Research shows that many patients offered psychotherapeutic treatment are ambivalent toward it and may not be ready to participate (Engle and Arkowitz, 2006; Hunter et al., 2014; Oliveira et al., 2021). Data on cognitive group therapy from the Danish TRACT-RCT indicates high dropout rates in Danish outpatient mental health service (MHS) clinics—an average of 35.5% (N = 291) (Reinholt et al., 2021). This is somewhat consistent with international meta-analytic findings, which show dropout rates of similar magnitude (21.9%) (Swift et al., 2017).

These dropout rates may result in a substantial waste of human resources, and is particularly regrettable in settings offering group therapy, where dropouts may result in both direct economic costs due to wasted resources and indirect costs due to disturbance of the group dynamics (Wierzbicki and Pekarik, 1993; Barrett et al., 2008). Further, they may potentially indicate that people who need help and have been seeking it are not receiving it.

Factors that presently inform the clinical choice of psychotherapy as a relevant treatment modality include the presence of interpersonal problems (Barkham et al., 2021), the patient’s recognized need for help and capacity to relate (Barkham et al., 2021), patient preference (Williams et al., 2016) and psychological-mindedness (McCallum and Piper, 1997), and the patient’s motivation or readiness for change (Vogel et al., 2006). Relevant screening tools might help detect patients who are not ready for psychotherapy and may require other treatment efforts instead—i.e., motivational enhancement pre-therapy or other treatment modalities.

A patient’s motivation for change plays a crucial role in the outcome of psychotherapy (Vogel et al., 2006). The overall concept of motivation relates to the intentional aspect of change, such as the internal drive preceding the initiation of therapy, as well as the patient’s ongoing engagement throughout treatment (Prochaska and DiClemente, 2005), including the desire to change (Truant, 1999), willingness to discuss personal matters (Krause, 1967), level of distress (Schneider and Klauer, 2001), and preparedness to make reasonable sacrifices (Hacker, 1962).

Theoretically, the concept of readiness for change is described as a core component in the “stages of change” cycle of the so-called Transtheoretical Model (TTM) of behavioral change set forward by Prochaska and DiClemente (1984). In the “stages of change” cycle, patients are assumed to vary in their overall motivation to change, being on different levels ranging from having no intention for therapeutic change (“precontemplation”), to being ambivalent about change (“contemplation”), to having intentions and a commitment to change (“preparation”), to starting changes and actively working on them (“action”), to consolidating changes in order to prevent relapse (“maintenance”). According to the model, patients continuously move through the individual stages in a spiral pattern, and relapses are integral to the cycle. Cycling back to an earlier stage of change gives the patient the opportunity to apply different and possibly more effective strategies to make the changes they were not successful with the first time (Prochaska and Norcross, 2014).

Several measures have been used to evaluate motivation for change, of which the University of Rhode Island Change Assessment (URICA) is one of the most commonly employed (DiClemente and Hughes, 1990). The URICA is designed to measure readiness for change in regards to any behavior (e.g., behavioral change aimed at weight loss, substance abuse, etc.). It assesses this motivation dimensionally, with four dimensions derived from the TTM: Precontemplation, Contemplation, Action, and Maintenance. It is commonly calculated as a Readiness To Change (RTC) score, which is a composite score of a patient’s overall readiness derived from the individual subscales. In addition to this, a Committed Action (CA) measure has been proposed by Pantalon et al. (2002), based on the reflection that many of the items on the Contemplation subscale reflect ambivalence about change, and that endorsement of these items may reflect a decreased likelihood of taking action to change (Pantalon et al., 2002).

Aside from the URICA, a few instruments exist that are specifically designed to assess readiness for psychotherapy. One is the Counselling Readiness Scale (CRS), a subscale of the Adjective Check List (Gough, 1980) designed to identify persons who are open to change and who seem likely to profit from psychotherapy. Herein, respondents are asked to endorse adjectives which they believe accurately describe them. The CRS includes different scales for men and women with different numbers of items. Only limited information exists about the instrument’s reliability and validity, though studies conducted by its developers (Heilbrun, 1966; Heilbrun, 1978) found it helpful in predicting premature termination of therapy. Another example is the Stages of Change Readiness and Treatment Eagerness Scales (SOCRATES), which was developed as an alternative to the URICA for measuring readiness for change with regard to problem drinking (Miller and Tonigan, 1996). The SOCRATES includes 19 items which together make up three scales, ambivalence, recognition, and taking steps, which are hypothesized to represent continuously distributed motivational processes.

