Análisis de datos

Proyecto para analizar la base de datos de tesis de Dra. Sandra Navarro

Primero, realizamos una limpieza de la base datos, así como recodificación de variables incluidas en la base de datos:

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──
## ✔ ggplot2 3.3.6     ✔ purrr   0.3.4
## ✔ tibble  3.1.7     ✔ dplyr   1.0.9
## ✔ tidyr   1.2.0     ✔ stringr 1.4.0
## ✔ readr   2.1.2     ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## 
## Attaching package: 'magrittr'
## The following object is masked from 'package:purrr':
## 
##     set_names
## The following object is masked from 'package:tidyr':
## 
##     extract
## 
## Attaching package: 'dlookr'
## The following object is masked from 'package:magrittr':
## 
##     extract
## The following object is masked from 'package:tidyr':
## 
##     extract
## The following object is masked from 'package:base':
## 
##     transform
## Warning: Unreplaced values treated as NA as `.x` is not compatible.
## Please specify replacements exhaustively or supply `.default`.

Tablas

Se diseñarón 2 tablas

La tabla 1 incluye las variables categoricas (Cualitativas)

DBsan %>% select(EDAD, age_group,SEXO, MINIMENTALCAT, KATZCAT, DIABETES, HIPERTENSION, CARDIOPATIA,
                 EVC, CA, PARKINSON, HIPOTIROIDISMO, ERC, TRASTORNO_DEL_ANIMO, 
                 TIPO,COMORBILIDADES,ESTATINAS, ANTICOL, ANTIPSICOTICO, BZD, POLIFARMACIA,
                 RIESGO,PAS, PAD, FRAMINGHAN,) %>% 
  tbl_summary(by=ESTATINAS) %>%  add_p() %>% add_overall()
## Warning: The `fmt_missing()` function is deprecated and will soon be removed
## * Use the `sub_missing()` function instead
Characteristic Overall, N = 1381 No, N = 701 Si, N = 681 p-value2
EDAD 79.0 (74.0, 82.8) 79.5 (74.2, 82.0) 79.0 (74.0, 83.0) 0.8
age_group >0.9
65-70 14 (10%) 8 (11%) 6 (8.8%)
71-75 30 (22%) 15 (21%) 15 (22%)
76-80 35 (25%) 17 (24%) 18 (26%)
81-85 59 (43%) 30 (43%) 29 (43%)
SEXO 0.3
Femenino 95 (69%) 51 (73%) 44 (65%)
Masculino 43 (31%) 19 (27%) 24 (35%)
MINIMENTALCAT 0.10
SIN DC 0 (0%) 0 (0%) 0 (0%)
DC LEVE 81 (59%) 35 (50%) 46 (68%)
DC MODERADO 27 (20%) 16 (23%) 11 (16%)
DC SEVERO 30 (22%) 19 (27%) 11 (16%)
KATZCAT 0.060
B-C 84 (61%) 36 (51%) 48 (71%)
D-E 27 (20%) 16 (23%) 11 (16%)
F-G 27 (20%) 18 (26%) 9 (13%)
DIABETES 0.9
No 78 (57%) 40 (57%) 38 (56%)
Si 60 (43%) 30 (43%) 30 (44%)
HIPERTENSION <0.001
No 47 (34%) 33 (47%) 14 (21%)
Si 91 (66%) 37 (53%) 54 (79%)
CARDIOPATIA <0.001
No 112 (81%) 65 (93%) 47 (69%)
Si 26 (19%) 5 (7.1%) 21 (31%)
EVC 0.010
No 115 (83%) 64 (91%) 51 (75%)
Si 23 (17%) 6 (8.6%) 17 (25%)
CA >0.9
No 130 (94%) 66 (94%) 64 (94%)
Si 8 (5.8%) 4 (5.7%) 4 (5.9%)
PARKINSON 0.2
No 130 (94%) 68 (97%) 62 (91%)
Si 8 (5.8%) 2 (2.9%) 6 (8.8%)
HIPOTIROIDISMO 0.8
No 105 (76%) 54 (77%) 51 (75%)
Si 33 (24%) 16 (23%) 17 (25%)
ERC 0.5
No 129 (93%) 64 (91%) 65 (96%)
Si 9 (6.5%) 6 (8.6%) 3 (4.4%)
TRASTORNO_DEL_ANIMO 0.040
No 73 (53%) 31 (44%) 42 (62%)
Si 65 (47%) 39 (56%) 26 (38%)
TIPO <0.001
1 46 (33%) 33 (47%) 13 (19%)
2 30 (22%) 8 (11%) 22 (32%)
3 49 (36%) 20 (29%) 29 (43%)
4 8 (5.8%) 5 (7.1%) 3 (4.4%)
5 5 (3.6%) 4 (5.7%) 1 (1.5%)
COMORBILIDADES 0.012
1 A 3 58 (42%) 38 (54%) 20 (29%)
4 A 5 58 (42%) 24 (34%) 34 (50%)
>6 22 (16%) 8 (11%) 14 (21%)
ANTICOL 0.9
No 68 (49%) 34 (49%) 34 (50%)
Si 70 (51%) 36 (51%) 34 (50%)
ANTIPSICOTICO 0.3
No 100 (72%) 48 (69%) 52 (76%)
Si 38 (28%) 22 (31%) 16 (24%)
BZD 0.008
No 96 (70%) 42 (60%) 54 (81%)
Si 41 (30%) 28 (40%) 13 (19%)
Unknown 1 0 1
POLIFARMACIA 0.092
<2 FARMACOS 16 (12%) 12 (17%) 4 (5.9%)
3 A 5 FARMACOS 59 (43%) 30 (43%) 29 (43%)
>6 FARMACOS 63 (46%) 28 (40%) 35 (51%)
RIESGO 0.7
ALTO 58 (42%) 31 (44%) 27 (40%)
BAJO 24 (17%) 13 (19%) 11 (16%)
MODERADO 56 (41%) 26 (37%) 30 (44%)
PAS 120 (110, 130) 120 (110, 130) 120 (110, 130) 0.6
PAD 70 (70, 80) 70 (70, 80) 70 (60, 80) >0.9
FRAMINGHAN 16 (10, 25) 16 (10, 22) 22 (12, 26) 0.043
1 Median (IQR); n (%)
2 Wilcoxon rank sum test; Pearson's Chi-squared test; Fisher's exact test

