TABLE 3.

Characteristics of included studies—Infectious disease.

Author; Year of publication; Trial registration	Study design; Country; Setting	Participant characteristics	Eligibility criteria	Interventions; Comparisons	Outcome measures (methods); Timepoints	Main findings	Safety	Funding; Conflict of interest
Infection control (i.e., hand disinfection)
Youn; 2021; Clinical Research Information Service No. KCT0003240 Youn et al. (2021)	RCT (parallel); South Korea; Community	N = 112 enrolled (n = 106 analysed) Age in years: not reported Female: not reported Note: Supplementary file 1 describing participant characteristics not available.	Inclusion criteria: aged 18–60 years. Exclusion criteria: skin disease affecting the hands or forearms; an open wound, hangnail, or other skin abnormality; immunosuppressant or antibiotic drug use; sensitivity to TTO confirmed by observation.	Intervention group: TTO disinfectant (10% TTO, Melaleuca alternifolia), brand and manufacturer not stated. Other ingredients: TTO added in ratio of 2:2:1:15 of TTO, solubiliser, glycerin and sterile distilled water. Dose of 5 mL applied to participant’s hands. Control group 1: Alcohol-based hand sanitiser (a gel containing 83% ethanol without water). Dose of 2 mL applied to participant’s hands. Control group 2: Benzalkonium chloride-based hand sanitiser (a foam containing benzalkonium chloride without water). Dose of 0.8 mL applied to participant’s hands. Control group 3: No treatment. To contaminate their hands, subjects applied 5 mL of Serratia marcescens (ATCC 14756) evenly on their hands for one minute and allowed their hands to dry for 2 min. This contamination procedure was then repeated a second time. Personnel then dispensed the allocated hand disinfectant into subjects’ hands which they rubbed in for ≥ 30 s.	Skin surface microbial counts (assessed by glove juice sampling procedure, US FDA-Tentative Final Monograph for Healthcare Antiseptics, and using MacConkey agar plate, Asan, Korea incubated at 25°C for 48 h). Skin surface organisms (assessed as relative light units by adenosine triphosphate concentration on skin using ATP Surface Test kit and a Clean-Trace Luminometer 3M Health Care). Timepoints: Baseline (i.e. in between 1st and 2nd hand contaminations) and post-treatment. Trans-epidermal water loss (assessed as g/m2/hr using gpskin Barrier probe, gpower, Korea, with normal values in the range 16–20 and higher values indicating greater water loss). Skin condition (patient-assessed under 3 categories: skin moistness, skin dryness and skin exfoliation, on 5-point scale: 1 = not at all to 5 = extremely high). Timepoints: Baseline (prior to 1st hand contamination) and post-treatment.	Skin surface microbial counts sig. decrease in TTO group -5.50 (1.90), compared with alcohol group: -2.33 (1.62), benzalkonium chloride group: -0.62 (0.57), and no treatment: +0.07 (0.73), all p < 0.05. Skin surface organisms sig. decrease in TTO group -0.46 (0.51) compared with no treatment -0.11 (0.32), p < 0.05. Trans-epidermal water loss no sig. difference between groups in pre- to post-test changes. Skin condition: →Skin moisture sig. difference between groups in pre- to post-test changes (TTO 0.96, alcohol 0.63, benzalkonium chloride 0.96, and no treatment 0.59). →Skin dryness sig. difference between groups in pre- to post-test changes (TTO −0.81, alcohol −0.26, benzalkonium chloride −0.85, and no treatment −0.59). →Skin exfoliation no sig. difference between groups in pre- to post-test changes. Note: Supplementary File S2 not available.	Not assessed.	Funding: National Research Foundation of Korea. Grant Number: NRF-2015R1A1A3A04001441. Authors state no conflicts of interest.
