Oral ivermectin versus 5% permethrin cream to treat children and adults with classic scabies: multicenter, assessor-blinded, cluster-randomised clinical trial
Data files
Oct 27, 2025 version files 218.39 KB
-
2025_10_24_IMPORT.txt
20.19 KB
-
2025_10_24_Primary_Outcome_Cluster.txt
10.66 KB
-
2025_10_24_Primary_Outcome_Index_Case.txt
8.64 KB
-
2025_10_24_Primary_Outcome_Individual.txt
6.67 KB
-
2025_10_24_SCRATCH_Format.txt
959 B
-
INCLUSION.csv
27.18 KB
-
ORDERCLU.csv
2.80 KB
-
ORDERIND.csv
12.30 KB
-
OUTCOME_AND_COMPLIANCE.csv
39.25 KB
-
RANDOMISATION.csv
8.99 KB
-
README.md
21.12 KB
-
RND.csv
25.47 KB
-
RNDCLU.csv
34.17 KB
Abstract
OBJECTIVE
With regards to the existing controversial data (Cochrane reviews, network meta-analysis) in classic scabies, to investigate whether oral ivermectin or topical 5% permethrin clinically cure scabies in index cases and his/her households.
DESIGN
Multicenter, assessor-blinded, cluster-randomised non-inferiority trial with a possible switch to superiority.
SETTING
28 French hospitals between January 2016 and December 2021.
PARTICIPANTS
Adults and children >15 kg, with dermoscopy-confirmed scabies—index cases—
INTERVENTIONS
Index cases were randomly assigned to either the ivermectin group or permethrin group (1:1 ratio). Each cluster member, defined as each index-case’s household, received the same treatment as the index case (except for children <15 kg, all prescribed 5% permethrin). All participants received on day (D)0 and D10, either oral 200-µg/kg ivermectin or 5% permethrin cream, applied head-to-toe.
MAIN OUTCOME MEASURES
Primary outcome was D28 clinical cure of the cluster, i.e., disappearance of scabies clinical signs/symptoms for all the cluster members. Secondary outcomes included index-case and individual-level analyses and safety. Blinded Dermatologists were used as assessors.
RESULTS
The main analysis included 507 ivermectin-treated participants, comprising 142 index cases-households, and 568 permethrin-treated participants, comprising 147 index cases-households. Their respective cluster-level cure rates were 71.8% vs. 88.5% (–16.7 percentage-point difference, [95% CI, –26.3 to –7.1]. Secondary outcome percentage-point differences also showed ivermectin inferiority to 5% permethrin for index cases** **76.6% vs. 91.5%; –14.9 percentage-point difference [–23.6 to –6.2] and participants: 85.3% vs. 94.2%; –9.2 percentage-point difference [–14.9 to –3.5]), respectively. Intraclass-correlation coefficients were higher for permethrin than ivermectin for all clusters (0.68 [0.61 to 0.75] vs. 0.46 [0.37 to 0.56] and cluster size >1 (0.67 [0.60 to 0.74] vs. 0.47 [0.37 to 0.56]). Cutaneous adverse events occurred in 11.9% and 15.6% of ivermectin and permethrin participants, respectively.
CONCLUSIONS
In this first cluster-randomised trial in dermoscopy-confirmed classic scabies designed, we did not demonstrate the non-inferiority of D0+D10 oral ivermectin to 5% permethrin cream to obtain D28 clinical scabies cures of index cases-households. Conversely, a statistical superiority of 5% permethrin was shown.
Dataset DOI: 10.5061/dryad.sn02v6xjd
Description of the data and file structure
Paper: Oral Ivermectin versus 5% Permethrin Cream to Treat Children and Adults with Classic Scabies: multicenter, assessor-blinded, cluster-randomised clinical trial
Authors: Franck Boralevi (deceased) et al
Corresponding author: Olivier Chosidow
Contact: Bruno Giraudeau bruno.giraudeau@univ-tours.fr
Description: This README file describes the data package accompanying the above publication.
The SCRATCH study was a French randomized trial. Data were collected using a case report form (CRF).
