StarHealing
Article

StarHealing's Progress Measures: DSM-5, SRS-2 and FHI

Plain definitions of every term and score we publish, including why a score can rise when a child gets better and which of our instruments have no normative data.

This page defines the words we use when we talk about progress. It exists so that a parent, a clinician or a search engine can check what a number on this site actually means before believing it.

What a milestone is

A milestone is one full cycle of work with one child and family: caregivers report current observations, our team reviews them together with any child video the family shares (and, where possible, time with the child directly), we interpret where development currently sits, we agree the next targets with the parents, and the cycle runs again. A milestone is not a session and not a month — it is a round of measurement and planning.

Milestones are numbered, and each one carries scores plus a written report for the family. Because every cycle's records are carried into the next one, a later score is read against that child's own earlier trajectory rather than against a population average.

Why a later score can be higher even when the child is better

This is the single most misread thing on our site, so it is worth stating plainly.

Before a child speaks, language-related items are not scorable — they register as absent. Once speech appears, the same items become observable, and a total can rise while the child gains a capacity. We call the moment this happens language opening. Similarly, when a child moves from a home-centred routine into a school or kindergarten setting, the environment starts demanding skills that were never being tested before — peer turn-taking, group routines, social rules, adapting to noise. The measurement context has changed, so the score is not an apples-to-apples comparison with the earlier one.

We treat a rise after such a transition as a new target becoming visible, not as regression. That is our interpretation, and it is falsifiable: if a score stays elevated across two further milestones after the transition, we read it as a real difficulty, not as a new window.

The DSM-5-domain tracking form

We use the DSM-5 autism criteria as a scoring frame, not as a diagnostic tool: four domains (social communication and interaction; restricted and repetitive behaviour; onset; functional impact), each item scored 0, 1 or 2 so the same child can be compared with themselves over time. Totals are then read against the familiar Level 1 / Level 2 support bands as a convenience, and "below threshold" simply means the items scored that way. A clinician diagnoses; this form tracks.

The 20-item SRS-2 short form

The Social Responsiveness Scale is a 65-item caregiver questionnaire. We use a custom 20-item short form covering five areas: social awareness, social cognition, social communication, social motivation, and restricted interests with repetitive behaviours. Each item is 0 to 3, so totals run 0 to 60 and higher means more difficulty. We band the totals as 0-19, 20-29, 30-39, 40-49 and 50-60 for planning conversations.

Two limits, stated deliberately: the short form is ours, so it has no published normative data and no validation study, and it is a tracking instrument — it is not a diagnostic instrument, and its bands are not clinical cut-offs.

The Family Happiness Index (FHI)

The FHI scores the family, not the child: caregiver stress (reverse-scored), caregiver-child emotional connection, caregiver confidence, how often relaxed or joyful moments happen, the caregiver's own emotional state, and outlook for the child's future. It is expressed as a percentage so one family's change can be compared across milestones.

It is a caregiver-reported, non-validated index. We publish it because family equilibrium is part of what we are trying to change, not because it is an outcome measure with authority. One caveat for readers comparing our two write-ups: the Stage 1 report described five sub-dimensions and reported raw scores, while the follow-up uses the six listed above and a percentage scale. The definition is being reconciled before we publish further series.

Free-expression capture

At each milestone caregivers write whatever they want to say, in their own words. Those narratives are read alongside the scores — our tooling uses language-model analysis to pick up shifts in tone and what parents foreground — and they routinely contradict the numbers. Where they do, the narrative goes into the report.

FAQ

Is StarHealing's assessment a diagnosis?

No. Nothing on this site diagnoses autism or rules it out. The forms are self-tracking instruments for a family already working with clinicians.

Are your outcome numbers peer-reviewed?

They are posted as OSF preprints, which means they have not been peer-reviewed, they have no control group, and the sample is three completed families in Stage 1 and two children in the follow-up. Cite them accordingly.

What would make you change your mind about a plan?

A named skill that has not moved across two consecutive milestones, or a score that stays elevated after a transition has settled. Both are read against the child's own history, not against an average.