In recent years, there has been a growing recognition of the complexity and diversity of gender identity and expression across the lifespan. Reference Agénor, Murchison, Najarro, Grimshaw, Cottrill and Janiak1 Increasing visibility of transgender and gender-diverse (TGD) individuals has brought attention to the need for appropriate clinical tools and research instruments capable of capturing the nuances of gender-related experiences. Reference Meneguzzo, Vozzi, Biscaro, Zuccaretti, Tenconi and Favaro2,Reference Shulman, Holt, Hope, Mocarski, Eyer and Woodruff3 Traditionally, gender identity has often been conceptualised in binary terms, with limited consideration for fluidity, exploration and the internal processes that characterise many individuals’ lived experiences. Reference Moolchaem, Liamputtong, O’Halloran and Muhamad4 However, contemporary models – such as the gender minority stress framework and embodied and developmental perspectives – emphasise that gender identity is not static and can be shaped by a dynamic interplay of biological, psychological and sociocultural factors. Reference Katz-Wise, Ranker, Kraus, Wang, Xuan and Green5,Reference Ben Hagai, Zurbriggen, Lugo, Ben Hagai and Zurbriggen6
Within this context, individuals may experience various degrees of preoccupation with their gender identity, or encounter moments of instability, uncertainty or shifts in self-perception. Reference Doyle7,Reference Levitt and Ippolito8 These experiences may or may not be associated with distress and can occur across different gender trajectories, including those of transgender, non-binary and cisgender individuals. While preoccupation with gender-related themes may be part of a broader process of identity consolidation, it may also reflect psychological suffering, particularly in contexts of social invalidation, stigma or lack of access to gender-affirming care. Reference Birkett, Newcomb and Mustanski9–Reference Timmins, Rimes and Rahman11
Despite the increasing clinical and theoretical interest in these dimensions, few standardised and validated tools are available to assess them. The Gender Preoccupation and Stability Questionnaire (GPSQ) was developed to address this gap by offering a brief, psychometrically sound instrument designed to measure two key constructs: the degree of cognitive and emotional preoccupation with gender identity, and the perceived stability of one’s gender identity over time. Reference Hakeem, Črnčec, Asghari-Fard, Harte and Eapen12 Although these two constructs were conceptually distinct, the original validation study found that they collapsed into a single factor, resulting in a unidimensional structure. The original validation study demonstrated good reliability and construct validity, and the GPSQ has since been employed in various clinical and research settings to explore the spectrum of gender experiences beyond diagnostic categories alone. Reference Bowman, Casey, McAloon and Wootton13,Reference Brokjøb and Cornelissen14
However, the use of such tools across different cultural and linguistic contexts requires careful adaptation and validation. Measures developed in English-speaking countries may not fully capture the lived realities of individuals in other sociocultural environments, where gender norms, access to care and societal attitudes towards gender diversity differ significantly. In Italy, TGD individuals continue to experience significant barriers to care despite the presence of specialised centres. National surveys highlight inadequate provider training and widespread discrimination. Reference Marconi, Brogonzoli, Ruocco, Sala, D’Arienzo and Manoli15,Reference Scala, Ceolin, Miscioscia, Basso, Campello and Camozzi16 While a few specialised units exist, regional disparities and long wait times remain, underscoring ongoing inequities in access, stigma, systemic discrimination and regional disparities in access to care. Reference Romani, Mazzoli, Ristori, Cocchetti, Cassioli and Castellini17,Reference Fisher, Marconi, Castellini, Safer, D’Arienzo and Levi18 The availability of validated instruments that are sensitive to these contextual factors is essential not only for accurate clinical assessment but also for advancing inclusive, evidence-based research and promoting the well-being of gender-diverse populations.
Therefore, the present study aims to provide a culturally and linguistically appropriate version of the GPSQ for the Italian context. By conducting a translation and psychometric validation process, we seek to contribute a valuable tool for use in both clinical practice and gender-focused research. Furthermore, the availability of such an instrument can help foster a more nuanced understanding of gender identity development, support early detection of distress related to gender incongruence and ultimately inform interventions that are attuned to the needs and narratives of TGD individuals in Italy.
