A UNICEF Data Report on Mortality Rates Among Adolescents Aged 15–19
Every year, hundreds of thousands of adolescents aged 15–19 lose their lives to preventable causes — a quiet, largely overlooked crisis unfolding across the developing world. While global attention often centres on infant or under-five mortality, the deaths of teenagers represent a distinct and equally urgent public health failure.
This report uses UNICEF data to examine how youth mortality rates vary across countries, regions, and time — and what those patterns reveal about the unequal world we live in.
“The death of a young person is not simply a statistic — it is a future stolen, a family shattered, and a community diminished.”
The analysis draws on two UNICEF datasets:
unicef_indicator_2.csv — Mortality rates (per 1,000 youths) for adolescents aged 15–19, broken down by sex, country, and year (1990–2023).unicef_metadata.csv — Country-level socioeconomic metadata including GDP per capita, life expectancy, birth rate, and population.import requests, zipfile, io, os
import pandas as pd
import numpy as np
import geopandas as gpd
from plotnine import *
import warnings
warnings.filterwarnings("ignore")
indicator = pd.read_csv("unicef_indicator_2.csv")
metadata = pd.read_csv("unicef_metadata.csv")
ind_total = indicator[indicator["sex"] == "Total"].copy()
ind_total["time_period"] = ind_total["time_period"].astype(int)The map below shows the most recent available mortality rate for each country. Darker shades indicate higher mortality — a stark visual reminder of geographic inequality.
Sub-Saharan Africa and parts of South Asia carry a disproportionate burden of youth mortality, while high-income countries in Europe and North America show rates close to zero.
url = "https://naciscdn.org/naturalearth/110m/cultural/ne_110m_admin_0_countries.zip"
r = requests.get(url)
z = zipfile.ZipFile(io.BytesIO(r.content))
z.extractall("/tmp/ne_countries")
world = gpd.read_file("/tmp/ne_countries/ne_110m_admin_0_countries.shp")
latest = (
ind_total
.sort_values("time_period", ascending=False)
.groupby("alpha_3_code", as_index=False)
.first()
)[["alpha_3_code", "obs_value", "country"]]
latest.columns = ["iso_a3", "mortality_rate", "country"]
world_merged = world.merge(latest, left_on="ADM0_A3", right_on="iso_a3", how="left")
(
ggplot(world_merged)
+ geom_map(aes(fill="mortality_rate"), color="white", size=0.15)
+ scale_fill_gradient(
low = "#fee8c8",
high = "#b30000",
na_value = "#cccccc",
name = "Deaths per\n1,000 youths"
)
+ coord_equal()
+ labs(
title = "Youth Mortality Rate (Age 15–19) — Most Recent Year",
subtitle = "Darker red = higher mortality | Grey = no data available",
caption = "Source: UNICEF Indicator Dataset"
)
+ theme_void()
+ theme(
plot_title = element_text(size=14, face="bold"),
plot_subtitle = element_text(size=10, color="#555555"),
legend_position = "right"
)
)
Not all countries are equal when it comes to protecting their young people. The chart below highlights the 15 nations with the highest average youth mortality rates over the most recent five years of available data.
Several of the most-affected countries are conflict-affected states or nations with severe poverty, demonstrating the tight link between political instability and child welfare.
max_year = ind_total["time_period"].max()
recent_5 = ind_total[ind_total["time_period"] >= max_year - 4]
top15 = (
recent_5
.groupby("country", as_index=False)["obs_value"]
.mean()
.sort_values("obs_value", ascending=False)
.head(15)
)
top15["country"] = pd.Categorical(
top15["country"],
categories = top15.sort_values("obs_value")["country"].tolist(),
ordered = True
)
(
ggplot(top15, aes(x="country", y="obs_value", fill="obs_value"))
+ geom_col()
+ coord_flip()
+ scale_fill_gradient(low="#fc8d59", high="#b30000")
+ guides(fill=False)
+ geom_text(
aes(label="obs_value.round(1).astype(str)"),
ha="left", nudge_y=0.3, size=8, color="#333333"
)
+ labs(
title = "Top 15 Countries: Highest Youth Mortality Rates",
subtitle = f"Average mortality per 1,000 adolescents aged 15–19 ({max_year-4}–{max_year})",
x="Country", y="Avg. Mortality Rate (per 1,000)",
caption = "Source: UNICEF Indicator Dataset"
)
+ theme_minimal()
+ theme(
plot_title=element_text(size=13, face="bold"),
plot_subtitle=element_text(size=9, color="#555555"),
axis_text_y=element_text(size=9),
panel_grid_minor=element_blank()
)
)
The scatterplot below plots each country’s GDP per capita against its youth mortality rate, with a regression line showing the overall trend.
