Univariate models

#mean

mean(master_matched$Magnesium_Baseline,na.rm=T)

## [1] 1.963158

sd(master_matched$Magnesium_Baseline,na.rm=T)

## [1] 0.3157326

mean(master_matched$BaselinePpi,na.rm=T)

## [1] 857.8932

sd(master_matched$BaselinePpi,na.rm=T)

## [1] 596.4267

#matched numbers

master_matched  %>% group_by(matched) %>% summarize(mean_mg=mean(Magnesium_Baseline,na.rm=T),sd_mg=sd(Magnesium_Baseline,na.rm=T),
                                         mean_ppi=mean(BaselinePpi,na.rm=T),sd_ppi=sd(BaselinePpi,na.rm=T),
                                         median_ppi=median(BaselinePpi,na.rm = T),
                                         Q1_ppi = quantile(BaselinePpi, 0.25, na.rm = TRUE),
                                         Q3_ppi = quantile(BaselinePpi, 0.75, na.rm = TRUE),
                                         death=sum(Outcome_at_6Wks_death == "dead"),
                                         total_count = n(),
                                         missing_ppi = sum(is.na(BaselinePpi)),
                                         missing_mg = sum(is.na(Magnesium_Baseline)))

## # A tibble: 2 × 12
##   matched mean_mg sd_mg mean_ppi sd_ppi median_ppi Q1_ppi Q3_ppi death
##     <dbl>   <dbl> <dbl>    <dbl>  <dbl>      <dbl>  <dbl>  <dbl> <int>
## 1       0    2.06 0.302     591.   342.       590.   361.   689.    27
## 2       1    1.9  0.315    1116.   677.      1090.   645.  1490.    28
## # ℹ 3 more variables: total_count <int>, missing_ppi <int>, missing_mg <int>

#models cases

master_cases <- master %>% filter(Group==1)

master_cases %>%
  select("Magnesium_Baseline","BaselinePpi","Amputation_at_6week"
  ) %>%
  tbl_uvregression(
    method = glm,
    y = Amputation_at_6week,
    method.args = list(family = binomial),
    exponentiate = TRUE,
    label = list(male = "Female"),
    pvalue_fun = ~style_pvalue(.x, digits = 3)
  )  %>%
  add_global_p() %>%
  
  bold_p() %>%        # bold p-values under a given threshold (default 0.05)
  bold_labels() %>% 
  modify_spanning_header(
    c(estimate, ci, p.value) ~  
      "**Amputation_at_6week Univariate logistic regression  **")

Characteristic	N	Amputation_at_6week Univariate logistic regression
Characteristic	N	OR¹	95% CI¹	p-value
Magnesium_Baseline	53	1.63	0.14, 19.9	0.694
BaselinePpi	68	1.00	1.00, 1.00	0.494
¹ OR = Odds Ratio, CI = Confidence Interval

master_cases %>%
  select("Magnesium_Baseline","BaselinePpi","Outcome_at_6Wks_death"
  ) %>%
  tbl_uvregression(
    method = glm,
    y = Outcome_at_6Wks_death,
    method.args = list(family = binomial),
    exponentiate = TRUE,
    label = list(male = "Female"),
    pvalue_fun = ~style_pvalue(.x, digits = 3)
  )  %>%
  add_global_p() %>%
  
  bold_p() %>%        # bold p-values under a given threshold (default 0.05)
  bold_labels() %>% 
  modify_spanning_header(
    c(estimate, ci, p.value) ~  
      "**Outcome_at_6Wks_death Univariate logistic regression  **")

Characteristic	N	Outcome_at_6Wks_death Univariate logistic regression
Characteristic	N	OR¹	95% CI¹	p-value
Magnesium_Baseline	53	0.11	0.01, 0.81	0.030
BaselinePpi	69	1.00	1.00, 1.00	0.005
¹ OR = Odds Ratio, CI = Confidence Interval

master_cases %>%
  select("Magnesium_Baseline","BaselinePpi","Hospitalizations_6wk"
  ) %>%
  tbl_uvregression(
    method = glm,
    y = Hospitalizations_6wk,
    method.args = list(family = binomial),
    exponentiate = TRUE,
    label = list(male = "Female"),
    pvalue_fun = ~style_pvalue(.x, digits = 3)
  )  %>%
  add_global_p() %>%
  
  bold_p() %>%        # bold p-values under a given threshold (default 0.05)
  bold_labels() %>% 
  modify_spanning_header(
    c(estimate, ci, p.value) ~  
      "**Hospitalizations_6wk Univariate logistic regression  **")

Characteristic	N	Hospitalizations_6wk Univariate logistic regression
Characteristic	N	OR¹	95% CI¹	p-value
Magnesium_Baseline	23	0.32	0.01, 4.79	0.413
BaselinePpi	29	1.00	1.00, 1.00	0.419
¹ OR = Odds Ratio, CI = Confidence Interval

#Correlation

shapiro.test(master_matched$Magnesium_Baseline) # => p = 0.1229

## 
##  Shapiro-Wilk normality test
## 
## data:  master_matched$Magnesium_Baseline
## W = 0.9691, p-value = 0.368

# Shapiro-Wilk normality test for wt
shapiro.test(master_matched$Magnesium_Baseline)

## 
##  Shapiro-Wilk normality test
## 
## data:  master_matched$Magnesium_Baseline
## W = 0.9691, p-value = 0.368

cor.test(master_matched$BaselinePpi, master_matched$Magnesium_Baseline,  method = "pearson", use = "complete.obs")

## 
##  Pearson's product-moment correlation
## 
## data:  master_matched$BaselinePpi and master_matched$Magnesium_Baseline
## t = -1.982, df = 36, p-value = 0.05515
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  -0.57562570  0.00669272
## sample estimates:
##        cor 
## -0.3136617

library("ggpubr")
ggscatter(master_matched, x = "Magnesium_Baseline", y = "BaselinePpi", 
          add = "reg.line", conf.int = TRUE, 
          cor.coef = TRUE, cor.method = "pearson",
          xlab = "BaselinePpi", ylab = "BaselinePpi")

## Warning: Removed 21 rows containing non-finite values (`stat_smooth()`).

## Warning: Removed 21 rows containing non-finite values (`stat_cor()`).

## Warning: Removed 21 rows containing missing values (`geom_point()`).