Method
Participants
Participants were recruited between 2023 and early 2025 at the Regional Reference Center for Gender Identity of the University Hospital of Padova (Italy), a specialised clinical and research service for TGD individuals. This recruitment strategy was chosen based on methodological considerations: clinical samples typically exhibit a broader range of gender-related distress and identity instability, which enhances the sensitivity and robustness of psychometric validation. However, this clinical sample design might limit generalisability to the broader population of TGD individuals. The inclusion criteria were: (a) age over 18 years; (b) self-identification as transgender, non-binary, genderqueer or any other non-cisgender identity within the TGD spectrum; (c) Italian as first language and (d) absence of cognitive impairment or conditions that could compromise the ability to understand and complete self-report measures.
All participants provided written informed consent after receiving a complete explanation of the study’s aims and procedures. Participation was entirely voluntary, and no financial compensation was provided. The study protocol was approved by the Ethics Committee of the Azienda Ospedale–Università di Padova and conducted in accordance with the Declaration of Helsinki and relevant national guidelines on research with human participants.
Procedure
The original English version of the GPSQ was translated into Italian using a standardised forward–backward translation procedure, in accordance with international guidelines for the cultural adaptation of psychological instruments. The process involved several steps to ensure linguistic accuracy, conceptual equivalence and cultural appropriateness.
First, two bilingual researchers (a clinical psychologist and a psychiatrist, both fluent in English and Italian) independently translated the GPSQ from English into Italian. These two versions were then reviewed and synthesised into a single, preliminary Italian version. Discrepancies were resolved through consensus, in consultation with a third expert in gender identity and psychological assessment.
Next, this synthesised version underwent back-translation by an English as their first-language speaker, who was fluent in Italian and unfamiliar with the original GPSQ. The back-translation was compared to the original English version to identify any semantic or conceptual inconsistencies, which were addressed by minor refinements to the Italian version.
The pre-final version was reviewed by an expert panel of clinicians and researchers with expertise in gender-affirming care. The panel evaluated each item for clarity, cultural relevance and conceptual accuracy, resulting in minor adjustments to improve cultural appropriateness. Notably, Item 13 was slightly reworded to better reflect the broader and more diverse rhythms of gender identity development – modifying the original reference to fluctuations ‘from one day to the next’ to a more inclusive formulation that captures variability over longer or less linear timelines.
To evaluate the clarity and acceptability of the final version, the translated GPSQ was pilot-tested on a small group of five TGD individuals from the target population. Participants were asked to comment on readability, item comprehension and perceived sensitivity of the language. Minor wording adjustments were made based on their feedback to improve comprehensibility and inclusivity.
The final Italian GPSQ was then administered as part of the broader study protocol. See supplementary material available at https://doi.org/10.1192/bjo.2025.10879 for the full Italian version and Table 1 for a comparison of the original and translated items.
Table 1 Original and translated versions of the Gender Preoccupation and Stability Questionnaire (GPSQ)
Measures
The GPSQ is a brief self-report instrument composed of 14 items, designed to assess information regarding preoccupation with gender identity and related cognitive-emotional content, as well as perceived stability or fluidity of gender identity over time. Reference Hakeem, Črnčec, Asghari-Fard, Harte and Eapen12 Responses are rated on a Likert scale (e.g. from 1 = strongly disagree to 5 = strongly agree), with higher scores indicating greater preoccupation and severity about gender identity. The original version demonstrated good psychometric properties and has been used in both clinical and research contexts focusing on gender-diverse populations.
The Symptom Checklist-58 (SCL-58) is a shortened version of the Symptom Checklist-90-R and is used to assess a broad range of psychological symptoms and distress. Reference Derogatis, Lazarus and Maruish19 It includes 58 items rated on a 5-point Likert scale (0 = not at all to 4 = extremely), covering multiple dimensions such as anxiety, depression, somatisation and interpersonal sensitivity. Higher scores reflect greater psychological symptom severity. The SCL-58 has been validated in the Italian context and widely used in both clinical and non-clinical populations. Internal consistency of the SCL-58 in the present sample was excellent for the total score (Cronbach’s α = 0.97), with subscale alphas ranging from good to excellent (α = 0.76 to 0.90).
