1 Introduction

Alexithymia is a trait involving a cluster of emotion processing deficits: difficulties identifying one’s own feelings (DIF), difficulties describing one’s own feelings (DDF), and an externally orientated thinking style (EOT) characterized by rarely focusing attention on emotions [1, 2]. As defined in the attention-appraisal model of alexithymia, together these alexithymic deficits reflect difficulties at the attention (EOT) and appraisal (DIF, DDF) stages of emotion processing [1]. Alexithymia was first coined in the 1970’s by Nemiah and Sifneos [3] based on their observations of these features in many of their psychiatric patients. Alexithymia appears to impair emotion regulation [4, 5] and is an important transdiagnostic risk factor for a range of psychopathologies, including anxiety disorders, depression, substance use disorders, personality disorders, eating disorders, and psychosomatic disorders (e.g., [5,6,7,8,9,10]). Thus, alexithymia assessments are of high importance in a wide range of clinical and research settings.

For assessing alexithymia across negative and positive emotions, the Perth Alexithymia Questionnaire (PAQ) was introduced in 2018 [11]. This 24-item questionnaire has since become established as one of the most widely used and psychometrically sound tools for assessing alexithymia (e.g., Preece et al. [11], Chan et al. [12], Fynn et al. [13], Greene et al. [14]), enabling the deriving of reliable subscale scores for all the facets of alexithymia across negative and positive emotions (five subscales: Negative-Difficulty identifying feelings, Positive-Difficulty identifying feelings, Negative-Difficulty describing feelings, Positive-Difficulty describing feelings and General-Externally orientated thinking), as well as a total scale score as an overall marker of alexithymia. Originally developed in English, the PAQ has since been translated and validated across over 15 languages (e.g., Becerra et al. [15], Kiskimska et al. [16], Larionow et al. [17], Mazidi et al. [18]), including the Russian version of the PAQ which has displayed good psychometric performance [19].

In order to further enable alexithymia assessments in time-pressured settings, a 6-item version of the PAQ was recently created, called the Perth Alexithymia Questionnaire-Short Form (PAQ-S) [20]. Its items span the DIF, DDF, and EOT facets of the construct. The PAQ-S is designed to provide a total score, representing an overall level of alexithymia [20]. Prior to the PAQ-S, most available alexithymia tools required 20–40 items to be administered (e.g., [21, 22]). In contrast, the brief format of the PAQ-S means it can optimize the practicality and accessibility of alexithymia assessments, thus enabling assessments in a much broader range of settings. For example, its format is well suited to performing quick screening assessments among clinical and normative groups. It is also a good fit for research designs requiring the tracking of alexithymia over time (i.e., repeated administrations), such as in the brief electronic surveys used in ecological momentary assessment designs, or for large-scale epidemiological surveys where alexithymia is just one of many constructs of interest and thus a measure is required that can capture alexithymia in a highly efficient manner with few items [20].

To date, the psychometrics of the original English [20] and Polish [23] language versions of the PAQ-S have been tested. These versions have shown good psychometric properties, with factor analysis supporting the intended 1-factor structure (i.e., all items being strong markers of the general alexithymia construct), good internal consistency reliability, and strong convergent and divergent validity, as well as good ability to predict psychopathology symptoms and well-being levels [20, 23]. Measurement invariance was also tested in the Polish study, finding the Polish PAQ-S was invariant across age and gender categories [23]. On present evidence, the PAQ-S, like its long-form, therefore seems to have good validity across different languages and cultures.

Our aim in this study was to introduce the first Russian version of the PAQ-S and assess its psychometric properties. There is presently no Russian version of the PAQ-S, so we derived one from the Russian long-form [19]. Based on past work with the PAQ-S [20, 23], we predicted that:

  1. 1.

    The theoretically informed 1-factor structure of the PAQ-S would be a good fit to the data;

  2. 2.

    The PAQ-S would have good internal consistency reliability;

  3. 3.

    Higher PAQ-S alexithymia levels would correlate positively with anxiety, depression and stress symptoms, and negatively with well-being;

  4. 4.