Se diseñarón 2 tablas

La tabla 2 incluye las variables continuas (Cuantitativas) y el análisis bioestadístico usando prueba chi cuadrada o exacta de fisher

DBsan %>% select(COLESTEROL, TRIGLICERIDOS, HDL, LDL, TSH, HB1AC, GLUCOSA,
                 SODIO, ALBUMINA,VITAMINAB12, ESTATINAS)%>% 
  tbl_summary(by=ESTATINAS,                                               # stratify entire table by outcome
              statistic = list(all_continuous() ~ "{median} ({IQR})"),   # stats and format for categorical columns
              digits = all_continuous() ~ 1) %>%  add_p() %>% add_overall
## Warning: The `fmt_missing()` function is deprecated and will soon be removed
## * Use the `sub_missing()` function instead
Characteristic Overall, N = 1381 No, N = 701 Si, N = 681 p-value2
COLESTEROL 182.0 (52.2) 176.0 (43.8) 189.5 (57.2) 0.049
TRIGLICERIDOS 125.5 (71.2) 119.5 (60.2) 134.0 (88.2) 0.059
HDL 50.5 (18.6) 51.5 (19.4) 49.5 (17.0) 0.6
LDL 104.5 (44.8) 99.0 (34.5) 109.5 (53.5) 0.11
TSH 2.5 (2.4) 3.0 (2.7) 2.4 (2.0) 0.5
HB1AC 5.9 (1.0) 5.7 (1.0) 6.1 (1.0) 0.026
Unknown 1 0 1
GLUCOSA 97.5 (28.0) 96.5 (27.2) 100.5 (29.8) 0.5
SODIO 140.0 (5.0) 140.0 (5.0) 141.0 (4.0) 0.8
ALBUMINA 3.7 (0.7) 3.7 (0.7) 3.6 (0.8) 0.7
Unknown 10 4 6
VITAMINAB12 394.3 (505.5) 373.0 (382.3) 422.0 (556.6) 0.8
Unknown 13 8 5
1 Median (IQR)
2 Wilcoxon rank sum test

