Building care plans

A care plan is the forward-looking half of a patient's record. A note says what happened; a plan says what happens next — the concerns you're treating, what you're aiming for, what the patient should do, and what they need to tick off before you see them again. It is the one clinical document written to be read by the patient rather than about them.

That changes how it behaves. A plan has a draft you edit privately and a published version the patient sees, and the two only meet when you decide they do. Everything in this article follows from that split.

Turning care plans on

Go to Settings → Care Plans. The toggle reads "Health concerns, goals, and recommendations a patient can follow between visits."

Only owners and admins can reach this page. There's no Save button — the setting autosaves.

With the feature off, care plans aren't half-available, they're invisible: Care Plans disappears from the Library section of the sidebar, the tab vanishes from every patient record, and the patient portal loses its Care Plan page. Going to any of those URLs directly sends you somewhere sensible rather than showing an error.

The Chronic Care Management block underneath is a separate decision — see Care Time and chronic care management. You can run care plans without it.

The template library

Go to Library → Care Plans — "Reusable care plan templates your team can apply to any patient." Your own templates sit at the top; below them is the Library panel, "Starter protocols ready to import and adapt to how your clinic works."

Four starter protocols ship, filterable by Chronic care, Weight management and Post-operative:

• Chronic care — hypertension. One concern (ICD-10 I10), two goals, three recommendations, two patient tasks. Marked CCM by default.
• Chronic care — type 2 diabetes. One concern (E11.9), three goals, three recommendations, three patient tasks. Also CCM by default.
• Weight management. A free-text concern rather than a diagnosis code, two goals, three recommendations, two patient tasks.
• Post-operative recovery. Two goals and three recommendations covering activity restrictions, wound care, and when to seek urgent help.

CCM by default means the imported template arrives with "Enroll in care management by default" switched on. It does nothing unless you've turned care management on.

Two things about the library are worth knowing before you start clicking.

You can't preview a starter protocol. The panel shows a name, a description and two counts, and that's deliberate — importing is how you read one. Import it, open it, and retire it if you don't want it.

Import is one-way. The button goes Import → Imported and stays there. You get one copy per starter protocol, and after that it's yours to edit. If you want a second variant, duplicate the one you imported.

These are starting points, not protocols. Two of the post-operative recommendations literally say "Replace this text with them before sharing the plan." Every target figure and review interval in all four is a placeholder a clinician is expected to overwrite. Read the whole thing before you apply it to anybody.

Building a template

New Care Plan opens the editor. The line under the title is the most important sentence on the page: "Applying this template snapshots its content onto a patient's draft plan — later edits here never change a plan a patient is already following."

A template is a rubber stamp, not a live link. Fix a typo in a template and the forty patients already on it keep the typo, because their plans are copies. That's the right trade — you don't want a clinic-wide edit silently rewriting a document a patient is following — but it means template edits only ever affect future applications.

Three fields sit above the content:

• Name — internal. What your team sees in the list. "e.g. Chronic care — hypertension (internal name)"
• Patient-facing title — what the patient reads at the top of their plan, like "Your blood pressure plan". Falls back to the internal name if you leave it blank.
• Description — internal, "Internal note about when to use this protocol." The patient never sees it.

Then four lists:

• Health concerns — "Why you're treating." Either an ICD-10 diagnosis or a plain-language concern. Whatever you type in the description is read by the patient, so write it as a person would say it.
• Goals — "What success looks like." Each takes a baseline, a target, and a term of short or long.
• Recommendations — "What the patient should do. Order here is the order the patient sees." Categorised as Activity, Nutrition, Supplement, Medication, Education or Custom, with a title and a body that supports Markdown.
• Patient tasks — "Checklist items created for the patient when this template is applied."

Dates in a template are day offsets, not calendar dates — "so the same template works for any patient." A goal with a target of 90 and a review of 30 resolves to ninety and thirty days after the plan's start date, whenever that turns out to be.

A template needs at least one health concern and one goal before it can be applied. Until it has both, the card warns "Needs a health concern and a goal before it can be applied." and the template won't appear in the apply menu at all.

Templates are never deleted, only Deactivated — "Template can no longer be applied to patients." They stay on the page greyed out, so the plans built from them still make sense. Duplicate gives you a "(Copy)" to fork from, and every save bumps the version number shown on the card.

Putting a plan on a patient

Open a patient and go to Care Plans. New plan offers Blank plan or From template.

Applying a template creates a draft. Nothing has happened to the patient yet — as the sheet says, "Creates a draft. The patient sees nothing until you activate it." Plans are grouped in the left rail as Active, Drafts and Past.

