library(DataExplorer)
library(janitor)##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testlibrary(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(tidyverse)## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ forcats 1.0.0 ✔ readr 2.1.5
## ✔ ggplot2 3.5.1 ✔ stringr 1.5.1
## ✔ lubridate 1.9.3 ✔ tibble 3.2.1
## ✔ purrr 1.0.2 ✔ tidyr 1.3.1## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(survival)
library(ggplot2)
library(survminer)## Loading required package: ggpubr
##
## Attaching package: 'survminer'
##
## The following object is masked from 'package:survival':
##
## myelomalibrary(gtsummary)
library(broom)
library(knitr)
library(GGally)## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2library(readxl)
library(lme4)## Loading required package: Matrix
##
## Attaching package: 'Matrix'
##
## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpacklibrary(cards)
library(cardx)
library(broom)
setwd(dir = "/Users/bellier_pro/Dropbox/Medecine/Anatomie/Projet_pedagogique/ECOS_VR/") # Define the directory
library(readxl)
ecos_vr <- read_excel("ecos_vr.xlsx") # Open the file
ecos_vr <- ecos_vr[,-1:-2]
ecos_vr <- ecos_vr[,-10]
ecos_vr$note_station <- as.numeric(ecos_vr$note_station)ecos_vr %>%
tbl_summary(
by = groupe,
statistic = list(
all_continuous() ~ "{mean} ({sd})",
all_categorical() ~ "{n} ({p}%)"
),
digits = all_continuous() ~ 1,
missing_text = "(Données manquantes)"
) %>%
add_p(pvalue_fun = ~style_pvalue(.x, digits = 3)) %>% # Remplace add_difference()
add_n() %>%
add_stat_label() %>%
modify_header(label ~ "**Variable**") %>%
modify_spanning_header(c("stat_1", "stat_2", "stat_3") ~ "**Groupes**") %>%
bold_labels()| Variable | N |
Groupes
|
p-value1 | ||
|---|---|---|---|---|---|
| Groupe 1 N = 40 |
Groupe 2 N = 30 |
Groupe 3 N = 32 |
|||
| id, Mean (SD) | 102 | 56.7 (29.9) | 45.0 (28.7) | 51.2 (29.7) | 0.262 |
| annee, n (%) | 102 | 0.439 | |||
| DFASM1 (D2) | 21 (53%) | 8 (27%) | 12 (38%) | ||
| DFASM2 (D3) | 10 (25%) | 10 (33%) | 10 (31%) | ||
| DFASM3 (D4) | 6 (15%) | 6 (20%) | 5 (16%) | ||
| DFGSM3 (D1) | 3 (7.5%) | 6 (20%) | 5 (16%) | ||
| vu_pl, n (%) | 102 | >0.999 | |||
| Non | 6 (15%) | 4 (13%) | 4 (13%) | ||
| Oui | 34 (85%) | 26 (87%) | 28 (88%) | ||
| cb_vu, n (%) | 89 | 0.584 | |||
| <3 | 20 (59%) | 12 (46%) | 18 (62%) | ||
| >5 | 7 (21%) | 4 (15%) | 4 (14%) | ||
| Entre 3 et 5 | 7 (21%) | 10 (38%) | 7 (24%) | ||
| (Données manquantes) | 6 | 4 | 3 | ||
| realise_pl, n (%) | 102 | 0.427 | |||
| Non | 24 (60%) | 14 (47%) | 15 (47%) | ||
| Oui | 16 (40%) | 16 (53%) | 17 (53%) | ||
| cb_realise, n (%) | 49 | 0.358 | |||
| <3 | 9 (56%) | 13 (81%) | 11 (65%) | ||
| >5 | 2 (13%) | 2 (13%) | 4 (24%) | ||
| Entre 3 et 5 | 5 (31%) | 1 (6.3%) | 2 (12%) | ||
| (Données manquantes) | 24 | 14 | 15 | ||
| type_aiguille, n (%) | 49 | 0.592 | |||
| Aiguille normal | 4 (25%) | 4 (25%) | 6 (35%) | ||
| Aiguilles atraumatique | 9 (56%) | 10 (63%) | 6 (35%) | ||
| Les deux | 3 (19%) | 2 (13%) | 5 (29%) | ||
| (Données manquantes) | 24 | 14 | 15 | ||
| cb_reussi, n (%) | 49 | 0.631 | |||
| <3 | 11 (69%) | 14 (88%) | 12 (71%) | ||
| >5 | 2 (13%) | 0 (0%) | 1 (5.9%) | ||
| Entre 3 et 5 | 3 (19%) | 2 (13%) | 4 (24%) | ||
| (Données manquantes) | 24 | 14 | 15 | ||
| stage_pl, n (%) | 102 | 0.273 | |||
| Non | 26 (65%) | 15 (50%) | 22 (69%) | ||
| Oui | 14 (35%) | 15 (50%) | 10 (31%) | ||
| formation_pl, n (%) | 102 | 0.658 | |||
| Non | 8 (20%) | 4 (13%) | 4 (13%) | ||
| Oui | 32 (80%) | 26 (87%) | 28 (88%) | ||
| ecos_pl, n (%) | 102 | 0.705 | |||
| Non | 31 (78%) | 22 (73%) | 22 (69%) | ||
| Oui | 9 (23%) | 8 (27%) | 10 (31%) | ||
| note_station, n (%) | 18 | 0.183 | |||
| 5 | 1 (13%) | 0 (0%) | 0 (0%) | ||
| 6 | 0 (0%) | 0 (0%) | 1 (17%) | ||
