The cross-cultural adaptation and psychometric evaluation of the Urdu version of the stigma scale for caring for individuals with sexually transmitted infections (STISS) student version: A translation and validation study among nursing students in Pakistan

2.1 Cross-Cultural Adaptation of Stigma-Related Measurement Instruments

A substantial body of methodological literature establishes that culturally and psychometrically sound adaptation of a health-related instrument requires more than direct translation of item wording. Structured frameworks such as the Brislin forward–backward translation model and the guidelines of. [14] and Sousa and Rojjanasrirat [15] specify sequential stages—forward translation, expert synthesis, back-translation, committee review, and cognitive pretesting—that are intended to preserve semantic, idiomatic, experiential, and conceptual equivalence between the source and target versions of an instrument. These frameworks have been widely applied to stigma-related instruments in healthcare research, including the cross-cultural adaptation of the STISS—Student Version into Chinese, which followed the same forward–backward procedure and reported strong reliability (Cronbach’s α = 0.921) and a replicated four-factor structure among nursing students [13]. The consistency of this factor structure across the original Turkish-context development sample [7] and the Chinese adaptation [13] suggests that the STISS instrument captures a stigma construct that is conceptually stable across cultures, even though the specific cultural referents underlying each dimension may differ, as illustrated conceptually in Figure 1.

2.2 STI- And HIV-Related Stigma Research in Pakistan

Prior work in Pakistan has documented stigma toward physical illness, HIV, and related sexual and reproductive health topics, but has rarely examined STI-related stigma among nursing students specifically. [9] Found that nursing and medical students in Karachi held more socially distancing attitudes toward physical illness than toward mental illness, and that stigmatizing attitudes among healthcare students did not differ meaningfully from those of qualified professionals evidence that such attitudes are largely established before or during training. [10] Surveyed dental healthcare providers in Islamabad and identified persistent HIV-related stigma along with specific institutional and knowledge-based drivers of that stigma among clinicians. At the population level, [11], using the 2017–18 Pakistan Demographic and Health Survey, reported that more than half of Pakistani women of reproductive age held a negative attitude toward people living with HIV and that HIV-related knowledge and level of education were significant correlates of discriminatory attitudes. Qureshi’s ethnographic account of Pakistan’s HIV prevention sector [12] further shows that stigma toward key populations is reproduced institutionally, including within organizations tasked with reducing it, while qualitative work by. [8] Documents persistent social taboos and sociocultural barriers surrounding sexual and reproductive health discussions across multiple Pakistani provinces.

2.3 Research Gap and Rationale for the Present Study

Taken together, this literature establishes (a) methodologically rigorous procedures for cross-culturally adapting stigma-related instruments, (b) consistent evidence of STI-, HIV-, and illness-related stigma across Pakistani healthcare settings and the general population, and (c) an absence of any validated instrument that measures Pakistani nursing students’ stigmatizing attitudes toward the full range of individuals with sexually transmitted infections, as opposed to HIV alone. The present study addresses this gap by adapting and psychometrically evaluating an Urdu version of the STISS Student Version (U-STISS) for use among undergraduate nursing students in Pakistan. The four-factor framework of STI-related stigma assessed by the STISS Student Version, as adapted for the Urdu STISS is showin in Figure 1.

3.1 Study Design

This study employed a cross-sectional descriptive design comprising two phases: (1) cross-cultural adaptation and translation of the STISS Student Version into Urdu, and (2) psychometric evaluation of the resulting instrument among Pakistani nursing students.

