Markdown: Solution

Author

Sten Willemsen

Published

June 6, 2024






---
title: "Solution Rmarkdown"
author: "Sten Willemsen"
date: "2024-06-06"
output: 
  html_document:
    code_folding: hide
---






## Setup





::: {.cell}

```{.r .cell-code}
dat <- read.csv("Data/R_data2.csv")
```
:::






## Introduction

## Data

We perform the following data transformation steps:

* We define the variable `pregnancy_length` by adding together. 

*  We define the variable `BMI_cat` by dividing the variable BMI into categories: <18.5 ("Underweight"), 18.5 - 24.9 ("Healthy weight"), 25 - 29.9 ("Overweight"), and >30 (Obesity).

* We log transform `homocysteine` and `vitaminB12`.

* We transform the categorical variables to factors. 

* We remove the original variables `pregnancy_length_weeks`, `pregnancy_length_days`,  `BMI` and `homocysteine` and `vitaminB12` from the data set.





::: {.cell}

:::







## Analysis and Results 

We show descriptives of all variables in the data set.


### Descriptives





::: {.cell}

```{.r .cell-code}
for(i in 1:length(names(dat))){
  if(class(dat[[i]]) == "numeric"){
    print(paste("Mean of", names(dat)[i], ":", mean(dat[[i]])))
    print(paste("Standard deviation of", names(dat)[i], ":", sd(dat[[i]])))
  } else if(class(dat[[i]]) == "factor"){
    print(paste("Frequency of", names(dat)[i], ":"))
    print(table(dat[[i]]))
  }
}
```

::: {.cell-output .cell-output-stdout}

```
[1] "Frequency of Status :"

normal brain development  intellectual disability 
                     108                       82 
[1] "Frequency of iodine_deficiency :"

 no yes 
113  77 
[1] "Frequency of educational_level :"

        high intermediate          low 
          63           82           45 
[1] "Frequency of alcohol :"

 no yes 
 84 106 
[1] "Frequency of smoking :"

 no yes 
151  39 
[1] "Frequency of medication :"

 no yes 
159  31 
[1] "Mean of SAM : 73.6157894736842"
[1] "Standard deviation of SAM : 17.5907251643386"
[1] "Mean of SAH : 17.5905263157895"
[1] "Standard deviation of SAH : 3.3569346588125"
[1] "Mean of cholesterol : 16.9436842105263"
[1] "Standard deviation of cholesterol : 0.93256039721745"
[1] "Mean of HDL : 26.4723157894737"
[1] "Standard deviation of HDL : 0.351814503286772"
[1] "Mean of triglycerides : 8.06752631578947"
[1] "Standard deviation of triglycerides : 0.517036413362704"
[1] "Mean of folicacid_serum : 32.7689473684211"
[1] "Standard deviation of folicacid_serum : 7.46542180292768"
[1] "Mean of folicacid_erys : 1295.03288526895"
[1] "Standard deviation of folicacid_erys : 207.393136290194"
[1] "Mean of pregnancy_length : NA"
[1] "Standard deviation of pregnancy_length : NA"
[1] "Frequency of BMI_cat :"

   Underweight Healthy weight     Overweight        Obesity 
             1            104             76              9 
[1] "Mean of log_homocysteine : 2.84472518390326"
[1] "Standard deviation of log_homocysteine : 0.189749427662033"
[1] "Mean of log_vitaminB12 : 5.93285200936397"
[1] "Standard deviation of log_vitaminB12 : 0.291889003071043"
```


:::
:::




For the continuous ones we also make a histogram.





::: {.cell}

```{.r .cell-code}
for(i in 1:length(names(dat))){
  if(class(dat[[i]]) == "numeric"){
    hist(dat[[i]], main = paste("Histogram of", names(dat)[i]))
  }
}
```

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-1.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-2.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-3.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-4.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-5.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-6.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-7.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-8.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-9.png){width=672}
:::

::: {.cell-output-display}
![](4-6_basic_markdown_practicals_solution_files/figure-html/unnamed-chunk-1-10.png){width=672}
:::
:::





### Unadjusted Analysis

We compare the mean of the logarithm of the Vitamin B12 for the two levels of `Status` (normal brain development or intellectual disability).





::: {.cell}

```{.r .cell-code}
t.test(log_vitaminB12 ~ Status,data = dat)
```

::: {.cell-output .cell-output-stdout}

```

    Welch Two Sample t-test

data:  log_vitaminB12 by Status
t = -1.1204, df = 168, p-value = 0.2642
alternative hypothesis: true difference in means between group normal brain development and group intellectual disability is not equal to 0
95 percent confidence interval:
 -0.13345843  0.03682312
sample estimates:
mean in group normal brain development  mean in group intellectual disability 
                              5.911999                               5.960317 
```


:::
:::





## Adjusted analysis

We now perform logistic regression analysis to investigate the association between `Status` and log `Vitamin B12` while adjusting for `medication`, `smoking` and `alcohol`.





::: {.cell}

```{.r .cell-code}
glm1_adjusted <- glm(Status ~ log_vitaminB12 +  medication + smoking + alcohol, data = dat, family = binomial)

summary(glm1_adjusted)
```

::: {.cell-output .cell-output-stdout}

```

Call:
glm(formula = Status ~ log_vitaminB12 + medication + smoking + 
    alcohol, family = binomial, data = dat)

Coefficients:
               Estimate Std. Error z value Pr(>|z|)    
(Intercept)    -5.19916    3.18418  -1.633 0.102510    
log_vitaminB12  0.78068    0.53318   1.464 0.143137    
medicationyes  -0.25830    0.42863  -0.603 0.546763    
smokingyes      1.45931    0.41006   3.559 0.000373 ***
alcoholyes      0.05375    0.31425   0.171 0.864181    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 259.83  on 189  degrees of freedom
Residual deviance: 243.72  on 185  degrees of freedom
AIC: 253.72

Number of Fisher Scoring iterations: 4
```


:::

```{.r .cell-code}
coef(glm1_adjusted)
```

::: {.cell-output .cell-output-stdout}

```
   (Intercept) log_vitaminB12  medicationyes     smokingyes     alcoholyes 
   -5.19915783     0.78068485    -0.25830270     1.45930700     0.05375298 
```


:::

```{.r .cell-code}
confint(glm1_adjusted)
```

::: {.cell-output .cell-output-stderr}

```
Waiting for profiling to be done...
```


:::

::: {.cell-output .cell-output-stdout}

```
                     2.5 %    97.5 %
(Intercept)    -11.5672460 0.9688055
log_vitaminB12  -0.2543216 1.8448461
medicationyes   -1.1217210 0.5712447
smokingyes       0.6784664 2.2969129
alcoholyes      -0.5649230 0.6700994
```


:::
:::








## Conclusion and Discussion

**Main points:**

* In the unadjusted analysis, we could not show that the mean of the logarithm of the Vitamin B12 is significantly different for the two levels of `Status`.

* In the adjusted analysis, we found that the log `Vitamin B12` is not significantly associated with `Status` while adjusting for `medication`, `smoking` and `alcohol`.