The compiling instructions informed the participants that the answers were anonymous and that there were no right or wrong answers. Informed consent was obtained from all subjects involved in the study. A privacy policy at the beginning of the questionnaire described to the participants the purposes and methods of treatment operated by the data controller on the personal data collected. The research was submitted to the ethics committee of the University Polyclinic “Paolo Giaccone” of Palermo and obtained a favorable opinion for its development.

The sociodemographic questionnaire included questions about gender, age, nationality, education, marital status, presence of children, and information about their experience with psychotherapy (whether they ever had it or not, frequency, number of sessions, and approach).

The functional anxiety assessment questionnaire (FQ) is a self-made questionnaire based on the psychotherapeutic functional theoretical model. It evaluated the manifestations of anxiety symptoms taking into account four different planes: cognitive-symbolic, emotional, physiological, and postural. The questionnaire consists of 12 items. For each item, subjects must express their degree of agreement in relation to a five-point Likert scale (1=never, 5=always). The questionnaire used for this study is available as Supplementary Material.

A Confirmative Factor Analysis was performed on all items corresponding to the four constructs measured as reported in Table 1. The final model scores well on the comparative fit index (CFI), Tucker – Lewis index (TLI), the Standardized Root Mean Square Residual (SRMR), and the root mean square error of approximation (RMSEA) equal to, 0.91, 0.88, 0.05, and 0.005, respectively.

The reliability analysis of the items of the final model by calculating the Cronbach alpha is acceptable (Cronbach’s Alpha=0.73). A completed detailed validity assessment of the FQ is actually under examination in a new study.

The coping orientation to the problems experienced-new Italian version-25 (COPE-NVI-25; Foà et al., 2015) is the reduced version of the COPE-NVI (Sica et al., 2008) developed to detect the coping strategies summarized by the authors in five categories: social support, avoidance, positive attitude, problem orientation, and transcendent orientation. The tool consists of 25 items that can be answered via a four-point Likert scale (1=I usually do not do this at all, 4=I usually do this a lot). The COPE-NVI-25 proved to be an instrument as valid as the original COPE, but easier to administer. Authors reported that Cronbach’s Alpha index is between 0.63 and 0.96.

A number of 432 participants answered the questionnaire, the demographic characteristics are presented in detail in Table 1. The data was processed to remove underage subjects and responses attributable to so-called careless responses (Meade and Craig, 2012), i.e., responses that do not reflect real scores. In particular, the number and mean of consecutive identical responses and the intra-individual variability of response were evaluated. Starting with the detection of these suspected cases, responses in which the scores obtained were greater than 1.9 SDs from the mean were removed as well as subjects who do not comply with the criterion of having done at least eight sessions of psychotherapy. Following the data cleaning process, the number of valid responses was 389, of which 76 were males and 313 females, with an age between 19 and 79years old.

Demographic characteristics are presented in detail in Table 2. The average age of this group was 44.38years (SD=11.66). With regards to the qualification, 75.19% of the subjects declared that they have obtained a qualification equal to or higher than the first level degree. About 185 subjects (42.70%) had psychotherapy sessions at least once in their life. Of these 185, 81 (42.70%) declared that they were continuing the therapy sessions at the time of completing the questionnaire. The subjects declared to have carried out an average of 87.19 therapy sessions (SD=72.84), these sessions took place in most cases on a weekly (62.34%) or biweekly (25.97%) frequency.

Multiple linear regressions were conducted to predict the scores obtained on the cognitive-symbolic, emotional, physiological, and postural dimensions of the FQ based on coping strategies measured by the COPE-NVI-25: positive attitude, avoidance, problem orientation, transcendent orientation, and social support. The regressions were conducted on both people who had psychotherapy sessions at least once in their life (clinical sample) and people who had not (non-clinical sample). Tables 3, 4 report the results of the regressions.

Starting from the results of the multiple regression analysis, a hierarchical multiple regression analysis was conducted to test the hypothesis that the psychotherapy moderates the relationship between the distress on:

The scores on the FQ dimensions were indicated as dependent variable, while the coping strategies and having had psychotherapy or not (−1 no psychotherapy, 1 psychotherapy) were modeled as predictors. The independent variable and predictors were introduced in step 1, the interaction between psychotherapy and coping strategies in step 2. To avoid the multicollinearity, the residual centering technique (Lance, 1988; Little et al., 2006; Geldhof et al., 2013) was applied with the support of the R pequod package (Mirisola and Seta, 2016).

