Van Tuyl et al., 2009

Defining remission in Rheumatoid Arthritis: Results of an Initial ACR Consensus Conference

Rheumatoid arthritis
Remission
Composites
Validity
Author

Simon Steiger

Published

June 10, 2024

At a glance
Objectives
To move towards a uniform definition of remission in RA.
Related articles
Maybe articles about the core set of variables. Lee et al., 2011 about imaging data and remission.
Link
DOI: https://doi.org/10.1002/art.24392

Background

  • At the time the manuscript was written (2009), the ACR definition of clinical remission was from 1981
  • Relatively few patients achieved remission at that time (1980s)
  • Authors report better correlation between (old) ACR criteria and full-joint-count DAS, not the DAS28

Methods

  • Qualitative discussion between expert RA clinicians.

Results

  • Should remission be defined differently for trials and in the clinic?

Defining remission

What is remission?

According to Maarten Boers, it is:

  • the absence of disease activity
  • neither cure nor arrest of the disease, the disease may return
  • a state, not a transition between states
  • independent of the time spent in the remission state1
  • Generally speaking, ACR, CDAI, SDAI and PAS/RAPID3 are “strict”, while DAS28 and mACR are “lax” (with remission in the latter group equal to minimal disease activity in the former)

The expert panel agreed on the need for a strict definition of remission, which can reliably differentiate low disease activity from remission.

This strict definition should include

  • no clinical disease (although entirely absent disease activity or pain may not always be possible)
  • no progression over time

Validity

Issues of validity in defining remission in RA
  1. Are 28 joints enough? Is it acceptable to label a patient as “in remission” if they have active joints in the feet?
  2. Should the component measures be restricted to the ACR core set?
  3. What’s the role of morning stiffness / fatigue in defining remission?
  4. What’s the role of imaging? Should remission imply a lack of radiographic progression? See also Lee et al., 2011.
  5. How should changes due to chronic disease be incorporated (or not) into remission?
  6. Should remission at some timepoint predict remission in the future? Has a subject achieved remission who is not in remission at time t but was at time t-1?
  7. Should remission predict outcomes such as joint damage, disability, death (i.e., have predictive validity)?

The expert panel felt that:

  • remission should be independent of long term outcomes such as radiographic damage
  • remission should signal no / reduced progression of joint damage and less deterioration / more improvement in functional status (HAQ)
  • from the previous point follows that remission should have predictive validity

Selectivity

Not sure if this is the right word. What the athors mean and I want to label is about discriminating well between two hypothetical patients who belong to different categories (in German, this is Trennschärfe, which might be related to statistical power but for a measure and not a hypothesis test).

Issues of discrimination in definitions of remission in RA
  • A measure of remisson should be able to discriminate well between patients in different disease states
  • If this is not the case, a larger sample would be needed to detect an existing difference between, e.g., placebo and treatment groups

Footnotes

  1. The time spent may however be relevant for defining a sustained state of remission↩︎