The Body Uneasiness Test (BUT) is a validated self-report measure assessing body image-related concerns and uneasiness. Reference Cuzzolaro, Cuzzolaro, Vetrone, Marano and Garfinkel20 It includes two parts: BUT-A (34 items) measuring general body uneasiness across dimensions such as weight phobia, body image concerns, avoidance, compulsive self-monitoring and depersonalisation; and BUT-B (37 items) assessing specific concerns about body parts and functions. Items are rated on a 6-point Likert scale, with higher scores indicating greater body image disturbance. The BUT is widely used in the assessment of body image issues in diverse clinical populations, including those with eating disorders, body dysmorphic disorder and gender dysphoria. Reference Castellini, Rossi, Cassioli, Sanfilippo, Ristori and Vignozzi21 As expected, the internal consistency of the total BUT-A score was excellent (α = 0.94), while subscale reliability values remained within acceptable to good ranges (α = 0.75–0.89), in line with prior studies on the Italian version.
Data analysis
Descriptive statistics were computed for sociodemographic variables and all questionnaire scores. In addition, descriptive item-level statistics (mean, s.d. skewness and kurtosis) were calculated for each GPSQ item to examine distributional properties and assess the presence of floor or ceiling effects. This step was used to verify the appropriateness of the items for factor analysis and to guide the choice of extraction method.
Internal consistency of the Italian version of the GPSQ was assessed using Cronbach’s α and McDonald’s ω. The factor structure was first evaluated using exploratory factor analysis (EFA) with oblimin rotation to explore a multifactorial structure. Sampling adequacy was examined using the Kaiser–Meyer–Olkin (KMO) index and Bartlett’s test of sphericity. With 151 participants and 14 items, the participant-to-item ratio was 10.8:1, exceeding the commonly recommended thresholds for EFA (5–10:1). Reference Kline22 Beyond sample size, adequacy also depends on the quality of the items, strength of factor loadings and factor overdetermination. Reference Costello and Osborne23 Evaluation of EFA model-fit was carried out at three levels: (a) sampling adequacy (KMO, Bartlett’s test), (b) factor retention (eigenvalues >1, scree plot inspection and parallel analysis) and (c) model adequacy (item communalities, salient loadings ≥0.40 with minimal cross-loadings, variance explained and global fit indices: root mean square error of approximation (RMSEA), Comparative Fit Index (CFI), Tucker–Lewis Index (TLI) and standard root mean squared residual (SRMR).
In the present study, all items showed salient loadings ≥0.40, communalities above 0.40, minimal cross-loadings and a conceptually coherent structure that accounted for 53.4% of total variance. Model-level fit indices were also extracted (RMSEA, CFI, TLI and SRMR), all of which supported the four-factor solution. These features are consistent with conditions under which samples in the 100–200 range yield reliable factor recovery. Reference Mundfrom, Shaw and Ke24–Reference de Winter, Dodou and Wieringa26 Test–retest reliability was assessed in a subsample of participants who completed the GPSQ at two time points, using both Pearson’s correlation coefficient and the intraclass correlation coefficient (ICC) based on a two-way mixed-effects model with absolute agreement.
To assess convergent validity, Pearson’s correlations were computed between the GPSQ scores and theoretically related subscales of the SCL-58 (e.g. depression, anxiety, interpersonal sensitivity) and the BUT (e.g. body image concerns, depersonalisation, avoidance). Additionally, known-groups validity was examined using an independent samples t-test to compare the GPSQ scores between participants undergoing gender-affirming hormonal therapy and those who were not. A multiple regression analysis was also conducted to explore which dimensions of psychological distress and body image disturbance predicted GPSQ scores.
All statistical analyses were conducted using SPSS (version 25 for Windows) and JASP (version 0.18 for Windows; JASP, Unversity of Amsterdam, The Netherlands; jasp-stats.org). Statistical significance was set at p < 0.05 for all tests.
