script.qmd

---
title: "Diabetes and suicide in the world"
format: dashboard
theme: superhero
---

```{python}
import pandas as pd
import numpy as np
import plotly.express as px
from itables import show
import country_converter as coco
```


```{python}
# load data
diabetes = pd.read_csv("data/sh_sta_diab_zs.csv")
suicide = pd.read_csv("data/suicide_per_100000_people.csv")

```


```{python}
# long both databases
suicide_long = suicide.melt(id_vars=["country"], var_name="year", value_name="mort")

diabetes_long = diabetes.melt(id_vars=["country"], var_name="year", value_name="mort")

```

```{python}
# filter suicide to include data since 2000 
suicide_long_ = suicide_long[suicide_long["year"].astype(int) >= 2000]
```

```{python}
# inspect values
suicide_un = suicide_long_["year"].unique()
suicide_long_un= suicide_long_["mort"].unique()
```

```{python}
#mortality by year
year_mort_counts = suicide_long.groupby("year")["mort"].count()
```

```{python}
# explore country
suicide_long_un = suicide_long_["country"].unique()
```

```{python}
# diabetes
diabetes_uni = diabetes["country"].unique()
```

```{python}
# diabetes long
diabetes_long_un = diabetes_long["year"].unique()
```

# Diabetes

:::{.valuebox title="Diabetes prevalence" color="#5ba3b6" icon="bar-chart-line-fill"}
Descriptive analysis using open data from https://www.gapminder.org/data/
:::

## Row 

Index 1: Diabetes prevalence (% of population ages 20 to 79) 2000-2021

## Row {height="80%"}

###  {.tabset width="70%"}

#### Descriptibe table 

```{python}
show(
    diabetes_long,
    paging=True, 
    pageLength=22,  
    lengthMenu=[[22, -1], [22, "All"]]
)
```

#### Figures 

```{python, message: false}
diabetes_long = diabetes_long.sort_values(by=["country", "year"])
diabetes_long = diabetes_long.dropna(subset=["year", "mort"])
import plotly.express as px

fig = px.line(
    diabetes_long,
    x="year",
    y="mort",
    color="country",
    title="Trends of diabetes prevalence by country, 2000-2021",
    labels={
        "year": "Year",
        "mort": "Prevalence (%)", 
        "country": "Country",
    },
)
```

```{python}
diabetes_long = diabetes.melt(id_vars=["country"], var_name="year", value_name="mort")

diabetes_long["country_code"] = coco.convert(diabetes_long["country"], to="ISO3")

fig_map = px.choropleth(
    diabetes_long,
    locations="country_code",
    color="mort",
    hover_name="country",
    animation_frame="year",
    title="Global prevalence of diabetes (2000-2021)",
    labels={
        "year": "Year",
        "mort": "Prevalence (%)",
        "country": "Country",
    },
)
```

::: {.panel-tabset}

## Chart

```{python}
fig.show()
```

## Map

```{python}
fig_map.show()
```

:::

# Suicide 

:::{.valuebox title="Suicide mortality" color="primary" icon="bar-chart-line-fill"}
Descriptive analysis using open data from https://www.gapminder.org/data/
:::

## Row 

Index2: Mortality due to self-inflicted injury, per 100 000 standard population, age ajusted, 2000-2016.

## Row {height="80%"}


###  {.tabset width="70%"}

#### Descriptibe table 

```{python}
show(suicide_long_, paging=True, pageLength=22, lengthMenu=[[22, -1], [22, "All"]])
```

#### Figures 

```{python, message: false}
# drop NaNs
suicide_long_ = suicide_long_.sort_values(by=["country", "year"])
suicide_long_ = suicide_long_.dropna(subset=["year", "mort"])

# Convert to numeric 
suicide_long_["year"] = pd.to_numeric(suicide_long_["year"], errors="coerce")
suicide_long_["mort"] = pd.to_numeric(suicide_long_["mort"], errors="coerce")

# Create the line chart
fig_suicide = px.line(
    suicide_long_,
    x="year",
    y="mort",
    color="country",
    title="Trends of suicide mortality by country, 2000-2016",
    labels={
        "year": "Year",
        "mort": "Mortality rate ajusted by age (per 100,000)",
        "country": "Country",
    },
)
```

```{python}
suicide_long = suicide.melt(id_vars=["country"], var_name="year", value_name="mort")
suicide_long_ = suicide_long[suicide_long["year"].astype(int) >= 2000]

suicide_long_["country_code"] = coco.convert(suicide_long_["country"], to="ISO3")

fig_map_suicide = px.choropleth(
    suicide_long_,
    locations="country_code",
    color="mort",
    hover_name="country",
    animation_frame="year",
    title="Global mortality rate of Suicide ajusted by age, 2000-2016)",
    labels={
        "year": "Year",
        "mort": "Mortality (per 100,000)",
        "country": "Country",
        "country_code": "ISO3",
    },
)
```

::: {.panel-tabset}

## Chart

```{python}
fig_suicide.show()
```

## Map

```{python}
fig_map_suicide.show()
```

:::

# Relationship

```{python}
# rename
diabetes_long_ren = diabetes_long.rename(columns={"mort": "mort_diabetes"})
suicide_long_ren = suicide_long_.rename(columns={"mort": "mort_suicide"})

# 'year' numeric
diabetes_long_ren["year"] = pd.to_numeric(diabetes_long_ren["year"], errors="coerce")
suicide_long_ren["year"] = pd.to_numeric(suicide_long_ren["year"], errors="coerce")

# merge
merged = pd.merge(
    diabetes_long_ren, suicide_long_ren, how="inner", on=["country", "year"]
)

# drop nas
merged_ = merged.dropna(subset=["mort_diabetes", "mort_suicide"])

merged_["year"] = merged_["year"].astype(int)

```

## Row

Exploring relationship between diabetes prevalence and suicide mortality rate ajusted by age worldwide.

## Row

### Column {.tabset}

#### All

```{python}
fig_all = px.scatter(
    merged_,
    x="mort_diabetes",
    y="mort_suicide",
    color="country",  
    hover_data=["year"],
    title="Relationship between diabetes prevalence and suicide mortality ajusted by age, 2000-2016",
    labels={
        "mort_diabetes": "Diabetes Prevalence (%)",
        "mort_suicide": "Suicide Mortality (per 100,000)",
        "country": "Country",
        "year": "Year",
    },
    trendline="ols",  
)
fig_all.show()
```