    PAQ-S scores would be significant positive predictors of these psychopathology symptoms and negative predictors of well-being (controlling for demographic characteristics).

2 Method

2.1 Procedure and participants

The study followed the Declaration of Helsinki Ethical Principles [24]. Between February 2023 to October 2023, participants from the general population in Russia were recruited via social media pages (i.e., V Kontakte, Instagram, and Telegram). Participants completed an online survey administered via the Google Forms platform. All participants provided informed consent for their data to be used. The study was anonymous and voluntary, and there was no reimbursement for the respondents.

Our sample consisted of 203 Russian-speaking adults, with 151 females (74.38% of the total sample) and 52 males (25.62%), with ages ranging from 18 to 74 years (mean = 23.58, standard deviation = 8.70). People with a higher education degree made up 55.17% (n = 112) of the respondents and with those with secondary education 44.83% (n = 91).

2.2 Measures

Our participants filled out a demographic questionnaire (age, gender, and education) and a short battery of psychometric self-report measures. Internal consistency reliability coefficients for all administered questionnaires are displayed in Table 1.

  1. 1.

    The Perth Alexithymia Questionnaire-Short Form (PAQ-S). The PAQ-S is a 6-item self-report measure of alexithymia [20]. Items are scored on a 7-point scale, ranging from 1 (strongly disagree) to 7 (strongly agree). The PAQ-S uses a total scale score (ranging from 6 to 42), which is calculated by summing the scores of all items. Higher scores indicate higher overall levels of alexithymia. For this study, the instructions and six translated items from the Russian PAQ [19] were the base for developing this Russian version of the PAQ-S. A copy of the PAQ-S is provided in Supplementary Materials.

  2. 2.

    The Patient Health Questionnaire-4 (PHQ-4). The PHQ-4 [25] is a 4-item self-report measure of anxiety and depression symptoms experienced over the previous two weeks. It comprises two 2-item subscales: anxiety (e.g., “Not being able to stop or control worrying”) and depression (e.g., “Little interest or pleasure in doing things”). Items are scored on a 4-point scale, ranging from 0 (not at all) to 3 (nearly every day). A total score can be also computed by summing the scores of all items. This total score constitutes an overall marker of psychological distress. Higher scores indicate higher levels of symptoms. In this study, the Russian version of the PHQ-4 was used [26, 27].

  3. 3.

    The Perceived Stress Scale-4 (PSS-4). The PSS-4 [28] is a 4-item self-report measure of perceived stress during the previous month. It comprises four items (e.g., “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?”), which are scored on a 4-point Likert scale from 0 (never) to 4 (very often). Higher scores indicate higher levels of perceived stress. In this study, the Russian version of the PSS-4 was used [29, 30].

  4. 4.

    The Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS). The WEMWBS [31] is a 14-item measure of positive well-being over the last two weeks. Items (e.g., “I’ve been feeling optimistic about the future”) are scored on a 5-point Liker scale from 1 (none of the time) to 5 (all of the time), and are summed into a total scale score. A higher score indicates a higher general level of well-being. In this study, the Russian translation of the WEMWBS was used [32].

Table 1 Descriptive statistics and internal consistency reliability coefficients for the study variables
Full size table

2.3 Analytic strategy

Statistical analyses were carried out using Statistica 13.3 and R 4.3.1 with the lavaan (for confirmatory factor analysis) and psych (for reliability analysis) statistical packages.

2.3.1 Factor structure

We examined the factor structure of the PAQ-S using confirmatory factor analysis (maximum likelihood estimation with robust standard errors and the Satorra–Bentler scaled test statistic). We tested the 1-factor model that was supported in the original PAQ-S study [20]. All items were specified to load on a single “general alexithymia” factor. This model also included several correlated error terms (between items 1 and 2; items 3 and 6; and items 4 and 5); this reflected the underlying theoretical structure of the alexithymia construct [11], accounting for the close connection between those two items designed to assess the appraisal of negative emotions, those two items on the appraisal of positive emotions, and those two items on EOT.

