Cochrane
Cochrane (formerly the Cochrane Collaboration) is an international network that produces systematic reviews of health interventions, published in the Cochrane Database of Systematic Reviews (Cochrane Library). Its dermatology arm, Cochrane Skin, was founded by Professor Hywel Williams at the University of Nottingham’s Centre of Evidence Based Dermatology and registered with Cochrane on 12 September 1997 (Cochrane news, 2017). Williams set it up partly because many dermatology trials “addressed unimportant questions that mean little to patients”. Several UK-based Cochrane review groups closed in 2023 when their funding ended. Cochrane Skin survived: at the end of March 2023 its editorial base moved to the Université Paris-Est Créteil, with Laurence Le Cleach as Co-ordinating Editor (Cochrane Skin news, 2023). It is still active in 2026, with a 2026 priority-setting update and a joint meeting with the European Dermato-Epidemiology Network in Paris in 2024 (Cochrane Skin news, 2026). For this vault, Cochrane reviews are the top tier of evidence. They decide most of the verdicts in the Treatment Outcomes Comparison between home remedies (Track A) and prescribed care (Track B).
Cochrane [defines] Evidence-Based Dermatology Cochrane [supports] Treatment Outcomes Comparison Cochrane [relates] NICE
Cochrane reviews this vault relies on. Each row gives the year and bottom line as recorded in the linked note.
| Review (CD number) | Year | Topic | Bottom line |
|---|---|---|---|
| van Zuuren et al. (CD012119) | 2017 | Emollients and moisturisers | 77 RCTs, 6,603 people. Moisturisers reduce severity and flares, and help TCS work better. No reliable evidence that one moisturiser beats another. Glycerol gives a modest benefit (high certainty). Urea gives a low-quality benefit with more adverse events. Oat creams made no difference to severity (Colloidal Oatmeal) |
| Kelleher et al. (CD013534) | 2022 | Infant skin care for prevention | Regular emollients in infancy “probably” do not prevent eczema. Includes BEEP and PreventADALL |
| Lax et al. (CD013356) | 2022 | Topical Corticosteroids strategies | 104 RCTs. Clear potency gradient. Once daily is probably as good as twice daily. Twice-weekly Weekend Therapy cut flares from about 576 to 248 per 1,000 (NNT ≈3, moderate certainty). 26 possible Skin Atrophy cases among 2,266 people, mostly with potent or very potent TCS |
| Topical anti-inflammatory network meta-analysis | 2024 | TCS, TCIs, others | Could not evaluate Topical Steroid Withdrawal (per Topical Corticosteroids note) |
| Topical tacrolimus review | 2015 | Tacrolimus | 20 RCTs, 5,885 people. Supports efficacy (cited by PCDS) |
| George et al. (CD003871) | 2019 | Anti-Staphylococcus aureus measures | Insufficient evidence. Oral antibiotics show no clear benefit in clinically infected eczema. Steroid–antibiotic combinations give a slight, low-quality benefit (Fusidic Acid). Bleach Baths and antiseptic additives show no benefit |
| Matterne et al. (CD012167) | 2019 | Oral Antihistamines | No reliable evidence that H1 antihistamines improve eczema; they may sedate |
| Bamford et al. (CD004416) | 2013 | Evening Primrose Oil / Borage Oil | 27 RCTs, 1,596 people. “Not effective treatments for eczema” |
| Gu et al. (CD008642) | 2013 | Chinese Herbal Medicine | 28 RCTs, 2,306 people. No conclusive evidence of benefit |
| Makrgeorgou et al. (CD006135) | 2018 | Probiotics for treatment | 39 RCTs, 2,599 people. “Probably make little or no difference” (moderate certainty) |
| Wang et al. (CD006475) | 2025 | Probiotics/synbiotics in infants for prevention | Possibly a slight reduction in eczema (low certainty). No serious adverse events |
| Osborn & Sinn (CD006474) | 2013 | Prebiotic infant feeds | 4 trials, 1,428 infants. Eczema RR 0.68 |
| Bath-Hextall et al. | 2012 | Dietary supplements | No convincing evidence for fish oil, Vitamin D, zinc, selenium and others |
| Nankervis et al. (CD008426) | 2015 | House dust mite avoidance | 7 RCTs, 324 people. No clear benefit (very low quality) |
| Ersser et al. (CD004054) | 2014 | Eczema Education Programmes | Limited evidence. Some benefit from nurse-led and multiprofessional programmes |
| Kramer & Kakuma (CD003517) | 2012 | Exclusive breastfeeding duration | 6 months versus 3–4 months shows no significant reduction in eczema |
Together, the reviews favour Track B. They support emollients and TCS (including proactive use), and they rate the commonest oral Track A supplements (EPO, borage oil, probiotics, Chinese herbs, fish oil) as ineffective or unproven.
Cochrane [supports] Emollient Therapy Cochrane [supports] Topical Corticosteroids Cochrane [supports] Weekend Therapy Cochrane [opposes] Evening Primrose Oil Cochrane [opposes] Chinese Herbal Medicine Cochrane [opposes] Probiotics Cochrane [opposes] Oral Antihistamines
How GRADE certainty works. Cochrane rates evidence with the GRADE approach (Grading of Recommendations Assessment, Development and Evaluation). Certainty is rated separately for each outcome, not for each study, and has four levels: high, moderate, low and very low (Cochrane Handbook ch. 14, Schünemann et al., 2023). A body of randomised trials starts at high and is rated down one level for each serious concern, or two for a very serious one. There are five such domains: risk of bias, inconsistency, indirectness, imprecision and publication bias. Non-randomised evidence starts at low. It can be rated up only for a large effect, a dose–response gradient, or plausible confounding that would have worked against the effect, and nothing can go below “very low”. The ratings appear in “Summary of findings” tables. Cochrane plain-language summaries map them to wording: high certainty = “reduces”, moderate = “probably reduces”, low = “may reduce”, very low = “uncertain”. This is why the vault quotes words like “probably” (Kelleher 2022; Makrgeorgou 2018) and “may”. Indirectness matters most for toddlers. Many eczema trials enrolled adults or mixed ages, so findings applied to 1–4-year-olds are often rated down.
