How much residual symptom is acceptable when the disease activity score says remission

Design

cross-sectional threshold derivation study with Patient Acceptable Symptom State as external anchor and ROC analysis

Population

370 patients with rheumatoid arthritis, including 213 in SDAI remission

Primary outcome

residual symptom score thresholds discriminating an acceptable symptom state

Effect

thresholds of 3 overall and 2 in SDAI remission (AUC 0.824 and 0.838); acceptable state reported by 95.2%, 83.9% and 46.8% across rising symptom categories, and 99.0%, 95.7% and 71.4% within the remission subgroup

A patient in Simplified Disease Activity Index remission who still reports symptoms presents a familiar problem: the score says stop escalating, the patient says otherwise, and there is no agreed way to decide how much residual symptom is acceptable. This cross-sectional study of 370 patients with rheumatoid arthritis, 213 of them in SDAI remission, used a patient-reported residual symptom instrument anchored against the Patient Acceptable Symptom State.

Two hundred and eighty-four of 370 patients reported an acceptable symptom state. The 75th percentile method identified a residual symptom threshold of 3 in the whole cohort and 2 among those in SDAI remission, and both discriminated reasonably (areas under the curve 0.824 and 0.838). The proportion reporting an acceptable state fell steeply as residual symptom scores rose - from 95.2% to 83.9% to 46.8% in the whole cohort, and from 99.0% to 95.7% to 71.4% among those already in remission. Lower scores also tracked with better physical function and lower patient global assessment.

The finding worth taking is inside the remission subgroup: nearly three in ten patients who were in SDAI remission and had higher residual symptom scores did not consider their state acceptable. That is a measurable gap between the target we treat to and the outcome the patient is living with. The instrument itself is Japanese and unlikely to be adopted elsewhere, and this is cross-sectional data with no outcomes. The transferable action is simpler than the instrument: ask the acceptable-symptom-state question directly in any patient you are calling remitted.

• Ask the acceptable symptom state question - whether the patient considers their current state satisfactory - in every patient you record as in remission
• Treat a no as a finding requiring exploration, not as non-concordance with a valid score
• Look for the usual explanations for residual symptoms in remission: fibromyalgia, osteoarthritis, damage, depression, poor sleep
• Do not escalate disease-modifying therapy on symptoms alone in a patient with no objective inflammation
• Record the answer, so the trajectory is visible across visits

Why it matters

It measures the distance between a remission score and a patient who feels well, and finds it is not small.

Don't overread it

Cross-sectional, single-instrument, with thresholds derived from this sample's own distribution and no outcome data.

The statistics, in plain English

An area under the curve of 0.82 to 0.84 is good discrimination for a patient-reported instrument, meaning the score separates acceptable from unacceptable states well in this cohort. The thresholds themselves come from the 75th percentile method, which defines the cutoff by the distribution of scores in this particular sample - so they will shift in a population with different disease severity, and should not be imported as fixed numbers. Being cross-sectional, the study cannot show that acting on a low score changes anything.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.