The plan itself reads top to bottom:

• Summary — the description, the first thing the patient reads. A plan applied from a template starts without one; write it for the individual.
• Health concerns and Goals — each goal carries a status (Not started, In progress, Achieved, Not achieved, On hold), a term, a target, a target date and a next-review date.
• Recommendations — in the order you set. A recommendation copied from a template is marked From template; one pulled in from a chart note is marked From chart note, and tells you if the note is unsigned or has been edited since.
• Attachments — photos and videos from the patient's chart.
• Tasks — the checklist.
• Versions — the publishing history.

Tasks are private by default. The switch reads "Show this to the patient" with the helper "Off by default. When off, only the care team sees this task.", and every row is labelled either Patient sees this or Care team only. Nothing is ever inferred here — a task is invisible to the patient until somebody explicitly says otherwise. Tasks that arrive from a template are the exception: they're patient-facing by definition, which is what makes them a checklist.

Activating and publishing

Activate turns a draft into the live plan. Three things happen at once: any plan the patient was already on is superseded, this plan becomes active, and version 1 is published automatically. A patient has exactly one active plan, so activating a second one always ends the first.

If Activate is unavailable, the reason is written underneath — "Add at least one health concern to activate." or "Add at least one goal to activate."

After that, editing and publishing are separate acts. Change anything a patient can see and an amber banner appears:

This plan has unpublished changes — "The patient is still seeing version 2. Publish to share your edits."

It can only be dismissed by publishing. It's driven by comparing the actual content of the plan against what was last published, not by timestamps, so it appears when something the patient reads has genuinely changed — and stays away when you tick a care-team task or edit something they never see.

Each published version is frozen: the content, a hash of it, and the exact PDF bytes the patient was given. That's what makes the history worth anything six months later when someone asks what the patient was actually told. Each row in Versions has its own PDF button that downloads that version.

You can download the plan as it stands at any point, but an unpublished working copy comes out stamped "Working copy — not the published plan" so it can't be mistaken for the document the patient holds.

Ending a plan

Deactivate… cancels a plan and requires a reason — Goals met, Patient request, Transferred, No longer appropriate, Created in error, or Other. The dialog is blunt: "will be cancelled and the patient will stop seeing it. This cannot be undone."

Superseded and Cancelled are terminal. There is no reopen. Clone is the way back: it copies the concerns, goals and recommendations into a fresh draft with today's start date and brand-new ids.

Two things a clone deliberately leaves behind: attachments, because those documents were shared under the old plan, and care management enrollment, because that needs its own consent decision. If the patient is enrolled in CCM, cloning is the moment that lapses — see Care Time and chronic care management.

Nothing is ever deleted. Archive hides a plan from the list without destroying it, and an active plan can't be archived at all — deactivate it first.

Sharing photos and videos

The Attachments section shares media from the patient's own chart: "Photos and videos already on this patient's chart. The patient sees what you attach once the plan is published."

Images and video only. There is no way to attach a PDF or a document to a care plan today. Upload new puts a file on the chart and then you select it — two steps on purpose, so nobody shares something with a patient by accident.

Removing an attachment takes it off the plan and leaves it on the chart. Adding or removing one counts as a change the patient can see, so the plan will need republishing before it reaches them.

What the patient sees

Patients get a Care Plan page in their portal, and it appears only when they actually have an active plan — not merely because your clinic turned the feature on. A patient with no plan sees no tab and no tile, rather than an empty page.

The plan is rewritten into their language. Sections are headed Your steps, What we're working towards, What to do, What this plan is for and Shared documents. The clinical vocabulary softens: a goal marked Not achieved reads as "Still working on it", Education recommendations appear under "Good to know", and Custom ones under "From your team".

Patients tick their own steps, and that flows straight back to the task list on your side. They can also Download PDF.

They only ever see the last published version. A half-finished edit is never visible, and neither is anything you kept to the care team — care-team-only tasks, your internal notes, which template the plan came from, and the CCM enrollment behind it are all structurally excluded from what gets sent.

Who can do what

Anyone on your clinical team — owner, admin, doctor or front-desk staff — can build templates, create plans, activate them and publish to a patient. Only owners and admins can turn the feature on or off in Settings.

See Roles, licences, and permissions.

Details that matter

• Templates are stamps, not links. Editing one never changes a plan already applied from it.
• A template needs a concern and a goal before it can be applied to anybody.
• Starter protocols import once, and can't be previewed first — import, read, retire if it's wrong.
• One active plan per patient. Activating a second supersedes the first, automatically.
• Activating publishes version 1. Every edit after that needs a deliberate Publish.
• Tasks are hidden from the patient by default. The switch is off until you turn it on.
• Superseded and Cancelled are terminal. Clone into a fresh draft — and remember the clone drops attachments and CCM enrollment.
• Attachments are photos and videos only. No PDFs, no documents.
• Nothing is deleted, ever. Plans archive, templates retire, versions are permanent.