Gnatta; 2021; trial not registered Gnatta et al. (2021)	RCT (cross-over); Brazil; Hospital workplace	N = 15 enrolled (n = 15 analysed) Age in years: not reported Female: not reported	Inclusion criteria: age 18–65 years; short fingernails. Exclusion criteria: visible dryness or lesions on hands; contact with antiseptic soap within 48 h prior to study; allergy to test substance(s); pregnant.	Intervention group: TTO soap (2% TTO, Melaleuca alternifolia), TTO met standard ISO 4730:2017. Other ingredients: lauryl ether sulfate sodium, cocoamidopropyl betaine, phenoxyethanol, and parabens). Dose of 1.5 mL used. Control group 1: 0.5% triclosan soap (Rioderm®, Rioquímica, Brazil). Dose of 1.5 mL used. Control group 2: 2.0% chlorhexidine soap (Riohex®, Rioquímica, Brazil). Dose of 1.5 mL used. Control group 3: Soft soap. Dose of 5 mL used. Hand hygiene: Prior to hand contamination, all subjects cleaned hands with 5 mL soft soap for 60 s, rinsed with mineral water and dried hands with paper towel. After hand contamination with E. coli, subjects washed hands with allocated soap.	Escherichia coli K12 (ATCC 14948; assessed as CFU/mL by rubbing fingertips in agar plates containing Tryptic Soy Agar with deoxycholate, which were then incubated at 37°C for 24 h and re-incubated for a further 24 h). Timepoints: Before contamination and after hand hygiene treatment.	Escherichia coli K12 CFU/mL log10 reduction factor sig. higher in TTO group 4.28 (0.50) compared with Riohex® (3.89 (0.62), p = 0.006) and soft soap (3.17 (0.55), p < 0.001). No sig. difference compared with Rioderm® wash (4.31 (0.35), p = 0.661).	Not assessed.	Funding: Foundation for Research Support of the State of São Paulo (Grant No.2013/23008-7). Authors state no conflicts of interest. Triclosan and chlorhexidine soaps donated by Rioquímica company.
Gnatta; 2013; trial not registered Gnatta et al. (2013)	RCT (cross-over); Brazil; Hospital workplace	N = 15 enrolled (n = 15 analysed) Age in years, mean (SD) 31.0 (7.7) Female 11/15	Inclusion criteria: age 18–55 years; clean, short fingernails. Exclusion criteria: visible dryness or lesions on hands; contact with antiseptic soap within 48 h prior to study; allergy to test substance(s); pregnant.	Intervention group: TTO soap (0.3% TTO, Melaleuca alternifolia), Doctornatu® liquid soap, Higinatu, Brazil). Dose of 1.5 mL used. Control group 1: 0.5% triclosan soap (Rioderm®, Rioquímica, Brazil). Dose of 1.5 mL used. Control group 2: Soft soap. Dose of 5 mL used. Control group 3: Soft soap + rinsing with 60% propan-2-ol Doses of 5 mL soft soap and 3 mL propan-2-ol used. Hand hygiene: Prior to hand contamination, all subjects cleaned hands with 5 mL soft soap for 60 s, rinsed with mineral water and dried hands with paper towel. After hand contamination with E. coli, subjects washed hands with allocated soap.	Escherichia coli K12 (ATCC 14948; assessed as CFU/mL Tryptone Soya Selective Agar (DIFCO®, BD®, Sparks, US) incubated at 37°C for 18-24 h and re-incubated for 24 h. Timepoints: Before contamination and after hand hygiene treatment.	Escherichia coli K12 CFU/mL log10 reduction factor sig. lower in TTO group (3.89 ± 0.69) compared with soft soap + propan-2-ol group (4.89 ± 0.91, p = 0.001). No sig. difference compared with soft soap alone (3.87 ± 1.13, p = 0.247). Triclosan group also no sig. difference compared with soft soap (3.59 ± 0.71, p = 0.2975)	Not assessed.	Funding: Foundation for Research Support of the State of São Paulo (Grant No.2011/18448-2). Authors state no conflicts of interest.
