Pseudosecure vs. individuated-secure client attachment to therapist: Implications for therapy process and outcome

We differentiated two hypothesized client subtypes: (a) Pseudosecure clients have high Client Attachment to Therapist Scale (CATS) Secure and high CATS Preoccupied scores, tend to idealize their therapist, and exhibit maladaptive dependency; (b) Individuated-secure clients combine high Secure with low Preoccupied scores and function more autonomously. Clients who, despite insecure attachment to others, "earn" individuated-secure attachment to their therapist benefit most from therapy.

We examined regression suppressor effects by reanalyzing raw data from four published studies. If pseudosecure attachment is present, when covariance between CATS Secure and Preoccupied scores is removed, residual Secure scores should be significantly better predictors of process/outcome indicators than raw Secure scores.

Suppressor effects were observed in eight of nine analyses. Two were statistically significant. Earned individuated-secure attachment predicted improvement in interpersonal relationship symptoms, but only for clients with Avoidant pre-therapy attachment patterns. Finally, significant meta-analytic effect size estimates were obtained for CATS subscales, Secure r = .274 (95% CI = .177, .366), Avoidant, r = -.296 (95% CI = -.392, -193), and Preoccupied, r = -.192 (95% CI = -.289, -.092).

Clients with pre-therapy Avoidant attachment who nevertheless "earn" individuated-secure attachment to their therapist appear to benefit more from therapy.