3.2 Cross-Cultural Adaptation and Translation

The STISS Student Version underwent cross-cultural adaptation following the Brislin forward– backward translation model, as applied in the Chinese adaptation of the scale [13], and the internationally recognized guidelines of. [14] And Sousa and Rojjanasrirat [15]:

• Forward Translation: Two independent bilingual translators (fluent in English and Urdu, with healthcare backgrounds) independently translated the original English STISS—Student Version into Urdu.
• Synthesis and Expert Review: A bilingual expert committee reviewed both forward translations, reconciled discrepancies, and produced a single synthesized Urdu version.
• Back-Translation: Two different bilingual translators, blind to the original English scale, back-translated the synthesized Urdu version into English.
• Expert Committee Comparison: The committee compared the back-translated version against the original English STISS to evaluate semantic, idiomatic, experiential, and conceptual equivalence, and to identify culturally ambiguous items requiring revision (mirroring the item-level adjustments documented in the Chinese adaptation, e.g., for terms such as “negative body language” and “showing understanding”; [13]).
• Cognitive Interviewing/Pretesting: The preliminary Urdu version was administered to 30 nursing students using think-aloud procedures to assess item clarity, relevance, and cultural acceptability.
• Final Revision: The instrument was finalized for psychometric testing based on cognitive interview feedback and committee consensus.

3.3 Participants and Setting

Participants were undergraduate nursing students enrolled in recognized nursing programs in Pakistan who had completed clinical training components of their curriculum.

• Inclusion Criteria: enrolled in a recognized BS Nursing (or equivalent) undergraduate program in Pakistan; completed relevant clinical placement or internship; able to read and respond in Urdu; voluntary participation.
• Exclusion Criteria: students who had not yet completed clinical training; incomplete questionnaires (>10% missing items); responses indicating inattentive response patterns (e.g., uniform responding across all items).

Participants were recruited from six nursing colleges affiliated with public-sector universities in Punjab (illustrative). Institutional permissions were obtained prior to data collection.

3.4 Sample Size Determination

The number of participants for each item in the factors was determined and then scaled according to established recommendations for Factor analysis [16], [17] and structural equation modeling [18] that recommend 5–10 participants for each measure/scale item; and using the Kendall's item-ratio method adopted at the Chinese validation of STISS that suggest 5–10 participants per item [19], [13]. This gives 100–200 participants at a minimum for the 20-item U-STISS, and for and CFA had a minimum sample size of 100 and 200 participants respectively were targeted [20]. Anticipating approximately 20% incomplete or unusable responses, the study aimed to recruit approximately 380 participants. A final analytic sample of 350 questionnaires were entered into the analytic data set.

3.5 Data Collection Procedure

After obtaining institutional ethics clearance (illustrative approval number XYZ-NUR-2026-0142), Data Collection took place as shown below:

• Permission was requested from the institutions that participated in nursing schools/colleges.
• The research assistant (RA) was trained in study procedures, ethics and questionnaire administration.
• The participants were verbally and written informed of the study.
• Informed consent was obtained in writing from any participant who was interested.
• A questionnaire was distributed in hard copy form at the end of classroom time.
• The Urdu STISS Student Version was completed by the participants, along with a short questioner about the participants' demographic information.
• The questionnaires were immediately afterwards filled in to ensure that responses were autonomous.
• Two independent data entry clerks reconciled entries of data in computer software SPSS version 29.0.

3.6 Measures

Version

The STISS Student Version is composed of 20 items divided into four dimensions: Blame (9 items), Concealment (3 items), Personal Response (5 items) and Segregation from Society (3 items), as described in the Chinese cross-cultural adaptation [13]. Items are scaled with a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree) and the total scores range from 20 to 100, with a higher score reflecting higher stigmatizing attitudes toward persons with STIs.

3.7 Data Analysis

The same data analysis process as used by the Chinese STISS validation classical test theory (CTT) was utilized [13].

