tesis-cicatrización.knit

library(readxl)
TES <- read_excel("C:/Users/Administrador/Downloads/TES.xlsx", 
    sheet = "R")
View(TES)

library(tidyverse)

## Warning: package 'ggplot2' was built under R version 4.4.3

## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   4.0.0     ✔ tibble    3.2.1
## ✔ lubridate 1.9.3     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── 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 errors

library(knitr)
library(lme4)

## Cargando paquete requerido: Matrix
## 
## Adjuntando el paquete: 'Matrix'
## 
## The following objects are masked from 'package:tidyr':
## 
##     expand, pack, unpack

library(sjPlot)

## Learn more about sjPlot with 'browseVignettes("sjPlot")'.
## 
## Adjuntando el paquete: 'sjPlot'
## 
## The following object is masked from 'package:ggplot2':
## 
##     set_theme

library(performance)

## Warning: package 'performance' was built under R version 4.4.3

library(lmerTest)

## Warning: package 'lmerTest' was built under R version 4.4.3

## 
## Adjuntando el paquete: 'lmerTest'
## 
## The following object is masked from 'package:lme4':
## 
##     lmer
## 
## The following object is masked from 'package:stats':
## 
##     step

#library(rworldmap) # no es necesario 
#library(countrycode) # no es necesario
library(dplyr)
library(performance)
library(DHARMa)

## Warning: package 'DHARMa' was built under R version 4.4.3

## This is DHARMa 0.4.7. For overview type '?DHARMa'. For recent changes, type news(package = 'DHARMa')

library(ggeffects)
library(naniar)

## Warning: package 'naniar' was built under R version 4.4.3

options(scipen=999, digits=2)

library(tidyverse)
library(marginaleffects)

## Warning: package 'marginaleffects' was built under R version 4.4.3

library(data.table)

## Warning: package 'data.table' was built under R version 4.4.3

## 
## Adjuntando el paquete: 'data.table'

## The following objects are masked from 'package:lubridate':
## 
##     hour, isoweek, mday, minute, month, quarter, second, wday, week,
##     yday, year

## The following objects are masked from 'package:dplyr':
## 
##     between, first, last

## The following object is masked from 'package:purrr':
## 
##     transpose

library(haven)
library(lme4)
library(geepack)

## Warning: package 'geepack' was built under R version 4.4.3

library(performance)
library(psych)

## 
## Adjuntando el paquete: 'psych'

## The following objects are masked from 'package:ggplot2':
## 
##     %+%, alpha

library(sjPlot)
library(lmerTest)
library(sjstats)

## 
## Adjuntando el paquete: 'sjstats'

## The following object is masked from 'package:psych':
## 
##     phi

## The following objects are masked from 'package:performance':
## 
##     icc, r2

vis_miss(TES)

TES <- TES %>%
  filter(!if_any(16:25, is.na))

TES[sapply(TES, is.character)] <- lapply(TES[sapply(TES, is.character)], as.factor)
TES$Cicatrización <- as.factor(TES$Cicatrización)


TES <- TES %>%
  mutate(across(16:25, as.factor))
TES <- TES %>%
  mutate(across(5:13, as.factor))
TES <- TES %>%
  mutate(across(OM, as.factor))

table(TES$Cicatrización)

## 
##   0   1 
## 124 175

TES$cicat=TES$Cicatrización
modelo1 <- glm(cicat ~ loca+ topo+nume+ isq+ infe+ ede+ neur+ tisu+ area+ cica , family = "binomial", data = TES)
t1<-tab_model(modelo1, show.se = T, show.dev = T, show.loglik = T, show.intercept = F,show.stat = F, show.reflvl = T, digits = 2, digits.p = 3, show.r2 = F,collapse.ci = F, title = "Cicatrización")
knitr::asis_output(t1$knitr)

