River
Y CombinatorBacked by Y Combinator
Log InStart Free
FREE TEMPLATE

Treatment Plan Documentation Template

Four documents and three sheets, computing every plan's review date from the interval and anchor its own service line uses.

Edit with AIDownload (.docx + .csv)

Free download  ·  No account needed

Review Due Tracking

The due date is computed, and the formula is not the same on every row

One row per plan. Anchor type decides which formula runs, and the authority column is on the row so the date can be checked instead of trusted.

Two anchor types, two formulas

Floating anchors add the interval to the last review. A fixed grid does not: the windows run from one date and reviewing early does not move the next one. Adding the interval to the last review is therefore wrong for a grid line in both directions, and only one of those directions is safe.

Why the ninth and tenth columns are both there

Days late compares badly across intervals. Sixteen days late on a fifteen-day interval means two review dates have gone by. Forty-four days late on a ninety-day interval means one. Sorting by days late puts the worse one second.

Signature is a separate test

Two signature columns, because a periodic signature and a signature on a mid-cycle change are different events. A plan current on its schedule with an unsigned change from six weeks ago is defective, and one status field calls it fine.

Two care plans opened on the same morning in two service lines are due for review 45 days apart. Home health plans have to be reviewed and revised no less frequently than once every 60 days, beginning with the start of care date. A hospice plan is reviewed no less frequently than every 15 calendar days. Outpatient rehabilitation recertification runs at 90 days. So a single quarterly or annual review policy is wrong about all three at once.

The interval is only half of it. The anchors differ too, and one of them is a fixed grid. Home health windows run from the start of care date. A plan reviewed a week early does not get its next date pushed a week out, so adding the interval to the last review is the wrong formula in both directions. At Kestrel Health Services that error would have been seven days conservative on one plan and five days optimistic on another, and the optimistic one is how a register loses its credibility.

So the review-due date is computed and the authority behind it sits on the row, which is what makes it checkable rather than believable. Signature is a second, independent test. A change to a rehabilitation plan has to be made in writing and signed, and incorporated in the plan immediately. So a plan current on its schedule with an unsigned change from six weeks ago is defective, and nothing about it looks wrong. Three of twelve plans were late and three had a signature problem, overlapping on two.

A third of the caseload carrying seven tenths of the review work

The computed due dates with the naive formula beside them, the per-payer register that explains why two identical plans differ, and exposure measured in review events rather than plans.

Review Due Tracking

Illustrative figures for a fictional provider, Kestrel Health Services, read on 31 March 2026. Twelve plans, four in each service line. Dates are 2026 unless the anchor precedes it. The eighth column is what the wrong formula would have said.

PlanService lineAnchor dateAnchor typeIntervalReviews recordedReview dueLast review plus intervalDays lateMissed cyclesSignatureChange signedAuthority
             
             
PlanService lineAnchorAnchor typeIntervalReviews recordedReview dueLast review + intervalDays lateMissed cyclesSignatureChange signedFinding
CP-101Outpatient rehab01-05Last cert, floating9001-0504-05Same formula——Yesn/aNone
CP-102Outpatient rehab01-12Last cert, floating9001-1204-12Same formula——YesNo, change 02-20Change unsigned
CP-103Outpatient rehab11-17Last cert, floating9011-1702-15Same formula441Non/aReview interval and signature missing
CP-104Outpatient rehab02-09Last cert, floating9002-0905-10Same formula——YesYes, change 03-02None
CP-201Home health01-26Start of care, grid6003-2005-2605-19——Yesn/aNone
CP-202Home health12-15Start of care, grid6002-11; 03-2506-1305-24——Yesn/aNone
CP-203Home health02-02Start of care, grid6003-1506-0205-14——Yesn/aNone
CP-204Home health11-03Start of care, grid6012-2803-0302-26281Non/aReview interval and signature missing
CP-301Hospice01-14Last review, floating1503-2004-04Same formula——Yesn/aNone
CP-302Hospice02-02Last review, floating1503-2404-08Same formula——Yesn/aNone
CP-303Hospice02-16Last review, floating1503-1804-02Same formula——Yesn/aNone
CP-304Hospice01-27Last review, floating1502-2803-15Same formula162Yesn/aReview interval
Twelve plans       88 Two missingOne unsignedFour with a finding

Compare the seventh and eighth columns on the four home health rows. CP-201 was reviewed a week inside its window, so the grid puts the next date at 26 May while last review plus 60 says 19 May: seven days of work the register would have asked for and did not need. CP-204 is the dangerous direction. Its window two deadline was 3 March and the wrong formula says 26 February, five days early, so the alarm would have gone off before the plan was actually late and the practice would have learned to distrust it. CP-304 is the row a days-late sort gets wrong: 16 days on a 15-day interval is two missed review dates, against CP-103 at 44 days on a 90-day interval, which is one.