Files:
Data bases: there are 7 data files. The first 3 data bases correspond do data collected via the CRF while the trial was ongoing. Some comments have to be made:
- only some of the collected data are in the enclosed data bases, i.e., those necessary to reproduce the results reported in Table 3 of the manuscript that is to say the results related to the outcome
- there is no direct identifiers in the datasets. Regarding indirect identifiers, we do need to keep the sex variable (which is used in the multiple imputation approach), we also do need to keep a cluster (i.e. household) identifier, which is necessary for the statistical analysis. Other indirect identifiers have been dropped out
- participants identifiers have been changed
The RANDOMISATION.csv file corresponds to the a priori randomisation sequences (i.e., which were generated before the beginning of the trial); the OUTCOME_and_COMPLIANCE.csv file corresponds to the a posteriori adjudicated primary outcome and compliance status (i.e., which were derived once the trial was finished). As for tables ORDERCLU.csv and ORDERIND.csv they are useful to re-run the statistical analysis. Indeed, subject and cluster identifications have been changed (to avoid any possibility to re-identify a participant), thus leading to changes in the way tables are ordered when used for the statistical analysis. However, the statistical analysis uses the SAS PROC MI which is sensitive to the order among participants. Differences are very small, but to recover the exact figures reported in the article, we considered it was better to have data sorted in the same way as they have been initially analysed. The two tables ORDERCLU.csv and ORDERIND.cvs allow achieving this.
Missing data were not replaced. This allows running the SAS program (enclosed) without any trouble.
Under "Format", the number after $ shows the maximum length (in characters) of the variable.
Files and variables
1. RND.csv: data associated to the index case characteristics
| # | Variable | Type | Format | Label |
| 1 | clePat | Texte | $6. | Patient identification number |
| 2 | rndNbPerClus | Num. | NBPERS. | Number of persons in the cluster (including the index case) |
| 3 | rndPatTyp | Num. | TYPPAT. | Type of patient |
| 4 | rndAgeMoi | Num. | 4. | Age of the index case (months) |
| 5 | rndAgeAnn | Num. | 3. | Age of the index case (years) |
| 6 | rndDelAppSigGal | Num. | 3. | Delay in onset of first scabies symptoms (days) |
| 7 | rndDelAppSigGalNa | Num. | COCHE. | Delay in onset of first scabies symptoms: NA (if asymptomatic) |
| 8 | rndTttGal12Moi | Num. | YESNO. | Scabies treatment in the past 12 months |
| 9 | rndTttGalNb | Num. | NBTRA. | Number of scabies treatments in the past 12 months |
| 10 | rndTttGalTypIver | Num. | COCHE. | Scabies treatments in the past 12 months: Ivermectin |
| 11 | rndTttGalTypPerm | Num. | COCHE. | Scabies treatments in the past 12 months: Permethrin |
| 12 | rndTttGalTypBenz | Num. | COCHE. | Scabies treatments in the past 12 months: Benzyl benzoate |
| 13 | rndTttGalTypEsde | Num. | COCHE. | Scabies treatments in the past 12 months: Esdepallethrin |
| 14 | rndTttGalTypCrot | Num. | COCHE. | Scabies treatments in the past 12 months: Crotamiton |
| 15 | rndTttGalTypAut | Num. | COCHE. | Scabies treatments in the past 12 months: Other |
| 16 | rndTttGalAutPrec | Texte | $255. | Scabies treatments in the past 12 months: Other (specify) |
| 17 | rndAtcdAllIve | Num. | YESNO. | History of allergy to ivermectin |
| 18 | rndAtcdAllPer | Num. | YESNO. | History of allergy to permethrin |
| 19 | rndAtcdInsHep | Num. | YESNO. | History of hepatic failure |
| 20 | rndAtcdInsRen | Num. | YESNO. | History of renal failure |
| 21 | rndExaPhyEcz | Num. | YESNO. | Eczema |
| 22 | rndExaPhyImp | Num. | YESNO. | Impetiginization |
| 23 | rndCriIncAge | Num. | YESNO. | Child or adult |
| 24 | rndCriIncPoi | Num. | YESNO. | Weight of the child (index case) greater than 15 kg |
| 25 | rndCriIncExaDerPos | Num. | YESNO. | Positive dermoscopic examination |
| 26 | rndCriIncAffSecu | Num. | YESNO. | Child covered by the national health insurance |
| 27 | rndCriIncProSuiEtu | Num. | YESNO. | Relatives willing to complete the full study |
