Keywords

This chapter outlines psychodynamic conceptualizations and treatments of anxiety disorders (social anxiety, generalized anxiety, panic and agoraphobia, and specific phobia) from Freud to the current day. Specifically, we trace psychodynamic theory from a one-person psychology to a two-person model.Footnote 1 We then describe contemporary psychodynamic theories of anxiety disorders. Next, we review and critically evaluate the empirical support for psychodynamic psychotherapies for anxiety disorders. We conclude by suggesting empirical and theoretical future directions for the psychodynamic treatment of anxiety disorders.

Overview of Psychodynamic Theory

Psychodynamic theory posits the existence of an unconscious mind, a core concept that differentiates it from other therapeutic orientations. According to psychodynamic theory, unconscious drives, wishes, and motivations influenced by early childhood experiences come into conflict with their conscious counterparts. Individuals, in an effort to find a compromise between their unconscious and conscious minds, use defense mechanisms (which can be understood as unconscious coping mechanisms) to keep conflict at bay. Psychological symptoms emerge from unconscious conflicts and insufficient defense mechanisms.

In the context of psychodynamic psychotherapy, the “presenting problem” (often symptoms) with which the patient presents is usually not the target of treatment. Instead, the psychodynamic psychotherapist considers the actual problem to be something of which the patient is not yet aware, such as an unconscious conflict. Thus, the goal of treatment is not symptom reduction per se, but rather the uncovering of the origins of the symptom. Symptom alleviation occurs as a consequence of identifying, understanding, and working through unconscious conflict.

Development of Psychodynamic Theory of Anxiety

Anxiety and Freud’s Topographic and Structural Hypotheses

The topographical hypothesis was Freud’s first attempt at modeling the psyche [1,2,3]. Freud conjectured that the human psyche comprises the unconscious, the preconscious, and the conscious. He imagined the mind as a seething cauldron of psychic forces pushing to be released. Over time, the excitation of these forces increases, ultimately moving toward discharge (i.e., cathexis). However, instead of being expressed, some psychic material (such as forbidden hate, sexual drives, or aggression) is kept unconscious by repression. The failure to eventually discharge these feelings/urges culminates in anxious symptoms, called neurotic anxiety. Thus, in the topographic theory, anxiety is a result of repression.

Ultimately, however, Freud rejected his original hypothesis and, in its stead, offered the structural hypothesis [4, 5], which marked the beginning of Freud’s tripartite model of the mind (i.e., the id, ego, and superego). According to Freud, the id is the part of the mind that contains basic drives such as sex and aggression, the superego contains the moral conscience, and the ego is part that is grounded in reality, responsible for mediating between the id and the superego. Freud no longer conceptualized the mind as a seething cauldron, but rather understood it as an entity in conflict and compromise. In the structural model, anxiety manifests in the ego. Initially, Freud theorized that conflicts between the ego and the id, the ego and the superego, and the ego and the external world yield distinct types of anxiety [4].

Freud eventually condensed the different types of anxiety into traumatic anxiety and neurotic anxiety [5]. Traumatic anxiety is automatically triggered in situations of (internal or external) danger that one’s ego is not equipped to handle. For example, immediately after birth, babies might automatically experience traumatic anxiety because their ego is not sufficiently developed to address the strangeness of their new reality. In contrast to traumatic anxiety, neurotic anxiety results from the ego feeling threatened by imagined dangers that are emblematic of earlier traumatic experiences. For instance, the experience of separation from one’s mother in infancy may lead to neurotic anxiety in adulthood in response to the threatened loss of a friend. In the context of neurotic anxiety, one experiences an affective signal of impending danger (i.e., signal anxiety), after which psychological defenses are engaged in an effort to protect the ego from reexperiencing this danger. In other words, in this model, anxiety may be an indication (i.e., signal) of potentially distressing threats to the ego such as shame, trauma, or narcissistic threat. Thus, the structural hypothesis posits that anxiety leads to repression (and other defense mechanisms), whereas in the earlier topographic theory, anxiety is a result of repressed forbidden impulses. Freud’s later work expanded into ego psychology, placing a greater emphasis on development, defenses, and the role of the ego.