Molluscum contagiosum
Markum; 2012; ACTRN12610000984099 Markum and Baillie, (2012)	RCT (parallel); United States; Not described	N = 53 enrolled (n = 53 analysed by ITT) Age in years, mean (SD) 6.3 (5.1) Female: not reported	Inclusion criteria: diagnosis of molluscum contagiosum; ≥ 50th percentile for height and weight; meeting all developmental milestones. Exclusion criteria: major disease.	Intervention group: TTO (75% TTO, Melaleuca alternifolia), terpinene-4-ol content 40.1% conforming to AS 2782-1985. Other ingredients: high selenium canola oil. Control group 1: TTO + iodine (the same TTO and canola oil as TTO intervention in addition to organically bound iodine. Total iodine concentration = 35 micromolar (US patent #7,311,928 from Naturopathix, Inc., US). Control group 2: iodine (the same organically bound iodine in a vehicle of high selenium canola oil). Subjects applied 4 µL topically to each lesion twice daily for 30 days (or until all lesions resolved).	> 90% reduction in lesions (assessment method not defined) Adverse events (parent or caregiver reported) Timepoints: Baseline and every 10 days until end of trial (30 days)	> 90% reduction in lesions occurred in sig. more children in TTO + iodine group (16/19) compared with TTO group (3/18, P<0.01) and iodine group (1/16, P<0.01) after 30 days. Note: conflict with text which states ‘…4 children [in TTO group] met the 90% reduction criterion’. No correlation between number of lesions at baseline and intervention group.	Redness ≤ 3 mm radius on lesion site: TTO + iodine = 1 TTO = 1 Iodine = 2 Warm sensation on application: TTO + iodine = 3 TTO group = 4	Funding: The Centre of Excellence in Biomedical Research Boise State University, Idaho, US. Authors state no conflicts of interest.
MRSA colonisation
Blackwood; 2013; ISRCTN65190967 Blackwood et al. (2013)	RCT (parallel); Northern Ireland; Hospital intensive care unit	N = 445 enrolled (n = 391 analysed) Age in years, mean (SD) TTO group 57.3 (17.9) Control group 57.1 (19.4) Female TTO group 88/195 Control group 68/196	Inclusion criteria: age ≥ 18 years; likely to remain in ICU > 48 hours. Exclusion criteria: pregnancy; colonised with MRSA on admission; known sensitivity to TTO; readmissions; enrolment in another Investigative Medicinal Product clinical trial within 30 days prior to study.	Intervention group: TTO body wash (50 mg/g TTO, Melaleuca alternifolia), Novabac®, Novasel Australia Pty Ltd. Control group: standard care (Johnson’s Baby Softwash®, Johnson & Johnson) Subjects received ≥ 1 full bed bath daily with their allocated body wash until detection of ICU-acquired MRSA, ICU discharge or death.	Incident MRSA colonisation (MRSA screening specimens obtained from the nose and groin) Timepoints: Baseline (on admission) and at end of study (i.e., ICU-acquired MRSA, discharge, or death). Incident MRSA bacteraemia Timepoints: colonisation and infection data measured daily Maximum increase in SOFA score Timepoints: Baseline (on admission), then daily.	Incident MRSA colonisation no sig. difference between TTO group (17/195, 8.7%) and control group (22/196, 11.2%) on discharge/death (2.5%, p = 0.50). Incident MRSA bacteraemia occurred in no patients during the study. Maximum increase in SOFA score no sig. difference between TTO group 1.28 (1.79) and control group 1.44 (1.92) during the study (P=0.85). Multiple regression analysis showed no sig. relationship between treatment arm and incident MRSA colonization (odds ratio 0.75, 95% CI 0.36–1.56, p = 0.445). Note: which confounding variables were included in this model is unclear from text and table but likely included SOFA score, length of stay and number of invasive device changes.	Rash TTO group n = 2 (both unrelated to bodywash).	Funding: The Northern Ireland Clinical Research Support Centre Authors state no conflicts of interest.