A third relevant instrument is the Readiness for Psychotherapy Index (RPI), developed by Ogrodniczuk et al. (2009) as a clinical tool to assess a patient’s readiness to enter into a psychotherapeutic intervention. Its creators amassed a pool of items from existing literature and clinical insight that were expected to reflect elements of readiness for psychotherapy. This resulted in a pool of 42 items that the authors believed to represent seven dimensions of readiness. From this pool, a final set of 20 items were selected and, through exploratory factor analysis, were found to be optimally described by a four-factor model. These factors encompass Disinterest, Perseverance, Openness, and Distress, and each one contains five items. An Overall Readiness score is calculated based on the four. High internal consistency was reported by the developers (Ogrodniczuk et al., 2009).

Trials have consistently found patients’ readiness to change to be an essential factor in predicting and moderating psychotherapy outcomes (Ogrodniczuk et al., 2009). Thus, a valid measure of this construct is critical to understanding its potential impact in specific contexts or populations, such as in the treatment of mood and personality disorders, eating disorders, and addiction. Readiness to change is particularly relevant in these populations since treatment consists primarily of psychotherapeutic interventions, which require much effort and persistence, and since there is an inherent low motivation or ambivalence toward change in these disorders (i.e., lack of energy and initiative in depression, ambivalence toward behavioral change in eating disorders and addiction).

The URICA has mainly been applied in research on substance (Abellanas and McLellan, 1993; Belding et al., 1995; Carney and Kivlahan, 1995; Belding et al., 1997; el-Bassel et al., 1998; Field et al., 2009) and alcohol use disorders (DiClemente and Hughes, 1990; Field et al., 2009; Willoughby and Edens, 1996; DiClemente et al., 1999; Edens and Willoughby, 1999; Carbonari and DiClemente, 2000; Edens and Willoughby, 2000; Hoyer et al., 2003), but has also been used in studies on other psychiatric disorders such as anxiety disorders (Dozois et al., 2004) and bulimia nervosa (Treasure et al., 1999). It has additionally been applied to patients with somatic pain (Kerns and Rosenberg, 2000), criminal offenders (Polaschek et al., 2010), tobacco smokers (DiClemente et al., 1991), and individuals wishing to lose weight (Ingledew et al., 1998).

Krebs et al. (2018) conducted a meta-analysis of the association between measures of readiness and psychotherapy outcomes. They included 76 studies, in which the most frequent readiness measures were the URICA (n = 46) and the SOCRATES (n = 10). Thirty-six studies provided data on the association between readiness and adherence to treatment/premature dropout. Among these reports, the authors found a medium mean effect size (d = 0.36) (95% CI = 0.26–0.47) (Krebs et al., 2018).

Two studies have investigated the use of RPI in populations with emotional and substance-related disorders awaiting psychotherapy (Ogrodniczuk et al., 2009; Kealy et al., 2017). Another has investigated the relationship between psychosis spectrum symptoms and psychotherapy readiness among inpatient adolescents in the mental health service (Thompson et al., 2021). None of these examined the instrument’s external validity regarding adherence to subsequent psychotherapy.

However, the utility of these two instruments for assessing readiness for psychotherapy engagement or adherence among patients with emotional disorders has only been analyzed in one study (Dozois et al., 2004). Further, they have yet to undergo psychometric validation in Danish. Hence, the objectives of the present study are to examine the psychometric properties of a public Danish translation of the URICA and our own Danish translation of the RPI and evaluate their predictive validity through their ability to predict early adherence to group psychotherapy.

We hypothesize that the Danish URICA and RPI both have a four-factor structure corresponding to the one suggested for the American (Dozois et al., 2004) and Canadian English originals (Ogrodniczuk et al., 2009). Additionally, we expect them to be able to adequately discriminate between patients who are ready or not ready for psychotherapy, as judged by their early adherence to at least three of their first four planned sessions of psychotherapy as a proxy measure of psychotherapy engagement.

Data collection took place between December 2021 and April 2023. A total of 155 patients gave informed consent and were included in the study. One hundred and five were recruited from the psychoeducational groups and 40 enrolled in the RCT.

