Dose, Cue, Follow Up: Home Programs Adults Actually Do

The home exercise program is the largest dose of therapy most adult rehab patients receive, and the least controlled. A patient on a twice-weekly schedule spends about two hours a week with a clinician and dozens of waking hours without one; whatever happens in those hours is the actual intervention. Yet the artifact that governs them is often a printed sheet of eight exercises, handed over in the last three minutes of a visit, that no one mentions again until discharge.

This article treats adherence as a design problem the clinician owns, not a character trait the patient lacks. It works through the three levers a prescriber actually controls — the dose, the cue, and the follow-up — and assembles them into a visit-by-visit process for physical and occupational therapists working with adult caseloads. For pediatric caseloads, where the program runs through a caregiver, see the companion piece on caregiver carryover; the mechanics here assume the person doing the exercises is the person in the room.

Non-adherence is not an edge case; in the research literature it is closer to the base rate. A systematic review of home-based physical therapy in Disability and Rehabilitation reports non-adherence reaching as high as 70%, and found that the strongest predictors of who follows through are not diagnosis or age but intention, self-motivation, self-efficacy, previous exercise adherence, and social support. Every one of those is a property of how the program lands in the patient’s life, which means every one of them is something the prescription itself can help or hurt.

Read those three findings together and a pattern appears. Programs fail in predictable, unglamorous ways: they ask for too much, they float free of the patient’s actual day, and nobody ever asks about them again. Each failure maps to a lever the clinician controls at the moment of prescription — dose, cue, and follow-up. The rest of this article takes them in order.

The eight-exercise handout usually happens by accretion: everything that seemed useful during the evaluation goes on the sheet, because leaving something off feels like withholding care. But a home program is not a record of everything that could help — it is a prescription for what will actually be done this week. The Henry study is worth knowing here despite its size: among fifteen community-dwelling adults over 65, the group prescribed two exercises performed them with better movement quality at follow-up than the group prescribed eight, while self-reported completion did not differ. More items did not mean more therapy; it meant the same claimed effort spread thinner and executed worse.

The practical discipline is to prescribe the minimum effective dose and say it out loud: the fewest exercises that advance the current goals, with sets, repetitions, and days per week stated as numbers rather than “a few times a day.” Then name the priority explicitly — “if you only do one thing, do the sit-to-stands” — so that a bad week degrades to the most important exercise instead of to nothing. Everything that did not make the cut is not discarded; it is the progression you add later, one item at a time, after the current dose is actually happening. Adding to a program that is being done is a progression. Adding to one that is not being done is a bigger pile of undone.

A program that needs a free half-hour competes with everything else that wants one, and loses. A program attached to an event that already happens every day does not need to win that fight. The mechanic is simple: for each exercise, pick an anchor — after the morning coffee, while the kettle boils, before turning on the evening news, right after parking the car at work — and write the anchor on the plan itself, in the patient’s words. “Heel raises, 2 × 15, at the kitchen counter while the coffee brews” is a different instruction from “heel raises, 2 × 15, daily,” even though the dose is identical. The first one has a time, a place, and a trigger; the second one has an intention.

The anchor conversation is also where understanding gets confirmed. Before the patient leaves, have them demonstrate each exercise once and explain the plan back — which exercises, what dose, attached to what. The Agency for Healthcare Research and Quality’s teach-back tool is the reference method: the point is to check how well you explained, not to quiz the patient, and “do you understand?” is not a teach-back question. With an adult rehab caseload this takes two or three minutes, and it routinely surfaces the exercise that was misunderstood in a way no printed diagram would have caught.

Dose, cue, and follow-up are not three tips; they are stations in one repeating loop that starts at the end of every visit and closes at the start of the next one. Run it the same way every time and the home program stops being a handout that exits the building and becomes a managed part of the plan of care.

1. 01

   Prescribe the minimum effective dose

   Choose the fewest exercises that advance the current goals — for many adult caseloads that is two to four, not eight. State sets, repetitions, and days per week as numbers, name which single exercise matters most, and hold the rest back as future progressions.

2. 02

   Anchor each exercise to the patient’s day

   For every item, agree on a trigger that already happens daily — a meal, a commute, a kettle, a television habit — and write the anchor into the program in the patient’s own words. An exercise without an anchor is a wish.

3. 03

   Teach it, then hear it back

   Have the patient perform each exercise once and say the whole plan back: exercises, dose, anchors. Fix what comes back wrong before they leave. This is teach-back, not a quiz — it checks the explanation, not the patient.

4. 04

   Send the plan through a channel they will open

   Printed sheet, patient portal, app, or photos of the plan on their own phone — the best channel is the one this patient will look at on a bad day. If your system supports reminders or remote check-ins between visits, use them; trial evidence modestly favors app delivery with remote support over paper alone.

5. 05

   Open the next visit with the program

   Ask about it first, before the table work: “Walk me through what the week actually looked like.” Asking first signals that the program is part of treatment; asking never signals the opposite. Get specifics — which exercises happened, on which days, and where it broke down.

6. 06

   Adjust the dose out loud, and record it

   Whatever the week was, respond by redesigning rather than re-issuing: shrink the dose, move the anchor, swap the exercise, or progress it. Say the change and the reason to the patient, and document what was reported and what changed — the note is where the program’s history lives.

The loop’s value compounds. By the third visit the patient knows the first question is coming, knows it is safe to answer honestly, and knows the program will shrink or shift in response to what they say. That expectation — not the sheet, not the app — is what adherence is made of.

Whatever channel delivers it, the written program should read like a plan for a specific person’s week, not a gallery of exercises. One page. Every line answers what, how much, and when — and the priority and the escape hatch are printed, not implied.

The follow-up conversation fails when it feels like an audit. “Did you do your exercises?” invites a yes, and a caseload full of yeses teaches you nothing. Ask for the shape of the week instead — what happened, which anchors held, where it broke down — and receive the answer as information about the program rather than a verdict on the person. A patient who reports two sessions out of seven and gets a redesigned program in response will keep reporting accurately. A patient who gets a disappointed pause will start reporting seven.

Honest reports are also what makes channel choice rational. The randomised trial by Lambert and colleagues found that people prescribed a four-week program through an app with remote support reported better adherence than those given paper handouts — a modest difference of 1.3 points on an 11-point scale, whose clinical importance the authors themselves call unclear. The sensible reading is not “apps fix adherence”; it is that the remote-support half of the intervention — someone checking in between visits — is doing real work, and that the delivery channel should be whichever one your specific patient will actually open. A portal message mid-week that asks “how are the step-downs going?” is a follow-up; a PDF is not.

Two or three honest cycles of the loop sometimes end in the same report: it is not happening. That is not a dead end; it is a differential. Work through it in order — dose (still too big for this life, this month), anchor (attached to a routine that turned out not to exist), exercise (painful, frightening, or pointless-feeling from the inside), and only then goals (the patient may not want what the plan of care assumes they want, and that conversation belongs in the open). Each hypothesis has a cheap test: shrink to the single priority exercise, move the anchor to the most stable routine the patient has, swap the offending exercise for one that trains the same capacity, or revisit the goals directly.

Document the trail as you go — the reported barrier, the change made, the result. That record is what separates “patient non-compliant with HEP” from a defensible clinical narrative: it shows skilled decision-making, it justifies the visit frequency the plan of care asks for, and if the answer eventually is that home programming is not viable for this patient right now, it shows the conclusion was earned rather than assumed. A single all-or-nothing line in the note cannot do any of that.