Gráficas

Las graficas que pueden usarse como opción a de las tablas

Antes hicimos un análisis descriptivo de estas variables continuas (Cuantitativas)

en general y despues agrupando por uso o no de estatinas

#Estadisticos descriptivos

DBsan %>%  select(COLESTEROL, TRIGLICERIDOS, HDL, LDL, TSH, HB1AC, GLUCOSA,
            SODIO, ALBUMINA,VITAMINAB12, PAS, PAD, FRAMINGHAN,ESTATINAS)%>%
  describe()
## # A tibble: 13 × 26
##    described_varia…     n    na   mean      sd se_mean     IQR skewness kurtosis
##    <chr>            <int> <int>  <dbl>   <dbl>   <dbl>   <dbl>    <dbl>    <dbl>
##  1 COLESTEROL         138     0 186.   4.17e+1 3.55e+0  52.2      0.500    0.172
##  2 TRIGLICERIDOS      138     0 145.   9.34e+1 7.95e+0  71.2      4.06    23.3  
##  3 HDL                138     0  51.4  1.51e+1 1.28e+0  18.6      0.964    1.53 
##  4 LDL                138     0 105.   3.59e+1 3.06e+0  44.8      0.446    0.502
##  5 TSH                138     0   3.61 6.10e+0 5.19e-1   2.39     9.19    97.0  
##  6 HB1AC              137     1   6.80 6.10e+0 5.21e-1   1       10.9    124.   
##  7 GLUCOSA            138     0 112.   4.18e+1 3.56e+0  28        2.70     8.36 
##  8 SODIO              138     0 140.   3.95e+0 3.37e-1   5       -0.119    0.263
##  9 ALBUMINA           128    10   3.66 5.83e-1 5.15e-2   0.725   -0.549    0.753
## 10 VITAMINAB12        125    13 809.   1.38e+3 1.23e+2 506.       4.21    18.3  
## 11 PAS                138     0 122.   1.71e+1 1.46e+0  20        0.417    0.746
## 12 PAD                138     0  72.2  9.09e+0 7.73e-1  10       -0.351    0.276
## 13 FRAMINGHAN         138     0  17.8  8.10e+0 6.90e-1  14.8      0.154   -1.26 
## # … with 17 more variables: p00 <dbl>, p01 <dbl>, p05 <dbl>, p10 <dbl>,
## #   p20 <dbl>, p25 <dbl>, p30 <dbl>, p40 <dbl>, p50 <dbl>, p60 <dbl>,
## #   p70 <dbl>, p75 <dbl>, p80 <dbl>, p90 <dbl>, p95 <dbl>, p99 <dbl>,
## #   p100 <dbl>
DBsan %>%
  group_by(ESTATINAS) %>% describe(COLESTEROL, TRIGLICERIDOS, HDL, LDL, TSH, HB1AC, GLUCOSA,
                                   SODIO, ALBUMINA,VITAMINAB12, PAS, PAD, FRAMINGHAN)
## # A tibble: 26 × 27
##    described_variab… ESTATINAS     n    na   mean     sd se_mean    IQR skewness
##    <chr>             <chr>     <int> <int>  <dbl>  <dbl>   <dbl>  <dbl>    <dbl>
##  1 ALBUMINA          No           66     4   3.67  0.597  0.0734  0.7     -0.793
##  2 ALBUMINA          Si           62     6   3.66  0.573  0.0727  0.850   -0.273
##  3 COLESTEROL        No           70     0 179.   38.4    4.59   43.8      0.613
##  4 COLESTEROL        Si           68     0 193.   44.0    5.34   57.2      0.344
##  5 FRAMINGHAN        No           70     0  16.6   7.89   0.943  11.5      0.412
##  6 FRAMINGHAN        Si           68     0  19.1   8.16   0.990  14.0     -0.105
##  7 GLUCOSA           No           70     0 112.   44.1    5.27   27.2      2.59 
##  8 GLUCOSA           Si           68     0 112.   39.7    4.81   29.8      2.90 
##  9 HB1AC             No           70     0   6.21  1.49   0.178   1.05     3.23 
## 10 HB1AC             Si           67     1   7.40  8.58   1.05    1        7.98 
## # … with 16 more rows, and 18 more variables: kurtosis <dbl>, p00 <dbl>,
## #   p01 <dbl>, p05 <dbl>, p10 <dbl>, p20 <dbl>, p25 <dbl>, p30 <dbl>,
## #   p40 <dbl>, p50 <dbl>, p60 <dbl>, p70 <dbl>, p75 <dbl>, p80 <dbl>,
## #   p90 <dbl>, p95 <dbl>, p99 <dbl>, p100 <dbl>