Cicatrización
	cicat
Predictors	Odds Ratios	std. Error	CI	p
area2	1.75	0.70	0.81 – 3.91	0.161
area3	2.00	1.32	0.55 – 7.48	0.294
cica2	0.15	0.18	0.01 – 1.04	0.101
cica3	0.34	0.38	0.02 – 2.27	0.341
ede1	1.83	0.82	0.76 – 4.42	0.177
ede2	1.13	0.55	0.44 – 2.93	0.796
ede3	0.53	0.35	0.14 – 1.88	0.334
infe1	0.48	0.27	0.15 – 1.44	0.191
infe2	0.36	0.21	0.11 – 1.14	0.087
infe3	0.21	0.15	0.05 – 0.87	0.033
isq1	0.61	0.24	0.28 – 1.32	0.205
isq2	0.28	0.11	0.13 – 0.61	0.001
isq3	0.17	0.08	0.07 – 0.43	<0.001
loca2	0.61	0.21	0.31 – 1.21	0.157
loca3	0.30	0.16	0.10 – 0.85	0.025
neur1	0.23	0.29	0.01 – 1.86	0.242
neur2	0.25	0.31	0.01 – 1.90	0.258
neur3	0.22	0.30	0.01 – 2.37	0.268
nume2	0.14	0.06	0.06 – 0.32	<0.001
nume3	0.16	0.10	0.05 – 0.51	0.003
tisu2	0.52	0.27	0.18 – 1.40	0.202
tisu3	0.68	0.43	0.19 – 2.36	0.542
topo2	1.42	0.59	0.63 – 3.26	0.402
topo3	1.09	0.46	0.48 – 2.50	0.844
Observations	299
Deviance	309.410
log-Likelihood	-154.705

modelo2 <- glm(cicat ~ loca , family = "binomial", data = TES)
modelo3 <- glm(cicat ~  topo , family = "binomial", data = TES)
modelo4 <- glm(cicat ~ nume , family = "binomial", data = TES)
modelo5 <- glm(cicat ~  isq , family = "binomial", data = TES)
modelo6<- glm(cicat ~  infe, family = "binomial", data = TES)
modelo7<- glm(cicat ~  ede , family = "binomial", data = TES)
modelo8<- glm(cicat ~  neur , family = "binomial", data = TES)
modelo9<- glm(cicat ~  tisu , family = "binomial", data = TES)
modelo10<- glm(cicat ~  area , family = "binomial", data = TES)
modelo6<- glm(cicat ~  cica, family = "binomial", data = TES)

library(broom)

## 
## Adjuntando el paquete: 'broom'

## The following object is masked from 'package:sjstats':
## 
##     bootstrap

library(dplyr)

# Crear una lista con los modelos
modelos <- list(
  modelo2, modelo3, modelo4, modelo5,
  modelo6, modelo7, modelo8, modelo9,
  modelo10, modelo6
)

# Convertir a tabla ordenada
tabla_coef <- lapply(modelos, tidy) %>%
  bind_rows(.id = "modelo")

# Ver la tabla
tabla_coef

## # A tibble: 33 × 6
##    modelo term        estimate std.error statistic       p.value
##    <chr>  <chr>          <dbl>     <dbl>     <dbl>         <dbl>
##  1 1      (Intercept)   0.595      0.162     3.67  0.000242     
##  2 1      loca2        -0.494      0.258    -1.91  0.0559       
##  3 1      loca3        -0.713      0.380    -1.88  0.0605       
##  4 2      (Intercept)   0.606      0.179     3.38  0.000730     
##  5 2      topo2        -0.0596     0.322    -0.185 0.853        
##  6 2      topo3        -0.703      0.267    -2.64  0.00836      
##  7 3      (Intercept)   0.877      0.151     5.80  0.00000000657
##  8 3      nume2        -1.62       0.317    -5.11  0.000000314  
##  9 3      nume3        -2.02       0.460    -4.40  0.0000108    
## 10 4      (Intercept)   0.820      0.173     4.73  0.00000220   
## # ℹ 23 more rows

table(TES$topo)

## 
##   1   2   3 
## 136  60 103

#Isquemia
TES%>%
  dplyr::group_by(isq) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = isq, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(isq) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = isq, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "isquemia", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(isq_factor = factor(if_else(isq %in% c(2, 3), "SI", "NO/leve")),
         isq_factor = factor(isq_factor, levels = rev(levels(isq_factor))))

table(TES$isq)

## 
##   0   1   2   3 
## 157  56  50  36

table(TES$isq_factor)

