📄 medicine

Improving National Drug List Harmonization Through an Eight-Pattern Analysis of WHO ATC/DDD Classifications

This paper introduces an 8-Pattern Classification Framework based on WHO ATC/DDD data to identify and resolve discrepancies in national drug lists, specifically addressing cases where single molecules have multiple classifications or inconsistencies in administration routes and defined daily doses to improve global harmonization and reporting accuracy.

Original authors: Mohammad Javad Rajabi, Mitra Mahmoudi-Meymand, Mehrdad Vesali-Sharif

Published 2026-07-02

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Mohammad Javad Rajabi, Mitra Mahmoudi-Meymand, Mehrdad Vesali-Sharif

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A Library with a Messy Filing System

Imagine the World Health Organization (WHO) runs a massive global library for medicines. To keep things organized, they use a special filing system called ATC/DDD.

ATC is like the "shelf location" (telling you what the drug does and where it goes in the body).
DDD is like the "standard serving size" (a fixed unit to measure how much people use, regardless of price or bottle size).

The problem is that some medicines are like shape-shifters. A single drug might have different names, different strengths, or be given in different ways (like a pill vs. an injection). In the WHO library, these shape-shifters get multiple "file cards" (codes).

The Problem: The "One-Size-Fits-All" Trap

Countries around the world try to copy this global library to make their own National Drug Lists (NDLs). However, because some drugs are so complex, countries often get confused. They might accidentally:

Miss a file card entirely.
Merge two different file cards into one.
Get the "serving size" wrong.

The authors say this is like trying to organize a closet where one shirt is listed as "Blue," "Cotton," and "Summer," but the country's list only says "Shirt." You lose the details that matter for buying the right amount and tracking usage accurately.

The Solution: The "8-Pattern" Sorting Game

To fix this, the authors created a new tool called the 8-Pattern Classification Framework.

Think of this framework as a traffic light system or a sorting hat for medicines. The authors looked at three specific things for every drug:

The Code (ATC): Is the file card the same or different?
The Serving Size (DDD): Is the standard dose the same or different?
The Delivery Method (Adm.R): Is it a pill, an injection, or an inhaler?

Since each of these three things can be either "Same" or "Different," the authors calculated that there are exactly 8 possible scenarios (or patterns) for how a drug can be listed.

What They Found: The "Olanzapine" and "Aspirin" Examples

The authors took a list of 2,714 medicines from the WHO and found that 913 of them were "shape-shifters" that fit into these 8 patterns.

The Rare Unicorn (Condition 1): They found only one drug that fit the very first pattern perfectly: Olanzapine. This drug has the same code and the same dose, but it comes in two forms: a regular injection and a "depot" (long-lasting) injection. It's the only one where everything matches except the specific type of injection.
The Common Chameleon (Condition 6): They found many drugs that fit a different pattern. For example, Aspirin (Acetylsalicylic Acid) has two different codes because it's used for two totally different things: preventing blood clots (heart) and stopping pain (headache). Even though the dose is the same, the "file card" changes because the job changes.

The Main Takeaway

The authors aren't saying we need to change the WHO's library rules. Instead, they are saying: "If you are a country trying to make your own drug list, you need a checklist."

By using their 8-Pattern Approach, countries can look at their list and ask: "Does this drug fit Pattern 1, Pattern 6, or one of the others?"

This ensures that when a country reports how much medicine they use, they aren't accidentally hiding the details. It helps them buy the right amount of medicine and report the data correctly, so the global library stays accurate.

In short: The paper provides a 8-step checklist to help countries sort out their "shape-shifting" medicines so they don't lose track of them when comparing data with the rest of the world.

Technical Summary: Improving National Drug List Harmonization Through an Eight-Pattern Analysis of WHO ATC/DDD Classifications

Problem Statement
The World Health Organization's Anatomical Therapeutic Chemical (ATC) classification and Defined Daily Dose (DDD) system serve as the international "platinum standard" for drug utilization research. However, a significant structural challenge exists in harmonizing National Drug Lists (NDLs) with these global standards: the phenomenon of "multiclassified" drugs. A single active substance is often assigned multiple ATC codes, varying DDD values, or different routes of administration (Adm.R) to reflect distinct therapeutic indications, strengths, or delivery methods.