Psychodynamics, Anxiety, and a Two-Person Psychology

In the 1920s, object relations theory developed, shifting the model of the mind from a one-person to a two-person psychology. Object relations theorists believed that in order to understand the human experience, one must consider not only the psychic drives within a single person but also how one’s relationships with others shape the psychic experience of the individual (e.g., [6, 7]). Psychoanalysts such as Melanie Klein and Donald Winnicott theorized that anxiety emerged from disrupted attachment and the loss, or threatened loss, of a relationship.

According to Melanie Klein [6,7,8], infants are born into a state of anxiety, seeking survival in a dangerous world. To cope with this anxiety, infants split good qualities from bad qualities, qualities in others (typically the primary caretaker), and the good qualities in themselves. Because infants locate the “bad” in others and desire to destroy them as a result, infants develop a sense of paranoia about others potentially seeking revenge on them. Thus, according to Klein, infants cope with a world they experience as uncertain and unsafe by dividing the good from the bad, which leads to anxiety. As infants develop, however, they begin to understand people as having both good and bad qualities, forming a more integrated sense of themselves and others. Consequently, the infant’s primary caretaker, albeit imperfect, can still be loved. The understanding that people can evoke anger and simultaneously be worthy of love leads the infant to develop a new kind of anxiety, characterized by guilt rather than paranoia. The infant feels guilty for the previous wish to destroy the object.

In sum, according to Klein, the early stages of development are characterized by fears of annihilation, and the later stages are characterized by anxiety about annihilating others. Thus, anxiety takes two forms: the expectation that something bad will happen to oneself and the expectation that something bad will happen to another. Crucially, the anxiety experienced in the context of early relationships is conjectured to repeat throughout the individual’s life when confronting similar relational circumstances.

Compared to Klein, Winnicott more greatly emphasized the infant’s attachment to the mother early in life [9]. For Winnicott, the bedrock of proper development is attuned caregiving, and anxiety emerges when there is a disruption in a child’s attachment to the primary caregiver. Winnicott believed that if infants could safely express their needs and expect such needs to be met, infants would establish a true (versus false) sense of self. Individuals with a true sense of self are able to have spontaneous and authentic experiences that are not contingent on others. In contrast, those with a false sense of self experience themselves only in relation to the expectations of others. Disrupted attachment sets the stage for numerous problems one encounters later in life, including a threatened sense of interpersonal safety and, subsequently, feelings of anxiety when expressing one’s emotions to others.

Following Winnicott’s work on a false self, Heinz Kohut launched an investigation into narcissism and ultimately developed a theory termed “self-psychology” [10, 11]. Kohut believed that, in order to establish a sense of self, children need to idealize their primary attachment figure and have that idealization mirrored back. Kohut conjectured that narcissism results from children growing up with poorly attuned parental attachment figures who failed to empathize with and idealize them, leading to anxious feelings of inadequacy that are ultimately defended against by narcissistic tendencies.

In the last several decades, relational (or intersubjective) psychodynamic theory has emerged as a technical extension of object relations theory (e.g., [12]). Relational theory and object relations theory share similar conceptualizations of the origins and nature of psychopathology but differ on the therapist’s treatment technique. Relational theory views psychotherapy as a shared process between therapist and patient, where meaning and understanding are co-constructed.

Contemporary Psychodynamic Theories of Anxiety

Common Themes Across Anxiety Disorders

A common feature of many anxiety disorders, per contemporary psychodynamic theory, is the conflict between the wish to be autonomous (separate from others) and the fear of separation (which may result in a strong dependency on others) [13]. Consistent with object relations theory, anxiety disorders are often thought to result from insecure attachments, leading to poorly internalized representations of the self and others that make it hard to feel connected to another person in their physical absence. Thus, anxious individuals often require the presence of another person to experience a sense of security [13].

Panic Disorder

According to the American Psychiatric Association [14], a panic attack is a sudden rush of intense fear or discomfort accompanied by several physical symptoms, and panic disorder (PD) is the presence of panic accompanied by the fear of having more panic attacks and/or subsequent maladaptive changes in behavior (e.g., avoidance of situations believed to provoke panic). Psychodynamic models of panic [15, 16] emphasize core conflicts related to dependency and anger as significant etiological factors. Individuals with panic are believed to have a genetic vulnerability to anxiety in unfamiliar situations, which leads them, in childhood, to seek safety and comfort from their parents more frequently than their non-anxious counterparts. Inevitably, these children come to experience their parents as failing to gratify all their dependent wishes, fostering feelings of anger. These children, however, inhibit the expression of anger due to a fear that it could lead to separation from their parents, which could result in feeling even more unprotected and vulnerable. Conflicts around dependency in PD may also include wishes for independence.