## 
##      SI NO/leve 
##      86     213

TES%>%
  dplyr::group_by(loca) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = loca, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(loca) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = loca, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "localizacion", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(loca_factor = factor(if_else(loca %in% c(2, 3), "Met/Tar", "Fal")),
         loca_factor = factor(loca_factor, levels = rev(levels(loca_factor))))

table(TES$loca)

## 
##   1   2   3 
## 166  99  34

table(TES$loca_factor)

## 
## Met/Tar     Fal 
##     133     166

TES%>%
  dplyr::group_by(topo) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = topo, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(topo) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = topo, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "topografia", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(topo_factor = factor(if_else(topo %in% c(1,2),  "dor/plan/lat", "Mas de dos")))
TES <- TES %>%
  mutate(topo_factor = factor(if_else(topo %in% c(1, 2), "dor/plan/lat", "Mas de dos")),
         topo_factor = factor(topo_factor, levels = rev(levels(topo_factor))))

table(TES$topo)

## 
##   1   2   3 
## 136  60 103

table(TES$topo_factor)

## 
##   Mas de dos dor/plan/lat 
##          103          196

TES%>%
  dplyr::group_by(nume) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = nume, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(nume) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = nume, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "numero de zonas", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(nume_factor = factor(if_else(nume %in% c(2, 3),  "Dos o mas", "Una")))
TES <- TES %>%  mutate(nume_factor = factor(nume_factor, levels = c("Una", "Dos o mas")))
TES$nume_factor <- factor (TES$nume_factor, levels= c ("Dos o mas","Una"))
table(TES$nume)

## 
##   1   2   3 
## 211  59  29

table(TES$nume_factor)

## 
## Dos o mas       Una 
##        88       211

TES%>%
  dplyr::group_by(infe) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = infe, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(infe) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = infe, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "infeccion", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(infe_factor = case_when(
    infe == 0 ~ "No",
    infe %in% c(1, 2) ~ "Leve/Mod",
    infe == 3 ~ "Grave",
    TRUE ~ NA_character_     # conserva los NA originales
  )) %>%
  mutate(infe_factor = factor(infe_factor, levels = c("No", "Leve/Mod", "Grave")))
TES$infe_factor <- factor (TES$infe_factor, levels= c ("Grave","Leve/Mod","No"))
table(TES$infe)

## 
##   0   1   2   3 
##  67  43 152  37

table(TES$infe_factor)

## 
##    Grave Leve/Mod       No 
##       37      195       67

TES%>%
  dplyr::group_by(ede) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = ede, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(ede) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = ede, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "edema", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(ede_factor = case_when(
    ede == 2 ~ "Unilateral",
    ede %in% c(0, 1) ~ "Sin/local",
    ede == 3 ~ "Bilateral",
    TRUE ~ NA_character_     # conserva los NA originales
  )) %>%
  mutate(ede_factor = factor(ede_factor, levels = c("Sin/local", "Unilateral", "Bilateral")))
TES$ede_factor <- factor (TES$ede_factor, levels= c ("Bilateral","Unilateral","Sin/local"))
table(TES$ede)

## 
##  0  1  2  3 
## 92 98 89 20

table(TES$ede_factor)

## 
##  Bilateral Unilateral  Sin/local 
##         20         89        190

TES%>%
  dplyr::group_by(neur) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = neur, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(neur) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = neur, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "Neuropatia", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(neur_factor = factor(
    if_else(neur %in% c(1, 2, 3), "Leve/mod/charcot", "No"),
    levels = c("No", "Leve/mod/charcot")
  ))
table(TES$neur)

## 
##   0   1   2   3 
##  10  57 216  16

table(TES$neur_factor)

## 
##               No Leve/mod/charcot 
##               10              289

TES%>%
  dplyr::group_by(tisu) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = tisu, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(tisu) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = tisu, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "Profundidad", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(tisu_factor = factor(if_else(tisu %in% c(2, 3),  "parcial/total", "Piel")))
table(TES$tisu)

## 
##   1   2   3 
##  52  98 149

table(TES$tisu_factor)

## 
## parcial/total          Piel 
##           247            52

TES%>%
  dplyr::group_by(area) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = area, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(area) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = area, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "Area", y = "Proporción de cicatrizados") +
  theme_minimal()

