jul 2023 Read the review: Vedolizumab for induction and maintenance of remission in Crohn's disease
- BMJ Open via PubMed Central Incidence and prevalence of inflammatory bowel disease in UK primary care: a population-based cohort study Study, Jul 2020 Read the paper: Incidence and prevalence of inflammatory bowel disease in UK primary care: a population-based cohort study
- Pediatric Gastroenterology, Hepatology & Nutrition Life-threatening lower gastrointestinal hemorrhage in pediatric Crohn's disease Case report, Mar 2013 Read the paper: Life-threatening lower gastrointestinal hemorrhage in pediatric Crohn's disease
- Gastroenterology STRIDE-II: An Update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) Initiative of the International Organization for the Study of IBD (IOIBD) Guideline, Apr 2021 Read on the Gastroenterology website: STRIDE-II: An Update on the Selecting Therapeutic Targets in Inflammatory Bowel Disease (STRIDE) Initiative of the International Organization for the Study of IBD (IOIBD)
- Cell Host & Microbe The treatment-naive microbiome in new-onset Crohn's disease Study, Mar 2014 Read the paper: The treatment-naive microbiome in new-onset Crohn's disease
- Gastroenterology Adalimumab for maintenance of clinical response and remission in patients with Crohn's disease: the CHARM trial Study, Jan 2007 Read the paper: Adalimumab for maintenance of clinical response and remission in patients with Crohn's disease: the CHARM trial
- New England Journal of Medicine Risankizumab versus Ustekinumab for Moderate-to-Severe Crohn's Disease (SEQUENCE) Study, Jul 2024 Read the paper: Risankizumab versus Ustekinumab for Moderate-to-Severe Crohn's Disease (SEQUENCE)
- New England Journal of Medicine Upadacitinib Induction and Maintenance Therapy for Crohn's Disease (U-EXCEL, U-EXCEED, U-ENDURE) Study, May 2023 Read the paper: Upadacitinib Induction and Maintenance Therapy for Crohn's Disease (U-EXCEL, U-EXCEED, U-ENDURE)
- Cochrane Biologic drugs for induction and maintenance of remission in Crohn's disease: a network meta-analysis Systematic review, Jun 2026 Read the review: Biologic drugs for induction and maintenance of remission in Crohn's disease: a network meta-analysis
- The Lancet Gastroenterology & Hepatology Laparoscopic ileocaecal resection versus infliximab for terminal ileitis in Crohn's disease (LIR!C) Study, Nov 2017 Read the paper: Laparoscopic ileocaecal resection versus infliximab for terminal ileitis in Crohn's disease (LIR!C)
- Gastroenterology Induction and maintenance infliximab therapy for the treatment of moderate-to-severe Crohn's disease in children (REACH) Study, Mar 2007 Read the paper: Induction and maintenance infliximab therapy for the treatment of moderate-to-severe Crohn's disease in children (REACH)
- The Lancet Effect of tight control management on Crohn's disease (CALM): a multicentre, randomised, controlled phase 3 trial Study, Dec 2017 Read the paper: Effect of tight control management on Crohn's disease (CALM): a multicentre, randomised, controlled phase 3 trial
- ISRCTN ISRCTN Registry Read on the ISRCTN website: ISRCTN Registry
- NICE Crohn's disease: management (NG129) - Evidence Systematic review, May 2019 Read on the NICE website: Crohn's disease: management (NG129) - Evidence
- NICE Crohn's disease: the care you should expect Web page, May 2019 Read on the NICE website: Crohn's disease: the care you should expect
- The Lancet Gastroenterology & Hepatology A biomarker-stratified comparison of top-down versus accelerated step-up treatment strategies for patients with newly diagnosed Crohn's disease (PROFILE) Study, May 2024 Read the paper: A biomarker-stratified comparison of top-down versus accelerated step-up treatment strategies for patients with newly diagnosed Crohn's disease (PROFILE)