According to psychodynamic models of PD [15, 16], these childhood conflicts (i.e., anger vs. fear of retaliation; dependency vs. autonomy) are repeated in the context of adult relationships; they may evoke panic symptoms when an individual experiences a threat to a current significant relationship. For example, individuals with panic are thought to react with anger when facing threats of separation or entrapment but simultaneously do not allow themselves to experience anger, as its expression could lead to the realization of their feared scenario. Thus, panic symptoms are understood as somatic manifestations of an unconscious emotion (typically anger). Importantly, by having a panic attack, individuals may provoke attention from others, potentially gratifying the original dependency wish.

Social Anxiety Disorder

Social anxiety disorder (SAD) is a chronic and impairing disorder characterized by fear of negative evaluation and/or humiliation, resulting in avoidance of feared social interaction or performance situations [14]. Psychodynamic theory of SAD [17] posits that the condition results from unresolved conflicts involving a wish, the response to the wish that one anticipates from other people, and a subsequent response from the self, which manifests as the patient’s symptoms. This wish-response triad comprises the “Core Conflictual Relationship Theme” (CCRT; [18]). Though the specific content of the CCRT varies from individual to individual, a typical person with social anxiety may have a core conflict related to closeness, with a CCRT which includes the wish “I want to be accepted by others,” the response “others might reject or humiliate me,” and the response from the self (i.e., the SAD symptoms such as avoidance) of trying to control the impression one makes on others and avoiding self-exposure or vulnerability. The CCRT in a socially anxious person is triggered by the perception of interpersonal threat. The activation of the CCRT brings about feelings of fear, loss of control, helplessness, and hopelessness—psychological experiences that are often accompanied by physiological anxiety symptoms.

Generalized Anxiety Disorder

Generalized anxiety disorder (GAD) is characterized by excessive anxiety and worry, accompanied by physical symptoms [14]. Contemporary psychodynamic theory posits that chronic worry may result from early relationships that predispose the individual to experience relationships as fragile [19]. Specifically, the understanding of relationships as fragile could stem from insecure attachment to primary caregivers, including feeling uncertain about whether one can depend on the parent in times of need, the perception that one may be rejected by the parent, or the experience of role reversal in which the child needs to care for an incompetent parent [20]. Insecure attachments to early caregivers may lead to a conflict between the wish for dependency and the need to become autonomous. Such insecure attachments can also contribute to the conflict between desire for closeness to attachment figures and fears of enmeshment or identity loss.

Busch and Milrod [21] suggest that, taken together, worry arises from efforts to maintain control (either actively or passively via avoidance) within relationships in order to prevent disrupted attachments. For instance, unconscious hostile or angry feelings can trigger anxiety and the fear of being rejected for such emotions, leading patients to inhibit their anger [21] and instead worry about various more superficial events or activities in their life. Worry is thus conceptualized as a defense mechanism that protects the anxious individual from thinking about difficult feelings or distressing issues [20, 22], such as feelings of loss of control or separations from loved ones. Worry and concomitant physical anxiety symptoms are activated to keep more threatening thoughts, feelings, or memories out of awareness [22].

Agoraphobia

Agoraphobia is characterized by anxiety about being in situations in which escape would be difficult and/or help is not readily available [14]. To date, no empirically established psychodynamic model of agoraphobia exists. Generally, agoraphobia, like panic, is believed to be an unconscious way of controlling the closeness of significant attachment relationships. For instance, individuals may avoid leaving a familiar environment in order to remain both physically and emotionally close to an attachment figure that they fear losing. The psychodynamic treatment of agoraphobia tends to focus on elucidating the connection between the patients’ current dependence on others and their childhood dependency wishes [23, 24].

Specific Phobia

A specific phobia is an intense and irrational fear of an object or situation [14]. To date, there are no empirically based psychodynamic theories of specific phobia. Phobias are believed to be symptomatic expressions of wishes and/or fears that a patient finds unacceptable and are thus kept unconscious [25]. For example, a specific phobia of flying on an airplane or other modes of travel may represent fears about the tenuousness of attachment relationships, insofar as travel may be unconsciously experienced as threatening an attachment.