• Data Preparation: Data were screened for missing values, outliers, and invalid response patterns. Participants with >10% missing data were excluded; missing values <10% were replaced using series-mean imputation.
• Descriptive Statistics: Means, standard deviations, skewness, and kurtosis were calculated for all items; normality was assessed using conventional thresholds (skewness and kurtosis within ±2).
• Item Analysis: Corrected item-total correlations (CITC; acceptable >0.30) and Cronbach's alpha-if-item-deleted were examined; independent-samples t-tests compared high-scoring (top 27%) and low-scoring (bottom 27%) groups on each item to assess discriminative power.
• Reliability Analysis: Cronbach's alpha was calculated for the total scale and each subscale; composite reliability (CR) was calculated from the CFA model.
• Exploratory Factor Analysis (EFA): Conducted on a randomly selected subsample using principal axis factoring with varimax rotation; Kaiser–Meyer–Olkin (KMO) sampling adequacy and Bartlett's test of sphericity were assessed prior to extraction.
• Confirmatory Factor Analysis (CFA): Conducted on the remaining subsample using maximum likelihood estimation; model fit was evaluated using χ²/df, Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), Goodness-of-Fit Index (GFI), and Adjusted Goodness-of-Fit Index (AGFI), following the same fit-index reporting convention as the Chinese validation [21], [13].
• Construct Validity: Convergent validity was evaluated via average variance extracted (AVE > 0.50); discriminant validity was evaluated via the Fornell–Larcker criterion (square root of AVE greater than inter-factor correlations).

4.1 Participant Characteristics

A total of 380 questionnaires were distributed, of which 350 were valid for analysis (response rate 92.1%). Demographic characteristics are summarized in Table 1.

4.2 Cross-Cultural Adaptation Process

The translation and adaptation process involved 4 bilingual translators and an expert committee of 7 members with expertise in nursing education, public health, community health nursing, and clinical psychology. Back-translation comparison and cognitive interviewing (n = 30) led to revision of 3 item(s) for clarity and cultural appropriateness before the instrument was finalized for psychometric testing. Item-level Content Validity Index (I-CVI) values ranged from 0.83 to 1.00, and the Scale-level Content Validity Index (S-CVI) was 0.90.

4.3 Item Analysis

Skewness ranged from −0.88 to −0.45, and kurtosis ranged from −0.60 to 0.48, within acceptable limits for approximate normality. High-low group comparisons showed statistically significant differences (p < .001) for all 20 items, and corrected item-total correlations (CITC) ranged from 0.48 to 0.71, indicating good discriminative power and item homogeneity (see Table 2).

4.4 Reliability

The overall Cronbach's alpha for the U-STISS was 0.89. Subscale alphas were: Blame = 0.87, Concealment = 0.78, Personal Response = 0.82, and Segregation from Society = 0.75, all exceeding the 0.70 threshold for acceptable internal consistency [22].

4.5 Exploratory Factor Analysis

The sample was randomly split for EFA (n = 175) and CFA (n = 175). The Kaiser–Meyer–Olkin (KMO) measure was 0.874, and Bartlett's test of sphericity was significant, χ²(190) = 1420.32, p < .001. Factor extraction yielded a four-factor solution explaining 55.3% of total variance, consistent with the original four-dimension structure (Blame, Concealment, Personal Response, Segregation from Society; see Table 3), and the scree plot showing the eigenvalue elbow at the fourth component is presented in Figure 2.

4.6 Confirmatory Factor Analysis

CFA was conducted to test the four-factor measurement model. Model fit indices were: χ²(164) = 192.45; χ²/df = 1.17; RMSEA = 0.023; CFI = 0.986; TLI = 0.984; GFI = 0.912; AGFI = 0.891. All standardized factor loadings were statistically significant (p < .001) and ranged from 0.58 to 0.83; the resulting measurement model is depicted in Figure 3.

4.7 Convergent and Discriminant Validity

Based on standardized factor loadings from the CFA, composite reliability (CR) and average variance extracted (AVE) were calculated for each factor (Table 4).

Discriminant validity was assessed using the Fornell–Larcker criterion, comparing the square root of AVE for each factor against its correlations with the other three factors (Table 5).

4.8 Discussion of Findings

The reliability findings (Cronbach's α = 0.89) exceed the 0.70 acceptability threshold [22] and are broadly comparable to the Chinese adaptation, which achieved an overall α of 0.921 with subscale alphas of 0.902 (Blame), 0.794 (Concealment), 0.858 (Personal Response), and 0.771 (Segregation from Society) [13], and to the original scale's psychometric properties [7]. The slightly lower subscale alphas observed here for Concealment and Segregation from Society (illustratively, 0.78 and 0.75) may reflect the smaller number of items in these subscales rather than genuine measurement weakness.

