Complete
Return to the defined pre-event baseline.
NATURAL HISTORY & STATE TRANSITIONS
Recovery & repeated measurement
Follow outcomes through recovery, repeat the same meaningful measure, and preserve no-change observations and competing explanations.
RECOVERY IS AN ENDPOINT
Do not stop the timeline at the worst point.
The return curve can distinguish a transient state, incomplete recovery, recurrence, and a lasting change.
Partial
Improvement in some domains while other burdens persist.
Delayed
Return occurs, but only after a prolonged interval.
Recurrent
The same or similar change returns under comparable or different conditions.
New baseline
A stable difference remains after the recovery window.
Unresolved
Follow-up is incomplete or the source cannot establish the outcome.
REPEATED MEASUREMENT WITHIN THE SAME PERSON
Use the same outcome before, during, and after a state window.
A context such as illness, poor sleep, pain, medication, or unusual demand may coincide with a change but does not prove its cause. Use the same outcome measure before, during, and after the window.
01
Usual baseline
Define usual function, active diagnoses and supports, and recent sleep, intake, pain, or illness using the same outcome measure.
02
During the change
Record the timing, context, and other active factors; add objective measurement when clinically indicated without changing the outcome definition.
03
Recovery observation
Record whether the context resolved, what support or intervention changed, and whether return was complete, partial, delayed, or unknown.
Important: A change seen after an intervention remains an association until the same outcome is measured repeatedly, other explanations are considered, and recovery is followed.
HOW TO MEASURE AN INTERVENTION OR EXPOSURE
Measure intervention response with planned follow-up.
Define the target, exposure, concurrent changes, follow-up window, and decision rule in advance; repeat the same meaningful outcome across baseline and recovery.
Predefine the target
Name the seizure, sleep, pain, bowel, communication, movement, regulation, or participation outcome before judging direction.
Summarize only people with follow-up
Count the people who received the intervention and have follow-up; do not divide by all 63 people unless all 63 were assessed.
Separate benefit from tolerability
An intervention can help one target, worsen another, or stop for reasons unrelated to efficacy.
Preserve no-change observations
No observed change is evidence within that window; it is not proof of universal tolerance or nonresponse.
Keep alternatives visible
Retain illness course, sleep, pain, nutrition, co-medication, expectation, and ordinary fluctuation.
Return decisions to clinicians
This framework does not recommend starting, stopping, avoiding, or changing medication, immunization, anesthesia, diet, or other care.
FROM PAST RECORDS TO A PROSPECTIVE STUDY
Repeat the same meaningful measure across state and time.
Rare-disease natural-history guidance emphasizes planned collection, clinically meaningful outcomes, common definitions, and reliable data. A prospective study would require participant consent, common measures, and clear rules for privacy and data use.
- 01
Predefine the outcome. Choose the function, event, or physiologic measure before examining direction.
- 02
Use the person as their own baseline. Record stability, active burdens, supports, and measurement conditions.
- 03
Capture the window. Time-lock the change to clinically appropriate measurements and concurrent conditions or demands.
- 04
Repeat through recovery. Use the same method and record periods when no change occurred and any missing follow-up.