Empirical Support for Psychodynamic Psychotherapy for Anxiety Disorders

Research Across Anxiety Disorders

The major goals of psychodynamic psychotherapy are to uncover unconscious core conflicts, to work through them, and to modify one’s maladaptive defense mechanisms, finding more adaptive ways to respond. Recently, some psychodynamically informed researchers and clinicians have made efforts to create manualized protocols to guide psychodynamic psychotherapy, both for specific disorders and for transdiagnostic emotional difficulties [26].

Generally, the extant empirical literature suggests that psychodynamic psychotherapies for a wide range of psychological disorders, including anxiety disorders, yield large effect sizes comparable to other psychotherapies that have been deemed “empirically supported” and “evidence based.” Furthermore, the “active ingredients” of other evidence-based treatments are elements that have been core to psychodynamic therapy since its origination (e.g., the establishment of a quality working alliance; [27]).

In a meta-analysis examining the efficacy of treatments for anxiety disorders [28], including GAD, PD with and without agoraphobia, and SAD, the average pre-post effect size for psychodynamic therapy was 1.17 (n = 5), compared to 1.56 (n = 4) for mindfulness therapies, 1.30 for individual cognitive-behavioral/exposure therapy (n = 93), and 0.83 for psychological placebos (n = 16). The authors did not test for significant differences in efficacies of theoretically divergent psychotherapies, though interpretation of the confidence intervals suggests that they are not significantly different. In another meta-analysis of 14 randomized controlled trials (RCTs) examining the controlled effects of psychodynamic psychotherapy for anxiety disorders [29], psychodynamic psychotherapy was significantly more effective than controls and was found to be as efficacious as active psychotherapies [e.g., cognitive-behavioral therapy (CBT)] at posttreatment and follow-up time points.

Panic Disorder with or Without Agoraphobia

Contemporary and manualized treatment protocols for PD [16, 17] are time limited and focused. The treatment has three main phases: (1) treatment of acute panic, (2) treatment of panic vulnerability, and (3) consolidation of gains and termination. In the first phase of psychodynamic treatment, the therapist works with patients to elucidate the meaning and function of their panic symptoms, aiming to attenuate panic through this awareness. Phase II uses the therapeutic relationship/transference to treat patients’ overall vulnerability to panic. In this phase, the therapist works to address patients’ conflicts between dependency and anger in their relationship with the therapist. In the final phase, the termination of treatment is utilized to extend gains by encouraging patients to directly address difficulties with separation and independence in the context of therapy’s termination.

Milrod and colleagues [30] conducted a pilot open trial of a manualized panic-focused psychodynamic psychotherapy (PFPP ; [31]) . Twenty-one patients with Diagnostic and Statistical Manual (DSM)-IV PD with or without agoraphobia received 24 sessions of twice-weekly PFPP. Sixteen of these patients achieved remission of panic and agoraphobia, and for those who began treatment with comorbid depression, their depression remitted. Gains were maintained at 6-month follow-up. Milrod and colleagues [32] next conducted a randomized controlled trial of 12 weeks of twice-weekly PFPP or applied relaxation training (ART) for PD. Patients were 49 adults with DSM-IV PD. Significantly more patients in PFPP (73%) than ART (39%) responded to treatment, and patients receiving PFPP demonstrated significantly greater reductions in functional impairment. This study indicated preliminary efficacy of PFPP.

In a larger, two-site [Weill Cornell Medical College (Cornell) and University of Pennsylvania (Penn)] RCT, Milrod and colleagues [33] randomized 201 patients with DSM-IV PD with and without agoraphobia to receive 19–24 sessions over 12 weeks of CBT, PFPP, or ART. There was a site by treatment interaction; response rates at Cornell significantly differed across treatments, such that CBT and PFPP were superior to ART. However, there were no significant differences in response rates between any of the three conditions at Penn. At Penn, patients in ART and CBT improved significantly faster and demonstrated greater panic symptom reduction than those in PFPP, whereas there were no such differences across groups at Cornell. All treatments yielded significant improvements in panic symptoms, but patients considered ART less acceptable. Notably, both CBT and psychodynamic psychotherapy had strong allegiance at both sites, but Cornell had more experience with the PFPP protocol. Additionally, psychodynamic supervision was conducted more frequently over the phone (rather than in person) at Penn compared to Cornell. This study suggests that, despite the effectiveness of this particular psychodynamic psychotherapy protocol, outcomes may be influenced by levels of experience and quality of supervision received at a given site. At Cornell, there were no differences between CBT and PFPP in terms of response rates, symptom improvement, or speed of improvement.