TES$area_factor <- with(TES, ifelse(area == 1, "Menor a 10",
                             ifelse(area %in% c(1, 2), "10 a 40",
                             ifelse(area == 3, "Mas de 40", NA))))
TES$area_factor <- factor(TES$area_factor, levels = c("Menor a 10", "10 a 40", "Mas de 40"))

table(TES$area)

## 
##   1   2   3 
## 169 103  27

table(TES$area_factor)

## 
## Menor a 10    10 a 40  Mas de 40 
##        169        103         27

TES%>%
  dplyr::group_by(cica) %>%
  dplyr::summarise(tar = mean (as.numeric(cicat)-1)) %>%
  ggplot(aes(x = cica, y = tar)) +
  geom_point() +
  geom_smooth(method = "lm", col = "blue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.5)

## `geom_smooth()` using formula = 'y ~ x'

TES %>%
  dplyr::group_by(cica) %>%
  dplyr::summarise(tar = mean(as.numeric(cicat)-1)) %>%
  ggplot(aes(x = cica, y = tar)) +
  geom_col(fill = "steelblue") +
  geom_text(aes(label = round(tar, 2)), vjust = -0.3) +
  labs(x = "Fase de cicatrizacion", y = "Proporción de cicatrizados") +
  theme_minimal()

TES <- TES %>%
  mutate(cica_factor = factor(if_else(cica %in% c(2, 3),  "Granu/infla", "Epit")))
TES$cica_factor <- factor (TES$cica_factor, levels= c ("Granu/infla","Epit"))
table(TES$cica)

## 
##   1   2   3 
##  14  48 237

table(TES$cica_factor)

## 
## Granu/infla        Epit 
##         285          14

mod_mixto <- glmer(cicat ~ isq_factor + (1 | NOMBRE),
                   family = binomial,
                   data = TES)

## boundary (singular) fit: see help('isSingular')

tab_model(mod_mixto)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.59	0.38 – 0.92	0.019
isq factor [NO/leve]	3.45	2.04 – 5.81	<0.001
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.00
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.087 / NA

mod_mixto2 <- glmer(cicat ~ isq_factor + infe_factor +(1 | NOMBRE),
                   family = binomial,
                   data = TES)


tab_model(mod_mixto2)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.20	0.08 – 0.50	0.001
isq factor [NO/leve]	3.77	2.15 – 6.60	<0.001
infe factor [Leve/Mod]	2.99	1.30 – 6.87	0.010
infe factor [No]	6.10	2.29 – 16.26	<0.001
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.24
ICC	0.07
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.152 / 0.209

mod_mixto3 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor +(1 | NOMBRE),
                   family = binomial,
                   data = TES)
tab_model(mod_mixto3)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.12	0.04 – 0.32	<0.001
isq factor [NO/leve]	4.17	2.34 – 7.45	<0.001
infe factor [Leve/Mod]	3.12	1.33 – 7.31	0.009
infe factor [No]	6.81	2.48 – 18.68	<0.001
loca factor [Fal]	2.17	1.28 – 3.67	0.004
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.23
ICC	0.07
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.190 / 0.244

table(TES$NOMBRE)

## 
##                                                              ABUDARA 
##                                                                   10 
##               Centro de rehabilitación, Obesidad y Diabetes. Formosa 
##                                                                    7 
##                                                                  CER 
##                                                                    6 
##                                                Clínica Centro Médico 
##                                                                    9 
##                                                 Hosp. Dr.  J.R.Vidal 
##                                                                   12 
##                          Hospital Centro de Salud Zenón J. Santillán 
##                                                                    8 
##                                                 Hospital de Clínicas 
##                                                                   16 
## Hospital de Día de Pie Diabético. Polo Sanitario Malvinas Argentinas 
##                                                                   61 
##                                                      Hospital Güemes 
##                                                                   11 
##                                                   Hospital Perrupato 
##                                                                   15 
##                                                     Hospital Posadas 
##                                                                   97 
##                                                  Hospital san Martin 
##                                                                   10 
##                                                      Hospital Sommer 
##                                                                   17 
##                                                                 HRRG 
##                                                                   10 
##                                                   Provincial Rosario 
##                                                                   10

