Tracking progress on Medicaid-funded ABA therapy in North Carolina

Author : Advanceable ABA | Published On : 24 Sep 2026

Six weeks before the authorisation period ends, a request goes in for the next one, and it is built almost entirely out of numbers that were collected on ordinary afternoons months earlier. A parent in Greensboro found this out when a reviewer asked about a goal her son had worked on since spring and nobody in the house could remember what the data looked like in May.

Progress tracking in a behavior program serves two audiences at once, and they do not want the same thing. The family wants to know whether life is getting easier. The funder wants documented evidence that services are producing change. A good program answers both without pretending they are the same question.

What actually gets counted

Most of it is collected in the moment by the technician. Trial by trial data records whether a response was correct, prompted or incorrect on each opportunity, which is what produces the percentage figures on the graphs. Frequency counts how often something happens, duration how long it lasts, and latency how long it takes a child to start after a request.

Independence is often the most informative measure of the lot. A child scoring ninety per cent with a gestural prompt and ten per cent without has a very different plan ahead of him than one scoring sixty per cent unaided, and a graph that does not distinguish prompt levels hides that completely.

Session notes carry the rest. They record what was worked on, for how long, any behavior of concern and anything unusual in the environment, which is how a reviewer can tell whether a flat month was a therapy problem or the month the family moved house.

How to read what you are sent

Look at the baseline first, the flat stretch at the left of the graph before teaching started, because everything after it is only meaningful relative to that. Then look for the vertical lines that mark a change in procedure. Those are the decisions the behavior analyst made, and each one should have a reason behind it.

Ask what the mastery criterion is for the goal you care about most and where the current data sits against it. Parents can also learn more about how funded services are documented in this state, which is worth doing before the first reauthorisation rather than during it.

North Carolina Medicaid publishes its clinical coverage policies openly at medicaid.ncdhhs.gov, and reading the criteria your child's plan is being measured against changes how you hear the whole conversation.

Where the numbers and real life disagree

Data can look excellent while a household feels no different, and that gap is worth raising rather than tolerating. It usually means skills are being mastered inside sessions and not turning up anywhere else, which is a generalization problem with a specific fix rather than a sign that nothing is working.

It can run the other way too. Families often notice change before it appears on a graph: a child who tolerates the supermarket, who waits ten seconds longer, who accepts a change of plan without escalating. Those observations are worth saying out loud, because a behavior analyst can turn them into a measurable target.

School data is a separate stream with its own rules, since IEP progress reporting runs on the school calendar and is meant to work hand in hand with the behavior plan rather than be replaced by it.

The most useful habit is also the cheapest. Keep a note on your phone, a dated line every week, about the one thing that was easier or harder than usual. It takes ten seconds, it survives better than memory, and when somebody asks in March what changed since October you will have an answer no graph was going to give you.