Half to three-quarters of rosacea patients develop ocular involvement. Most trackers have nowhere to log it.

The eye component most trackers miss

Rosacea is not only a skin disease. Roughly half to three-quarters of cutaneous rosacea patients develop ocular involvement at some point in their disease course, and in a meaningful subset the ocular component shows up first, before the centrofacial erythema, the papules, or the visible vessels.

This is not a fringe observation. The ROSCO 2017 phenotype framework (Tan J, et al. Br J Dermatol. 2017;176(2):431-438) lists ocular manifestations as one of the four major phenotypes of rosacea, on equal footing with flushing, telangiectasia, and inflammatory papules. The NRS 2017 expert committee statement (Gallo RL, et al. J Am Acad Dermatol. 2018;78(1):148-155) reaches the same conclusion. The phenotype is real, it is documented, and it has diagnostic weight in the current framework.

The weird gap is that consumer rosacea-tracker apps largely do not capture it. The five rosacea-named apps available on iOS in our category-audit window have no field for ocular symptoms in their daily log. The trackers are operating on the assumption that rosacea is a face-redness disease. Half to three-quarters of patients eventually find out it is not, and at that point the tracker that recorded a year of careful face data has no field for the symptom they actually need help noticing.

What ocular rosacea looks like

The published clinical descriptions of ocular rosacea (see Vieira ACC, Mannis MJ. Ophthalmology, and downstream reviews) cluster around a small number of symptoms patients experience but often do not connect to their skin condition:

- Burning, stinging, or foreign-body sensation in the eye. Not the watery-eye reaction to wind; a persistent low-grade irritation that worsens with reading, screen time, or air conditioning. - Redness of the eyelid margins (blepharitis). A specific kind of redness centered on where the lashes meet the lid, not the broader conjunctival redness of allergy or infection. - Recurrent styes or chalazia. Inflamed lumps on the eyelid that come back even after they resolve. The meibomian gland involvement that drives them is the same gland dysfunction the dermatology literature implicates in ocular rosacea. - Dry eye disease symptoms. Tearing that does not relieve the dryness (a paradox patients find confusing), grit-feeling, photophobia, fluctuating vision after blinks. - Lid telangiectasia. Visible vessels along the lid margin, the eye analogue of the centrofacial telangiectasia of cutaneous rosacea. - Corneal involvement in advanced cases. Less common but more serious, the literature describes it as the reason ocular rosacea matters clinically: vision-affecting corneal damage is the avoidable worst case.

The symptom pattern is often diagnosed and treated as "dry eye" by an optometrist or primary-care visit without the connection to the skin disease being made. Patients can spend years cycling through artificial tears and warm compresses while their dermatologist works on the face redness, neither side knowing the two are the same disease.

What the prevalence numbers actually say

Reported prevalence of ocular involvement among rosacea patients is wide. The literature commonly cites figures in the 50 to 72 percent range, with variation depending on how aggressively the cohort was screened for ocular signs (Vieira and Mannis 2013; Awais et al. 2015; Webster et al. 2017 review; Geng F, et al. JEADV Clin Pract. 2024).

The Geng 2024 piece (JEADV Clinical Practice) is one of the more recent systematic looks at the prevalence question and emphasizes that careful screening with slit-lamp examination produces higher numbers than self-report or unaided clinical exam. A patient survey will miss ocular rosacea in patients who have it but have not connected the symptoms; a slit-lamp screen catches more. The reported prevalence is therefore both a measurement of how often the disease co-occurs and a measurement of how thoroughly the cohort was screened.

The key clinical fact is that the ocular and cutaneous components co-occur far more often than either appears alone, and in roughly 20 percent of patients the ocular symptoms precede the cutaneous signs by months to years (Vieira and Mannis, and downstream summaries). A patient who has been having mild burning eyes for two years and now starts noticing centrofacial flushing has not had two separate problems; they have had one disease with a longer-than-usual lead-in on the ocular side.

The leading-edge framing matters for a tracker. The patient who would benefit most from logging the eye symptoms is the one who does not yet have facial signs to log, and who might therefore go years without seeing a dermatologist who could connect the picture.

