Meet us at eHealth Canada 2026See the full Tessera platform live, and ask about a scoped pilotBook a demo
Clinical data concepts, explained

Care pathway vs clinical registry: what is the difference?

The two terms get used interchangeably, but they answer different questions. Here is each one in plain language, how they relate, and why running both on one platform is where the value is.

A care pathway is a roadmap

A care pathway is the expected route through care for a given condition or procedure. It sets out the steps, the decisions, and the timing that a programme expects for a typical patient: the assessments to complete, the interventions to offer, the milestones to reach, and roughly when each should happen.

A pathway is prospective. It looks forward and says, in effect, this is what good care should look like for this population. It is a plan against which an individual patient's actual care can be compared.

A clinical registry is a logbook

A clinical registry is a structured, longitudinal record of a defined patient population. It captures who is in the cohort, what was done, and what outcomes followed, consistently and over time. Where a pathway describes the intended route, a registry records what actually happened across many patients and many years.

A registry is longitudinal and largely retrospective. Its value grows with time and consistency: the longer and the more uniformly a population is tracked, the more reliably you can measure outcomes, compare sites, and see trends.

Where they meet: variance

Pathways and registries are most useful together. When you hold the expected route (the pathway) against what was actually recorded (the registry), the gap between them is a variance: a deviation from the plan. A missed milestone, a delayed intervention, an out-of-range result, a step skipped.

Variance is where a lot of the quality signal lives. Some deviations are clinically appropriate and well justified; others point to a problem worth acting on. Either way, you cannot see variance unless you have both the plan and the record, captured in a way that lets you compare them.

A worked example

Take a surgical recovery programme. The care pathway sets the expected route: a pre-operative assessment, the procedure, a set of recovery milestones, a follow-up at a defined interval, and a discharge criterion. That is the plan, the same for every patient who enters the programme.

The registry, meanwhile, records what actually happened for each of those patients over time: the dates, the results, the complications, the outcomes. After a year, the registry is the evidence base for how the programme is really performing.

Now suppose the follow-up that the pathway expects at six weeks is happening at ten weeks for a cluster of patients. The pathway alone would not show it, because the pathway is only the plan. The registry alone would record the dates without flagging them as a problem. Holding one against the other is what turns a delayed follow-up into a visible, actionable variance.

Why the distinction matters

Confusing the two leads to predictable gaps. A programme that builds a pathway but never keeps a registry has a plan and no evidence of whether it is being followed. A programme that keeps a registry but never defines a pathway has a record of what happened with nothing to measure it against.

The useful position is to treat them as two halves of the same picture: the intended route and the recorded reality, with variance as the signal that connects them. That is the lens for quality improvement, for outcomes analysis, and for showing a funder or a regulator that care is being delivered as designed.

Doing both on one platform

Tessera runs care pathways and clinical registries on a single platform and a single data contract. Because the pathway definition and the registry record share the same governed concepts, comparing intended care against delivered care is built in rather than bolted on.

That is what makes automated variance detection possible: the platform already knows the expected route and the recorded reality in the same structure, so it can surface deviations as they happen instead of waiting for a downstream audit. One platform, both views, with the variance signal in between.

Keep reading

Meet us at eHealth Canada 2026

See the full platform, then start with a scoped pilot

Book a personalised demo. We'll show you the platform live and scope a low-risk pilot, one registry, one pathway, or one surveillance stream, to prove the model in your environment.

Low-risk entry point. We'll work with your EMR/EHR, your data residency, and your team.