The evidence file

What's actually been measured — and what hasn't

Two of the four things in this kit are measurements, not ingredients. Tear osmolarity and a scored symptom questionnaire are both used in eye clinics and in published dry eye research, and the literature on how well they perform — including where they perform badly — is public. We've put the real papers below, including the ones that don't flatter the method.

The other two are the actives in the starter drops: carboxymethylcellulose sodium 0.5% and glycerin 0.9%. These are long-standing lubricant ingredients with published randomised comparisons behind them. What the research consistently shows is comfort and surface-lubrication effects that are broadly comparable to other well-made artificial tears — not a dramatic advantage over them.

Everything below is evidence for those four things individually. None of it is a trial of this kit, this bottle, or this subscription. This product has never been trialled, and we won't imply otherwise.

These are published studies of the individual markers and ingredients — tear osmolarity, a scored symptom questionnaire, carboxymethylcellulose 0.5% and glycerin 0.9%. None of them tested this kit or this drop formula, and several report results that limit what the measurement can honestly claim.

Tear osmolarity

Tears are mostly water with salts dissolved in it. When the water evaporates too quickly, or too little is produced, the salts concentrate and the reading climbs. A higher number is the recognised signal that the tear film is under strain; readings around and below roughly 300–308 mOsm/L are treated as the normal range in the literature.

Described in the TFOS DEWS II Diagnostic Methodology report (2017) and in the diagnostic-performance literature comparing osmolarity with break-up time, staining and Schirmer testing.

Validated symptom questionnaire

A fixed set of scored questions about symptoms and the situations that trigger them — screens, wind, air conditioning, night driving — converted into a single number. It captures what you actually experience, which surface tests on their own don't.

Described in the original reliability-and-validity paper for the Ocular Surface Disease Index (Arch Ophthalmol, 2000) and in the DEWS II diagnostic sequence, where symptom screening comes first and triggers the objective tests.

Carboxymethylcellulose sodium

A viscous, water-binding polymer. It raises the drop's ability to stay on the eye rather than draining away in a blink or two, which is why lubricant comfort tends to last longer than with plain saline.

Described in randomised artificial-tear comparisons using tear break-up time, corneal and conjunctival staining and symptom scores as endpoints.

Glycerin

A humectant that attracts and holds water. Paired with a polymer, it's included to keep water in the film and to buffer surface cells against the osmotic stress of over-concentrated tears.

Described in randomised comparisons of the preservative-free carboxymethylcellulose 0.5% / glycerin 0.9% combination against other artificial tear formulations.

Tear osmolarity

Lemp MA, Bron AJ, Baudouin C, et al. 'Tear Osmolarity in the Diagnosis and Management of Dry Eye Disease.' American Journal of Ophthalmology, 2011;151(5):792–798

Multicentre diagnostic-accuracy study comparing six common dry eye tests in the same subjects

314

Single assessment per subject

Of the six tests compared, osmolarity had the best overall diagnostic performance, with the highest area under the ROC curve (0.89). The most sensitive threshold separating normal from mild/moderate subjects was 308 mOsm/L and the most specific was 315; at a 312 mOsm/L cutoff, sensitivity was 73% and specificity 92%. Other routine tests performed poorly in one direction or the other — corneal staining 54% and conjunctival staining 60% sensitivity, break-up time 45% and Schirmer 51% specificity.…

Tear osmolarity

'Utility of Tear Osmolarity Measurement in Diagnosis of Dry Eye Disease.' Scientific Reports, 2020 (author list not captured in the sources reviewed)

Cross-sectional comparison of osmolarity against OSDI, break-up time, staining, Schirmer and meibomian gland assessment in clinically diagnosed patients vs…

786

Single assessment per subject

Unflattering and important: osmolarity was the only test that failed to separate patients from controls, at any cutoff tested (>308, >316, or an inter-eye difference >8 mOsm/L). Using the 308 mOsm/L cutoff, 58% of patient right eyes exceeded it — but so did 50% of control right eyes. The authors' conclusion is that a single osmolarity number cannot be read in isolation as a diagnosis. This is the main reason a reading here is paired with a symptom score and a human review rather than treated as…

Tear osmolarity

Wolffsohn JS, Arita R, Chalmers R, et al. 'TFOS DEWS II Diagnostic Methodology report.' The Ocular Surface, 2017;15(3):539–574

International expert-consensus evidence review of dry eye diagnostic tests and the recommended testing sequence

Not applicable — review, no enrolled…

The recommended sequence starts with symptom screening (DEQ-5 or OSDI); a positive screen then triggers break-up time, osmolarity and ocular surface staining. Crucially for anyone hoping one number sorts everything: sub-classification into predominantly evaporative or aqueous-deficient requires meibomian gland, lipid layer and tear volume assessment — osmolarity alone does not do it. The report also states plainly that normal and dry eye distributions overlap significantly for all current…