| 28 | rndCriIncConSign | Num. | YESNO. | Written informed consent signed by persons with parental authority |
| 29 | rndCriExAge | Num. | YESNO. | Child (index case) younger than 2 months |
| 30 | rndCriExAllIver | Num. | YESNO. | Known allergy to ivermectin or any of its components |
| 31 | rndCriExAllPerm | Num. | YESNO. | Known allergy to permethrin or any of its components |
| 32 | rndCriExTttAntSca | Num. | YESNO. | Antiscabies treatment less than 4 weeks ago |
| 33 | rndCriExEczImpDif | Num. | YESNO. | Diffuse eczema or impetiginization |
| 34 | rndCriExInsHepRen | Num. | YESNO. | Known severe hepatic or renal failure |
| 35 | rndCriIncClusNb | Num. | YESNO. | Cluster composed of no more than 6 persons |
| 36 | rndCriIncClusAge | Num. | YESNO. | Youngest member aged at least 2 months |
| 37 | rndCriIncClusAffSecu | Num. | YESNO. | National health insurance coverage for all cluster members |
| 38 | rndCriIncClusConSign | Num. | YESNO. | Written informed consent signed by all adult cluster members |
| 39 | rndCriIncClusSuiEtu | Num. | YESNO. | Willingness of all cluster members to complete the full study |
| 40 | rndCriExClusAge | Num. | YESNO. | One of the members younger than 2 months |
| 41 | rndCriExClusAllIver | Num. | YESNO. | Known allergy to ivermectin or any of its components |
| 42 | rndCriExClusAllPerm | Num. | YESNO. | Known allergy to permethrin or any of its components |
| 43 | rndCriExClusTttAntSca | Num. | YESNO. | Antiscabies treatment less than 4 weeks ago |
| 44 | rndCriExClusEczImpDif | Num. | YESNO. | Diffuse eczema or impetiginization |
| 45 | rndCriExClusInsHepRen | Num. | YESNO. | Known severe hepatic or renal failure |
| 46 | rndAttBraTtt | Num. | TTT. | Treatment arm assignment |
2. RNDCLU.csv: data associated to the cluster members characteristics
| # | Variable | Type | Format | Label |
| 1 | clePat | Texte | $6. | Patient identification number |
| 2 | rndcluPatPresMem | Num. | PRESABS. | Presence of cluster member at visit |
| 3 | rndcluPatTyp | Num. | TYPPAT. | Type of patient |
| 4 | rndcluNum | Texte | $6. | Cluster member identification number |
| 5 | rndcluAgeAnn | Num. | 3. | Age (in years) |
| 6 | rndcluAgeMoi | Num. | 2. | Age (in months) |
| 7 | rndcluDelAppSigGal | Num. | 3. | Time to onset of first scabies symptoms (in days) |
| 8 | rndcluDelAppSigGalNa | Num. | COCHE. | Time to onset of first scabies symptoms – not applicable (if asymptomatic) |
| 9 | rndcluTttGal12Moi | Num. | YESNO. | Treatment for scabies in the past 12 months |
| 10 | rndcluTttGalNb | Num. | NBTRA. | Number of treatments for scabies in the past 12 months |
| 11 | rndcluTttGalTypIver | Num. | COCHE. | Treatments for scabies in the past 12 months – Ivermectin |
| 12 | rndcluTttGalTypPerm | Num. | COCHE. | Treatments for scabies in the past 12 months – Permethrin |
| 13 | rndcluTttGalTypBenz | Num. | COCHE. | Treatments for scabies in the past 12 months – Benzyl benzoate |
| 14 | rndcluTttGalTypEsde | Num. | COCHE. | Treatments for scabies in the past 12 months – Esdepallethrin |
| 15 | rndcluTttGalTypCrot | Num. | COCHE. | Treatments for scabies in the past 12 months – Crotamiton |
| 16 | rndcluTttGalTypAut | Num. | COCHE. | Treatments for scabies in the past 12 months – Other |
| 17 | rndcluTttGalAutPrec | Texte | $255. | Treatments for scabies in the past 12 months – Other (specify) |
| 18 | rndcluAtcdAllIve | Num. | YESNO. | History of allergy to ivermectin |
| 19 | rndcluAtcdAllPer | Num. | YESNO. | History of allergy to permethrin |
| 20 | rndcluAtcdInsHep | Num. | YESNO. | History of hepatic insufficiency |
| 21 | rndcluAtcdInsRen | Num. | YESNO. | History of renal insufficiency |
| 22 | rndcluExaPhyEcz | Num. | YESNO. | Eczematization |
| 23 | rndcluExaPhyImp | Num. | YESNO. | Impetiginization |
3. INCLUSION.csv: data collected during the inclusion visit for both the index cases and cluster members
| # | Variable | Type | Format | Label |
| 1 | ClePat | 10 | $6. | Patient identification number |
| 2 | IncSex | 8 | SEX. | Sex |
| 3 | IncCluPresMem | 8 | PRESABS. | Household member present or absent |
| 4 | Age | 8 | BEST12. | Age |
| 5 | Prurit | 8 | YESNO. | Presence of prurit |
4. RANDOMISATION.csv: this file displays the randomisation sequences
| # | Variable | Type | Format | Label |
