Need help now? Support now: crisis lines by country.
Questions about pain · Mathematical Institute for Research of Pain
I started something new during a flare and my pain eased. Did it work?
Maybe, but easing after a flare is expected even when nothing new was started. An unusually bad day tends to be followed by a less extreme one; this is called regression to the mean.
A small pain I carry; a great one carries me.
The short answer
People often try something new when the pain is at its worst. That timing matters. If the starting point is a flare, an unusually high value, later values tend to be lower even when nothing changed. The Institute’s Pain Bias Codex names this step “A flare becomes the usual baseline”: an unusually high starting value is treated as the person’s ordinary level when judging later change.
This does not mean the new thing did nothing. It means the easing alone cannot tell the two explanations apart.
What is known
Englund and Turkiewicz (2023) applied the entry criteria typical of knee osteoarthritis trials to people in a large population-based study in the United States, in which no trial treatment was given. Regression to the mean alone explained an improvement of about 1 point on a 0 to 10 pain scale. A methods tutorial by Barnett, van der Pols and Dobson (2005) explains how unusually high or low repeated measurements tend to move toward the mean, and how study design can allow for it.
What is not yet known
The 2023 analysis applied hypothetical trial criteria to an existing cohort and was published as a short comment; it was not a randomised trial. It shows the size of the effect in a group. It does not establish why one person’s pain changed.
What would tell us
A comparison that starts from the same kind of day: other flares in the same person’s record, with and without the new thing, or a trial with a control group selected the same way.
A worked example
Take two measurements of pain at different times, with the same average and the same spread, and correlation between them. If the relationship is linear, the expected later value, given a first value , is
.
Here is the usual level, is how far the first day was from it, and , between 0 and 1 for most repeated pain ratings, is how strongly the two days move together. With invented numbers: a usual level of 4, a flare day of 6 and give an expected next value of . The pain is expected to ease by 1 point with nothing new at all.
A fair judgement marks how the starting point was chosen and looks at the earlier range when it is available.
These pages explain reasoning and evidence. Nothing here is medical advice or a forecast about any person.
Sources
- Englund M, Turkiewicz A. Pain in clinical trials for knee osteoarthritis: estimation of regression to the mean. Lancet Rheumatol. 2023;5(6):e309–e311. doi:10.1016/s2665-9913(23)00090-5
- Barnett AG, van der Pols JC, Dobson AJ. Regression to the mean: what it is and how to deal with it. Int J Epidemiol. 2005;34(1):215–220. doi:10.1093/ije/dyh299
Related questions
Pain words: Regression to the mean · Placebo · Flare · N-of-1 trial · All words
Related, in Nociflux, the Institute’s ecosystem for living with pain: the N-of-1 Lab.
Inside the Institute
The Pain Bias Codex gathers checks on conclusions about pain, in four families. With an invite code you can read this check there beside the others, with its full evidence notes. Pain math explains what the Institute does.
No code yet? Ask for an invite. A person reads every request and writes back by hand with an invite code, so it may take a few days.
Need help now? Support now: crisis lines by country.
Except where a page says otherwise, the Institute’s own text and figures are shared under CC BY-NC 4.0. Pictures and quotations from others keep their own credits and licences.