Your ophthalmologist mentioned something called AdaptDx — or a "dark-adaptation test," or "the rod-intercept test." Maybe it came up because you are over 60, or because someone in your family has macular degeneration, or because you mentioned that night driving has gotten harder. The recommendation landed without much explanation, and now you are weighing twenty extra minutes in the chair, a self-pay charge that probably is not covered by insurance, and a test you have never heard of.
This post is the practical walk-through. What the test measures, what the appointment actually feels like in the room, what the number means, what it cannot tell you, and how it sits next to the rest of the eye exam. Audience: you, between the recommendation and the appointment.
What a dark-adaptation test measures
Walk from a sunlit parking lot into a dim restaurant and for a long moment the room is a black wall. A minute later shapes return. Five minutes later you can read the menu. Twenty minutes in — say, on a moonlit walk — you can see stars you didn't know were there. That recovery is dark adaptation, and the curve it traces is one of the more elegant measurements in vision science. The longer version of that story lives in our dark-adaptation primer. The short version: your rod photoreceptors — the dim-light receptors blanketing the retina outside the macula — depend on a small molecule (11-cis-retinal) that is recycled in the tissue layer beneath them, the retinal pigment epithelium (RPE). The cycle is slow. After a bright bleach, the rods need many minutes to rebuild their light-detection pigment and reach full sensitivity.
A clinical dark-adaptation test measures how long that recovery takes. The most common instrument in US optometry practices is the AdaptDx, made by a company called MacuLogix (now part of LumiThera) — a tabletop device about the size of a small printer with a chin rest, a viewing aperture, and a hand-held response button. There is a faster successor, the AdaptDx Pro, that runs in a head-mounted goggle. Either way, the device runs a standardised protocol and reports a single number: the rod-intercept time, abbreviated RIT.
Rod-intercept time is the number of minutes between the end of a controlled bleaching flash and the moment your rod-mediated threshold falls to a defined sensitivity criterion. In healthy young eyes it lands around four to five minutes. The clinical cutoff the device uses, validated against age-matched cohorts, is 6.5 minutes — anything longer is flagged as impaired dark adaptation. That cutoff and the device behind it received FDA 510(k) clearance for assisting in the assessment of dark adaptation function.
What makes the test interesting clinically is what slow rod recovery is associated with. Delayed rod-mediated dark adaptation is one of the earliest known functional signals of age-related macular degeneration (AMD) — measurable in some patients before drusen are visible on a standard fundus exam, and associated with the development of early AMD over a few years of follow-up (Owsley et al., 2016). The mechanism is plausible. The RPE that recycles rod pigment is exactly the layer AMD damages first. When that supply chain weakens, the rods slow down before the macula looks abnormal on imaging.
The protocol, step by step
The room is dimly lit when you walk in. The technician seats you, explains the test, and confirms which eye is being tested first. One eye is patched. You rest your chin on a support and look into an aperture at a faint fixation target.
The bleach. A brief, bright white flash — about a quarter of a second, bright enough that you blink reflexively. It is uncomfortable for the moment it happens; it is not painful and there are no after-effects beyond the test. The flash bleaches a controlled fraction of your rhodopsin so the recovery clock starts from a known starting point.
The test phase. Then you sit. The aperture shows nothing for a while except the fixation target. Periodically a faint test stimulus appears at a defined location off the centre of your gaze — the rod-rich zone outside the fovea. The instant you see it, you press the button. The device steps the stimulus brighter or dimmer based on whether you caught the previous one (an adaptive staircase, the same general method we use in our home test) and tracks your threshold across the minutes as your rods recover.
The result. As soon as your threshold falls below the device's criterion the test ends. The full protocol takes around five to twenty minutes per eye depending on how quickly your rods recover; the faster AdaptDx Pro variant uses a "rapid test" mode that completes in roughly six and a half minutes maximum. If your rods are recovering on schedule, you may be done in five minutes; if they are slow, the device runs longer and reports the longer time.
The experience itself is undramatic. The bleach is brief. The dark-room sit is the long part. The button press becomes automatic. Most patients describe it as boring more than anything else — which is the right shape for a test like this: nothing to perform, no judgement of right and wrong answers, just press as soon as you see a faint dot. You can blink normally. You can move your eyes if the technician says it is fine. You will not feel sleep-deprived afterward; the bleach is not bright enough to leave a lasting after-image once you are back in normal light.
The technician will likely repeat the test on the second eye. With both eyes, total chair time is typically twenty to forty minutes including setup, conversation, and the actual measurement.