| 1 | listRanTtt | Texte | $12. | Treatment A or B |
| 2 | listRanProd | Texte | $11. | Treatment name |
| 3 | ClePat | Texte. | $4. | Patient identification number |
5. OUTCOME_AND_COMPLIANCE.csv: this file displays the adjudicated primary outcome as well as the compliance status
| # | Variable | Type | Format | Label |
| 1 | ClePat | Texte | $6. | Patient identification number |
| 2 | Cluster | Num. | BEST12. | Cluster identification number |
| 3 | Compliance_Cluster | Texte | $4. | Compliance status - Cluster level |
| 4 | Compliance_Individuel | Texte | $9. | Compliance status - Individual level |
| 5 | Outcome_Cluster | Texte | $5. | Adjudicated primary outcome - Cluster level |
| 6 | Outcome_Individuel | Texte | $5. | Adjudicated primary outcome - Individual level |
6. ORDERCLU.csv: this file displays the original order among clusters, thus allowing to recover the exact result reported in the article. Indeed, the SAS MI procedure is slightly sensitive to the order
| # | Variable | Type | Format | Label |
| 1 | ClePat | Texte | $4. | Patient identification number |
| 2 | Order | Num. | BEST12. | Initial cluster order |
7. ORDERIND.csv: this file displays the original order among individuals, thus allowing to recover the exact result reported in the article. Indeed, the SAS MI procedure is slightly sensitive to the order
| # | Variable | Type | Format | Label |
| 1 | ClePat | Texte | $4. | Patient identification number |
| 2 | Order | Num. | BEST12. | Initial individual order |
Code/software
The statistical analysis of the SCRATCH trial has been performed using SAS. Below are the SAS formats of the variables previously listed. These formats are also in the 2025_10_24_SCRATCH_Format.txt SAS program.
| value yesno (NOTSORTED) | |
| 0 = "No" | |
| 1 = "Yes" | |
| . = "Unknown"; | |
| value ttt (NOTSORTED) | |
| 1 = "Ivermectin" | |
| 2 = "Permethrin" | |
| . = "Unknown"; | |
| value coche (NOTSORTED) | |
| 1 = "Yes" | |
| . = "Unchecked"; | |
| value nbPers (NOTSORTED) | |
| 2 = "2" | |
| 3 = "3" | |
| 4 = "4" | |
| 5 = "5" | |
| 6 = "6" | |
| 7 = "7" | |
| 8 = "8" | |
| 9 = "9" | |
| 10 = "10" | |
| . = "Unknown"; | |
| value typPat (NOTSORTED) | |
| 1 = "Child" | |
| 2 = "Adult" | |
| . = "Unknown"; | |
| value nbTra (NOTSORTED) | |
| 1 = "1" | |
| 2 = "2" | |
| 3 = "3 - 4" | |
| 4 = "5 and more" | |
| . = "Unknown"; | |
| value presAbs (NOTSORTED) | |
| 0 = "Absent" | |
| 1 = "Present" | |
| . = "Unknown"; | |
| value sex (NOTSORTED) | |
| 1 = "Male" | |
| 2 = "Female" | |
| . = "Unknown"; |
Finally, the following SAS programs are provided:
1. 2025_10_24_IMPORT.txt: this program allows creating SAS databases using the 12 data bases and the program 2025_10_24_SCRATCH_Formats.txt
2. 2025_10_24_Primary_Outcome_Cluster.txt
2025_10_24_Primary_Outcome_Index_Case.txt
2025_10_24_Primary_Outcome_Individual.txt
These 3 files report the SAS statistical program used to obtain the main results of the referenced publication (reported in Table 3)
To reproduce the statistical analysis:
1. Create a directory in which you upload all the programs
2. Change the extension of the 2025_10_24_SCRATCH_Format.txt file such that it becomes a SAS program
3. Change the extension of the 2025_10_24_IMPORT.txt file such that it becomes a SAS program. Run it, such that the SAS tables are created
/!\ Be aware that directories have to be changed in the program
4. Change the extension of the 2025_10_24_Primary_Outcome_Cluster.txt and run it to obtain the cluster level analysis result
/!\ Be aware that directories have to be changed in the program
5. Change the extension of the 2025_10_24_Primary_Outcome_Index_Case.txt and run it to obtain the index case analysis result
/!\ Be aware that directories have to be changed in the program
6. Change the extension of the 2025_10_24_Primary_Outcome_Individual.txt and run it to obtain the individual level analysis result
/!\ Be aware that directories have to be changed in the program
Human subjects data
We confirm that we received informed consent from all participants. There is no identifiable information in the data sets which is a requirement to conduct clinical research in France. We changed identification numbers and drop all the days from the data bases.