For clinicians
What your client did between sessions, written down.
Meloa is a self-monitoring app your client owns. It attaches distress ratings to the actual events on their calendar — before, during, and after — so the thing you talk about on Tuesday isn’t reconstructed from memory a week later. If they want you to see it, they export a PDF and bring it.
What the client is actually collecting
SUDS at three timestamps
0–100 subjective units of distress logged before an event (T−1), during it (T0), and after (T+1) — anchored to a real, dated thing on the client's calendar rather than a free-floating daily prompt.
A four-part breakdown
Each check-in optionally splits into somatic, cognitive, anticipatory, and autonomic subscores (0–25 each), so a flat 70 can be read as a body-led 70 or a rumination-led 70.
Context, not just numbers
Every log carries the event's category, whether it happened at home, and whether it sat in a prep, recovery, or decompression window — so a score arrives with the situation attached.
Periodic standardised measures
GAD-7, LSAS, and PHQ-9 self-report snapshots, each independently skippable. An unassessed scale is stored as null, never as a zero, so a skipped measure can't be misread as a floor score.
The report
Your client generates a PDF and picks which sections go in it. Nothing is included by default that they haven’t chosen.
- GAD-7 / LSAS / PHQ-9 scores over time
- Daily baseline check-ins, independent of any event
- SUDS logs with anticipation and recovery metrics
- Event history for anything they rated 7+ for anticipated distress
One pattern comes up often enough to be worth naming: clients who log both a before and an after score frequently find the anticipation ran hotter than the event. Meloa doesn’t interpret that for anyone — it just makes the pair of numbers exist, dated, so the two of you can look at it together.
What it’s built on
Meloa didn’t invent any of these. It just puts them on a calendar.
Prospective forecast vs. outcome
The app asks for an anticipated distress rating before an event and an actual one after, then keeps the running record. LaFreniere and Newman's RCT found that logging worries alongside their predicted and actual outcomes over ten days reduced trait worry in GAD — the mechanism being repeated disconfirmation of one's own predictions rather than reassurance from anyone else.
Affective forecasting / impact bias
People systematically over-predict the intensity and duration of negative feeling, and anxiety amplifies it. That's why the before/after pair is stored and surfaced as a record: the gap is the intervention, and it only works if it's the client's own dated data rather than a therapist's reassurance.
Affect labelling
Check-ins ask for a number and, optionally, a four-part split of what the distress was made of. Lieberman et al. (2007) found that putting affect into words dampens amygdala response and recruits right ventrolateral prefrontal cortex — naming is doing something, not just recording.
Ecological momentary assessment
Ratings are captured in the moment and in context rather than reconstructed in session a week later, which is the whole rationale for EMA. Retrospective recall of distress is biased toward peak and most recent experience.
Psychological detachment and recovery
The decompression buffer after a hard event is a scheduling expression of the recovery literature — Sonnentag and Fritz's work on detachment as a distinct recovery experience, and boundary theory's account of transitions between roles. Meloa's contribution is making the transition a defended block rather than something a person is supposed to remember to take.
SUDS in graded exposure
The 0–100 scale is Wolpe's, unchanged, and remains the standard unit for pacing exposure work. Meloa holds the numbers between sessions; you decide what they mean and what happens next.
Where the evidence stops. These mechanisms are established; Meloa is not. It has not been trialled, and the overall effectiveness of EMA as a mental-health intervention is still unsettled — some randomised trials show symptom improvement and others don’t. There is also a documented risk worth weighing for specific clients: self-monitoring without an acceptance or mindfulness component can increase emotional reactivity in some people, and for a client whose presentation is checking or reassurance-seeking, another thing to log may feed the compulsion rather than loosen it. Meloa has schedule-locking safeguards intended to reduce that, but you are better placed than we are to judge whether tracking helps a given person.
What Meloa is not
- Not a medical device. Meloa has no FDA clearance and makes no treatment claim. It records self-report; it does not assess, diagnose, or treat anything.
- Not a validated instrument in itself. SUDS, GAD-7, LSAS, and PHQ-9 are established measures, but Meloa’s particular way of collecting them has not been separately validated, and its scores should not be read as equivalent to administration in session.
- Not a monitoring service. Nobody is watching the data in real time. Meloa does not alert you, page anyone, or escalate. It is not appropriate for risk monitoring, and it does not replace emergency services.
- Not a clinician portal. There is no therapist login and no dashboard of your caseload. You see a client’s data only when that client hands you a file.
- Not structured ERP. Meloa can hold the record around exposure work you direct. It does not build hierarchies, set exposures, or pace anyone through them.
Full terms in our clinical disclaimer.
Whose data it is
The client’s, entirely. Every table is row-level isolated per user, so no other account — including ours in ordinary operation, and including anyone on a shared billing plan — can read their calendar or health data. Meloa’s group plan is a billing bundle only: five people, five separate private accounts, no shared visibility of any kind.
Sharing is a deliberate act. The client chooses the sections, generates the PDF, and decides who gets it. There is no background sync to a clinician, because there is no clinician account for it to sync to.
AI-assisted features are off until a client separately opts in, distinct from paying for any tier. See our privacy policy and consumer health data notice.
Pointing a client to it
Send them to meloa.io. The parts that matter clinically — the before/during/after check-ins, the buffer windows, the calendar itself — are free and not time-limited, so a client can start without a card and without a trial running out mid-treatment. The paid tier adds the longitudinal charts, the standardised assessments, and the PDF export.
Questions about the data model, or something you need it to do? support@meloa.io — a person reads these.