What the number can tell you
A rod-intercept time at or below 6.5 minutes is reassuring — your rod system is recycling pigment within the range typical for healthy older adults. A reading above 6.5 minutes, particularly one well above it, is a real signal that the rod–RPE complex is slower than expected.
In the AMD research literature, that signal earns attention. Owsley and colleagues at the University of Alabama at Birmingham followed older adults with healthy maculas at baseline and showed that those with delayed rod-mediated dark adaptation were several times more likely to develop early AMD over three years compared with peers whose dark adaptation was normal (Owsley et al., 2016). That is what "functional biomarker" means in this context: a measurement that predicts the development of structural disease before the structure looks obviously abnormal.
What the number is good for, in clinical practice:
- A baseline when AMD risk is on the table — family history, age, smoking history, light irises — and the standard exam is still clean.
- A prompt for closer follow-up. A delayed RIT does not commit you to a diagnosis; it commits the practice to looking more carefully at the macula on OCT and scheduling sooner-than-annual follow-up.
- A quantitative anchor for the conversation about lifestyle changes (smoking cessation, AREDS2 supplementation when indicated, UV protection) when the rest of the exam is borderline.
What the number cannot tell you
Worth being explicit about, because this is where well-meaning tests get oversold.
A delayed RIT is not a diagnosis of AMD. It is a functional signal that something in the rod–RPE pathway is slower than typical. AMD is the most-discussed reason a clinician might pursue that signal, but the rod–RPE story has other causes — and the AdaptDx itself does not distinguish among them.
Other conditions that delay dark adaptation include vitamin A deficiency (rare in well-fed populations, more common in malabsorption syndromes — post-bariatric surgery, severe Crohn's disease, advanced liver disease, highly restrictive diets), retinitis pigmentosa and related rod dystrophies, Stargardt disease, certain medications (isotretinoin, amiodarone), and the slow drift of normal aging itself. A reading above 6.5 minutes is a prompt to investigate, not an answer.
A normal RIT does not rule out AMD. The signal is sensitive but not perfectly so; some patients with early AMD have RITs within the normal range, and the dark-adaptation pathway can be intact while other macular function (cone-driven detail, contrast sensitivity, central acuity) is already affected. A normal AdaptDx result is reassuring on one specific axis; it does not exonerate the macula.
The test does not image anything. It is a functional measurement. It cannot show drusen, geographic atrophy, intraretinal fluid, or any other structural feature. The tools for those are the dilated fundus exam, OCT, and (when wet AMD is suspected) fluorescein angiography. AdaptDx adds an early-functional layer to that toolkit; it does not replace any of it.
Single-test variability is real. Sleep, time of day, hydration, recent bright-light exposure, and concentration all introduce some noise. A single delayed reading in an otherwise low-risk patient is usually repeated before any clinical decision is made on it.
How it fits next to contrast sensitivity
Dark adaptation and contrast sensitivity are siblings — not twins. Both are functional measurements that catch parts of vision the eye chart misses; they measure different parts of vision through different photoreceptor systems.
Contrast sensitivity is typically tested in normal room lighting on a stationary pattern. It probes the cone system at steady state — the daytime, detail-and-edges workforce concentrated in the macula. The classic clinical tools (Pelli-Robson, FACT, CSV-1000) and at-home tests like ours all live in this photopic regime.
Dark adaptation is a temporal measurement of the rod system — the dim-light receptors outside the macula — and how quickly it recovers after a bleach. It is fundamentally about transitions into low light: walking into a dim hallway, the first minutes of dusk driving, the recovery from oncoming headlights at night.
In AMD specifically, both signals have been documented as early. A delayed RIT and a reduced cone-mediated contrast sensitivity both appear before standard visual acuity falls in many patients — and they often appear independently of each other. Either can be present without the other. Together they cover more of the early functional window than either alone, because they ask the macular tissue two different questions.
For a patient using both: the AdaptDx number, run in the clinic perhaps once a year, gives the rod-side picture. A monthly at-home contrast sensitivity track gives the cone-side trend on a tighter cadence. Neither replaces the dilated exam and OCT that remain the diagnostic tools. Both add texture the eye chart misses.
Cost and access
Dark-adaptation testing is generally not covered by US medical insurance for screening purposes. Most practices that run it charge self-pay, typically somewhere in the $50 to $150 range per visit, with some variation by region and by whether one or both eyes are tested. A small number of practices bundle it into a "macular health screening" package; ask in advance what the charge is and whether your AMD risk profile changes the answer.
