Zalaznik, D., Weiss, M., & Huppert, J. D. (2019). Improvement in adult anxious and avoidant attachment during cognitive behavioral therapy for panic disorder. Psychotherapy Research, 29(3), 337-353.
Summary by: Sofija Miljovic
One of the psychological treatments for panic disorder that has been shown to be effective and is well established in the literature is cognitive behavioral therapy (CBT). It is generally described as a time-limited, structured therapy that addresses dysfunctional thought patterns and avoidance behaviors. Attachment theory, however, has traditionally emphasized the relative stability of internal working models of relationships developed in early life. Zalaznik, Weiss, and Huppert’s (2019) study is a valuable addition to the clinical literature, as it asks whether CBT, although not specifically designed to address attachment, might positively influence adult attachment patterns. Using this study as an illustrative example, this article critically discusses the theoretical background, methodology, findings, and implications of the research and offers a personal reflection on its scientific and clinical relevance.
The primary objective of the study was to determine whether attachment anxiety and avoidance decrease during CBT for panic disorder with agoraphobia. The researchers used a longitudinal design involving 31 adults diagnosed with panic disorder and agoraphobia. Participants received structured CBT, an approach that typically involves psychoeducation about the physiology of panic, cognitive restructuring of catastrophic interpretations of bodily sensations, interoceptive exposure to feared bodily sensations, and gradual in-vivo exposure to feared situations. Before each session, participants completed self-report questionnaires assessing attachment anxiety and avoidance, anxiety sensitivity, agoraphobic avoidance behaviors, emotion regulation, and the therapeutic alliance. The repeated measurements allowed the authors to examine changes throughout treatment and investigate the temporal relationships between different psychological processes.
For the statistical analysis, the researchers employed multilevel modeling to account for repeated measurements nested within individuals. Moreover, they performed time-lagged analyses to examine whether changes in attachment preceded or followed changes in panic-related symptoms and psychological processes. The findings showed improvements in both anxious and avoidant attachment during therapy. Changes in anxiety sensitivity were associated with changes in both attachment dimensions. Improvements in avoidance behaviors when participants were alone, but not when accompanied by another person, were also associated with changes in attachment. Importantly, reductions in anxiety sensitivity and avoidance when alone predicted reductions in attachment anxiety in the following session, but not vice versa. Similarly, reductions in avoidance behaviors and emotional suppression preceded improvements in attachment avoidance. The therapeutic alliance also improved, although its changes were associated only with simultaneous changes in attachment anxiety. Overall, these temporal patterns suggest that improvements in attachment may follow changes in panic-related psychological processes rather than act as an initial mechanism of symptom reduction.
These results have several theoretical implications. First, they suggest that adult attachment patterns might be more flexible than traditionally assumed, at least in clinical populations receiving structured psychological treatment. Second, they challenge the distinction commonly drawn between symptom reduction and deeper changes in personality or relational functioning. Rather, the findings support a more integrative perspective in which cognitive, emotional, and interpersonal processes may influence one another. CBT methods, including exposure and cognitive restructuring, might help improve a person's sense of mastery, emotional safety, and confidence in their coping skills. These improvements could potentially extend to interpersonal situations and contribute to reduced attachment insecurity. However, this possible mechanism remains a theoretical interpretation rather than a directly established finding.
Despite its contribution, the study has notable limitations. The sample size was relatively small, which limits statistical power and the ability to generalize the findings to broader populations. Also, the lack of a control group makes it challenging to determine whether changes in attachment resulted specifically from CBT rather than the passage of time, expectancy effects, or the general therapeutic relationship. The reliance on self-report measures of attachment also raises concerns about possible response biases, including socially desirable responding. Furthermore, although the time-lagged analyses provide information about the sequence of changes, they cannot establish causal relationships. Without long-term follow-up data, it is also unclear whether these improvements represent lasting changes in attachment patterns. These shortcomings suggest that the findings should be considered promising but preliminary.
As a student, I find this work especially compelling because it brings together two traditions in psychology that are often introduced as distinct approaches. I appreciate that the authors did not present their findings as definitive evidence of causation, but instead highlighted associations between changes in attachment and other psychological processes. The most useful contribution of this study, in my opinion, is not that it proves CBT alters attachment, but that it introduces an important line of inquiry into the broader effects of evidence-based psychological interventions. At the same time, I remain skeptical about the strength of the evidence. These changes cannot yet be established as genuine, lasting changes in attachment patterns without experimental control and long-term follow-up data demonstrating that the improvements extend beyond temporary changes in self-reported relational functioning.
To sum up, the article by Zalaznik, Weiss, and Huppert (2019) provides valuable empirical evidence that CBT for panic disorder with agoraphobia is associated with reductions in attachment anxiety and avoidance. The study offers an important starting point for future research, although its methodological limitations restrict the strength of the conclusions. Larger controlled studies with long-term follow-up are needed to determine whether these attachment changes are stable and clinically significant over time. Overall, this study contributes to a more nuanced understanding of how psychological interventions may influence not only symptoms but also underlying relational patterns.