Results
Sample characteristics
The final sample consisted of 151 transgender and gender-diverse (TGD) adults who accessed a specialised gender clinic in Northern Italy between 2022 and 2023. Participants were aged between 18 and 54 years (mean 26.9, s.d. = 6.8); 63.6% identified as trans men, 27.2% as trans women and 9.2% as non-binary or gender-fluid. At the time of assessment, 52.3% were undergoing Gender-Affirming Hormone Therapy (GATH), while the remaining 47.7% had not yet initiated GATH. Educational levels ranged from lower secondary school to postgraduate degrees, with most participants (67.5%) reporting a high school diploma or higher. The majority were Italian-born (92.1%), and 17.8% were unemployed or in precarious employment. All participants self-identified within the TGD spectrum and reported Italian as their first language.
Descriptive statistics for psychological and body image-related distress indicated moderate levels of symptomatology across the sample. The mean scores for the SCL-58 subscales were as follows: somatisation (mean 1.47, s.d. = 0.90, [0.00–3.00]), obsessive–compulsive (mean 1.56, s.d. = 0.99, [0.00–3.25]), interpersonal sensitivity (mean 1.48, s.d. = 0.90, [0.00–3.67]), depression (mean 1.61, s.d. = 0.92, [0.09–3.00]), anxiety (mean 1.58, s.d. = 0.88, [0.00–3.17]) and the Global Severity Index (mean 2.65, s.d. = 0.85, [0.47–4.32]). Scores on the BUT subscales were: weight phobia (mean 2.94, s.d. = 1.03, [0.50–5.00]), body image concerns (mean 3.35, s.d. = 1.01, [0.56–4.89]), avoidance (mean 2.21, s.d. = 1.21, [0.00–5.00]), compulsive self-monitoring (mean 1.89, s.d. = 1.11, [0.20–4.60]) and depersonalisation (mean 2.27, s.d. = 1.06, [0.00–4.33]).
Exploratory factor analyses
Descriptive statistics for the 14 items of the GPSQ are reported in Table 2. Mean scores ranged from 2.13 to 3.60 (on a 5-point Likert scale), with a s.d. between 1.07 and 1.41, suggesting a moderate spread of responses. Skewness values ranged from −0.56 to 0.85 and kurtosis ranged from −1.03 to 0.21, indicating that item distributions were approximately symmetric and mesokurtic. Floor and ceiling effects were generally limited; the highest floor effect was 32.5% (Item 12), and the highest ceiling effect was 25.2% (Item 9), with no item exceeding thresholds typically associated with problematic response distribution (>50%). These findings support the suitability of the items for EFA using continuous estimators.
Table 2 Descriptive statistics and distribution characteristics of Gender Preoccupation and Stability Questionnaire (GPSQ) items (N = 151)
GPSQ, Gender Preoccupation and Stability Questionnaire. Mean, skewness and kurtosis are reported for each item. Floor and ceiling effects reflect the percentage of participants selecting the lowest (1) and highest (5) response options, respectively. No item showed extreme skewness (>|2|) or kurtosis (>|7|), suggesting acceptable distributional properties for factor analysis.
In line with the original GPSQ validation, we first tested a one-factor model using principal axis factoring without rotation. This solution accounted for only 25.5% of the total variance. Although the Kaiser–Meyer–Olkin (KMO) index indicated adequate sampling adequacy (KMO 0.72), the low variance explained and weak communalities suggested that a single factor did not capture the construct’s complexity.
We therefore proceeded with EFA to investigate the dimensionality of the Italian GPSQ. Principal axis factoring with oblimin rotation supported a four-factor solution, which explained 53.4% of the total variance. Sampling adequacy remained acceptable (KMO 0.72), and Bartlett’s test of sphericity was significant (χ 2(91) = 826.31, p < 0.001). The four-factor solution was conceptually coherent and showed minimal cross-loadings.
Factor retention was based on the scree plot, eigenvalues >1 and interpretability. The rotated factor loadings are presented in Table 3. All items loaded saliently (≥0.40) on a single factor with minimal cross-loadings. Factor 1 (‘Gender Identity Instability’) reflected fluctuations and inconsistency in gender identity. Factor 2 (‘Cognitive-Affective Salience’) captured the centrality of gender-related thoughts and their interference with daily life. Factor 3 (‘Preoccupation’) reflected intrusive and repetitive gender-related thoughts. Factor 4 (‘Distress and Intervention-Oriented Reflection’) encompassed discomfort with gender and considerations of gender-affirming interventions. Internal consistency was acceptable for the full scale (Cronbach’s α = 0.78; McDonald’s ω = 0.73), with factor-level reliability ranging from adequate to excellent (see Table 3).