mod_mixto4 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + (1 | NOMBRE),
                   family = binomial,
                   data = TES)
tab_model(mod_mixto4)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.09	0.03 – 0.26	<0.001
isq factor [NO/leve]	4.18	2.33 – 7.50	<0.001
infe factor [Leve/Mod]	2.88	1.22 – 6.79	0.016
infe factor [No]	5.30	1.88 – 14.98	0.002
loca factor [Fal]	2.23	1.31 – 3.78	0.003
topo factor [dor/plan/lat]	1.68	0.94 – 3.01	0.080
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.20
ICC	0.06
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.205 / 0.250

mod_mixto5 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ (1 | NOMBRE),
                   family = binomial,
                   data = TES)
tab_model(mod_mixto5)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.06	0.02 – 0.19	<0.001
isq factor [NO/leve]	4.03	2.20 – 7.40	<0.001
infe factor [Leve/Mod]	2.38	0.96 – 5.92	0.062
infe factor [No]	3.93	1.31 – 11.75	0.014
loca factor [Fal]	1.79	1.03 – 3.11	0.038
topo factor [dor/plan/lat]	0.87	0.44 – 1.71	0.678
nume factor [Una]	5.01	2.55 – 9.85	<0.001
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.16
ICC	0.05
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.290 / 0.323

mod_mixto6 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ ede_factor +(1 | NOMBRE),
                   family = binomial,
                   data = TES)
tab_model(mod_mixto6)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.02	0.00 – 0.12	<0.001
isq factor [NO/leve]	4.01	2.16 – 7.45	<0.001
infe factor [Leve/Mod]	2.45	0.95 – 6.30	0.062
infe factor [No]	3.88	1.23 – 12.25	0.021
loca factor [Fal]	1.84	1.04 – 3.23	0.035
topo factor [dor/plan/lat]	0.83	0.41 – 1.68	0.605
nume factor [Una]	4.84	2.43 – 9.64	<0.001
ede factor [Unilateral]	2.82	0.81 – 9.76	0.102
ede factor [Sin/local]	3.29	1.02 – 10.67	0.047
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.24
ICC	0.07
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.314 / 0.360

mod_mixto7 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ ede_factor + neur_factor + (1 | NOMBRE),
                   family = binomial,
                   data = TES)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00389862 (tol = 0.002, component 1)

tab_model(mod_mixto7)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.10	0.01 – 1.63	0.106
isq factor [NO/leve]	4.07	2.18 – 7.61	<0.001
infe factor [Leve/Mod]	2.43	0.94 – 6.27	0.066
infe factor [No]	3.78	1.18 – 12.10	0.025
loca factor [Fal]	1.85	1.04 – 3.27	0.035
topo factor [dor/plan/lat]	0.83	0.41 – 1.69	0.613
nume factor [Una]	4.76	2.37 – 9.54	<0.001
ede factor [Unilateral]	2.80	0.81 – 9.72	0.104
ede factor [Sin/local]	3.27	1.01 – 10.59	0.049
neur factor [Leve/mod/charcot]	0.21	0.02 – 1.93	0.166
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.29
ICC	0.08
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.329 / 0.384

mod_mixto8 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ ede_factor + neur_factor + tisu_factor + (1 | NOMBRE),
                   family = binomial,
                   data = TES)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.0034586 (tol = 0.002, component 1)

tab_model(mod_mixto8)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.09	0.01 – 1.49	0.093
isq factor [NO/leve]	3.69	1.96 – 6.97	<0.001
infe factor [Leve/Mod]	2.23	0.88 – 5.70	0.093
infe factor [No]	2.76	0.82 – 9.30	0.101
loca factor [Fal]	1.85	1.04 – 3.28	0.035
topo factor [dor/plan/lat]	0.81	0.40 – 1.64	0.561
nume factor [Una]	4.78	2.39 – 9.58	<0.001
ede factor [Unilateral]	2.86	0.83 – 9.91	0.097
ede factor [Sin/local]	3.14	0.97 – 10.16	0.056
neur factor [Leve/mod/charcot]	0.25	0.03 – 2.37	0.226
tisu factor [Piel]	2.08	0.80 – 5.39	0.132
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.24
ICC	0.07
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.341 / 0.386

