nucleation

THEORY-903: In a 2026 meta-analysis of standalone safety-behaviour manipulations outside exposure, mostly with young non-clinical or subclinical adults and mostly on appearance concerns and social anxiety, reducing safety behaviours lowered symptoms relative to control (d = 0.43, about 0.31 after adjustment for publication bias), and instructed increases raised them (d = 0.65, five studies)

2 min read

Properties6
StatusActive
Version2
Tagspsychopathology-and-treatment
SourceLIT-1491 — The standalone effect of safety behavior manipulations: A systematic review and meta-analysis
Bears onQUESTION-326 — Does reducing safety behaviours lower symptoms, and instructed increases raise them, in diagnosed anxiety disorders, in pre-registered trials with active controls and masked rationale or a larger meta-analysis, once publication bias and expectancy are accounted for?
SupportsCLAIM-446 — Some persistent anxiety may be governance-maintained rather than belief-maintained: anticipated threat grants defensive processes authority over action, attention and evidence acquisition; under the avoidance that authority selects, the hypotheses that the situation is dangerous and that it is safe predict the same observation, so repeated safety need not correct the threat estimate; relief reinforces the defense without being evidence of danger; and the regime can outlast a corrected belief, its pathology, where present, being defensive authority insensitive to corrective evidence at avoidable cost

Graph View

Source

Goodson, Patel, Zech, Sala & Cougle, LIT-1491, read in full in NOTE-1128 from the NIH author manuscript (PMC12997458): Results, Discussion and Table 1. The version of record is closed and was not read.

What was actually shown

  • Reduction lowers symptoms. Against control, d = 0.43 (95% CI 0.27–0.59; k = 16). Egger’s test was significant (p = .012), and trim-and-fill gave d = 0.31 (95% CI 0.12–0.51).
  • Dose-response, across studies. The between-group effect on safety behaviour moderated the effect on symptoms (B = 0.23, R² analogue 0.85). No other planned moderator did.
  • Increase raises symptoms. In five non-treatment experiments with healthy college samples, instructed increases raised symptoms against control (d = 0.65, 95% CI 0.46–0.83, I² = 0%). Bias could not be tested.
  • The samples. 2,048 participants, mean age 24.9, 26.6% male. Most studies concerned appearance concerns or social anxiety. One concerned PTSD and two checking.

What could have come out otherwise: once exposure is removed, cutting safety behaviours could have changed nothing, as Meulders et al. (2016) nearly found for safety-behaviour instructions within exposure.

What this does not say

  • That safety behaviours maintain anxiety disorders. Diagnosed samples are rare. The authors say only that their results “align” with the maintenance theory.
  • How they act. No included study is reported to measure threat estimates, approach, or what was learned once safety behaviours stopped. Expectancy effects are possible, since every treatment trial gave a rationale.
  • That the process mediates the outcome. The moderator is between studies.

Connections

  • CLAIM-446. It supports the premise that protective behaviour does causal work on symptoms. It is silent on the claim’s distinctive thesis that defensive authority, not belief, is the maintaining link.
  • THEORY-053. The moderator is process-outcome covariation of the kind that theory asks for, but between studies, not within them.
  • THEORY-058. Both directions of the manipulation move symptoms, which fits a loop through behaviour without showing that it closes.

Backlinks