| 8 | 1 (13%) | 0 (0%) | 2 (33%) | ||
| 8.5 | 0 (0%) | 0 (0%) | 1 (17%) | ||
| 9 | 2 (25%) | 0 (0%) | 0 (0%) | ||
| 12 | 2 (25%) | 0 (0%) | 1 (17%) | ||
| 14 | 2 (25%) | 3 (75%) | 0 (0%) | ||
| 17.9 | 0 (0%) | 1 (25%) | 0 (0%) | ||
| 20 | 0 (0%) | 0 (0%) | 1 (17%) | ||
| (Données manquantes) | 32 | 26 | 26 | ||
| sexe, n (%) | 102 | 0.571 | |||
| Femme | 29 (73%) | 22 (73%) | 20 (63%) | ||
| Homme | 11 (28%) | 8 (27%) | 12 (38%) | ||
| stai_trait, Mean (SD) | 101 | 40.3 (8.8) | 44.6 (8.3) | 41.9 (8.6) | 0.080 |
| (Données manquantes) | 0 | 1 | 0 | ||
| stai_etat, Mean (SD) | 101 | 37.1 (9.5) | 39.6 (8.7) | 38.7 (10.5) | 0.362 |
| (Données manquantes) | 0 | 1 | 0 | ||
| note_ecos, Mean (SD) | 102 | 15.1 (2.7) | 15.0 (3.6) | 13.5 (4.4) | 0.287 |
| stai_etat2, Mean (SD) | 102 | 37.4 (9.8) | 41.0 (10.6) | 39.9 (8.6) | 0.063 |
| delta_stai_etat, Mean (SD) | 102 | 0.3 (11.9) | 2.8 (9.3) | 1.3 (8.3) | 0.456 |
| 1 Kruskal-Wallis rank sum test; Fisher’s exact test; Pearson’s Chi-squared test | |||||
### Pour la note aux ECOS tests
library(lmerTest) ##
## Attaching package: 'lmerTest'## The following object is masked from 'package:lme4':
##
## lmer## The following object is masked from 'package:stats':
##
## step# Filtrage des lignes complètes sur les variables du modèle
ecos_model <- ecos_vr %>%
filter(!is.na(note_ecos),
!is.na(groupe),
!is.na(vu_pl),
!is.na(realise_pl),
!is.na(formation_pl),
!is.na(stage_pl),
!is.na(type_aiguille),
!is.na(annee)) %>%
mutate(across(c(groupe, vu_pl, realise_pl, formation_pl, stage_pl, type_aiguille, annee), as.factor))
# Modèle de régression à ordonnée à l'origine aléatoire
lrm_note_ecos <- lmer(note_ecos ~ groupe + formation_pl + stage_pl +
type_aiguille + sexe +
(1 | annee), data = ecos_model)
summary(lrm_note_ecos)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: note_ecos ~ groupe + formation_pl + stage_pl + type_aiguille +
## sexe + (1 | annee)
## Data: ecos_model
##
## REML criterion at convergence: 218.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.68701 -0.58963 -0.03005 0.53473 1.98386
##
## Random effects:
## Groups Name Variance Std.Dev.
## annee (Intercept) 8.635 2.938
## Residual 6.316 2.513
## Number of obs: 49, groups: annee, 4
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) 12.2284 1.9875 6.5866 6.153
## groupeGroupe 2 0.5617 0.9151 38.0208 0.614
## groupeGroupe 3 -0.2965 0.9451 37.6536 -0.314
## formation_plOui 2.0535 0.8990 37.8856 2.284
## stage_plOui 0.8978 0.8663 37.6857 1.036
## type_aiguilleAiguilles atraumatique -1.9500 0.9499 39.7689 -2.053
## type_aiguilleLes deux 0.3831 1.2069 39.4389 0.317
## sexeHomme 1.4256 0.8929 37.9856 1.597
## Pr(>|t|)
## (Intercept) 0.000592 ***
## groupeGroupe 2 0.542940
## groupeGroupe 3 0.755418
## formation_plOui 0.028045 *
## stage_plOui 0.306642
## type_aiguilleAiguilles atraumatique 0.046717 *
## type_aiguilleLes deux 0.752589
## sexeHomme 0.118638
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) grpGr2 grpGr3 frmt_O stg_pO typ_Aa typ_Ld
## groupeGrop2 -0.192
## groupeGrop3 -0.245 0.469
## formatn_plO -0.316 -0.155 -0.128
## stage_plOui -0.307 0.080 0.252 -0.079
## typ_gllAgla -0.309 -0.009 0.089 0.081 -0.058
## typ_gllLsdx -0.177 0.077 -0.113 -0.056 -0.194 0.561
## sexeHomme -0.222 0.079 -0.137 -0.012 0.163 0.075 0.245anova(lrm_note_ecos)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## groupe 5.308 2.654 2 37.837 0.4203 0.65991
## formation_pl 32.956 32.956 1 37.886 5.2181 0.02804 *
## stage_pl 6.783 6.783 1 37.686 1.0740 0.30664
## type_aiguille 46.514 23.257 2 38.845 3.6824 0.03432 *
## sexe 16.100 16.100 1 37.986 2.5492 0.11864
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Tests post-hoc
library(emmeans)## Welcome to emmeans.