mod_mixto9 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ ede_factor + neur_factor + tisu_factor + area_factor + (1 | NOMBRE),
                   family = binomial,
                   data = TES)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00309462 (tol = 0.002, component 1)

tab_model(mod_mixto9)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.05	0.00 – 0.92	0.044
isq factor [NO/leve]	3.74	1.98 – 7.07	<0.001
infe factor [Leve/Mod]	2.55	0.97 – 6.67	0.057
infe factor [No]	3.35	0.97 – 11.63	0.057
loca factor [Fal]	1.92	1.08 – 3.42	0.027
topo factor [dor/plan/lat]	0.91	0.44 – 1.88	0.803
nume factor [Una]	5.79	2.72 – 12.32	<0.001
ede factor [Unilateral]	2.63	0.76 – 9.12	0.127
ede factor [Sin/local]	3.10	0.95 – 10.10	0.060
neur factor [Leve/mod/charcot]	0.27	0.03 – 2.52	0.249
tisu factor [Piel]	2.20	0.85 – 5.69	0.105
area factor [10 a 40]	1.68	0.80 – 3.55	0.172
area factor [Mas de 40]	1.81	0.51 – 6.41	0.360
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.21
ICC	0.06
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.346 / 0.385

mod_mixto10 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor + topo_factor + nume_factor+ ede_factor + neur_factor + tisu_factor + area_factor + cica_factor + (1 | NOMBRE), family = binomial,
                   data = TES)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00374972 (tol = 0.002, component 1)

tab_model(mod_mixto10)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.05	0.00 – 0.98	0.048
isq factor [NO/leve]	3.63	1.93 – 6.84	<0.001
infe factor [Leve/Mod]	2.57	0.98 – 6.73	0.055
infe factor [No]	3.02	0.86 – 10.53	0.084
loca factor [Fal]	1.91	1.07 – 3.39	0.028
topo factor [dor/plan/lat]	0.92	0.45 – 1.89	0.823
nume factor [Una]	5.71	2.69 – 12.14	<0.001
ede factor [Unilateral]	2.42	0.70 – 8.44	0.164
ede factor [Sin/local]	2.97	0.91 – 9.65	0.070
neur factor [Leve/mod/charcot]	0.27	0.03 – 2.49	0.247
tisu factor [Piel]	1.96	0.75 – 5.14	0.172
area factor [10 a 40]	1.72	0.82 – 3.62	0.153
area factor [Mas de 40]	1.89	0.53 – 6.69	0.325
cica factor [Epit]	3.57	0.40 – 31.99	0.256
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.16
ICC	0.05
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.357 / 0.387

##parece que este funciona
mod_mixto11 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor +  nume_factor+ ede_factor +  (1 | NOMBRE),
                   family = binomial,
                   data = TES)
tab_model(mod_mixto11)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.02	0.00 – 0.11	<0.001
isq factor [NO/leve]	4.01	2.16 – 7.44	<0.001
infe factor [Leve/Mod]	2.44	0.95 – 6.25	0.063
infe factor [No]	3.72	1.19 – 11.60	0.023
loca factor [Fal]	1.87	1.07 – 3.28	0.028
nume factor [Una]	4.51	2.40 – 8.48	<0.001
ede factor [Unilateral]	2.88	0.83 – 9.97	0.094
ede factor [Sin/local]	3.25	1.00 – 10.49	0.049
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.23
ICC	0.06
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.313 / 0.358

AIC(mod_mixto, mod_mixto2, mod_mixto3, mod_mixto4, mod_mixto5, mod_mixto6, mod_mixto7, mod_mixto8, mod_mixto9, mod_mixto10, mod_mixto11)

##             df AIC
## mod_mixto    3 389
## mod_mixto2   5 379
## mod_mixto3   6 372
## mod_mixto4   7 371
## mod_mixto5   8 349
## mod_mixto6  10 349
## mod_mixto7  11 348
## mod_mixto8  12 348
## mod_mixto9  14 350
## mod_mixto10 15 350
## mod_mixto11  9 347

performance::icc(mod_mixto7);performance::icc(mod_mixto8);performance::icc(mod_mixto9); performance::icc(mod_mixto11)