Table 3 Exploratory factor analysis rotated factor loadings for the Italian version of the Gender Preoccupation and Stability Questionnaire items (N = 151)
Loadings ≥0.40 are considered salient and reported. Extraction method: Principal Axis Factoring. Rotation: oblimin with Kaiser normalisation. Uniqueness represents 1 – communality. Cronbach’s α and McDonald’s ω indicate internal consistency for each factor. Factors were labelled as follows: Factor 1 – Gender Identity Instability; Factor 2 – Preoccupation; Factor 3 – Cognitive-Affective Salience; Factor 4 – Distress and Intervention-Oriented Reflection.
To evaluate the optimal factor structure, we compared alternative EFA models. The three-factor solution explained 55.3% of the total variance, whereas the four-factor solution accounted for 64.6%. Parallel analysis also supported retention of the four factors. Model-level fit indices further confirmed that the four-factor model provided superior fit (RMSEA 0.06; CFI 0.95; TLI 0.93; SRMR 0.04) compared to the three-factor solution (RMSEA 0.09; CFI 0.88; TLI 0.85; SRMR 0.07). Beyond statistical indices, the four-factor model yielded a more conceptually coherent structure, with all items showing salient loadings ≥0.40, communalities >0.40 and minimal cross-loadings.
Test–retest reliability
Test–retest reliability was assessed in a subsample who completed the GPSQ twice over a two-week interval. The total score showed excellent stability (Pearson’s r = 0.98, p < 0.001, 95% CI [0.97, 0.99]). Similarly, the intraclass correlation coefficient (ICC [1,1], two-way mixed-effects, absolute agreement) was 0.98 (95% CI [0.98, 0.99]), confirming the high temporal consistency of the measure.
Convergent validity and group differences
To evaluate convergent validity, GPSQ total and factor scores were correlated with measures of psychological distress (SCL-58) and body image concerns (BUT). As expected, the GPSQ total score was positively associated with SCL-58 subscales, particularly obsessive–compulsive (r = 0.37, p < 0.001), interpersonal sensitivity (r = 0.37, p < 0.001) and depression (r = 0.35, p < 0.001). Similar correlations emerged with BUT subscales such as body image concerns (r = 0.26, p = 0.001) and avoidance (r = 0.27, p < 0.001). These associations remained robust after controlling for age and sex at birth (partial r range = 0.18–0.36). At the factor level, Preoccupation correlated most strongly with obsessive–compulsive symptoms (r = 0.42), while Cognitive-Affective Salience showed stronger links with body image concerns (r = 0.37) and depersonalisation (r = 0.37). Gender Identity Instability correlated mainly with depression (r = 0.38), whereas Distress and Intervention-Oriented Reflection was associated with avoidance (r = 0.32). These differentiated patterns support the multidimensional validity of the scale. These differentiated patterns highlight the clinical relevance of assessing gender-related cognitive-affective experiences in a nuanced manner, and they support the validity of the four-factor solution. Moreover, the intercorrelations among the four GPSQ factors ranged from r = 0.33 to r = 0.63 (all p < 0.001), supporting their conceptual distinctiveness while confirming they assess related aspects of gender-related cognitive and emotional experiences. See Fig. 1 for complete correlation coefficients.
Fig. 1 Pearson correlation matrix displaying associations between Gender Preoccupation and Stability Questionnaire (GPSQ) subscales, psychological distress (SCL-58) and body image concerns (BUT). Colour intensity represents the strength of the correlations. *p < 0.05; **p < 0.01; ***p < 0.001. SCL-58, Symptoms Checklist-58; BUT, Body Uneasiness Test; SOM, somatisation; OCB, obsessive–compulsive behaviours; IPS, interpersonal sensitivity; D, depression; A, anxiety; BUT WP, weight phobia; BUT BIC, body image concerns; BUT A, avoidance; BUT CSM, compulsive self-monitoring; BUT D, depersonalisation.
