Sudden gains
On change in therapy that proceeds in jumps more often than we tend to think.
In practice and in research it has been described for some three decades: patients do not always improve gradually. Sometimes, between two sessions, they suddenly make a leap forward, without our being able to point precisely to what brought it about.
The term is sudden gains: sudden, substantial and lasting symptom improvements between two consecutive treatment moments. The article usually referred to is that of Tang and DeRubeis from 1999. They described the leap as large in absolute terms, large relative to the preceding level, and stable: the gain does not quickly disappear again. In depression they found such a leap in a considerable proportion of patients, and those who had made it did better in the longer term.
Since then the phenomenon has been described repeatedly, in depression, anxiety disorders, PTSD, eating disorders and OCD. It therefore does not seem to be an artefact of a single measurement moment or a single type of treatment. In our view, people sometimes really do improve in jumps.
What makes it hard to place
In the prevailing images of therapy, change proceeds gradually. A patient learns skills, applies them better and better, and the complaints slowly diminish. Cognitive schemas shift step by step, the effects of exposure accumulate. It is a classical dose-response model: more therapy, more effect.
Part of the improvement does not fit that picture well. It can proceed in a jump, sometimes even before the central intervention of the protocol has been reached, and not infrequently without patient or clinician being able to identify which moment tipped the balance.
What it shows
Various explanations have been put forward. Some researchers point to cognitive restructuring that tips at a specific moment. Others point to the building of hope and trust early in treatment. Still others think of self-organisation: a system that, after a period of increasing variability, comes to rest in a new pattern.
Schiepek and colleagues study the phenomenon from this last perspective. With daily measurements, they often see a period of heightened variability in mood, complaints and behaviour preceding a sudden gain, as if the system briefly becomes restless before it reorders itself.
"Discontinuous progress seems to be the rule rather than the exception." (Schiepek et al., freely translated)
For practice
What does this mean for the consulting room? Put cautiously, we arrive at three considerations.
The first is that it may be worthwhile to normalise this pattern. Patients, too, often expect progress to be gradual. If things suddenly go much better, they wonder whether it is real and fear a relapse. If progress fails to appear, they take this as a sign that the treatment is not working. Neither reaction seems to us entirely justified: change in jumps is, in our impression, a common pattern, and it may already be in the making while little is yet visible from the outside.
The second is that, in a slow course, we need not automatically assume that more protocol is the best intervention. Sometimes a patient benefits more from conditions in which her system can reorder itself than from additional technique. Which conditions those are differs from person to person.
The third concerns measurement. A weekly questionnaire shows the leap, but usually misses the restlessness that precedes it. Daily measurements, or several a day, often give a fuller picture.
Modesty fits here too
The literature on sudden gains is extensive, but much remains unclear about the underlying mechanisms. We know that the phenomenon occurs and that it is associated with a better outcome in the longer term. We have some insight into signals that seem to precede a shift. How such a leap might be reliably brought about, we do not know.
Perhaps that is also what the idea of self-organisation teaches us: what arises in another system cannot always be planned. It sometimes comes, often at a moment an outsider does not immediately understand, and only in hindsight do we believe we see the connection.