The sample had a mean age of 31.1 (SD = 10.8) years. Most (79.3%) were female and were on sick leave (43.9%) or unemployed (14.8%). Most were awaiting treatment for avoidant personality disorder (32.9%) or borderline personality disorder (30.2%). Further, the 86 patients who were offered therapy during the study waited for a mean 10.0 (SD = 5.9) months from the time of their first appointment in the clinic to their first psychotherapy session. See Table 2 for further demographic and diagnostic details of the included sample.

For eight patients (5.2%), there was a reason documented in the electronic health records for why the patient could not attend therapy. One had a child with a health condition, four had a somatic health condition of their own, and three could not attend due to lack of time.

The present study examined the psychometric characteristics of the Danish URICA and RPI in a clinical sample of patients with PTSD, mood, and personality disorders who were awaiting therapy. To the best of our knowledge, it is the first time the RPI has been investigated as a possible screening instrument. Our sample was heterogeneous regarding diagnosis, but all were patients in the publicly funded Danish Mental Health Services and were, therefore, all patients with complex psychopathology who had not sufficiently responded to previous attempts at treatment. The sample was highly motivated overall and had URICA scores of comparable magnitude on all subscales to populations with anxiety (Dozois et al., 2004), substance use (Pantalon et al., 2002), and bulimia nervosa (Treasure et al., 1999). Of previous research, only Dozois et al. (2004) study on patients with anxiety explored predictive validity regarding adherence to therapy. However, this study simply noted that patients completed the URICA before beginning therapy, not how long beforehand, and no information was given on the duration between completion of the URICA and the beginning of therapy.

We found only a poor-to-reasonable fit to the four-factor model which had previously been suggested for the URICA. This was, however, the best fit of the two different models evaluated. This reasonable fit is consistent with other studies on the URICA, which have likewise investigated the factor structure of the entire instrument with CFA (Pantalon et al., 2002; Field et al., 2009).

Similarly, we carried out CFA for the RPI and found a poor to reasonable fit to the four-factor model which had previously been suggested for this instrument. We are unaware of any other studies that have reported CFA fit measures for the RPI, as Ogrodniczuk et al. (2009) did not. This could suggest that the instruments do not, in fact, assess four distinctly different dimensions as has been theoretically suggested.

Neither the URICA, the RPI nor any of their subscales or composite measures predicted early adherence to therapy in our study. Nor did the Revised four factor model 1 or 2. This contrasts with findings by Dozois et al. (2004), who examined the English URICA’s ability to predict retention to cognitive behavior therapy in a sample of 81 patients recruited from a teaching hospital in Canada. They showed that the Action subscale predicted retention to therapy [F_{(4, 76)} = 2.66, p < 0.05] (Dozois et al., 2004). We are not aware of any studies that have examined the predictive power of RPI.

Our results suggest that there is no or insignificant association between readiness for therapy and early adherence to group psychotherapy for this patient group. If that is the case, other factors (e.g., demands from the social system that dictates that patients should seek and attend treatment to receive welfare benefits; relationships with other members of the therapeutic group; patients not wanting to lose access to their contact person in the clinic) could explain why some patients attend group psychotherapy despite low readiness. Some patients may also decline the offer to start group psychotherapy due to other obligations which do not allow them to attend in the daytime (e.g., employment or schooling conflicts; having to move to another city to start schooling; living with a somatic health condition). Hence, patients in full-time occupation may not receive an appropriate/feasible treatment offer with current clinical practice, which is designed for individuals who can attend therapy in the daytime and has little to no flexibility regarding time scheduling. The findings could thus signal that some patients should have been offered psychotherapy in the evenings, or flexible individual psychotherapy. Due to the naturalistic design of the study, patients were not systematically asked for the reason why they did not attend therapy. Future research should systematically investigate and record patients’ reasons for non-adherence or for declining therapy.