Access is uneven. The device is concentrated in AMD-focused optometry and ophthalmology practices and in some retinal-specialist offices. Smaller general optometry practices and many large chains do not have one. If your ophthalmologist mentioned the test, they likely have access — either in-house or through a partner. If you are looking for testing without a direct referral, calling local optometry practices and asking specifically about "AdaptDx" or "dark adaptation testing" is the most direct route.
Some practices include an AdaptDx reading in their standard older-adult workup at no extra charge as part of a broader macular health assessment. The economics of that vary, and the device is an investment for the practice, so a self-pay add-on is the more common model.
What to ask the doctor
A few questions that earn their keep before saying yes:
- Given my specific risk factors — age, family history, smoking history, current eye exam findings — what would you do differently if this number came back delayed? The answer should be concrete. If the answer is "nothing specific" the test is not going to add much for you.
- If the number is delayed, what is the follow-up protocol? A reasonable answer involves a closer look at the macula on OCT, possibly sooner-than-annual follow-up, and a conversation about modifiable risk factors. An evasive answer is worth a second opinion.
- Will this result be repeated before any clinical decision is made on it? Single-test variability is real; a one-off delayed reading in an otherwise low-risk patient is usually retested.
- Are there other reasons my dark adaptation might be slow that we should consider? Vitamin A status, medications, and family history of rod dystrophies all belong in the differential when a number comes back delayed.
- How much does this cost, and is any of it covered by my insurance? Always worth asking explicitly before the test runs.
A clinician who can answer these clearly is using the device the way it is meant to be used: as one input among several, in a specific population where the answer changes what happens next.
Take a baseline you can bring back
Note. A dark-adaptation test measures the rod-driven recovery from a bleach. Our home test measures cone-mediated contrast sensitivity in normal room lighting. They are different parts of vision through different photoreceptor systems. A normal contrast sensitivity result does not mean your dark adaptation is normal; a normal AdaptDx result does not mean your contrast sensitivity is normal. The two sit alongside each other, and neither replaces the dilated eye exam and OCT that remain the diagnostic tools for AMD.
If your ophthalmologist has flagged AMD risk and an AdaptDx appointment is on the calendar, a few weeks of contrast sensitivity tracking on the same device in similar lighting is one of the more useful inputs you can show up with. It does not replace the clinical testing. It does give the conversation a longitudinal record the appointment otherwise would not have, on a part of vision the AdaptDx does not directly probe.
Take the test now, save the share link, retake in a week or two. Bring the links to your appointment alongside whatever your clinician has scheduled. The AdaptDx asks one question of your retina; the contrast sensitivity track asks another. Together they are more informative than either alone — and both sit beside, not instead of, the dilated exam and OCT that are doing the actual diagnostic work.
References
- Owsley, C., McGwin, G., Jr., Clark, M. E., Jackson, G. R., Callahan, M. A., Kline, L. B., Witherspoon, C. D., & Curcio, C. A. (2016). Delayed rod-mediated dark adaptation is a functional biomarker for incident early age-related macular degeneration. Ophthalmology, 123(2), 344–351. The load-bearing paper for dark adaptation as an early-AMD functional biomarker. Older adults with healthy maculas at baseline but delayed rod-mediated dark adaptation were several times more likely to develop early AMD within three years of follow-up — the result that motivates the clinical use of the AdaptDx for AMD risk assessment.
- Jackson, G. R., Scott, I. U., Kim, I. K., Quillen, D. A., Iannaccone, A., & Edwards, J. G. (2014). Diagnostic sensitivity and specificity of dark adaptometry for detection of age-related macular degeneration. Investigative Ophthalmology & Visual Science, 55(3), 1427–1431. Multi-site validation study reporting that a rod-intercept-time cutoff of 6.5 minutes yielded high diagnostic sensitivity and specificity for AMD versus healthy older controls — the empirical basis for the cutoff the AdaptDx device reports.
- Lamb, T. D., & Pugh, E. N., Jr. (2004). Dark adaptation and the retinoid cycle of vision. Progress in Retinal and Eye Research, 23(3), 307–380. Molecular and biophysical review of dark adaptation. Establishes that after a large bleach, recovery time is set by the rate of 11-cis-retinal delivery to opsin in the rod outer segments — the mechanism the AdaptDx is, indirectly, measuring.
- Pelli, D. G., Robson, J. G., & Wilkins, A. J. (1988). The design of a new letter chart for measuring contrast sensitivity. Clinical Vision Sciences, 2, 187–199. Methodology anchor for clinical contrast sensitivity measurement and the broader argument that contrast sensitivity is a clinically meaningful complement to standard visual acuity. Cited here for the cone-side functional measurement that sits alongside the rod-side AdaptDx result.