Determinamos la normalidad en cada una de las variables con la prueba de Shapiro wilks

DBsan %>%
  group_by(ESTATINAS) %>%
  normality(COLESTEROL, TRIGLICERIDOS, HDL, LDL, TSH, HB1AC, GLUCOSA,
            SODIO, ALBUMINA,VITAMINAB12, PAS, PAD, FRAMINGHAN)
## # A tibble: 26 × 5
##    variable      ESTATINAS statistic  p_value sample
##    <chr>         <chr>         <dbl>    <dbl>  <dbl>
##  1 COLESTEROL    No            0.965 5.01e- 2     70
##  2 COLESTEROL    Si            0.988 7.78e- 1     68
##  3 TRIGLICERIDOS No            0.832 1.93e- 7     70
##  4 TRIGLICERIDOS Si            0.638 1.15e-11     68
##  5 HDL           No            0.906 6.81e- 5     70
##  6 HDL           Si            0.962 3.45e- 2     68
##  7 LDL           No            0.934 1.15e- 3     70
##  8 LDL           Si            0.993 9.66e- 1     68
##  9 TSH           No            0.288 1.02e-16     70
## 10 TSH           Si            0.718 4.05e-10     68
## # … with 16 more rows

Las variables tuvierón una distribución no normal (p>0.05), por ello usaremos estadística no parametrica para el análisis estadístico

Se desarrollarón gráficas de boxplot para las variables continuas y sus respectivas significancia estadísticas usando la prueba wilcoxon

Te presento las opciones como te lo prefieras, en esta primer opción es todas las variables cualitativas comprando entre usuarios de estatinas y su respectivo valor p

## # A tibble: 1,632 × 3
##    ESTATINAS variables  level
##    <chr>     <chr>      <dbl>
##  1 Si        COLESTEROL   208
##  2 Si        COLESTEROL   192
##  3 Si        COLESTEROL   258
##  4 No        COLESTEROL   174
##  5 Si        COLESTEROL   231
##  6 Si        COLESTEROL   126
##  7 No        COLESTEROL   194
##  8 No        COLESTEROL   158
##  9 Si        COLESTEROL   197
## 10 No        COLESTEROL   188
## # … with 1,622 more rows

Posteriormente, dividi estas graficas en grupos: Lipidos, quimica sanguínea y otros laboratorios

Te muestro lípidos:

#COLESTEROL

col<-DBsan %>% select(ESTATINAS,COLESTEROL)

col
## # A tibble: 138 × 2
##    ESTATINAS COLESTEROL
##    <chr>          <dbl>
##  1 Si               208
##  2 Si               192
##  3 Si               258
##  4 No               174
##  5 Si               231
##  6 Si               126
##  7 No               194
##  8 No               158
##  9 Si               197
## 10 No               188
## # … with 128 more rows
colfig<-ggboxplot(col, x = "ESTATINAS", y = "COLESTEROL",width = 0.7, ylab="Colesterol (mg/dL)",
          fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 10) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

colfig

#TRIGLICERIDOS

trig<-DBsan %>% select(ESTATINAS,TRIGLICERIDOS)


trigfig<-ggboxplot(trig, x = "ESTATINAS", y = "TRIGLICERIDOS",width = 0.7, ylab="Trigliceridos (mg/dL)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

trigfig

#HDL
HDL<-DBsan %>% select(ESTATINAS,HDL)
HDL
## # A tibble: 138 × 2
##    ESTATINAS   HDL
##    <chr>     <dbl>
##  1 Si         29  
##  2 Si         34  
##  3 Si         58  
##  4 No         57  
##  5 Si         46  
##  6 Si         32  
##  7 No         84  
##  8 No         65.3
##  9 Si         45  
## 10 No         48.9
## # … with 128 more rows
HDLfig<-ggboxplot(HDL, x = "ESTATINAS", y = "HDL",width = 0.7, ylab="HDL (mg/dL)",
                   fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

HDLfig

#LDL

LDL<-DBsan %>% select(ESTATINAS,LDL)
LDL
## # A tibble: 138 × 2
##    ESTATINAS   LDL
##    <chr>     <dbl>
##  1 Si         78  
##  2 Si        111  
##  3 Si        164  
##  4 No         88  
##  5 Si        166  
##  6 Si         31.6
##  7 No         86  
##  8 No         76  
##  9 Si        116  
## 10 No        114  
## # … with 128 more rows
LDLfig<-ggboxplot(LDL, x = "ESTATINAS", y = "LDL",width = 0.7, ylab="LDL (mg/dL)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

LDLfig

library(cowplot)
## 
## Attaching package: 'cowplot'
## The following object is masked from 'package:ggpubr':
## 
##     get_legend
library(dplyr)
library(gridExtra)
## 
## Attaching package: 'gridExtra'
## The following object is masked from 'package:dplyr':
## 
##     combine
#FIGURA DE LIPIDOS
FIGLIPID<-plot_grid(colfig, trigfig, LDLfig,HDLfig, labels=c("A", "B","C","D"), ncol = 2, nrow = 2)

FIGLIPID

valores de química sanguínea

#FIGURA DE QS
#ALBUMINA
alb<-DBsan %>% select(ESTATINAS,ALBUMINA)

alb
## # A tibble: 138 × 2
##    ESTATINAS ALBUMINA
##    <chr>        <dbl>
##  1 Si             3.5
##  2 Si            NA  
##  3 Si             3.8
##  4 No             3.4
##  5 Si             4.4
##  6 Si             3.1
##  7 No             3.6
##  8 No             4.4
##  9 Si             4.2
## 10 No             4.8
## # … with 128 more rows
albfig<-ggboxplot(alb, x = "ESTATINAS", y = "ALBUMINA",width = 0.7, ylab="Albumina (g/dL)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

albfig
## Warning: Removed 10 rows containing non-finite values (stat_boxplot).
## Warning: Removed 10 rows containing non-finite values (stat_signif).

#glucosa
glu<-DBsan %>% select(ESTATINAS,GLUCOSA)

glu
## # A tibble: 138 × 2
##    ESTATINAS GLUCOSA
##    <chr>       <dbl>
##  1 Si             87
##  2 Si            125
##  3 Si             95
##  4 No            129
##  5 Si             94
##  6 Si            113
##  7 No            129
##  8 No             84
##  9 Si             92
## 10 No             95
## # … with 128 more rows
glufig<-ggboxplot(glu, x = "ESTATINAS", y = "GLUCOSA",width = 0.7, ylab="Glucosa (mg/dL)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