To examine whether the observed associations were influenced by demographic characteristics, we computed partial correlations controlling for age and sex at birth. The results showed that the pattern of associations remained stable, with only minimal attenuation of effect sizes. For instance, the GPSQ total score remained significantly correlated with depression (partial r = 0.29, p < 0.01), interpersonal sensitivity (partial r = 0.34, p < 0.001) and with body image concerns (partial r = 0.27, p < 0.01). Across all significant associations, partial correlation coefficients ranged from 0.18 to 0.36, closely mirroring the zero-order correlations.
Independent sample t-tests showed that participants not receiving gender-affirming hormone therapy reported significantly higher scores on Gender Identity Instability (t(149) = 2.17, p = 0.031, d = 0.36), Preoccupation (t(149) = 2.29, p = 0.023, d = 0.40) and Distress/Intervention-Oriented Reflection (t(149) = 2.22, p = 0.028, d = 0.38), with a trend for Cognitive-Affective Salience (t(149) = 1.89, p = 0.061, d = 0.32).
Predictive validity
Multiple regression analyses indicated distinct psychological predictors for the four GPSQ factors. Gender Identity Instability was predicted by interpersonal sensitivity (β = 0.25, p = 0.018; model: F(8,142) = 2.75, p = 0.008, adj. R 2 = 0.08). Preoccupation was associated with obsessive–compulsive symptoms (β = 0.24, p = 0.022) and depression (β = 0.21, p = 0.041; F(8,142) = 3.16, p = 0.002, adj. R 2 = 0.10). Cognitive-Affective Salience was predicted by depression (β = 0.25, p = 0.015; F(8,142) = 3.46, p = 0.001, adj. R 2 = 0.11). Distress/Intervention-Oriented Reflection was predicted by anxiety (β = 0.23, p = 0.037; F(8,142) = 2.37, p = 0.020, adj. R 2 = 0.07).
Discussion
This study aimed to validate the Italian version of the GPSQ in a clinical sample of TGD individuals. Employing a comprehensive psychometric approach – including EFA, internal consistency evaluation, test–retest reliability and convergent validation – we sought to determine whether the GPSQ adequately captures the nuanced cognitive and emotional experiences related to gender in the Italian context.
Our findings challenge the unidimensional structure proposed in the original version of the GPSQ. The one-factor solution explained a limited proportion of the variance and showed poor fit in confirmatory analysis, indicating that a singular latent construct is insufficient to capture the complexity of gender preoccupation and stability among Italian-speaking TGD individuals. This is perhaps unsurprising given that gender-related distress is shaped by a multilayered interplay of psychological, social and cultural forces, particularly in a country like Italy where sociopolitical narratives about gender diversity remain contested. Reference Bariola, Lyons, Leonard, Pitts, Badcock and Couch27,Reference Brennan, Irwin, Drincic, Amoura, Randall and Smith-Sallans28 Instead, the EFA revealed a four-factor solution encompassing identity instability, cognitive-affective salience, preoccupation and distress linked to intervention-oriented reflection. Together, these findings point to a multidimensional conceptualisation of gender-related distress, aligning with clinical literature that frames gender preoccupation not as a single symptom but as a constellation of experiences involving cognitive rumination, emotional strain, interpersonal difficulty and somatic incongruence. Reference Briggs, Hayes and Changaris29–Reference Bowman, Hakeem, Demant, McAloon and Wootton31
The reliability of the GPSQ was supported by internal consistency estimates that were acceptable across the full scale and individual factors. Most strikingly, the instrument demonstrated exceptional test–retest reliability, with both Pearson’s r and the intraclass correlation coefficient indicating near-perfect stability over a two-week period. These findings suggest that the GPSQ taps into relatively enduring psychological processes, making it a robust tool not only for cross-sectional assessment but also for longitudinal research and clinical monitoring.
Convergent validity analyses further strengthened the instrument’s credibility. GPSQ scores were significantly associated with psychological distress, particularly obsessive–compulsive symptoms, interpersonal sensitivity, depression and anxiety – dimensions frequently implicated in gender dysphoria and related mental health concerns. Reference Inderbinen, Schaefer, Schneeberger, Gaab and Garcia Nuñez32,Reference Bouman, Claes, Brewin, Crawford, Millet and Fernandez-Aranda33 Associations were also observed with body image concerns, depersonalisation and avoidance, consistent with existing research suggesting that body-related discomfort often coexists with gender identity distress, particularly in pre-GATH or non-affirmed phases. Reference Meneguzzo, Zuccaretti, Tenconi and Favaro34 Importantly, participants who had not initiated GATH reported significantly higher scores on several GPSQ dimensions, especially in Preoccupation and Gender Identity Instability, confirming the scale’s sensitivity to known-group differences and reinforcing the role of gender-affirming medical care in mitigating psychological distress.
