Feature insights into Somalia’s severe acute malnutrition admissions (SAM): a time series analysis spanning from January 2019 to December 2024

👋 Hello, and welcome to this repository!

This repository contains an R implementation of a time series analysis designed to extract feature insights from Somalia’s severe acute malnutrition (SAM) admission data, providing a basis for evidence-based decision-making for the overall country nutrition information system and programming.

Note

The analysis was made possible thanks to the collaboration of the Somalia Nutrition Cluster coordination team.

All in all, the analysis sought to:

• Identify trend patterns: a long-term direction or movement in admissions that persists across the series - the underlying patterns removing short-term fluctuations The analysis described:

  • The direction of the trend along the series:
    • upward: a general increase over time.
    • downward: a general decrease/decay over time.
    • flat: relatively constant over time.
  • The Shape of the trend:
    • Linear trend: a straight line best fits the data along the series, indicating a constant rate of change over time.
    • Nonlinear trend: the trend follows a curved line, indicating acceleration and deceleration of the rate of change.
  • The Stability of the trend:
    • Stable trend: one that remains consistent over the entire time series.
    • Changing trend: one that evolves over time, possibly with abrupt shifts.

The average rate of change (ARC) was calculated for each periods that exhibited a notable directional shift.

• Identify seasonality patterns: recurring patterns that occur in a fixed and specific period every year. Seasonal patterns can be fixed or time-varying. The analysis explored:
  • Amplitute of changes: when the strength and intensity of the peak or nadir vary over time.
  • Phase shifts: when the timing of the peak season and nadir changes over time.
  • Irregular patterns: when the periodicity of the seasonal patterns are irregular.
• Visualize data insights: creating intuitive plots for clearer interpretation and communication of results.

The above objectives were addressed by utilizing time series analyses techniques.

A glance at the results

The time series plot

The components

Note

The analysis results were presented to the Somalia Nutrition Cluster and partners on 8 July 2025, during a two-hour virtual session attended by approximately 60 participants. Attendees included representatives from local and international NGOs, UN agencies, and donors. The presentation was followed by a Q&A session, during which participants provided very positive and constructive feedback on the gleaned insights, and how useful they will be to shape the cluster’s programming and information system.

Repository Structure

The repository is structured in the following way:

• data/: a data.frame of class tsibble containing the admissions of SAM cases over time. Data is reported on a monthly basis, with a reporting rate >= 80%, as advised by the data owner. The reporting rate is defined as the number of catchment areas that submitted their reported in a given month, divided by the overall number of catchment areas that are expected to report.

• R/: some handy user-defined functions for the project.

• reports/: Analysis report and presentation.

• scripts/: A set of R scripts used for the analysis. These are split into different files, based on the specific task they address:

  • read-in-data.R: read input data and shapefiles.
  • data-wrangling.R: prepare the admission data for downstream worflow.
  • maps.R: some ilustrative maps.
  • eda-graphics.R: graphical exploratory data analysis.
  • decomposition.R: decompose the time series into trend, seasonal effect, and ramainder.
  • arc.R: calculate the average rate of change of the trend component

  The following workflow is recommended:

    flowchart LR
    A(Retrieve secret key for decryption)
    B(Load project-specific functions.R)
    C(Run read-in-data.R)
    D[Run data-wrangling.R] 
    E(Run maps.R)
    F(Run eda-graphics.R)
    G(Run decomposition.R)
    H(Run arc.R)
    
    A --> B --> C --> D --> E --> F --> G --> H

Loading

:::

The above flowchart can be implemented simply by running the scrip.R file found in the root directory.

Reproducibility information

The repository was created in R version 4.4.3. This project uses the {renv} framework to record R package dependencies and versions. Packages and versions used are recorded in renv.lock and code used to manage dependencies is in renv/ and other files in the root project directory. On starting an R session in the working directory, run renv::restore() to install R package dependencies.

Data encryption

This project uses {cyphr} to encrypt the raw data that lives in data-raw/ directory. In order to be able to access and decrypt the encrypted data, the user will need to have created their own personal SSH key and make a request to be added to the project. An easy-to-grasp guide on how to make a request will be found here

License

This repository is licensed under a GNU General Public License 3 (GPL-3).

Feedback

If you wish to give feedback, file an issue or seek support, kindly do so here.

Author

Tomás Zaba