Model goodness-of-fit was judged based on the following fit index values: root mean square error of approximation (RMSEA) with 90% confidence intervals (90% CI), standardized root mean square residual (SRMR), comparative fit index (CFI), and Tucker–Lewis index (TLI). RMSEA and SRMR values ≤ 0.08 indicate acceptable fit and values ≤ 0.06 excellent fit. CFI and TLI values ≥ 0.90 indicate acceptable fit and values ≥ 0.95 excellent fit [33].

2.3.2 Internal consistency reliability

McDonald’s omega values (ω) and Cronbach’s alpha coefficients (α) with 95% confidence intervals were calculated. For these coefficients, values ≥ 0.70 were judged as acceptable, ≥ 0.80 as good, and ≥ 0.90 as excellent [34].

2.3.3 Convergent and divergent validity

For assessing convergent validity of the PAQ-S, we calculated Pearson correlations between alexithymia (PAQ-S scores) and anxiety and depression symptoms (PHQ-4 scores), as well as stress (PSS-4 scores). For assessing divergent validity, we calculated Pearson correlations between PAQ-S scores and well-being (WEMWBS scores).

2.3.4 Predictive role of alexithymia in psychopathology symptoms and well-being

We ran five hierarchical multiple regression analyses to explore whether PAQ-S scores predicted significant variance in various clinically relevant criterion variables: PHQ-4 anxiety scores, PHQ-4 depression scores, PHQ-4 total scores, PSS-4 scores, and WEMWBS well-being scores. In the first step of the regression, we controlled for demographic factors (gender, age, education). In the second step, the PAQ-S total score was added to the model.

2.3.5 Demographic differences

Pearson correlations between PAQ-S scores and age were calculated. We compared the PAQ-S scores between females and males, as well as between people with higher education levels and secondary education levels using Student's t-tests. For these tests, we calculated Cohen's d effect size using the Psychometrica calculator [35].

3 Results

3.1 Descriptive statistics

Table 1 presents descriptive statistics for all study variables. In the total sample, skewness ranged from − 0.41 to 0.73 across the variables, whereas kurtosis ranged from − 0.33 to − 0.04, indicating that all the study variables were reasonably normally distributed.

3.2 Factor structure

Our confirmatory factor analysis indicated that the PAQ-S was well represented by the 1-factor model. All items loaded well on the “general alexithymia” factor (loadings = 0.55–0.58; see Table 2), and the model had all fit index values in the excellent range: RMSEA = 0.051 [90% CI 0.000; 0.129], SRMR = 0.035, CFI = 0.991, and TLI = 0.977.

Table 2 Descriptive statistics for the PAQ-S items and their standardized factor loadings on the “general alexithymia” factor from confirmatory factor analysis (n = 203)
Full size table

3.3 Internal consistency reliability

As reported in Table 1, the PAQ-S total score showed acceptable internal consistency reliability (ω and α = 0.78).

3.4 Convergent and divergent validity

PAQ-S scores were positively correlated (all ps < 0.001) with anxiety symptoms (r = 0.29), depression symptoms (r = 0.30), the PHQ-4 Total score (r = 0.34), stress (r = 0.33), as well as negatively correlated with well-being (r = − 0.48), thus supporting good convergent and divergent validity.

3.5 Regression analyses

Across our five regression analyses, the PAQ-S was a significant predictor of clinically relevant variables (see Table 3). After accounting for the variance explained by demographic factors, PAQ-S scores accounted for a significant additional 5.5% of the variance in anxiety, 5.5% in depression, 7.2% in overall psychological distress (PHQ-4 Total scores), 9.7% in stress, and 22.4% in well-being. PAQ-S scores were a positive predictor of ill-being (i.e., anxiety, depression, overall distress, and stress) and a negative predictor of well-being.

Table 3 Regression models for predicting psychopathology symptoms and well-being (n = 203)
Full size table

3.6 Demographic differences

Pearson correlations between PAQ-S scores and age were significant and negative (r = − 0.23, p < 0.001), indicating that younger people tended to have higher levels of alexithymia. Our Student’s t-test revealed no statistically significant differences in PAQ-S scores between females and males (t = − 0.81, df = 201, p = 0.420, Cohen’s d = − 0.13). We also compared levels of alexithymia between people with a higher education degree (n = 112) and people with secondary education (n = 91). There were statistically significant differences between people with higher education levels (M = 15.27, SD = 6.68) compared to people with secondary education levels (M = 18.82, SD = 6.91), with people with higher education reporting lower levels of alexithymia (t = 3.71, df = 201, p < 0.001, Cohen’s d = 0.52).

