Dr. Jason Tan
Psychiatrist
Specializes in adult psychiatry, neuropsychiatry, anxiety, depression and phobia treatments. “I am passionate about improving the integration of physical and mental health services.”
We're Claro — a calm companion for the antidepressant journey
Millions of people are prescribed antidepressants every year and then left to work out the hardest part on their own. Claro exists to close that gap between what your prescription says and what actually happens to you.
- discontinue antidepressants within 30 days
42%
discontinue antidepressants within 30 days
Olfson et al., 2006
- of Claro users come to understand what's happening on their antidepressant
62%
of Claro users come to understand what's happening on their antidepressant
App onboarding — “What brought you to Claro”
- clinical reviewers
8
clinical reviewers
Psychiatrists, nurse practitioner, neuroscientist, and specialist advisors
What Claro is
Claro is a self-monitoring and communication app for people taking antidepressants. It helps you track what is changing — mood, sleep, side effects, energy, how you are functioning — and turns that into something you can actually understand, and something your doctor can actually use.
It is built specifically around how antidepressant treatment works: week by week, dose by dose. The first two weeks of a new prescription behave differently from month twelve. Tapering behaves differently again. Claro knows which of those phases you are in and changes what it asks and what it shows you accordingly.
The gap
There is a blind spot in mental healthcare, and it sits exactly where treatment actually happens: between appointments.
An antidepressant is not a painkiller. Side effects usually peak in the first one to two weeks, while the therapeutic effect takes four weeks to become noticeable and six to twelve to fully arrive. Someone who started medication for anxiety and feels more anxious by day five draws the obvious conclusion — this isn’t working — and stops.
Around 42% of people discontinue within the first 30 days and 72% within 90. In a UK survey of 752 people, only 9% of those who stopped had done so in agreement with their doctor. Roughly one in five people who show no response by week four will respond if they keep going.
Patients don't know whether what they're feeling is normal. Clinicians don't know what happened between visits. Both problems have the same root — nobody is capturing the weeks in between. That is the only problem Claro is trying to solve.
How Claro works
Most mental health apps are diaries. You tick boxes, then try to assemble the meaning yourself. Claro collects weak, scattered signals and tries to turn them into a coherent picture.
What Claro pays attention to
Side effects
Structured checklists and free description, plus how much distress each one is causing and whether it is following a typical early-treatment trajectory.
Mood and affect
Daily scores tracked as direction and trend, with PHQ-9 and GAD-7 as a moving line rather than a single reading.
Sleep
Duration, quality, and consistency, including Apple Health data. Sleep often shifts before mood does.
Functioning
Work, study, relationships, basic self-care.
Medication chronology
Every dose taken, missed, or delayed, and every dose change with its date.
Safety signals
Recognised across languages by meaning rather than keyword, with every flag mapped to a clinical source and an escalation route.
Rather than a flat good-or-bad reading, Claro continually re-checks which recognisable phase of treatment you appear to be in. These are working hypotheses based on trajectories, not labels applied to a single bad day. Clinical logic lives in reviewable decision tables our advisors can inspect — not buried inside a model. When there isn’t enough data to say something useful, the system is designed to say less rather than guess.
What we deliberately do not do
This is the most important part of how we are built, so we state it plainly rather than bury it in a disclaimer.
- No diagnosis. Claro never tells you what condition you have.
- No medication advice. It does not recommend doses, changes, switches, or taper schedules. Anything that looks like one is filtered out before it reaches you.
- Not a therapist. Claro does not perform therapy and does not present itself as a substitute for it.
- No streaks, no gamification. A “7-day streak” is a poor fit for depression — losing it on a bad day turns an app into another source of failure.
- No community feed, for now. Unmoderated peer advice about medication can be dangerous; doing it responsibly is a separate product.
- Nothing hidden from your clinician. The doctor-facing summary is structural, not optional.
If an AI product starts acting as a parallel clinician, it doesn’t close the gap in care — it adds another layer the doctor cannot see. Claro is designed to strengthen the treatment you already have, never to stand in for it.
Our clinical advisors
Our clinical advisors
Claro is reviewed at every layer by practising clinicians. They shape the tapering methodology, the patient-facing copy, our safety boundaries, and how we explain the biology.
Get in touch
If you use Claro
We read everything people send us, and a meaningful amount of the product exists because someone wrote in. Questions and problems go to our support team.
[email protected]
If you're a clinician
Claro is designed to be transparent to the people treating our users — you can see what your patient sees. To understand how it works, or talk about using it in your practice, start with our page for professionals.
Claro for professionals →
If you're a journalist
We're happy to talk, including about the parts that are still unfinished.
[email protected]Claro is operated by Individual Roman Romanov, trading as Claro, established in the Kingdom of Thailand. It is developed with Pink Elephant, the studio behind the product.
References
Sources for the clinical claims on this page. Unconfirmed figures are not published here.
- [1]Olfson M et al. Continuity of antidepressant treatment for adults with depression in the United States. Archives of General Psychiatry 2006. doi:10.1001/archpsyc.63.8.84842% discontinue within 30 days; 72% within 90 days.
- [2]Mind. Survey of 752 people on antidepressant treatment (UK). Mind. Link34% had stopped; 36% tried and could not; 9% stopped in agreement with a doctor.
- [3]Cipriani A. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. The Lancet 2018. doi:10.1016/S0140-6736(17)32802-7
- [4]Trivedi MH, Rush AJ. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. American Journal of Psychiatry 2006. doi:10.1176/ajp.2006.163.1.2840% of remitters reached remission only after week 8.
- [5]Godlewska BR. Early changes in emotional processing as a marker of clinical response to SSRI treatment in depression. Psychological Medicine 2016. doi:10.1017/S003329171600102X
- [6]Goodwin GM. Emotional blunting with antidepressant treatments: a survey of over 1000 participants. Journal of Affective Disorders 2017. doi:10.1016/j.jad.2017.05.048Emotional blunting reported by 46%.
- [7]Read J, Williams J. Adverse effects of antidepressants reported by a large international cohort: emotional blunting, suicidality, and withdrawal. Current Drug Safety 2018. doi:10.2174/157488631366618060509574071% emotional blunting; under 5% were warned.
- [8]NICE. Depression in adults: treatment and management (NICE guideline NG222). NICE 2022. LinkReview at 2–4 weeks; 1 week for ages 18–25 or where suicide risk is present.
- [9]Horowitz MA, Taylor D. The Maudsley Deprescribing Guidelines. Wiley 2024. doi:10.1002/9781394291052
Ready for calmer clarity between appointments?