Regression analyses of the four GPSQ dimensions revealed nuanced relationships between specific psychological variables and different aspects of gender-related distress. Interpersonal sensitivity emerged as a significant predictor of Gender Identity Instability, while obsessive–compulsive symptoms and depression were uniquely associated with Preoccupation. Depression also predicted Cognitive-Affective Salience, and anxiety significantly contributed to Distress and Intervention-Oriented Reflection. These differentiated patterns emphasise that the multidimensional structure of the GPSQ captures distinct facets of the gender-related experience, each linked to specific psychological profiles. The fact that different psychological variables independently predicted different GPSQ dimensions underscores the utility of a multidimensional assessment. For instance, the strong association between interpersonal sensitivity and gender identity instability highlights the social embeddedness of gender-related suffering – suggesting that these experiences are not merely intrapsychic but are also shaped by relational dynamics such as perceived judgment, social exclusion or identity invalidation. Reference MacCallum, Widaman, Zhang and Hong25 These findings contribute to a growing body of evidence supporting a more complex, socially and emotionally nuanced understanding of gender-related distress.
These findings carry important implications for both clinical practice and research. Clinicians working with TGD populations may benefit from using the GPSQ not only as a screening tool but as a guide to explore different facets of gender-related distress – identity fluctuation, ruminative thinking, emotional burden and readiness for medical gender-affirming care. Elevated scores on specific factors could inform personalised intervention strategies, such as enhancing emotion regulation, addressing cognitive rigidity or developing coping skills for social invalidation. Moreover, the scale’s stability and multidimensionality render it suitable for evaluating treatment outcomes and tracking change over time, particularly in the context of gender-affirming interventions.
Limits
Nonetheless, several limitations must be acknowledged. A key limitation is that participants were recruited exclusively from a specialised gender clinic, potentially reducing the representativeness of our findings for the broader TGD population in Italy, particularly for individuals who may not have access to or seek specialised gender-affirming care. Future studies employing community-based samples are needed to confirm the generalisability and applicability of the Italian GPSQ beyond clinical contexts. While the inclusion of a broad spectrum of gender identities strengthens ecological validity, the sample size precluded subgroup analyses, which could have revealed differences in preoccupation and instability between binary and non-binary individuals.
Future studies should replicate the CFA in an independent sample. Additionally, although convergent validity was well supported, future studies should evaluate discriminant validity by including constructs less proximally related to gender identity, such as personality traits or physical health measures.
In conclusion, the Italian version of the GPSQ emerges as a reliable and valid tool for assessing gender-related preoccupation and identity instability in TGD individuals. Departing from a unidimensional model, our findings illuminate a richer, multifaceted structure that reflects the layered and often conflicting nature of gender experience. The scale offers valuable insight into the emotional, cognitive and social components of gender identity-related distress, contributing to a more nuanced and empathetic approach in both clinical care and gender-affirmative research.
Supplementary material
The supplementary material is available online at https://doi.org/10.1192/bjo.2025.10879
Data availability
The datasets used during the current study are available from the corresponding author on reasonable request due to ethical restrictions imposed by the Ethics Committee to protect participant confidentiality.
Author contributions
P.M., M.M., A.G., D.D.B., E.T. and A.F. designed the study and wrote the protocol. P.M., A.S. and M.B. collected the data. P.M. undertook the statistical analysis. P.M. wrote the first draft of the manuscript. All authors contributed to the manuscript’s revision and approved the final version for submission.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.
Declaration of interest
None.
Ethical standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2013. All procedures involving human subjects/patients were approved by the local ethics committee for clinical studies, the Comitato Etico Territoriale (CET) Area Centro-Est Veneto (approval number 6011/AO/24). All individuals included in this study provided informed consent.
Open access
eLetters
No eLetters have been published for this article.