What national digital-health programmes actually fail at

After four decades of national digital-health programmes on four continents, the failure modes are surprisingly consistent, and almost none of them are technical.

The programmes that fail do not fail because a vendor could not implement a standard, or because a database schema was wrong, or because a cloud region was down. They fail at the handoff between the political horizon and the operational one. That is a design problem that the field, collectively, has still not learned how to solve.

The two horizons

Every national digital-health programme sits between two clocks that run at very different speeds.

The political horizon is the clock of ministers, mandates, elections, and budget windows. It runs on one-to-five-year cycles. Its instruments are strategies, announcements, milestones, and press releases. Its risk tolerance is asymmetric: political actors are much more punished for visible failures than they are rewarded for invisible successes.

The operational horizon is the clock of clinicians, records departments, hospital IT teams, laboratory information systems, and the several thousand small integrations that a national platform actually consists of. It runs on ten-to-thirty-year cycles. Its instruments are user training, workflow change, configuration drift, and slow institutional learning. Its risk tolerance is symmetric: an operational team that makes a change tends to be judged on whether the change worked, not on whether it was announced.

What breaks at the handoff

The failure mode is almost always at the handoff between these two clocks. Specifically:

1. The political horizon designs on the assumption that the operational one will move at its speed. A three-year rollout plan is built as if a hospital IT department can absorb a new integration in a quarter. It cannot. The plan misses. The programme is described as behind schedule. Political capital is spent explaining the delay rather than solving the underlying pace mismatch.
2. The operational horizon designs on the assumption that the political one will not change. A workflow team designs for the current strategy. The strategy changes with a change of government. The workflow team is left half-finished, either abandoned or asked to pivot. The institutional memory of “we tried that once and it didn’t work” is corrosive.
3. Neither horizon writes down its assumptions about the other. The political side does not write down what pace it expects the operational side to move at. The operational side does not write down what assumptions it is making about political stability. When the assumptions turn out to be wrong (which they always do), there is no shared document to go back to.

The pattern that works

The programmes that survive the two-clock problem tend to share three habits:

1. They have an explicit horizon-brief document. Something short (under twenty pages) signed by both the political sponsor and the operational lead, that says: these are the political milestones we expect; this is the operational cadence we can sustain; here is where we know they don’t line up. It sounds trivial. Almost no programme writes it.
2. They design in operational tranches, not political ones. A tranche is scoped to what the operational horizon can actually deliver in the cadence it operates at. Political milestones are attached to tranches, not the other way around. This tends to look, from the political side, like a slower rollout, but it is a rollout that actually happens.
3. They protect institutional memory across political transitions. The programme’s core artefacts (reference architecture, standards portfolio, principles document) are held by a governance body that is deliberately insulated from ministerial change. The specific programme comes and goes; the artefacts survive. Successor programmes start where the last one finished, not where the previous strategy paper started.

Why the field still fails at this

The field, collectively, still fails at this because the incentives push the other way. Political sponsors are rewarded for announcing ambition; they are not rewarded for pacing an announcement to operational reality. Operational leaders are rewarded for delivering against the plan they were given; they are not rewarded for pushing back on a plan that they know is impossible.

The consulting industry, honestly, does not help. There is a strong commercial incentive to write plans that match the political ambition, because those are the plans that get funded. There is very little commercial incentive to write the counter-paper: the one that says the current plan cannot land at the current pace.

The programmes I have worked with that survived tended to be the ones where a senior operational leader had the standing to push back on a political timeline and where a senior political sponsor had the patience to listen. Both are rare. When both are in the room, the programme tends to work.

What to do about it

If you are running a national digital-health programme right now, the highest-leverage thing you can do this quarter is probably not technical. It is probably to write the horizon brief: the document that names your political milestones, your operational cadence, and the places where they do not line up. Get it signed by both the political sponsor and the operational lead.

If you cannot get it signed, that is the finding. Act on it.

---

This essay is one of a series drawn from four decades of national digital-health practice. The patterns described are anonymised generalisations from programmes on four continents.