glufig

#HB1AC

#FIGURA DE QS

hb1<-DBsan %>% select(ESTATINAS,HB1AC)

hb1
## # A tibble: 138 × 2
##    ESTATINAS HB1AC
##    <chr>     <dbl>
##  1 Si          5.3
##  2 Si          6.4
##  3 Si          5.8
##  4 No          6.5
##  5 Si          6.1
##  6 Si          6.8
##  7 No          7.2
##  8 No          6.5
##  9 Si          6.3
## 10 No          5.5
## # … with 128 more rows
hb1fig<-ggboxplot(hb1, x = "ESTATINAS", y = "HB1AC",width = 0.7, ylab="HB1AC (%)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

hb1fig
## Warning: Removed 1 rows containing non-finite values (stat_boxplot).
## Warning: Removed 1 rows containing non-finite values (stat_signif).

#SODIO

#FIGURA DE QS

Na<-DBsan %>% select(ESTATINAS, SODIO)

Na
## # A tibble: 138 × 2
##    ESTATINAS SODIO
##    <chr>     <dbl>
##  1 Si          142
##  2 Si          140
##  3 Si          141
##  4 No          142
##  5 Si          144
##  6 Si          139
##  7 No          141
##  8 No          143
##  9 Si          132
## 10 No          144
## # … with 128 more rows
Nafig<-ggboxplot(Na, x = "ESTATINAS", y = "SODIO",width = 0.7, ylab="Sodio (mEq/L)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

Nafig

#TSH

TSH<-DBsan %>% select(ESTATINAS, TSH)

TSH
## # A tibble: 138 × 2
##    ESTATINAS   TSH
##    <chr>     <dbl>
##  1 Si         8.22
##  2 Si         4.25
##  3 Si         1.85
##  4 No         3.33
##  5 Si         1.22
##  6 Si         1.18
##  7 No         1.74
##  8 No        10.5 
##  9 Si         3.4 
## 10 No         2.07
## # … with 128 more rows
TSHfig<-ggboxplot(TSH, x = "ESTATINAS", y = "TSH",width = 0.7, ylab="TSH (nmol/L)",
                 fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

TSHfig

#vitamina B12

VB12<-DBsan %>% select(ESTATINAS, VITAMINAB12)

VB12
## # A tibble: 138 × 2
##    ESTATINAS VITAMINAB12
##    <chr>           <dbl>
##  1 Si               971.
##  2 Si                NA 
##  3 Si               485.
##  4 No               776.
##  5 Si               764 
##  6 Si               182.
##  7 No               240.
##  8 No               261 
##  9 Si               993 
## 10 No              5765.
## # … with 128 more rows
VB12fig<-ggboxplot(VB12, x = "ESTATINAS", y = "VITAMINAB12",width = 0.7, ylab="VITAMINA B12 (pg/mL)",
                  fill = "ESTATINAS",palette = "jco") + 
  stat_compare_means(comparisons=list(c("Si","No")),label = "p.signif",label.y.npc = 0.4, size= 6) +theme(legend.position='none')+
  scale_y_continuous(expand = expansion(mult = c(0, 0.1)))

VB12fig
## Warning: Removed 13 rows containing non-finite values (stat_boxplot).
## Warning: Removed 13 rows containing non-finite values (stat_signif).

FIGQS <- plot_grid(albfig,glufig,hb1fig,Nafig,TSHfig,VB12fig,labels=c("A", "B","C","D", "E", "F"), ncol = 2, nrow = 3)
## Warning: Removed 10 rows containing non-finite values (stat_boxplot).
## Warning: Removed 10 rows containing non-finite values (stat_signif).
## Warning: Removed 1 rows containing non-finite values (stat_boxplot).
## Warning: Removed 1 rows containing non-finite values (stat_signif).
## Warning: Removed 13 rows containing non-finite values (stat_boxplot).
## Warning: Removed 13 rows containing non-finite values (stat_signif).
FIGQS