How the WHO Classifies Digital Health, and Why a Shared Vocabulary Matters

The World Health Organization sorts digital health tools by a plain question: whom does this technology serve. Its Classification of Digital Health Interventions organizes tools into groups defined by the user, clients, health workers, health system managers, and data services. The purpose is not to rank tools or approve them. It is to give people from very different backgrounds a shared vocabulary so they can describe what a tool actually does, compare it against alternatives, and see where effort is duplicated or missing.

Why a taxonomy was needed at all

Digital health draws in an unusually wide mix of people. A government official, a software engineer, a clinician, a program funder, and a field implementer can sit at the same table and use the same words to mean different things. One person's "patient app" is another person's "decision-support system" is another's "surveillance platform." When the words drift, the analysis drifts with them. Gaps go unnoticed because two teams each assume someone else is covering them. Duplication goes unnoticed because two tools sound different but do the same job.

The WHO published the first version of this classification (v1.0) in 2018, and a second edition followed in 2023 under a broader title covering digital interventions, services, and applications in health. The anchoring idea stayed the same across both. The basic unit is a discrete function, a specific thing the technology does to meet a health objective, rather than the brand, the platform, or the marketing category. That choice matters. A single app might send appointment reminders, track a stock of medicines, and feed data to a national registry. Those are three functions serving three different users, and the framework lets you name each one instead of collapsing them into "the app."

The four groups, by who is served

The framework's first cut is by user. That is what makes it portable across countries and disease areas.

Interventions for clients

Clients are current or potential users of health services, including caregivers. Tools here reach people directly: appointment and medication reminders, health education and behavior-change messaging, ways to report a problem or give feedback, and channels for a person to access their own records. The organizing question is whether the tool touches the individual seeking care.

Interventions for health workers

This group covers the health workforce delivering care. It includes decision support that surfaces a guideline at the point of care, digital checklists and protocols, tools for registering and following clients over time, telemedicine consultation between providers, and training delivered on a device. The common thread is a tool that helps a worker do the clinical or public-health task in front of them.

Interventions for health system managers

Managers run and oversee the system rather than deliver care hands-on. Their tools handle the logistics of health: supply-chain and stock management, human-resource scheduling, facility and equipment tracking, and the financial and reporting machinery that keeps services running. A stock-out prevented by a good supply dashboard is invisible to the patient, but it is exactly the failure that empties a clinic shelf.

Data services

The fourth group is cross-cutting. Data services are the shared plumbing: collection, storage, exchange, and analysis that the other three groups depend on. Registries, data interchange between systems, geolocation, and the ability to analyze and visualize what has been collected all sit here. It is the least visible group and often the most consequential, because a tool for a client or a worker is only as trustworthy as the data underneath it.

Why the common language earns its keep

Sorting by user does something subtle. It separates the health job to be done from the technology used to do it. That separation is what makes comparison honest.

When two programs each claim to "improve maternal health with mobile technology," the sentence tells you almost nothing. Run both through the framework and the picture sharpens. One might be client reminders plus worker decision support; the other might be a manager-facing logistics tool plus a data registry. Now you can ask the right question of each: does the reminder change behavior, does the decision support change what the worker does, does the logistics tool reduce stock-outs. Different functions call for different evidence, and naming the function is the first step toward choosing the right yardstick.

The shared vocabulary also travels. A classification anchored to universal functions, reminding, supporting a decision, managing a supply, exchanging data, describes a tool in Malmö and one in a rural district clinic with the same words. That is what lets a health ministry, a researcher, and a funder in different countries look at a catalog of tools and reason about coverage and duplication without first arguing over definitions.

The limits are worth stating plainly. A classification tells you what a tool is meant to do. It does not tell you whether the tool works, whether it is safe, or whether it should be paid for. Those are questions for evidence and for regulation, and they use their own frameworks.

Where classification meets regulation

The WHO scheme is descriptive. Regulators ask a different question: is a given software function a medical device, and if so, what oversight applies. The two systems complement each other, and a shared vocabulary helps the handoff between them.

Consider clinical decision support, which sits squarely in the health-worker group. In the United States, the 21st Century Cures Act set out criteria that can place certain decision-support software outside the medical-device definition, turning on whether it analyzes signals or images, whether it draws on recognized medical information, whether it advises a professional rather than a patient, and, critically, whether the professional can independently review the basis for the recommendation and not simply defer to it. Other regulators, through frameworks for software as a medical device and real-world evidence, ask their own versions of the same underlying questions. None of that replaces the WHO's classification. It builds on the same instinct: describe the function precisely, then decide how much scrutiny it warrants.

For anyone evaluating a digital health tool, the sequence is useful. Name the function and the user with the WHO vocabulary. Then ask what evidence would show it works, and what regulatory category it falls into. Getting the description right first makes the harder questions answerable.

This article is educational and is not medical or legal advice; for decisions about your own care, speak with your own clinician.