However, it is also possible that the lack of predictive validity reflects methodological shortcomings in the present study, which could have several explanations:

The URICA and RPI were administered to patients at the psychoeducational groups. For patients to reach these groups, they had already needed to take several steps (see Figure 1). As such, all the patients included in the sample had, at this point, shown a significant amount of persistence in their pursuit of treatment and had, on many occasions, had the chance to drop out of it. This could explain the homogeneous high readiness observed in the current sample, and the homogeneity of readiness in the sample plus the low number of no-shows could be the reason for the scales’ inability to predict adherence. In addition to this, only a limited number of the included patients were offered therapy within the duration of the study, and of these, only a few dropped out. Future studies should administer the instruments to patients at their first appointment in the clinic or at the start of therapy. Future research could also be conducted in a primary care setting, administering the instruments to patients who are referred to private practice psychologists from general practitioners, and investigate any association with adherence to private practice psychotherapy. On a related note, it is possible that patients’ motivation fluctuated over time, and that participants were either less or more motivated to change at the point in their treatment journey where the instruments were administered as compared to later. Administration of the instruments on several occasions could investigate whether this is the case. Further, the Transtheoretical Model (TTM) of behavioral change has received substantial criticism for not addressing the fact that a person may be in different stages of change at the same time for one single type of behavior (e.g., in the action stage regarding adherence to therapy, but in the contemplation stage regarding therapy homework). Two individuals may therefore be placed in the same stage despite having different perspectives regarding the behavior (e.g., in precontemplation, one patient may know that she is inactive but not be interested in changing her course of behavior, while another may be entirely unaware that he is inactive and needs to change). These two kinds of precontemplators might score the same on the URICA but still be entirely different. Another issue is that the TTM attempts to define stages with sequential transitions between them, but this phenomenon might be a continuum of change (e.g., a patient might be contemplating change but simultaneously reading about therapy online, i.e., taking small action steps). Such a patient’s stage on the TTM would be hard to define (Sussman et al., 2022). It might therefore be the case that the TTM and, in turn, the URICA lacks precision and does not capture the constructs necessary to assess readiness for behavioral change in a way that is relevant for psychotherapy.

Next, we examined the internal consistency of the Danish URICA and RPI. The internal consistency of the action subscale was satisfactory, while the remaining subscales were found to be either borderline satisfactory or unsatisfactory. These measures of internal consistency were similar to early research on the original URICA (DiClemente and Hughes, 1990) (alpha for the individual subscales: Precontemplation = 0.69; Contemplation = 0.75; Action = 0.82; Maintenance = 0.80), and in patients with anxiety (alpha for the individual subscales: Precontemplation = 0.73; Contemplation = 0.79; Action = 0.90; Maintenance = 0.81) (Dozois et al., 2004) and cocaine use (alpha for the individual subscales: Precontemplation = 0.75; Contemplation = 0.79; Action = 0.83; Maintenance = 0.78) (Pantalon et al., 2002). However, they were moderately poorer than in research utilizing the original URICA in patients with other substance use (alpha between 0.80 and 0.84) (Carney and Kivlahan, 1995). We are unaware of any studies that have published an omega value for the URICA subscales. The poorer alpha values could possibly be explained by the fact that the original URICA was explicitly developed to assess motivation for change among alcohol and substance users, and such users might understand the individual items differently than individuals with anxiety and other psychiatric disorders.

Internal validity of the Danish RPI was broadly comparable to the alpha values in Ogrodniczuk et al. (2009) for subscale Openness (0.77 vs. 0.79), but was lower for subscales Disinterest (0.46 vs. 0.87), Perseverance (0.66 vs. 0.80), and Distress (0.63 vs. 0.71). Ogrodniczuk et al. (2009) did not publish omega values in their paper (Ogrodniczuk et al., 2009). These findings might suggest that the items in the Danish RPI related to disinterest, perseverance and distress are understood differently by Danish patients as compared to the original sample.

Limitations to the present trial include the fact that its results may generalize only to the included diagnoses and outpatient mental health services. In addition, the composition of the sample limits the generalizability of the results to males and to people older than 40 years. A limitation of the present study is the limited sample size, and especially the limited subsection of patients which did actually proceed to psychotherapy during the study period. Due to the limited sample size, we did not explore the possible impact of diagnosis or motivation on readiness for change. The study’s primary strength is its public mental health setting and inclusion criteria similar to those used in clinics, which adds to the external validity of the findings.

In conclusion, the predictive validity of the URICA and RPI are limited regarding early adherence to group psychotherapy in the Danish Mental Health Services. Our results suggest that they may not be useful as screening instruments; however, further study is needed. Future research in this area should administer the URICA and RPI upon intake to the clinic or at the start of therapy, or upon referral to the clinic from a general practitioner. Future research could also explore any possible influence of diagnosis and motivation on patients’ readiness for change.