Social Anxiety Disorder

Leichsenring and colleagues [17] created a psychodynamic treatment manual for SAD based on Luborsky’s [18] supportive-expressive psychotherapy. In the early phase of this treatment, the therapist works to establish goals and a therapeutic alliance, identifies the CCRT through observing patterns in patients’ descriptions of interpersonal interactions or relationships in their life, and discusses the connection between the CCRT and the patient’s symptoms. The next phase of treatment involves exploring the ways in which the CCRT influences patients’ current relationships, including the therapist-patient relationship. The patients’ social avoidance (i.e., the response from the self or the symptom) is conceptualized as a defense mechanism that leads to the self-fulfillment of feared outcomes, such as negative evaluation from others or social isolation. Thus, the therapist encourages patients to respond differently to their internal conflict, instructing patients to engage in self-exposure to feared situations outside of the therapy room. In the last phase, the therapist works with patients to synthesize what has been learned and discuss how the CCRT might manifest in symptom resurgence during the process of termination due to the anticipated loss of the therapist. Booster sessions may be used to monitor and support continued progress, encourage self-guided exposure, and connect relapse with the CCRT [18].

In a multicenter controlled noninferiority trial, Leichsenring et al. [34] randomly assigned adults with DSM-IV [35] SAD to receive up to 25 sessions of CBT ([36]; n = 209), psychodynamic psychotherapy (n = 207), or a wait-list control (WL; n = 79). Though significantly more patients achieved remission in CBT (36%) than in psychodynamic psychotherapy (26%), the treatments did not yield significantly different response rates (60% in CBT vs. 52% in psychodynamic). Both active treatments demonstrated superior response and remission rates compared to WL. CBT patients showed significantly greater improvement than psychodynamic patients on measures of social anxiety (SA) and interpersonal problems, but not depression. However, at 6-month follow-up, there were no differences in response or remission rates between treatments. At a 2-year follow-up [37], no differences existed between the treatments as both had response rates of about 70% and remission rates of about 40%. In sum, there were small effects suggesting the superiority of CBT on some, but not every symptom measure at posttreatment, but the efficacy of the two treatments was not different at follow-up.

In another randomized trial [38], patients with SAD (N = 47) received up to 36 sessions of psychodynamic psychotherapy or CBT. Both treatments exhibited large within-subject effect sizes. There were no significant differences between the two treatments at posttreatment or follow-up. Sixty-three percent of patients in psychodynamic psychotherapy and 64% of patients in CBT demonstrated clinically significant change in SA [39]. At 1-year follow-up, 75% of patients in psychodynamic psychotherapy, and 65% of patients in CBT, had achieved clinically significant change, corroborating Leichsenring et al.’s [37] finding that neither response nor remission rates differ between the two treatments at follow-up. However, psychodynamic psychotherapy required slightly more sessions than CBT to produce comparable outcomes.

Research also suggests that Internet-based psychodynamic psychotherapy is efficacious for SAD. Johansson et al. [40] conducted an RCT comparing a 10-week Internet-based affect-focused psychodynamic therapy (IPDT; n = 36) for SAD to WL (n = 36). IPDT involved helping patients conceptualize their difficulties as an “internal affect phobia,” identify underlying affects that are more adaptive, gain insight into defensive behaviors, and work to resolve internal conflicts arising in current interpersonal contexts. IPDT was superior to WL at posttreatment, and this effect was large (Cohen’s d = 1.05). SA symptom levels continued to decline significantly at 2-year follow-up for the treated group.