Caelli; 2000; trial not registered Caelli et al. (2000)	RCT (parallel); Australia; Hospital	N = 30 enrolled (n = 30 analysed by ITT analysis) Age in years, TTO group mean 58, range 28–82 Control group mean 74, range 45–87 Female TTO group 10/15 Control group 8/15	Inclusion criteria: infected or colonised with MRSA. Exclusion criteria: none.	Intervention group: TTO regimen (4% TTO nasal ointment + 5% TTO bodywash), both supplied by Australian Bodycare Pty Ltd. Control group: standard care regimen (2% Mupirocin nasal ointment + Triclosan body wash) Subjects followed allocated topical regimen for at least three days. All patients infected with MRSA received intra-venous vancomycin in addition to the decolonization regimen.	MRSA decolonisation (MRSA screening at nostrils, perianal region, and any sites previously MRSA-positive) Timepoints: 48 and 96 h after the cessation of topical treatment.	MRSA decolonisation achieved in 5/15 (33%) patients in TTO group compared with 2/15 (13%) in control group (note: no statistical comparison). MRSA infection remained in 3/15 (20%) of patients in TTO group, compared with 8/15 (53%) in the control group (note: no statistical comparison).	TTO nasal ointment: mild swelling of nasal mucosa and acute burning upon application (numbers not reported). Triclosan body wash: skin tightness (reported by one patient). No adverse events reported for TTO bodywash or mupirocin nasal ointment.	Funding: (1) Cavell Trust, (2) the Rural Industries Research and Development Corporation and (3) Australian Bodycare Pty Ltd. Conflict of interest: no data provided. TTO products provided by study funder (Australian Bodycare Pty Ltd.).
Dryden; 2004; trial not registered Dryden et al. (2004)	RCT (parallel); United Kingdom; Hospital	N = 236 enrolled (n = 224 analysed) Age: not reported Female: not reported	Inclusion criteria: age ≥ 16 years; colonisation with MRSA. Exclusion criteria: known sensitivity to TTO; pregnant or lactating.	Intervention group: TTO regimen (10% TTO cream applied to nostrils three times daily and to skin lesions, wounds and ulcers once a day + 5% TTO body wash used all over body at least once daily), both supplied by Ord River Tea Tree Oil Pty Ltd, Australia. Note: TTO cream also applied to axillae and groin areas as an alternative to the TTO bodywash. Control group: standard care regimen (2% mupirocin nasal ointment i.e., Bactroban®, applied to nostrils three times daily + 4% chlorhexidine gluconate soap used all over body at least once daily + 1% silver sulfadiazine cream applied to open skin lesions and ulcers once daily). Subjects followed allocated regimen for five days.	MRSA decolonisation i.e., cleared of MRSA carriage (defined as an absence of MRSA at nostrils, throat, axillae, groin creases and any open skin lesions at both post-treatment assessment timepoints, i.e., at 2-days and 14-days post-treatment) Timepoints: Baseline, 2-days after treatment and 14-days after treatment.	MRSA decolonisation (carriage clearance) no sig. difference between TTO group 46/110 (41%) compared with control group 56/114 (49%), at 14 days post-treatment (p = 0.0286). Nasal carriage clearance sig. lower in TTO group who used 10% TTO cream (36/76) compared with control group who used mupirocin nasal ointment (58/74) at 14 days post-treatment (p = 0.0001).	None reported verbally or written in patient medical records.	Authors state no funding and no conflicts of interest. Acknowledgment that Ord River Tea Tree Oil Pty Ltd, Australia, supplied the tea tree preparations.