Cochrane [defines] GRADE GRADE [part-of] Cochrane
Limits. (1) Few toddler-specific data. The TCS prevention trials mostly lasted 16–20 weeks and enrolled few 1–4-year-olds, so the absence of skin thinning cannot be assumed for years of use in toddlers (Weekend Therapy). (2) Out-of-date searches. Some reviews are more than ten years old: the supplements and EPO reviews used searches from about 2010–2012, and newer meta-analyses post-date them (Fish Oil and Omega-3). (3) “No evidence of benefit” is not “evidence of no benefit”. Very-low-certainty null results, as with house dust mites, mean the question is still open. (4) Disagreement with other syntheses. A 2022 GRADE meta-analysis (Bakaa) found bleach baths probably help, while Cochrane 2019 found no benefit (Bleach Baths). Cochrane’s finding that oat creams had no effect also conflicts with industry-linked paediatric oat trials (Colloidal Oatmeal). (5) TSW is outside its scope. The trials were too short to capture topical steroid withdrawal. (6) Institutional fragility. The 2023 closure of UK review groups shows that updates depend on funding. Cochrane Skin’s move to France kept the group going, but update speed is unknown.
Conflict: [Cochrane, George et al., 2019] found no benefit from bleach baths. [Bakaa et al., Ann Allergy Asthma Immunol, 2022, GRADE] found they probably improve clinician-rated severity by about 22%. Unresolved — add to open_questions.
Cochrane [contradicts] Bleach Baths Cochrane [relates] Topical Steroid Withdrawal
Connections
- Emollient Therapy — reviews evidence for emollients, source: CD012119; CD013534
- Topical Corticosteroids — Lax 2022: once daily ≈ twice daily; weekend therapy prevents flares, source: CD013356
- Herbal and Home Remedies for Eczema — reviews: EPO/borage (2013), probiotics (2018), Chinese herbal medicine (2013), source: Cochrane
- Staphylococcus aureus — George et al. 2019 (CD003871): insufficient evidence for anti-staph interventions in eczema, source: Cochrane 2019
- Urea Creams — van Zuuren 2017: 4 urea studies, low-quality benefit, more adverse events (RR 1.65), source: Cochrane 2017
- Glycerol — van Zuuren 2017: glycerol improved SCORAD vs control (high certainty), source: Cochrane 2017
- Probiotics — treatment review null (2018); infant prevention review low-certainty slight benefit (2025), source: Makrgeorgou 2018; Wang 2025
- Prebiotics and Synbiotics — prebiotic infant feeds: eczema RR 0.68, source: Osborn & Sinn 2013
- Vitamin D — Bath-Hextall 2012 dietary supplements review, no convincing evidence, source: Cochrane 2012
- Fish Oil and Omega-3 — two small trials, possible modest benefit, not convincing, source: Cochrane 2012
- House Dust Mites — 2015 review: no clear benefit from mite avoidance, source: Nankervis 2015
- Eczema Education Programmes — Ersser 2014 CD004054: limited evidence, some benefit from multiprofessional and nurse-led education, source: Cochrane
- BEEP Trial — included in Kelleher 2022 infant skin-care review, source: Cochrane
- PreventADALL Trial — consistent with Kelleher 2022 null prevention finding, source: Cochrane 2022
- Weekend Therapy — proactive TCS flares 576→248/1,000, NNT ≈3, source: Lax 2022
- Skin Atrophy — 26 possible cases in 2,266, mostly potent/very potent TCS, source: Lax 2022
- Tacrolimus — 2015 review, 20 RCTs, supports efficacy, source: Cochrane 2015 via PCDS
- Oral Antihistamines — no reliable evidence of benefit, source: Matterne 2019
- Evening Primrose Oil — 27 RCTs, not effective, source: Bamford 2013
- Borage Oil — 8 RCTs, no benefit over placebo, source: Bamford 2013
- Chinese Herbal Medicine — 28 RCTs, no conclusive benefit, source: Gu 2013
- Fusidic Acid — steroid–antibiotic combinations slight low-quality benefit, source: George 2019
- Bleach Baths — no benefit in Cochrane 2019; conflicts with Bakaa 2022, source: George 2019
- Colloidal Oatmeal — oat moisturisers no difference in severity (low quality), source: van Zuuren 2017
- Breast Milk — 6 vs 3–4 months exclusive breastfeeding no eczema reduction, source: Kramer & Kakuma 2012
- Topical Steroid Withdrawal — 2024 network meta-analysis could not evaluate TSW, source: Topical Corticosteroids note
- Treatment Outcomes Comparison — Cochrane evidence underpins Track A vs B verdicts, source: vault synthesis
- GRADE — method Cochrane uses to rate certainty, source: Cochrane Handbook ch. 14
- NICE — UK guideline body that draws on systematic reviews, source: vault synthesis
- Evidence-Based Dermatology — Cochrane Skin founded at Nottingham’s Centre of Evidence Based Dermatology, source: Cochrane news 2017