Generalized Anxiety Disorder

Treatment for GAD [19, 22] is based on the CCRT and the attachment themes described above. In GAD, a patient’s perceptions of current relationships are believed to be distorted because of earlier relational experiences. The CCRT wish in the patient with GAD often involves the wish to be protected or cared for, with the anticipation of a negative response from others, which triggers the patient’s response of worry and anxiety. The therapist helps patients identify the CCRT and the links between the patients’ relationship patterns, symptoms, and maladaptive methods of coping, while helping patients revise their perceptions of their relationships. During the termination phase, the therapist links the CCRT and the patient’s relational fears to the anticipated loss of the therapist. For instance, if worry reemerges during termination, the therapist helps the patient understand the symptom as a means of defending against distress about losing the therapist, which mirrors distress associated with losing an attachment figure.

Crits-Cristoph et al. [19] developed a brief, 16-session supportive-expressive psychotherapy based on Luborsky’s CCRT model , adapted specifically for GAD, which focused on understanding anxiety and worry in the context of interpersonal and intrapsychic conflicts. In an open trial [41], the treatment resulted in significant changes on all outcome measures of anxiety, worry, and depression. Ratings of therapist competence and adherence suggested that the manual could be implemented with fidelity and indicated discriminant validity of the treatment (i.e., that it could be differentiated from other treatments). Critz-Christoph et al. [42] followed this open trial with a pilot randomized controlled trial comparing supportive-expressive psychotherapy for adults with DSM-IV GAD (n = 15) to supportive therapy (n = 16). Remission rates were significantly higher in the supportive-expressive (46%) compared the to the supportive (12.5%) group.

Leichsenring et al. [43] conducted an RCT comparing CBT to short-term psychodynamic psychotherapy, based on Crits-Christoph’s [19] supportive-expressive therapy manual. Treatment focused on modifying the patient’s CCRT by using the therapeutic alliance to facilitate corrective emotional experiences. Further, the treatment focused on encouraging new behaviors in an effort to modify the CCRT. Patients received up to 30 sessions of CBT (n = 29) or psychodynamic psychotherapy (n = 28). The mean number of sessions did not differ between the two groups. Patients in both treatments demonstrated significant and large improvements in symptoms of anxiety and depression at posttreatment and 6-month follow-up. The two treatments did not differ at posttreatment or follow-up on the primary outcome measure of anxiety or two corroborating anxiety symptom measures. However, CBT was superior on measures of worry, trait anxiety, and depression. At 6-month follow-up, the two treatments did not differ on most symptom measures; however, CBT yielded superior improvements in trait anxiety and worry. At 12-month follow up [44], both treatments maintained large improvements on the primary outcome measure of anxiety and three corroborating self-report measures, with no differences between groups. After 1 year, CBT was superior to psychodynamic psychotherapy with regard to trait anxiety and worry, but not depression.

In a randomized controlled superiority trial [45], individuals with DSM-IV GAD received Internet-based CBT (ICBT; n = 27), Internet-based psychodynamic treatment (IPDT; n = 27), or WL (n = 27). Each Internet-based treatment included weekly written communication with a therapist related to the specific treatment themes. Both treatments were superior to WL at posttreatment on the primary outcome measure assessing worry, with moderate effect sizes, and did not differ from each other. Both treatments had moderate to large within-group effect sizes at 3- and 18-month follow-ups on the primary measure and did not differ significantly from each other. The two treatments were only marginally better than WL at 3-month follow-up. Overall, the treatments were similarly efficacious on the primary worry measure and on secondary outcome measures of anxiety and depression.

Agoraphobia

To our knowledge, there is only one study examining treatments for patients with PD with (and not without) agoraphobia [46]. Patients with DSM-III-R [47] PD with agoraphobia were hospitalized for 11 weeks and received either an integrative treatment that integrated exposure therapy and psychodynamic group therapy and was developed specifically to target agoraphobia (IT; n = 37) or psychodynamic group therapy as usual that was conducted with people with a range of diagnoses (AU; n = 32). Fifty-eight percent of patients in IT had significantly improved at posttreatment (as determined by at least 50% improvement on posttreatment relative to pretreatment scores), and 61% had improved at follow-up, compared to 41% and 26% at posttreatment and follow-up, respectively, for those in AU. The proportion of responders was significantly greater for IT than PT at follow-up, but not at posttreatment. Though interesting, the lack of a controlled design in this study precludes the possibility of determining superiority between general psychodynamic group psychotherapy or an integrated therapy for agoraphobia; however, the integrated treatment specific to agoraphobia demonstrated promise.