The regression line confirms a strong negative relationship — as GDP per capita rises, youth mortality falls sharply. This underscores the need to address economic inequality as a root cause of preventable youth deaths.
latest_mortality = (
ind_total
.sort_values("time_period", ascending=False)
.groupby("alpha_3_code", as_index=False)
.first()
)[["alpha_3_code", "obs_value", "country"]]
gdp_latest = (
metadata
.dropna(subset=["GDP per capita (constant 2015 US$)"])
.sort_values("year", ascending=False)
.groupby("alpha_3_code", as_index=False)
.first()
)[["alpha_3_code", "GDP per capita (constant 2015 US$)"]]
gdp_latest.columns = ["alpha_3_code", "gdp_per_capita"]
scatter_df = latest_mortality.merge(gdp_latest, on="alpha_3_code", how="inner")
scatter_df = scatter_df.dropna(subset=["gdp_per_capita", "obs_value"])
scatter_df["log_gdp"] = np.log10(scatter_df["gdp_per_capita"])
(
ggplot(scatter_df, aes(x="log_gdp", y="obs_value"))
+ geom_point(aes(color="obs_value"), alpha=0.75, size=2.5)
+ geom_smooth(method="lm", color="#b30000", se=True, size=1)
+ scale_color_gradient(low="#fdbb84", high="#7f0000", name="Mortality\nRate")
+ labs(
title = "Wealth vs Youth Mortality: A Clear Inverse Relationship",
subtitle = "Each dot = one country | X-axis is log-scaled GDP per capita (constant 2015 USD)",
x="Log GDP per Capita (USD)", y="Youth Mortality Rate (per 1,000)",
caption = "Source: UNICEF Indicator & Metadata Datasets"
)
+ theme_minimal()
+ theme(
plot_title=element_text(size=13, face="bold"),
plot_subtitle=element_text(size=9, color="#555555"),
panel_grid_minor=element_blank()
)
)
The final chart tracks global progress in reducing youth mortality from 1990 to 2023, showing trends separately for male and female adolescents.
While there has been meaningful progress since 1990, male adolescent mortality consistently exceeds female mortality — a gap that reflects higher exposure of young men to:
ts_data = indicator[indicator["sex"].isin(["Male", "Female"])].copy()
ts_data["time_period"] = ts_data["time_period"].astype(int)
ts_avg = (
ts_data
.groupby(["time_period", "sex"], as_index=False)["obs_value"]
.mean()
)
(
ggplot(ts_avg, aes(x="time_period", y="obs_value", color="sex", group="sex"))
+ geom_line(size=1.2)
+ geom_point(size=1.8, alpha=0.7)
+ scale_color_manual(
values={"Male": "#b30000", "Female": "#1f78b4"},
name="Sex"
)
+ scale_x_continuous(breaks=list(range(1990, 2025, 5)))
+ labs(
title = "Global Youth Mortality Trend (1990–2023)",
subtitle = "Global average deaths per 1,000 adolescents aged 15–19, by sex",
x="Year", y="Average Mortality Rate (per 1,000)",
caption = "Source: UNICEF Indicator Dataset"
)
+ theme_minimal()
+ theme(
plot_title=element_text(size=13, face="bold"),
plot_subtitle=element_text(size=9, color="#555555"),
panel_grid_minor=element_blank(),
legend_position="top"
)
)
The data tells a clear and urgent story:
UNICEF and its partners must continue to direct resources toward the nations where adolescent mortality remains tragically high.
Report prepared using UNICEF open data. Visualisations created with plotnine in Python/Quarto.