Requirement Register

One row per service line and payer, because a payer can shorten a federal interval and none can lengthen one. All three federal intervals are written as no less frequently than, which is what makes the contract column meaningful.

Service linePayerIntervalAnchor typeAuthoritySignature required fromChange needs its own signatureContract shortens itInterval used
Outpatient rehabMedicare Part B90Last certification, floating42 CFR 424.24(c)(4)(i)Treating physician with knowledge of response to treatmentYesNo90
Outpatient rehabCascade Mutual90Last certification, floatingPlan medical policy 31-208 adopts the Medicare floorTreating physicianYesNo90
Outpatient rehabRidgeline Health90Last certification, floatingProvider agreement schedule C shortens the review to 60Treating physicianYesYes60
Home healthMedicare Part A60Start of care, fixed grid42 CFR 484.60(c)(1)Physician or allowed practitioner responsible for the planYesNo60
Home healthState Medicaid60Start of care, fixed gridState plan adopts the federal intervalPhysician or allowed practitionerYesNo60
HospiceMedicare Part A15Last documented review, floating42 CFR 418.56(d)Interdisciplinary group with the attending physicianYesNo15

Row three is why this sheet is per payer rather than per service line. Outpatient rehabilitation is a 90-day interval everywhere except under one provider agreement that shortens it to 60, so two identical plans in the same service line have due dates a month apart and only the payer column explains it. Note also that the permitted signer differs between the two outpatient rows and the recertification requirement is narrower still, so "signed" is not one test. The authority column is the load-bearing one: a due date whose authority reads "our policy" cannot be defended, and everybody stops believing the register.

Finding Register

The same twelve plans arranged by service line, with exposure computed before findings are counted. Exposure is review events per year rather than plan count, which is the figure that decides how the work gets staffed.

Service linePlansIntervalReview events per plan per yearLine exposureShare of review eventsShare of plansFindingsInterval onlySignature onlyBoth
Hospice41524.3397.3270.6%33.3%11——
Home health4606.0824.3217.6%33.3%1——1
Outpatient rehab4904.0616.2411.8%33.3%2—11
Total12  137.88100%100%4112
Findings per hundred review events      2.9   

Four plans in each line and hospice carries 70.6% of the review work, because 365 over 15 is 24.33 events a plan a year against 4.06 for rehab, a ratio of exactly 6.0. That is the sentence to take away: review work is not proportional to caseload and staffing it as though it were is how a fifteen-day interval quietly becomes a sixteen-day one. The finding columns then separate what an audit finds from what a practice never finds on its own. The two both-columns entries are the obvious cases, a late review on an unsigned plan. The signature-only entry is CP-102, which is current on schedule and correctly signed for its certification, and has a change from 20 February that nobody signed. Nothing about that plan looks wrong until you ask the second question.

Edit with AI

What's in the pack

01

How the Review Date Is Derived

The two anchor forms written out as procedures, the missed-cycles formula, and the exposure calculation. The method note that has to be read before the sheets are worth anything.

02

Plan Format by Service Type

Three formats differing exactly where the regulations differ. Rehabilitation carries the six required contents, home health a window table rather than a date list, hospice an interdisciplinary review record.

03

Requirement Register

One row per service line and payer, carrying the interval, the anchor type, the authority and the permitted signers, plus whether a contract shortens the federal floor. Practice-wide policies live in the practice policy pack.

04

Review Due Tracking

The computed due date with last-review-plus-interval printed beside it, so the difference between a grid anchor and a floating one is visible on every home health row.

05

Review and Update Procedure

Six steps, including reminders tiered as a fraction of the interval rather than a fixed lead time, and recording a review that concluded nothing needed changing. Patient-facing material has its own derived review date in the patient education material pack.