Specific Phobia

To our knowledge no studies examining the efficacy of psychodynamic psychotherapy for specific phobia are available. Treatment of CCRTs may lead to the resolution of the phobic symptom, although this has not yet been empirically established. Generally, although there is value in treating and understanding specific phobias psychodynamically, exposure is a more economical model of treating the symptoms of the condition.

Summary

Extant RCTs suggest that psychodynamic psychotherapy is an efficacious treatment for SAD, defined as having at least two RCTs conducted in independent research settings in which the treatment of interest is superior to no treatment, placebo, or alternative therapies or equivalent to a treatment already deemed efficacious [48]. With regard to GAD and PD, there have not been enough studies to form concrete conclusions regarding direct comparisons of psychodynamic psychotherapy and CBT [49]; however, the research we reviewed provides support for the possible efficacy of psychodynamic psychotherapy for GAD and PD with and without agoraphobia when compared to CBT.

In a systematic review, Leichsenring and colleagues [50] concluded that psychodynamic psychotherapy can be considered efficacious for SAD and possibly efficacious for GAD and PD. Our review corroborates this conclusion. Research employing rigorous study designs, examining treatments adapted for specific disorders, suggests that psychodynamic psychotherapy is as efficacious for anxiety disorders as other empirically supported treatments and generally continues to exert its effects from posttreatment to follow-up.

Future Directions

Research on psychodynamic psychotherapy for anxiety disorders is still scant compared to other psychotherapies, such as CBT. More RCTs are necessary to clarify the relative efficacy of psychodynamic psychotherapy compared to other active treatments. Furthermore, the continued use of meta-analytic methods is necessary to compare psychodynamic psychotherapy to other empirically supported psychotherapies [49]. Additionally, controlled studies that examine psychodynamic psychotherapy for specific phobias should be conducted.

Attention needs to be directed not just toward the quantity but also toward the quality of RCTs examining psychodynamic psychotherapy. In a quality-based review of RCTs of psychodynamic psychotherapy, Gerber and colleagues [51] concluded that many promising trials of psychodynamic psychotherapy compared the treatment to an inactive comparator rather than an empirically supported psychotherapy. Thus, more trials are needed that compare psychodynamic psychotherapy to an active treatment and that utilize quality designs and adequate sample sizes. In addition, researchers need to continue examining site-by-treatment interactions [33]. Such interactions are rarely examined but do occur, and exporting therapy protocols to new sites, as well as differences between sites in supervision, can influence outcomes [33, 52]. It is unclear whether existing single-site trials that compared psychodynamic psychotherapy to another active treatment were negatively impacted by site-specific variables.

Additionally, given the centrality of the unconscious in psychodynamic models, future research should seek to utilize methods capable of observing change on this level. Implicit associations and functional neurological correlates of unconscious conflict might offer promise. For instance, one study found differences between self-reported self-esteem and implicit self-esteem in a socially anxious sample [53]. Furthermore, another study showed quantitative differences in the neural substrates of conscious versus unconscious control [54]. These studies illustrate the potential of applying contemporary research methodology to constructs relevant to psychodynamic theory.

There is also a glaring and problematic discrepancy between what psychodynamic psychotherapy aims to achieve and the variables that outcome studies typically assess [27]. The goal of psychodynamic psychotherapy extends beyond the reduction of acute symptoms, to the development of inner capacities that enable people to live life more flexibly and with a greater sense of possibility [27]. The outcome variables in the reviewed trials are generally constrained to the symptom level rather than tapping other constructs such as resilience, emotion regulation, fantasy life, increased self-reflectiveness, or increased self-understanding.

The restriction of outcome measures in RCTs to conscious, symptom-level variables is largely a result of the DSM, which defines psychiatric disorders based on symptoms. Because CBT aims to alleviate symptoms, CBT researchers initiated their outcome research using symptom reduction as a measure of treatment efficacy. Psychodynamic psychotherapy researchers, in an effort to model their studies after CBT RCTs and compare dynamic therapy to therapies already established as empirically supported, utilized the same outcome measures as their CBT research predecessors. However, whereas CBT aims to alleviate symptoms within a DSM framework, psychodynamic psychotherapy targets unconscious conflicts in addition to other outcomes (such as those detailed above). Such outcomes are not captured (or even addressed) by DSM diagnoses and may be more complicated to assess. It would be fruitful to learn whether psychodynamic psychotherapy indeed produces this kind of inner growth.