## Caution: You lose important information if you filter this package's results.
## See '? untidy'##
## Attaching package: 'emmeans'## The following object is masked from 'package:GGally':
##
## pigsemmeans_ecos <- emmeans(lrm_note_ecos, pairwise ~ groupe, adjust="bonferroni") # Estimer les moyennes marginales par groupe
emmeans_ecos$contrasts # Résultats des comparaisons par paires avec correction de multiplicité (par Bonferroni)## contrast estimate SE df t.ratio p.value
## Groupe 1 - Groupe 2 -0.562 0.918 38.4 -0.612 1.0000
## Groupe 1 - Groupe 3 0.297 0.946 38.1 0.314 1.0000
## Groupe 2 - Groupe 3 0.858 0.961 38.3 0.893 1.0000
##
## Results are averaged over the levels of: formation_pl, stage_pl, type_aiguille, sexe
## Degrees-of-freedom method: kenward-roger
## P value adjustment: bonferroni method for 3 testsemmeans_ecos$emmeans # Afficher les moyennes ajustées## groupe emmean SE df lower.CL upper.CL
## Groupe 1 13.9 1.69 3.97 9.17 18.6
## Groupe 2 14.5 1.76 4.42 9.75 19.2
## Groupe 3 13.6 1.67 3.73 8.81 18.4
##
## Results are averaged over the levels of: formation_pl, stage_pl, type_aiguille, sexe
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95### STAI
library(reshape2)##
## Attaching package: 'reshape2'## The following object is masked from 'package:tidyr':
##
## smithsecos_vr_long <- melt(ecos_vr,
id.vars=c("stai_trait","groupe","formation_pl","stage_pl", "annee","sexe", "id"),
measure.vars=c("stai_etat", "stai_etat2"),
variable.name="temps_stai_etat",
value.name="stai_etat") # Passer en format long plutôt que large
lrm_stai_etat <- lmer(stai_etat ~ groupe*temps_stai_etat +
formation_pl + stage_pl + sexe + (1 | id),
data = ecos_vr_long)
anova(lrm_stai_etat)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## groupe 181.960 90.980 2 87.365 1.6562 0.19680
## temps_stai_etat 70.524 70.524 1 89.892 1.2838 0.26020
## formation_pl 262.826 262.826 1 86.809 4.7845 0.03140 *
## stage_pl 34.534 34.534 1 87.348 0.6287 0.43000
## sexe 192.615 192.615 1 87.053 3.5064 0.06449 .
## groupe:temps_stai_etat 24.121 12.061 2 89.861 0.2196 0.80331
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# Tests post-hoc
emmeans_stai <- emmeans(lrm_stai_etat, pairwise ~ groupe, adjust="bonferroni") # Estimer les moyennes marginales par groupe## NOTE: Results may be misleading due to involvement in interactionsemmeans_stai$contrasts # Résultats des comparaisons par paires avec correction de multiplicité (par Bonferroni)## contrast estimate SE df t.ratio p.value
## Groupe 1 - Groupe 2 -3.241 1.97 96.4 -1.648 0.3080
## Groupe 1 - Groupe 3 -2.702 1.92 95.5 -1.409 0.4861
## Groupe 2 - Groupe 3 0.539 2.07 96.3 0.260 1.0000
##
## Results are averaged over the levels of: temps_stai_etat, formation_pl, stage_pl, sexe
## Degrees-of-freedom method: kenward-roger
## P value adjustment: bonferroni method for 3 testsemmeans_stai$emmeans # Afficher les moyennes ajustées## groupe emmean SE df lower.CL upper.CL
## Groupe 1 37.8 1.48 95.6 34.8 40.7
## Groupe 2 41.0 1.70 96.2 37.6 44.4
## Groupe 3 40.5 1.66 95.6 37.2 43.8
##
## Results are averaged over the levels of: temps_stai_etat, formation_pl, stage_pl, sexe
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95