Lee; 2014; trial not registered Lee et al. (2014)	RCT (parallel); Hong Kong; Residential aged care facility	N = 32 enrolled (n = 32 analysed) Age in years, mean (SD) TTO group 81.0 (7.6) Control group 79.4 (6.9) Female TTO group 13/16 Control group 12/16	Inclusion criteria: ≥ stage II MRSA-colonized wounds (assessed using pressure ulcer categorization from the National Pressure Ulcer Advisory Panel); open chronic wounds with positivity in MRSA wound culture and a break of skin >6 weeks without progression to healing through normal repair processes. Exclusion criteria: peripheral vascular disease; use of systemic or topical antimicrobial agents; clinical signs of infection; more than 105 MRSA bacteria/gram of wound tissue; wounds with undermining or tunnelling; known sensitivity or allergy to TTO.	Intervention group: TTO topical solution (10% TTO, Melaleuca alternifolia), NOW Foods Company, US. Other ingredients: 90% medical grade paraffin oil. Control group: saline gauze wound dressing only. Subjects’ wound dressings were changed daily by a trained nurse. The wound was cleaned with a 0.9% saline solution. Subjects allocated to the intervention group then had the TTO solution applied. The wound was then covered with a non-adhesive pad.	MRSA decolonisation (measured as CFU/mL after incubating wound swabs aerobically at 35°C for 18-24 h on MRSASelect™ (BioRad, US) agar plates). Wound healing (assessed using the Pressure Ulcer Scale for Healing tool 3.0 with scores ranging from 0 to 16, where 0 = completely healed). Timepoints: Baseline and weeks 1, 2, 3 and 4	MRSA decolonisation – wound CFU/mL sig. lower in TTO group compared with control group at: Week 1 (4531 vs 8125, p ≤ 0.001) Week 2 (2375 vs 8937, p ≤ 0.001) Week 3 (468 vs 9875, p ≤ 0.001) Week 4 (93 vs 10312, p ≤ 0.001) Wound healing sig. better in TTO group compared with control group at: Week 1 (5.5 vs 7.6, p = 0.005) Week 2 (5.4 vs 6.9, p = 0.000) Week 3 (1.0 vs 5.5, p = 0.000) Week 4 (0.0 vs 4.6, p = 0.000)	No adverse events reported from use of 10% TTO preparation.	Authors state no funding and no conflicts of interest.
Oral Candida infection
Maghu; 2016; trial not registered Maghu et al. (2016)	RCT (parallel); India; Outpatient clinic	N = 36, although this number omits subjects who dropped out of the study (n = 36 analysed) Age in years, mean TTO group: 64.9 Control group 1: 63.2 Control group 2: 64.5 Female TTO group: Control group 1: 4/13 Control group 2: 2/10	Inclusion criteria: age 20–60 years; oral fungal infection (diagnosis confirmed by histopathological screening or lesion and any prosthesis used). Exclusion criteria: anti-fungal medication use; HIV-positive; high risk patients (e.g., leukemia or lymphoma); radiation therapy.	Intervention group: TTO mouthwash (0.25% or 5 mL TTO, botanical species not stated). Other ingredients: 50 mL water. Subjects swished with mouthwash (55 mL) three times daily after meals and were instructed to avoid eating or drinking for following 30 min. Control group 1: clotrimazole ointment; subjects applied ointment three times daily after meals. Control group 2: conservative treatment (regular cleaning and washing of the prosthesis on a daily basis and removal of prosthesis during the night in case fungal infection turned out to be denture-induced). All subjects instructed not to use any other mouthwashes or oral cleaning aids during the study.	Erythema (assessment method not defined) Inflammation (assessment method not defined) Fungal hyphae (assessment method not defined) Burning sensation (patient-assessed on VAS) Timepoints: Baseline and after weeks 1, 2 and 3.	Erythema 89% reduction in TTO group compared with 71% with clotrimazole ointment and 40% with conservative treatment. Inflammation 86% reduction in TTO group compared with 80% with clotrimazole ointment and 67% with conservative treatment. Fungal hyphae 86% reduction in TTO group compared with 100% with clotrimazole ointment and 100% with conservative treatment. Burning sensation 100% reduction in TTO group compared with 100% with clotrimazole ointment and 50% with conservative treatment. Note: no statistical comparisons were performed.	No patient reported any adverse effects in relation with the TTO mouthwash.	Funding: no data provided Authors state no conflicts of interest.

Abbreviations: ATCC, american type culture collection; ATP, adenosine triphosphate; CFU, colony forming units; CI, confidence interval; FDA, food and drug administration; ICU, intensive care unit; ISO, international organization for standardization; ITT, intention-to-treat (analysis); MRSA, methicillin-resistant Staphylococcus aureus; RCT, randomised controlled trial; SD, standard deviation; SOFA, sequential organ failure assessment; TTO, tea tree oil; US, United States; VAS, visual analogue scale.