05 SIC caries Index Epidemiology Latvia 2023

Published

Feb 4, 2025, 00:00

Modified

Feb 4, 2025, 18:00

PACKAGES & DATASETS

[1] "/home/sergiouribe/Insync/sergio.uribe@gmail.com/Google Drive - Shared with me/Epidemiologijas petijumi/2023_Publication_epidemiology/2023_analysis_epidemiology_Latvia_ilze_12_15"

Dataset for EPi

Load data for Epidemiological Analysis

Now subset only ages 12 and 15

Labelling

SIC caries index - Significant Caries Index

SIC 12 years old caries

SIC 15 years old caries

Table SIC results

SiC Index for 12 and 15-Year-Olds
Caries Threshold	12-Year-Olds	15-Year-Olds
D1MFS	17.37	25.76
D1MFT	10.58	14.75
D3MFS	8.24	14.79
D3MFT	5.08	8.93
D5MFS	6.67	12.38
D5MFT	3.86	7.18

--- title: "05 SIC caries Index Epidemiology Latvia 2023" date: 2025-02-04 date-modified: last-modified date-format: "MMM D, YYYY, HH:mm" theme: default format: # docx: default # pdf: default html: toc: true toc-location: left embed-resources: true toc-expand: 1 code-fold: true code-tools: true editor: source execute: echo: false cache: true warning: false message: false --- # PACKAGES & DATASETS ```{r} pacman::p_load(tidyverse, performance, # Assessment of Regression Models Performance # expss, # for the labels # haven, # to keep the labels names # Hmisc, # for the labels # sjlabelled, # for the labels labelled, # for the labels using purrr, grid, # for arrow results in the SIC section gt, # fot the tables knitr, tidymodels, sjPlot, scales, gtsummary, irr, # for agreement calculations patchwork, # for several plots viridis, # colour palette janitor, here) ``` ```{r} theme_set(theme_minimal()) ``` ```{r} here::here() ``` ## Dataset for EPi Load data for Epidemiological Analysis ```{r} df <- read_rds(here("data", "df.rds")) # the rds file with the correct levels ``` ```{r} # Will subset the dataset, only works with the csv file, not the rds # df <- df |> # select(-c(`17 [Distal]`:`Sealants[ 47]`)) ``` Now subset only ages 12 and 15 ```{r} df <- df |> filter(Age %in% c("12", "15")) ``` ## Labelling ```{r} ## For labelled package df <- set_variable_labels( df, Count = "Skaits", `Examen date` = "Egzāmena datums", `Examen time` = "Egzāmena laiks", Age = "Vecums", `Examiner code` = "Egzaminatora kods", Region = "Reģions", `School code` = "Skolas kods", `Child code` = "Bērna kods", Examination = "Pārbaude", Gender = "Dzimums", `Date of birth` = "Dzimšanas datums", `Language spoken iat home` = "Mājās runātā valoda", `Place of living` = "Dzīvesvieta", `Toothbrushing frequency` = "Zobu sukošanas biežums", Toothpaste = "Zobu pasta", `Daily sugary drinks` = "Ikdienu saldētie dzērieni", `Daily sweets` = "Ikdienu saldumi", `Annual dental / dental hygiene visits` = "Ikgadējās zobārstniecības / zobu higiēnas vizītes", `Smoking or other tobacco at least once per week` = "Smēķēšana vai citu tabakas izstrādājumu lietošana vismaz reizi nedēļā", `Visible plaque` = "Redzams zobu aplikums", Smoking = "Smēķēšana" ) ``` ## SIC caries index - Significant Caries Index ```{r} caries <- df |> select(Gender, Age, D1MFS:D5MFT) ``` ```{r} # Function to calculate and plot SiC Index for a given age group and caries threshold plot_sic_index <- function(data, age, caries_column) { # Ensure caries_column is a valid column name if (!caries_column %in% names(data)) { stop("Invalid caries column name. Choose from: D1MFS, D3MFS, D5MFS, D1MFT, D3MFT, D5MFT") } # Filter data for selected age and caries column df <- data |> filter(Age == age) |> select(all_of(caries_column)) |> arrange(.data[[caries_column]]) # Rename column dynamically for consistent referencing colnames(df) <- c("Caries_Score") # If there are no valid values, return an error if (nrow(df) == 0) { stop("No data available for the selected age group.") } # Calculate cumulative frequency distribution df <- df |> mutate(Cumulative_Percent = (row_number() / n()) * 100) # Identify the SiC Index threshold (top 1/3) sic_threshold <- quantile(df$Cumulative_Percent, probs = 2/3, na.rm = TRUE) # Calculate the SiC Index as the mean of the highest 1/3 sic_index <- df |> filter(Cumulative_Percent >= sic_threshold) |> summarise(SiC = mean(Caries_Score, na.rm = TRUE)) |> pull(SiC) # Generate the plot p <- ggplot(df, aes(x = Cumulative_Percent, y = Caries_Score)) + # Shaded area below the step function geom_ribbon(aes(ymin = 0, ymax = Caries_Score), fill = "black", alpha = 0.2) + # Step plot for cumulative distribution geom_step(size = 1.2, color = "black") + # Vertical line marking the SiC Index threshold geom_vline(xintercept = sic_threshold, linetype = "solid", color = "black", size = 1.2) + # Arrow to indicate the SiC range annotate("text", x = sic_threshold + 5, y = max(df$Caries_Score, na.rm = TRUE) - 1, label = "SIC", fontface = "bold", hjust = 0) + annotate("segment", x = sic_threshold, xend = 100, y = max(df$Caries_Score, na.rm = TRUE) - 2, yend = max(df$Caries_Score, na.rm = TRUE) - 2, arrow = arrow(type = "closed", ends = "both", length = unit(0.3, "cm")), color = "black") + # Labels and theme labs(title = paste("Cumulative Distribution of", caries_column, "for", age, "Year-Olds"), x = "Cumulative Percentage of Group", y = paste(caries_column, "Score")) + theme_minimal() return(p) # Explicitly return the plot } # <-- This closing bracket was missing! ``` ### SIC 12 years old caries ```{r} plot_sic_index(caries, age = 12, caries_column = "D1MFS") ``` ```{r} plot_sic_index(caries, age = 12, caries_column = "D3MFS") ``` ```{r} plot_sic_index(caries, age = 12, caries_column = "D5MFS") ``` ```{r} plot_sic_index(caries, age = 12, caries_column = "D1MFT") ``` ```{r} plot_sic_index(caries, age = 12, caries_column = "D3MFT") ``` ```{r} plot_sic_index(caries, age = 12, caries_column = "D5MFT") ``` ### SIC 15 years old caries ```{r} plot_sic_index(caries, age = 15, caries_column = "D1MFS") ``` ```{r} plot_sic_index(caries, age = 15, caries_column = "D3MFS") ``` ```{r} plot_sic_index(caries, age = 15, caries_column = "D5MFS") ``` ```{r} plot_sic_index(caries, age = 15, caries_column = "D1MFT") ``` ```{r} plot_sic_index(caries, age = 15, caries_column = "D3MFT") ``` ```{r} plot_sic_index(caries, age = 15, caries_column = "D5MFT") ``` ## Table SIC results ```{r} # Function to calculate SiC Index (without plotting) calculate_sic_index <- function(data, age, caries_column) { # Ensure caries_column exists in the dataset if (!caries_column %in% colnames(data)) { stop("Invalid caries column name.") } # Filter data for selected age and caries column df <- data |> filter(Age == age) |> select(all_of(caries_column)) |> arrange(.data[[caries_column]]) # Rename column dynamically colnames(df) <- c("Caries_Score") # If there are no valid values, return NA if (nrow(df) == 0) { return(NA) } # Calculate cumulative frequency distribution df <- df |> mutate(Cumulative_Percent = (row_number() / n()) * 100) # Identify the SiC Index threshold (top 1/3) sic_threshold <- quantile(df$Cumulative_Percent, probs = 2/3, na.rm = TRUE) # Calculate the SiC Index as the mean of the highest 1/3 sic_index <- df |> filter(Cumulative_Percent >= sic_threshold) |> summarise(SiC = mean(Caries_Score, na.rm = TRUE)) |> pull(SiC) return(sic_index) } ``` ```{r} # Define caries thresholds caries_levels <- c("D1MFS", "D3MFS", "D5MFS", "D1MFT", "D3MFT", "D5MFT") ``` ```{r} # Compute SiC Index for each age (12 & 15) and caries level sic_table <- expand.grid(Age = c(12, 15), Caries_Level = caries_levels, stringsAsFactors = FALSE) |> mutate(Caries_Level = as.character(Caries_Level)) |> # Convert factor to character rowwise() |> mutate(SiC_Index = calculate_sic_index(caries, Age, Caries_Level)) |> ungroup() ``` ```{r} # Reshape the table: Caries Thresholds in rows, Age in columns sic_table_wide <- sic_table |> pivot_wider(names_from = Age, values_from = SiC_Index) |> arrange(Caries_Level) # Display the transformed table sic_table_wide |> gt() |> tab_header(title = "SiC Index for 12 and 15-Year-Olds") |> cols_label(Caries_Level = "Caries Threshold", `12` = "12-Year-Olds", `15` = "15-Year-Olds") |> fmt_number(columns = c("12", "15"), decimals = 2) |> tab_options( table.font.size = "medium", heading.title.font.size = "large" ) ```