While the WHO assigns a single code based on predominant global use to ensure statistical stability, this approach frequently misaligns with national clinical patterns where drugs are used for multiple indications or via different routes. This discrepancy leads to "information loss" during data mapping, where national registries oversimplify complex pharmacological profiles into single code entries for administrative convenience. Such disharmonization compromises statistical integrity, hinders cross-national comparisons, complicates reimbursement policies, and risks arbitrary decision-making in drug procurement and utilization reporting.

Methodology
To address the inherent heterogeneity of the ATC/DDD system, the authors developed a novel 8-Pattern Classification Framework. This methodology is based on the combinatorial permutations of three critical parameters:

ATC Code
DDD Value
Route of Administration (Adm.R)

By analyzing these three parameters, the authors constructed a decision tree to categorize all potential scenarios where a single pharmaceutical entity might exhibit uniformity or divergence across records. Mathematically, this results in $2^3 = 8$ distinct situations (patterns). The framework was applied to a dataset of 913 drug pairs identified as belonging to multiple classifications within the WHO source data (out of a total of 2,714 medications with assigned DDDs and Adm.R). The study utilized WHO primary data to systematically assign each medication to one of these eight predefined patterns, aiming to eliminate selection bias and ensure high internal validity during the data harmonization phase.

Key Results
The analysis of the 913 drug pairs revealed specific distributions across the eight conditions:

Condition 1 (Similar ATC, Similar DDD, Different Adm.R): Only one molecule, Olanzapine, satisfied this condition. Specifically, the pair consisting of immediate-release parenteral olanzapine and depot parenteral olanzapine shared the same ATC code (N05AH03) and DDD (10 mg) but differed in administration route.
Condition 2 (Similar ATC, Similar DDD, Different Adm.R): The majority of the analyzed pairs, including 402 medications, fell into this category. This includes the other Olanzapine combinations (e.g., oral vs. parenteral).
Condition 6 (Different ATC, Similar DDD, Different Adm.R): This condition captures drugs with distinct therapeutic classifications but similar dosing metrics. Examples provided include:
- Acetylsalicylic Acid: Classified as B01AC06 (antiplatelet) and N02BA01 (analgesic) with a DDD of 1, differing by route (Oral vs. Parenteral).
- Betamethasone: Classified as H02AB01 and R01AD06 with a DDD of 0.4.
- Budesonide, Colistin, Vancomycin, and others were also identified with different ATC codes but identical DDDs and varying administration routes.

The study confirms that 913 drug pairs satisfy at least one of the eight classification conditions, highlighting the prevalence of multiclassified drugs that require specific attention during list harmonization.

Significance and Claims
The paper posits that the 8-Pattern Framework offers a robust, evidence-based reference for national regulatory authorities to achieve rigorous and transparent harmonization of their drug lists. The authors claim the following significance:

Methodological Rigor: Unlike previous models focusing solely on ATC code discrepancies, this framework integrates three dimensions (ATC, DDD, Adm.R), ensuring that no therapeutic identity or variant is overlooked.
Data Integrity: By providing a systematic taxonomy for these scenarios, the framework prevents the "suppression of valid codes" for administrative simplification, thereby preserving the biological and therapeutic reality of drugs in national registries.
Practical Utility: The framework is presented as a tool to mitigate statistical distortion in drug utilization research. It aids in accurate procurement planning, pharmacovigilance (by allowing granular signal detection for adverse drug reactions based on specific variants), and pharmacoeconomics (preventing distorted cost-effectiveness models caused by aggregated consumption volumes).
Stability: The authors argue that while radical structural reforms to the ATC system (e.g., mechanism-based classification) are risky for longitudinal data continuity, the 8-Pattern Approach offers a pragmatic, stable pathway to international harmonization within the existing framework.

The study concludes that adopting such predefined identification protocols is essential for policymakers to bridge the gap between traditional classification structures and modern pharmacology until global structural reforms are implemented.