When to Retest: Timing a CPT Around Stimulant Dosing

There’s no single correct time to give a continuous performance test (CPT) relative to a stimulant dose — the right timing depends on the clinical question. If you’re asking “does this dose control symptoms during the hours that matter to this patient,” test at that time of day, on medication. If you’re asking “how does this patient perform off medication,” test during a documented washout, mindful of what that means for the patient’s day. What matters most is testing under the same conditions each time, and recording exactly what those conditions were, so a comparison across visits means what you think it means. This guide walks through the timing decisions that affect a CPT’s usefulness during titration.

Photo by Behnam Norouzi on Unsplash

Why does testing timing matter for a CPT?

A CPT measures sustained attention, response inhibition, and reaction-time variability at the moment the patient takes it — not an average of how they function across a day or a week. That’s part of its value: it’s a controlled, repeatable snapshot. It’s also the reason timing changes what the snapshot shows. Stimulant medications have a pharmacokinetic profile — an onset, a peak, and a duration of action — and where in that curve a patient sits when they test will affect the result independent of anything about their underlying attention capacity [1]. Two tests given at different points in a dosing cycle aren’t really the “same” measurement, even if the same instrument was used both times.

This is the same logic behind using objective data in ADHD medication titration: a trend across sessions is only informative if the sessions are comparable. Timing is one of the biggest levers you have over comparability.

On medication or off medication — which should you test?

Both are legitimate, and they answer different questions.

Testing on medication, timed to the period the medication is meant to be covering (for example, mid-morning for a patient who takes an extended-release dose at breakfast to cover a school or work day), tells you whether the current dose and formulation are controlling attention and impulse control during the hours that matter functionally. This is the more common approach during active titration, since the clinical question is usually “is this dose working,” not “what does this patient look like unmedicated.”

Testing off medication (at diagnostic baseline, or during a deliberate medication-free window) tells you how the patient performs without pharmacological support, which is useful for establishing a true baseline before starting treatment, or for a periodic reassessment of whether ongoing treatment is still indicated. Off-medication testing during ongoing treatment means asking a patient to go without their medication for a defined period, which has its own considerations for daily functioning, safety, and any interim symptoms — discuss it with the patient in advance rather than treating it as a routine step.

Whichever you choose, the record should say which one it was, and future retests should match it. A titration trend that mixes on-medication and off-medication sessions without noting which is which is much harder to interpret.

How stimulant duration of action affects same-day timing

Stimulant formulations differ substantially in onset, peak, and duration, which is why “test in the afternoon” means something different depending on the formulation:

Category	Typical onset	Typical duration	Testing implication
Immediate-release methylphenidate	About 1–2 hours to peak	Roughly 3–6 hours	Coverage gaps between doses are common; testing timing relative to the last dose matters a lot
Extended-release methylphenidate (e.g., OROS formulations)	Initial rise within about an hour, gradual peak over several hours	Approximately 10–12 hours	A single morning dose may or may not still be active by a late-afternoon test
Amphetamine extended-release	Onset within 1–2 hours	Roughly 8–12 hours	Similar mid-to-late-day coverage question
Lisdexamfetamine (prodrug)	Slower onset, up to about 2 hours	Often 10–14 hours	Longest typical coverage window among common stimulants

These are general ranges drawn from product labeling and pharmacology references, not a substitute for the specific product label you’re prescribing [1][2][3]. The clinical point isn’t the exact numbers — it’s that “on medication” can mean very different things depending on the formulation and how many hours have passed since the dose. Record the formulation, dose, and time since last dose alongside every test result.

How does practice effect change your retest schedule?

Separate from medication timing, repeating any cognitive task introduces a practice effect — performance can improve simply from familiarity with the task, independent of any real change in attention or impulse control. Research on continuous performance tasks and related cognitive measures generally finds that practice effects are largest between the first and second administrations and diminish with further repeats, and that longer intervals between tests are associated with smaller practice effects than very short intervals [4][5]. A 2023 study of open-source attention tasks specifically found practice effects on repeat CPT administration [6].

Practically, that means:

• Interpret the first retest cautiously. Some of an improvement from baseline to the first follow-up may reflect familiarity with the task, not just the medication.
• Space retests out where clinically reasonable. Testing every few days for a fast-moving titration schedule will show more practice-effect noise than testing at each dose plateau, typically weeks apart.
• Watch trends, not single-session jumps. A study that tracked children and adolescents on ADHD medication with a CPT at baseline, one month, and 12 months found symptom reductions on both the objective measure and an investigator-rated scale across that longer timeline [7] — the kind of multi-point trend that’s more informative than any two-point comparison alone.

Try it with your next patient. AxonCPT gives you 2 free tests every month — no card required. Send a link, your patient finishes in ~16 minutes from home, and you get a norm-referenced report instantly. Create your free account →

What should you standardize between sessions?