## # Intraclass Correlation Coefficient
## 
##     Adjusted ICC: 0.081
##   Unadjusted ICC: 0.054

## # Intraclass Correlation Coefficient
## 
##     Adjusted ICC: 0.068
##   Unadjusted ICC: 0.045

## # Intraclass Correlation Coefficient
## 
##     Adjusted ICC: 0.059
##   Unadjusted ICC: 0.038

## # Intraclass Correlation Coefficient
## 
##     Adjusted ICC: 0.065
##   Unadjusted ICC: 0.045

##mejor performance
mod_mixto12 <- glmer(cicat ~ isq_factor + infe_factor + loca_factor +  nume_factor+ ede_factor + evol +  (1 | NOMBRE),
                   family = binomial,
                   data = TES)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00214939 (tol = 0.002, component 1)

tab_model(mod_mixto12)

	cicat
Predictors	Odds Ratios	CI	p
(Intercept)	0.02	0.00 – 0.14	<0.001
isq factor [NO/leve]	3.96	2.10 – 7.46	<0.001
infe factor [Leve/Mod]	2.46	0.94 – 6.39	0.066
infe factor [No]	4.47	1.38 – 14.42	0.012
loca factor [Fal]	1.68	0.95 – 2.99	0.076
nume factor [Una]	4.71	2.47 – 8.99	<0.001
ede factor [Unilateral]	2.93	0.83 – 10.29	0.094
ede factor [Sin/local]	3.49	1.05 – 11.52	0.041
evol	0.99	0.99 – 1.00	0.015
Random Effects
σ²	3.29
τ₀₀ _NOMBRE	0.39
ICC	0.11
N _NOMBRE	15
Observations	299
Marginal R² / Conditional R²	0.358 / 0.427

ranef(mod_mixto12)$NOMBRE

##                                                                      (Intercept)
## ABUDARA                                                                    0.273
## Centro de rehabilitación, Obesidad y Diabetes. Formosa                     0.493
## CER                                                                        0.512
## Clínica Centro Médico                                                      0.469
## Hosp. Dr.  J.R.Vidal                                                      -0.245
## Hospital Centro de Salud Zenón J. Santillán                               -0.628
## Hospital de Clínicas                                                       0.174
## Hospital de Día de Pie Diabético. Polo Sanitario Malvinas Argentinas      -0.584
## Hospital Güemes                                                           -0.092
## Hospital Perrupato                                                         0.444
## Hospital Posadas                                                           0.274
## Hospital san Martin                                                       -0.434
## Hospital Sommer                                                           -0.567
## HRRG                                                                       0.255
## Provincial Rosario                                                        -0.461

efectos_centros <- ranef(mod_mixto12)$NOMBRE
efectos_centros_df <- data.frame(
  Centro = rownames(efectos_centros),
  Intercepto_aleatorio = efectos_centros[, "(Intercept)"]
)
head(efectos_centros_df)

##                                                   Centro Intercepto_aleatorio
## 1                                                ABUDARA                 0.27
## 2 Centro de rehabilitación, Obesidad y Diabetes. Formosa                 0.49
## 3                                                    CER                 0.51
## 4                                  Clínica Centro Médico                 0.47
## 5                                   Hosp. Dr.  J.R.Vidal                -0.25
## 6            Hospital Centro de Salud Zenón J. Santillán                -0.63

intercepto_fijo <- fixef(mod_mixto12)["(Intercept)"]

efectos_centros_df$Intercepto_total <- intercepto_fijo + efectos_centros_df$Intercepto_aleatorio

library(ggplot2)

ggplot(efectos_centros_df, aes(x = reorder(Centro, Intercepto_total), y = Intercepto_total)) +
  geom_point() +
  coord_flip() +
  labs(x = "Centro", y = "Intercepto total", title = "Variación del intercepto por centro") +
  theme_minimal()

##El gráfico que te propuse (de los interceptos por centro) es una forma visual de entender cómo varía la probabilidad basal de cicatrización entre los centros, después de ajustar por las variables clínicas del modelo.
##🔹 Qué representa cada punto

#Cada punto del gráfico corresponde a un centro (tu variable NOMBRE), y el valor en el eje Y es el intercepto total para ese centro.

#El intercepto total = intercepto fijo global + efecto aleatorio del centro.

#Es decir, refleja la tendencia basal del centro a que los pacientes cicatricen, antes de considerar los predictores clínicos.