4 Discussion

The aim of this study was to examine the psychometric properties of the first Russian version of the PAQ-S. Overall, the Russian version of the PAQ-S appeared to maintain the strong psychometrics of the other language versions, demonstrating good validity and reliability.

The Russian PAQ-S had a factor structure well represented by the intended 1-factor model, thus supporting the summing of all items into a total scale score, and supporting the validity of a strong general alexithymia factor. These findings are consistent with past work with the original English [20] and Polish [23] versions of the PAQ-S, thus supporting cross-cultural consistency in the measure and structure of alexithymia. These findings with the PAQ-S further demonstrate that a robust marker of overall alexithymia can be extracted using just six items, thus providing the field with a highly practical and accessible option for alexithymia assessments across a range of settings.

The convergent and divergent validity of the Russian PAQ-S was also supported. We revealed that higher alexithymia levels (higher PAQ-S scores) were correlated with higher ill-being scores (i.e., anxiety, depression and stress symptoms), with the size of these correlations being similar to the correlations presented in the original English [20] and the Polish PAQ-S validation studies [23]. Higher alexithymia levels were also robustly associated with lower well-being scores. These results are also in line with the past work [23], supporting the high clinical relevance of the alexithymia construct. Indeed, after controlling for demographic factors, our regression analyses indicated that PAQ-S scores were positive and moderate predictors of anxiety, depression, and stress symptoms, as well as a negative and strong predictor of well-being. Our data therefore support the status of alexithymia as an important transdiagnostic risk factor for psychopathology [36,37,38]. The alexithymia field has often focused more so on relations with ill-being compared to well-being, and thus our findings of strong relationships with low well-being have good novelty in demonstrating that alexithymia may also play a role in reduced positive experiences. Given that alexithymia impairs emotion regulation [4, 5], and that emotion regulation is important in up-regulating positive experiences to facilitate well-being [39], these relationships between alexithymia and well-being are in line with the specifications of conceptual models linking alexithymia, emotion regulation, and affective outcomes, namely the attention-appraisal model of alexithymia [1] and the process model of emotion regulation [40].

As for demographic differences, in this data-set, there were no statistically significant gender differences in alexithymia levels. People with higher education levels had lower alexithymia than people with secondary education, supporting previous conclusions [41]. We also found that younger people tended to report higher alexithymia. In general, these results correspond with the those presented in the Polish validation study of the PAQ-S [23] and several other reports in the alexithymia field (e.g., [42]). However, there are some inconsistencies across the alexithymia literature with respect to links with demographic factors, regardless of the alexithymia measure used. For example, some data have found higher alexithymia levels in males [41, 43], or have found that alexithymia can increase with age [41]. Further work with larger more diverse samples, and ideally longitudinal designs, will be required to determine how (or whether) PAQ-S scores differ across different demographic groups. Conclusions on demographic influences in this study should be considered tentative.

4.1 Limitations and future directions

Whilst our sample size was sufficient for the psychometric analyses we conducted, our analyses were carried out on a relatively small sample from the general community sample of Russian people, and no clinical samples were included. We also did not examine some psychometric properties, such as test–retest reliability, and we did not directly compare the performance of the PAQ-S with the longer PAQ. All our concurrent validity markers were self-report questionnaires, and future work with performance-based or behavioral markers of relevant constructs would be useful. Our study is cross-sectional, therefore, the temporal order of alexithymia and its correlates cannot be established here. As aforementioned, future research in more diverse samples, with longitudinal research designs, will therefore be beneficial.

5 Conclusions

Our data suggest that the Russian version of the PAQ-S has good psychometric properties as an overall measure of alexithymia. As a brief and freely available questionnaire, the PAQ-S should therefore help to substantially enhance the accessibility of robust alexithymia assessments in a variety of settings.