Recently, however, there has been movement away from outcomes based on symptoms alone. For example, new emphases on the network analysis and neurobiological underpinnings of psychological disorders (e.g., the National Institute of Health Research Domain Criteria (RDoC), [55]) seek to conceptualize outcomes beyond symptoms and DSM diagnoses. Recent interest in outcomes that are not constrained to the symptom level are more closely aligned with the changes sought in psychodynamic psychotherapy. In addition, the Patient-Centered Outcomes Research Institute funds studies in which outcomes are centered around what patients believe is important to them and not just symptoms. For instance, patients with panic disorder may want to understand the meaning and origins of their panic attacks in addition to feeling less anxious. Psychodynamic psychotherapy may be at least as efficacious compared to CBT on these outcomes and potentially more efficacious when compared to pharmacotherapy.

Future research focused on creating a unified psychodynamic treatment for anxiety disorders, comparing it to transdiagnostic CBTs, and assessing transdiagnostic psychic processes such as those outlined in RDoC or by Shedler [27] would be informative. Leichsenring and Salzer [26, 56] proposed a transdiagnostic protocol that consists of seven modules addressing topics such as socialization of the patient to therapy, establishing treatment goals and a secure helping alliance, identifying the core conflict, modifying underlying defenses and the underlying response of the self, and processing termination. This treatment, which integrates the most efficacious methods of psychodynamic psychotherapy into a broadly applicable protocol, parallels Barlow and colleague’s unified protocol for emotional disorders [57]. Leichsenring and colleagues plan to test this protocol using a controlled multi-site design.

Future theory development and research might also consider integrating psychodynamic and cognitive-behavioral conceptualizations and treatments. Although psychodynamic and cognitive-behavioral conceptualizations and interventions are different, they are far from incompatible. For instance, avoidance of feared situations can be viewed as a safety behavior (CBT) or a defense mechanism (psychodynamic). Furthermore, embedded in cognitive-behavioral models is the theory akin to unconscious psychic conflict [58]. For example, in their gold standard treatment for SAD, Hope et al. [59] describe core beliefs as something often outside the patient’s awareness: “Even though the layers underneath have always been there, you may not be able to see them until you remove the outer layers” (p. 207). In CBT, as in psychodynamic theory, patients are believed to be largely unaware of central parts of their psyche. Although CBT does not give as much emphasis to changing these deeper psychic structures (compared to conscious automatic thoughts, for instance), there are conceptual parallels between this component of the therapy and psychodynamic psychotherapy.

Empirical results suggest that integrative treatments may have promise [46]. Psychodynamic clinicians may consider using cognitive-behavioral techniques (rather than pharmacotherapy) to reduce acute anxiety symptoms before delving into conflict-related work aimed at reducing patients’ vulnerability to symptoms. Psychodynamic conceptualizations may also help guide exposures in CBT. For example, the understanding that attachment fears underpin panic symptoms may enhance CBT for PD with agoraphobia. When guiding an interoceptive exposure for panic, for instance, the therapist could temporarily leave the patient alone so as to potentiate habituation to the feared separation in addition to the physiological arousal. Though a full integrative model is out of the scope of this chapter, future theories conceptualizing anxiety disorders could expand upon these ideas, and proposed integrative approaches should ultimately be subjected to scientific inquiry.

Conclusions

Over the last three decades, academic psychologists have largely contended that psychodynamic psychotherapy is unscientific. However, the literature reviewed in this paper suggests that this narrative is misguided, at least as it pertains to anxiety disorders. In addition, as the field moves away from using symptom-driven criteria to define treatment outcomes, psychodynamic therapy may demonstrate even greater efficacy.

The increasing balkanization of psychology limits the advancement of psychotherapy. Though empirically supported treatments for anxiety disorders help people, a large proportion of patients do not remit. Thus, there is much work to be done, and dogmatic allegiance to particular theoretical approaches stifles conversation that could advance the common project of helping those who suffer. There is a need to shift the field of psychology away from its current state of divisiveness, toward productive collaboration and a new commitment to pragmatism.