To keep repeat CPT results comparable, hold as many of these constant as possible across visits:

• Time of day. Test at a similar clock time relative to the patient’s schedule and dosing.
• Time since last dose (or confirmed off-medication status). Record it every time, not just at baseline.
• Formulation and dose, and note any change since the last test.
• Environment. For remote testing, a quiet room, a consistent device, and no phone nearby reduce noise unrelated to medication. Our guide to remote CPT testing validity covers the evidence on environment controls.
• Sleep and recent substances. Poor sleep the night before, or caffeine and alcohol use, can shift performance independent of the stimulant dose.

Is repeat testing during titration medically necessary — and billable?

APA Services’ billing guidance lists monitoring response to intervention among appropriate uses of objective testing, but also flags repeat testing ordered without a documented clinical reason as not reasonable and necessary [8]. The fix is the same one that makes the clinical data useful in the first place: document why you’re retesting now (a dose change, a new symptom report, a discrepancy between a rating scale and observed function) and what the result changed about the plan. For code selection on a repeat automated test, see our guide to CPT 96146, and for the cost side of a multi-test titration course, see the cost of repeat CPT testing during titration.

How this looks in practice

A psychiatrist titrating an adult patient on an extended-release stimulant might structure retesting like this:

1. Baseline, before starting medication or at a documented off-medication window, with the conditions recorded.
2. First on-medication retest, timed to the hours the current dose is meant to cover, several weeks after starting (not days), to let the practice effect settle and give the dose time to show a stable response.
3. Retest at each subsequent dose change, at the same time of day relative to dosing, with formulation, dose, and hours since last dose documented each time. With objective ADHD testing with AxonCPT, the patient completes the standardized test remotely in about 16 minutes, and the report — five metrics against age- and sex-matched norms — is ready immediately, so you can review it before the follow-up visit rather than after.
4. Read the trend, not any single session, alongside the rating scale and the patient’s report of function during the hours the dose is supposed to cover.

You can see what the report looks like to judge how a dated, condition-labeled report fits your documentation.

Frequently asked questions

Should I always test patients on medication? Not necessarily. Test on medication when the question is whether the current dose and timing control symptoms during functional hours. Test off medication for a true baseline or periodic reassessment of ongoing need. Either is valid as long as the condition is documented and held consistent across retests.

How soon after starting a new dose should I retest? There’s no fixed interval, and this is a clinical judgment, not a rule this article can set. Many clinicians wait until the dose has reached a stable, clinically observed effect — often several weeks — rather than retesting within days, partly to let any initial adjustment settle and partly to reduce practice-effect noise on an immediate second session.

Does testing on medication mean the result isn’t a “real” baseline? It means the result reflects performance under that specific condition — on this dose, at this time since the last dose. That’s exactly the condition you’re usually trying to measure during titration. It becomes a problem only if you later compare it to an off-medication result without accounting for the difference.

Can a patient’s sleep or caffeine intake the night before affect the result? Yes. Poor sleep, caffeine, alcohol, and other substances can all shift attention-task performance independent of stimulant dosing. Ask about these factors before the test and record anything unusual alongside the result.

Does AxonCPT tell you when to test relative to a patient’s dose? No — AxonCPT is a testing platform, not a titration protocol. The timing decisions described here are clinical judgments you make based on the patient’s formulation, dosing schedule, and the question you’re trying to answer.

Start testing your patients with AxonCPT

A titration trend is only as good as the consistency behind it. AxonCPT lets you send a standardized test whenever your clinical timing calls for it, with the patient completing it from home in about 16 minutes, no app or login required. You get 2 free tests every month, with no card required. Create your free AxonCPT account or learn more at axoncpt.com.

References

1. U.S. Food and Drug Administration. Concerta (methylphenidate HCl) extended-release tablets prescribing information. Accessed September 28, 2026.
2. U.S. Food and Drug Administration. Adderall XR prescribing information. Revised October 2023.
3. Psychiatrist.com. Stimulant Formulations for the Treatment of Attention-Deficit/Hyperactivity Disorder. Accessed September 28, 2026.
4. Chen H, et al. Practice Effects and Test-Retest Reliability of the Continuous Performance Test, Identical Pairs Version in Patients with Schizophrenia over Four Serial Assessments. PubMed. Archives of Clinical Neuropsychology. 2020.
5. Bergman H, et al. Test–retest reliability, practice effects and estimates of change: A study on the Mindmore digital cognitive assessment tool. Scandinavian Journal of Psychology. 2025.
6. Langner R, Scharnowski F, Ionta S, et al. Evaluation of the reliability and validity of computerized tests of attention. PLoS One. 2023;18(1):e0281196.
7. Cedergren K, Östlund S, Åsberg Johnels J, Billstedt E, Johnson M. Monitoring medication response in ADHD: what can continuous performance tests tell us? European Archives of Psychiatry and Clinical Neuroscience. 2022.
8. APA Services Inc. 2026 Psychological and Neuropsychological Testing Billing and Coding Guide. 2026.