06

Signature Requirements Note

Why the permitted-signer list for a plan change is wide and for a recertification is narrow, and the four columns that separate nobody signed from the wrong role signed.

07

Finding Register

Interval findings, signature findings and both, per service line, with exposure computed first so a finding rate is per review event rather than per plan.

08

Where the documentation itself gets improved

The plan's clinical content stays the clinician's. A documentation improvement review reads what is there, the payer audit pack handles it when a reviewer asks for the plans, and a telehealth encounter has its own non-clinical elements to get right first.

How to use it

  1. 1

    Open in River, or download it

    Open the pack in River and let the agent build the requirement register from your service lines and payer agreements, or download the blank Word and CSV files instantly and start with one line.

  2. 2

    Fill the anchor type before the interval

    It decides which formula runs. Confusing a fixed grid with a floating one is the single most common error here, and it produces wrong dates in both directions.

  3. 3

    Put the authority on every row

    The regulation, the contract section or the standard, cited precisely enough to go and read. A due date whose authority reads "our policy" is an assertion and gets treated as one.

  4. 4

    Run both tests, separately

    Reviewed on time, and signed by a permitted role including any mid-cycle change. Collapsing them into one status field reports the second failure as fine, permanently.

Frequently asked questions

Is this template free?

Yes. The four documents and three sheets download as Word and CSV files with no signup and no card. "Edit with AI" is the optional path for providers that want the register built from their own service lines and payer agreements. The rest of the library is at the template library.

What format are the downloaded files?

Word documents for the derivation note, the plan formats, the review procedure and the signature note, plus CSV for the three sheets, zipped into one file. They open in Word, Pages, Google Docs, Excel, Numbers and Sheets with nothing to convert.

Why not just review every plan quarterly?

Because a quarterly cycle is roughly six times too slow for a hospice plan and slightly too fast for outpatient rehabilitation recertification. In the worked example the same twelve plans generate 137.88 review events a year, and 70.6% of them belong to a third of the caseload.

What is the difference between a fixed grid and a floating anchor?

A floating anchor restarts from the last review, so reviewing early moves the next date out. A fixed grid runs from one date and does not move. Home health is a grid measured from the start of care; hospice and rehabilitation certification float.

Why track signature separately from the review date?

Because they fail independently. A change to a rehabilitation plan has to be signed and incorporated immediately, not at the next review. One plan in the worked quarter is on schedule, correctly signed for its certification, and carries an unsigned change from February. Nothing looks wrong until you ask.

Can a payer contract change the interval?

It can shorten one. All three federal intervals are written as "no less frequently than", so a contract or accreditation standard can require more frequent review and none can require less. The register carries both, with the regulation on the row as the floor.

Does the pack write or assess the care plan?

No. Every format ships with its clinical fields marked and empty, and the rules file forbids drafting them, judging whether a plan is appropriate, or stating a medical necessity or coverage position. What it tracks is whether the plan was reviewed and signed by its own deadline.

Written by

Chandler Supple

Co-founder & CEO, River

Chandler has spent six years working in AI, building tools for millions of users and Fortune 100 companies along the way. He now leads River as co-founder and CEO.

LinkedInGitHubX

Reviewed by

Brian Joseff, Co-founder, River

LinkedInX

Related

Documentation improvement review

Reads the documentation that exists and names what a reviewer would query, which is the clinical half this register deliberately leaves alone.

Payer audit and RAC response pack

For when a reviewer asks for the plans themselves, where a missing signature and a late review are two separate arguments.

More free templates

HIPAA Medical Records Request Template

Routes every records request by legal basis, because the 30-day clock, the fee cap, minimum necessary and the six-year log attach to different pathways.

Patient Education Handout Template

Free patient education pack: every handout carries a named clinician's approval and a review date derived from what it depends on, not a blanket annual policy.

Patient Intake Form Template

The intake template that gets shorter. Every field names the requirement that reads it, and a field that was right last year is measured as stale, not filled.

Telehealth Encounter Documentation Template

Audits telehealth encounters against five non-clinical elements, priced by element and by payer, because a signed consent isn't always enough.

Find out which of your due dates were computed with the wrong formula

Send the plan list however it exists, an EHR export or a spreadsheet. The first thing back is every due date recomputed on the right anchor, with the naive figure beside it.

Edit with AI