The study explored the relationship between coping strategies and a multidimensional assessment of anxiety symptoms measured during the COVID-19 pandemic in people who were having psychotherapy sessions, while participating in the study and those who have never had them. Moreover, the role of psychotherapy as moderator between anxiety symptoms and coping strategies has been evaluated. Research results show that there is a substantial difference in functioning between patients in psychotherapy and those who have never had it. As predicted, psychotherapy moderates the relationship between multidimensional assessment of the anxiety and coping strategies. In particular, the moderation of psychotherapy influences the relationship between the cognitive level and social support, between the cognitive level and avoidance, and, finally, between the muscular one and the positive attitude toward life. In the first two cases, the subjects in the clinical sample who use the coping strategies of social support and avoidance the most obtain high scores on the cognitive dimension for the presence of negative fantasies, confusion, difficulty making decisions, rumination, and reduced planning. Social support can therefore also be a risk factor if the subject’s network of relationships causes stress, for example, discouraging the expression of emotions, making critical observations, or not providing the promised help (Lincoln, 2000). This result appears, at first sight, dissonant with a long tradition of research (e.g., Blazer, 1982; Cohen, 1988; House et al., 1988; Taylor, 2011), which has highlighted the positive aspects of social support, highlighting the positive consequences from the point of view of the health of the subjects who receive it, but it is consistent with what was found by Babore et al. (2020) on health professionals.

The risk associated with the avoidance coping strategy is known and consistent with the scientific literature. Avoidance, described as a person’s tiring attempt to deny, minimize, and evade direct confrontation with stressful requests, is generally linked to anxiety and depression (Cronkite and Moos, 1995; Blalock and Joiner, 2000; Penley et al., 2002; Marchand and Hock, 2003). Some authors (Holahan et al., 2005) have shown that the avoidance coping strategy can play a key role in generating a wide range of additional stressors. For example, avoidance can lead to financial or health problems, or aggravate tensions at work or within the family. Especially among women, hypertrophic rumination is associated with more severe and long-lasting depressive symptoms (Nolen-Hoeksema, 1998).

The third and final analysis of moderation that has obtained significant results relates to subjects who use the coping strategy of a positive attitude toward life. In these subjects, if belonging to the clinical sample, the higher their level of positive attitude, the more their perception of the rigidity of their posture, limbs, neck, and back muscles decreases. It is known that high levels of positive attitude are positively correlated with almost all dimensions of psychological well-being (Sagone and De Caroli, 2014) and also have a positive impact on the salary received (Mohanty, 2009a,b). This result is, once again, consistent with the study of Babore et al. (2020) carried out during the first peak of the pandemic in Italy on healthcare workers, in which the higher positive attitude of the subjects is, the lower is their perceived stress. Promoting a positive attitude toward life is therefore configured as a powerful protective factor from stress and malaise.

An additional result of the study is the identification of the orientation to the transcendent as predictor of physiological scores in the non-clinical sample. Currently, it is known that rigid adherence to religious beliefs can interfere with adaptive coping. The worst score obtained to the physiological dimension can be related, for example, to an external locus of control, delegating a legitimate healthcare to the will of a supernatural being (Carone and Barone, 2001).

The results of this study are promising but there are several limitations to mention that may jeopardize both internal and external validity. The sample involved in the study cannot be considered representative of the Italian people who have access to psychotherapeutic services. Considering that the percentage of people who have had psychotherapy sessions in Italy is relatively small (Vecchia, 2020), authors decided to use a convenience sampling, asking psychotherapists and schools of psychotherapy to involve their patients in the study. As a consequence, the number of people who have experience of psychotherapy is over-represented and this introduces a selection bias that limits the generalizability of the findings. The patients involved in the study had a different number of psychotherapy sessions and the reasons they asked for help are very heterogeneous. Although, an inclusion criterion of having done at least eight psychotherapy sessions (according to Lambert, 2013) has been established, the uneven composition of the sample can introduce a confounding effect that may influence the outcomes. For example, being in different phases of the therapy can influence the use of coping strategies (Kramer et al., 2010; Starrs and Perry, 2018; Conceição et al., 2019).

Moreover, considering that the psychotherapists of the patients involved in the study use many different approaches, multiple interventions are carried out to face the broad range of mental problems mentioned above. Authors relied on the functional equivalence of the various form of treatment according to many meta-analyses reporting a lack of significant differences among psychotherapeutic treatments (e.g., Shapiro and Shapiro, 1982; Blatt and Zuroff, 2005; Wampold, 2013; Barth et al., 2016; Cuijpers, 2017). This functional equivalence is called Dodo bird effect (Luborsky et al., 2002), a label inspired by Alice in Wonderland in which a Dodo states that “Everyone has won, and all must have prizes.” However, although, the studies reporting small and non-significant differences between treatments are very numerous, the Dodo bird effect is still controversial (Budd and Hughes, 2009). For example, meta-analyses have to take into account the role of the therapist’s allegiance that is highly predictive of the efficacy of interventions (Luborsky et al., 1999; Wampold, 2013). In addition, researchers highlight that the use of randomized control trials to identify specific effects of psychological therapy should be questioned in favor of systematic single case designs (Hallam, 2015). According to the considerations outlined above, a replication study should consider the interaction between the independent variables “diagnosis” and “type of treatment.”

With regards to the measure used in the study, the validity of the FQ must be improved for example accompanying it with another instrument or predictor. Therefore, although the confirmatory factor analysis of the FQ is acceptable, the questionnaire requires the application of sophisticated methodologies both quantitative (Kunnan, 1998) and qualitative (Banerjee and Luoma, 1997) in order to be validated.

Finally, the lack of longitudinal data cannot capture changes in psychological distress during both the psychotherapeutic treatment and the course of the COVID-19.