#🔹 Cómo interpretar el eje vertical (valores)

#Si el Intercepto total está más alto (más positivo), significa que ese centro tiene una mayor probabilidad basal de cicatrización comparado con el promedio de todos los centros.

#Si el Intercepto total está más bajo (más negativo), ese centro tiene una menor probabilidad basal de cicatrización (sus pacientes, en promedio, tienden a cicatrizar menos, ajustando por las covariables).

efectos_centros_df$Probabilidad_basal <- plogis(efectos_centros_df$Intercepto_total)
# 1️⃣ Cargar paquete
library(lme4)
library(ggplot2)
library(dplyr)

# 2️⃣ Extraer efectos aleatorios del modelo
efectos_centros <- ranef(mod_mixto12, condVar = TRUE)$NOMBRE

# 3️⃣ Convertir a data.frame
efectos_centros_df <- data.frame(
  Centro = rownames(efectos_centros),
  Intercepto_centro = efectos_centros[ , "(Intercept)"]
)

# 4️⃣ Sumar el intercepto fijo global
intercepto_fijo <- fixef(mod_mixto12)[["(Intercept)"]]
efectos_centros_df <- efectos_centros_df %>%
  mutate(
    Intercepto_total = intercepto_fijo + Intercepto_centro,
    Probabilidad_basal = plogis(Intercepto_total) # convierte log-odds a probabilidad
  )

# 5️⃣ Graficar la probabilidad basal de cicatrización por centro
ggplot(efectos_centros_df, aes(x = reorder(Centro, Probabilidad_basal), y = Probabilidad_basal)) +
  geom_point(size = 3, color = "steelblue") +
  geom_hline(yintercept = mean(efectos_centros_df$Probabilidad_basal), 
             linetype = "dashed", color = "red") +
  coord_flip() +
  labs(
    x = "Centro",
    y = "Probabilidad basal de cicatrización (ajustada)",
    title = "Variación entre centros en la probabilidad basal de cicatrización"
  ) +
  theme_minimal(base_size = 13)

library(DHARMa)

sim_res <- simulateResiduals(fittedModel = mod_mixto12, n = 1000)
plot(sim_res)

# Test formales
testUniformity(sim_res)

## 
##  Asymptotic one-sample Kolmogorov-Smirnov test
## 
## data:  simulationOutput$scaledResiduals
## D = 0.05, p-value = 0.3
## alternative hypothesis: two-sided

testDispersion(sim_res)

## 
##  DHARMa nonparametric dispersion test via sd of residuals fitted vs.
##  simulated
## 
## data:  simulationOutput
## dispersion = 1, p-value = 0.6
## alternative hypothesis: two.sided

testZeroInflation(sim_res)

## 
##  DHARMa zero-inflation test via comparison to expected zeros with
##  simulation under H0 = fitted model
## 
## data:  simulationOutput
## ratioObsSim = 1, p-value = 1
## alternative hypothesis: two.sided

library(performance)

# Colinealidad
check_collinearity(mod_mixto12)

## # Check for Multicollinearity
## 
## Low Correlation
## 
##         Term  VIF   VIF 95% CI adj. VIF Tolerance Tolerance 95% CI
##   isq_factor 1.08 [1.01, 1.40]     1.04      0.93     [0.72, 0.99]
##  infe_factor 1.22 [1.11, 1.45]     1.11      0.82     [0.69, 0.90]
##  loca_factor 1.07 [1.01, 1.42]     1.03      0.94     [0.70, 0.99]
##  nume_factor 1.09 [1.02, 1.38]     1.04      0.92     [0.73, 0.98]
##   ede_factor 1.24 [1.12, 1.47]     1.11      0.81     [0.68, 0.89]
##         evol 1.05 [1.01, 1.50]     1.03      0.95     [0.67, 0.99]

# Bondad de ajuste (R² marginal y condicional)
r2_nakagawa(mod_mixto12)

## # R2 for Mixed Models
## 
##   Conditional R2: 0.427
##      Marginal R2: 0.358

# Varianza de efectos aleatorios
VarCorr(mod_mixto12)

##  Groups Name        Std.Dev.
##  NOMBRE (Intercept) 0.628