What we caught means in practice

If a tracker logs ocular symptoms, what does the patient actually get out of it?

First: a longitudinal record of the eye symptoms before any clinical visit. The patient who shows up at a dermatologist with three months of "burning eyes 4 out of 7 days, worse with screen time" data, plus the parallel record of centrofacial flushing episodes, walks into the visit with the diagnostic picture pre-assembled. The dermatologist does not have to extract that history under time pressure; it is already on the page. This is the patient-reported-outcome "vital sign" framing the JAAD 2022 piece on PROMs makes in dermatology generally.

Second: an early-warning signal for the worst-case complication. Untreated ocular rosacea, in a small subset of patients, progresses to corneal involvement that affects vision. The literature is clear that this is uncommon but not negligible, and that the difference between an outcome that does and does not involve vision changes is often early identification. A tracker that has been recording "persistent eye irritation, gradually worsening, three months running" is exactly the kind of signal that drives a faster referral to an ophthalmologist familiar with ocular rosacea.

Third: a clearer picture of which interventions actually move which symptoms. The cutaneous and ocular responses to a given treatment are not always aligned. A topical metronidazole gel may improve papules and have no measurable effect on lid margin inflammation. A patient who is not tracking the ocular dimension cannot tell that the treatment is working on one phenotype and missing the other; a patient who is tracking both sees the divergence and can bring it to the next visit.

The single common factor across all three is the same: the tracker exists to make the patient a better-informed participant in their own care. The ocular dimension is the place this matters most, because it is the dimension most often missed in clinical practice and the one the patient is least likely to connect on their own.

What Skinframe captures

Skinframe ships an ocular phenotype toggle on the first daily-log surface, alongside the cutaneous toggles.

The ocular log records, on each day the patient flags eye symptoms:

- Burning or stinging frequency. A 0-to-3 rating of how present the sensation has been over the day. - Tear film quality. A patient-language indicator (eyes feel dry, eyes feel grainy, eyes feel normal). The literature term is tear film dysfunction; the in-product copy translates. - Lid margin awareness. A simple flag for whether the lid margins look red or irritated in the mirror, with an optional photo capture for the dermatologist handoff. - Visual disturbance. A flag for any blurring or vision-affecting symptom that does not clear on blinking. This is the one we surface most prominently in patient-facing copy, because it is the symptom that most warrants escalation to an ophthalmologist. - Stye or chalazion in the last week. A binary plus a count, because recurrent meibomian gland inflammation is itself a diagnostic clue.

The dashboard surfaces the ocular trajectory on its own, alongside the cutaneous trajectories. The derm-handoff PDF includes the ocular log on the second page so the dermatologist can scan the eye history without prompting.

None of this is exotic engineering. It is a list of fields the disease has, recorded as the patient experiences them, on a tracker that does not pretend rosacea is only a skin problem. The reason this is unusual is that the consumer-app category has not done the reading. The reason Skinframe does it is that the literature is unambiguous about what the disease covers.

How to check whether your tracker is treating eye involvement seriously

If you are evaluating a rosacea tracker, three quick checks for whether the ocular dimension is part of the product.

1. Is there a field for eye symptoms on the daily log? Not buried under notes; a first-class field on the same surface as the redness and bumps log. If not, the tracker is implicitly defining rosacea as a face disease.
2. Does the onboarding or in-app education mention ocular involvement? This is the cheap signal that the team has read the ROSCO 2017 framework. A tracker that surfaces the phenotype name somewhere in the experience is one that knows the disease has four major phenotypes, not three.
3. Does the derm-handoff (PDF, CSV, summary) include the eye data? This is the test of whether the tracker treats the ocular component as data your dermatologist needs to see or as a note you might mention. The literature consistently identifies the structured-summary handoff as what dermatologists actually want; the ocular component belongs in that summary if the tracker is recording it.

A tracker that fails all three is operating on the 2002 four-subtype model where ocular was a discrete bucket most patients were assumed not to be in. The 2017 update made it a major phenotype that affects more than half of cutaneous patients. Trackers that have not caught up are missing the part of the disease most likely to need an ophthalmologist.