Validated symptom questionnaire

Schiffman RM, Christianson MD, Jacobsen G, Hirsch JD, Reis BL. 'Reliability and validity of the Ocular Surface Disease Index.' Archives of Ophthalmology, 2000;118(5):615–621

Psychometric validation study; dry eye patients and normal controls completed the OSDI alongside other questionnaires and a full ophthalmic examination

139

Single visit plus test–retest

Internal consistency was good to excellent (Cronbach's alpha 0.92) and test–retest reliability was good to excellent, supporting the OSDI as a repeatable severity measure with three subscales: ocular symptoms, vision-related function and environmental triggers. The honest limitation, noted in later analyses of this same data: OSDI scores did not correlate well with clinical tests — correlations with break-up time, Schirmer, lissamine green and fluorescein staining ranged from −0.21 to +0.19 and…

Carboxymethylcellulose sodium

Lee JH, Ahn HS, Kim EK, Kim TI. 'Efficacy of sodium hyaluronate and carboxymethylcellulose in treating mild to moderate dry eye disease.' Cornea, 2011;30(2):175–179

Prospective, randomised, blinded comparison of preservative-free unit-dose 0.5% carboxymethylcellulose vs 0.1% sodium hyaluronate, dosed six times daily

67

8 weeks

Both preservative-free formulations improved corneal and conjunctival staining, tear break-up time and subjective symptoms, with safety maintained at frequent dosing. The efficacies of the two were judged equivalent — meaning carboxymethylcellulose 0.5% performed as well as a common comparator, not better than it. Anyone told a specific polymer is decisively superior to the others is being oversold.

Glycerin

Roth HW, Conway T, Hollander DA. 'Evaluation of carboxymethylcellulose 0.5%/glycerin 0.9% and sodium hyaluronate 0.18% artificial tears in patients with mild to moderate dry eye.' Clinical Optometry…

Multicentre, investigator-masked, randomised, parallel-group, active-controlled comparison of preservative-free CMC 0.5%/glycerin 0.9% vs sodium hyaluronate…

2 weeks

Both preservative-free formulations produced statistically significant improvements from baseline in OSDI symptom scores, tear break-up time, corneal and conjunctival staining and conjunctival hyperemia — but there were no statistically significant differences between the two groups on any measure. Non-significant trends went in both directions: sodium hyaluronate trended toward less blurring after instillation (P = 0.055), while CMC/glycerin trended toward longer symptom control and a better…

What has not been studied

This kit has never been trialled. There is no study of this test strip, this read card, this bottle, this subscription, or this sequence of test-then-match. The evidence above is for the individual markers and the individual lubricant ingredients, studied separately, mostly in clinic settings by trained examiners.

Specifically untested: whether an osmolarity reading taken by a person at their own kitchen table performs the way clinic readings perform in the published literature. Sampling technique, reflex tearing, recent drop use, eye makeup, room temperature and time of day are all real confounders, and the papers above were run under conditions that control for them. We have no published data showing our at-home workflow reproduces clinic-grade numbers.

Also untested: whether matching a drop format to a phenotype produces better outcomes than simply using a good preservative-free lubricant consistently. That's the premise of the product, and it is a reasonable premise built on the DEWS II sub-classification logic — but reasonable is not proven, and no trial has compared a matched-drop routine against an unmatched one.

And nothing here has studied the combination itself: a symptom score plus a strip reading plus a telehealth conversation, feeding into a specific formula. Each part has its own literature. The assembly does not.

This kit is a screening and self-monitoring aid, not a diagnosis and not a medical device substitute for an eye examination. It does not detect glaucoma, corneal disease, infection, retinal problems, or any of the other conditions that can cause eye discomfort and blurring. The starter drops are over-the-counter lubricant eye drops for temporary relief of dryness and irritation — they are not a prescription therapy and are not claimed to treat, cure or prevent any disease.

See an eye care professional — promptly, and before relying on any reading from this kit — if you have eye pain, sudden or worsening vision changes, marked light sensitivity, discharge, an injury, a red eye that doesn't settle, or symptoms that make driving feel unsafe. Also check in with a professional before use if you wear contact lenses, have had recent eye surgery, use prescription eye medication, or have an autoimmune or thyroid condition. If your reading and your symptoms disagree, that is a reason to see someone, not a reason to self-treat harder.

Format caveats, honestly: artificial tears can sting briefly and can blur vision for a minute or two after instillation — don't drive until vision clears. Discontinue and seek advice if irritation, redness or discomfort worsens or persists, or if the drops change how your eyes feel in a way you didn't expect. The bottle is preservative-free and depends on the valve tip staying clean: never let the tip touch your eye, eyelid or any surface, keep it capped, and respect the in-use and expiry dates printed on the unit. Test strips are single-use and for tear sampling only — never place a strip in the eye, and discard any strip whose foil sachet is damaged, or any strip that has been exposed to…