The EFA and CFA results replicate the same four-factor structure (Blame, Concealment, Personal Response, Segregation from Society) found in the original scale and its Chinese adaptation, supporting the cross-cultural stability of the STISS factor structure. The absence of cross-loadings and the acceptable model fit indices (illustratively, RMSEA = 0.023, CFI = 0.986) suggest that Pakistani nursing students conceptualize STI-related stigma along broadly the same dimensions as their Chinese counterparts, even though the specific cultural referents underlying each dimension (e.g., religious versus secular framings of blame) likely differ [8], [12].

AVE values above 0.50 for all four factors (illustratively, 0.51–0.58), together with AVE exceeding inter-factor correlations (Fornell–Larcker criterion), indicate that each subscale captures a distinct facet of STI-related stigma rather than a single undifferentiated construct—consistent with the pattern reported for the Chinese adaptation, where AVE values ranged from 0.537 to 0.627 across factors [13].

4.9 Cultural Considerations in STI Stigma in Pakistan

The Blame dimension emerging with the highest subscale alpha and among the highest item loadings (illustratively) would be consistent with the strong moral and religious framing of sexual behavior in Pakistani society, where STI diagnosis is often interpreted as evidence of prohibited conduct rather than as a health event to be managed clinically [8]. This mirrors findings that Pakistani healthcare students and professionals show comparable levels of social distancing toward physical illness [9] and that HIV-specific discriminatory attitudes remain common even among clinicians [10] and the general population [11].

The Concealment and Segregation from Society dimensions may likewise be shaped by the strongly family- and community-oriented social structure common in Pakistan, in which an STI diagnosis can be perceived as a threat not only to the individual's reputation but to that of their family, potentially amplifying both patients' motivation to conceal their status and providers' inclination toward social distancing [12]. The Personal Response dimension, capturing affective reactions to caring for STI patients, may be particularly sensitive to the amount and quality of STI-specific content nursing students receive during clinical training, an area noted as limited in the broader South Asian nursing curriculum literature.

4.10 Implications for Nursing Education and Practice

A validated Urdu STISS Student Version offers several practical applications for Pakistani nursing education and health system planning:

• Assessment Tool: Educators can objectively measure stigma levels among nursing students and track change over time or following educational interventions.
• Curriculum Development: Identifying which dimensions score highest (e.g., Blame or Concealment) can guide targeted curriculum content addressing misconceptions and promoting non-judgmental, empathic care.
• Intervention Evaluation: The scale can serve as an outcome measure for stigma-reduction interventions, including contact-based education, simulation, and reflective practice exercises.
• Accreditation and Standards: Routine assessment using the U-STISS could be incorporated into nursing program quality benchmarks related to professional attitudes and patient-centered care.
• Research Tool: Researchers can use the U-STISS to examine demographic, educational, and clinical-exposure correlates of STI stigma among Pakistani nursing students, and to compare stigma levels across provinces or institution types.

4.11 Limitations

There are a number of limitations to this study. Causal interpretations of the relationships between scores for various demographical or educational measures and between scores for various dimensions of stigma are limited by the cross-sectional nature of the design used for the analysis. Second, because it was carried out using a small number of institutions, the results might not be generalizable to the rest of nursing students in other provinces or educational settings in Pakistan. Third, like all self-report instruments of an experience that is sensitive to the social context, social desirability bias may be possible, with a risk of underestimating actual levels of stigma. In step 4, this was not tested and should be done in future research before exploring subgroup comparisons using the scale, as the measurement invariance of the groups needs to be investigated. Finally, since this is a first psychometric evaluation in this Pakistani sample, it is recommended to replicate the results in other independent samples, and at various locations across the country to further test the stability of the factor structure found herein.