Turkish Digital Stressors Scale Validated for Healthcare Professionals in 605-Person Study

As hospitals digitize everything from patient records to clinical decision-making, the pressures that digital tools place on the people who use them have become harder to ignore. A new study published in BMC Digital Health reports that a Turkish-language version of the Digital Stressors Scale, a questionnaire designed to capture the multidimensional nature of digital workplace stress, performs well among healthcare professionals in Türkiye, offering a locally validated instrument for a population that spends its working hours navigating electronic health records, remote monitoring systems, and constant digital communication.

The study, led by Kübra Karabulut Meral, Emel Filiz, and Ramazan Kıraç, was a cross-sectional methodological investigation conducted in Konya, a province in Central Anatolia. The researchers approached 682 healthcare professionals working in three public and four private hospitals between June and July 2025, administering the questionnaire face to face after obtaining written informed consent. After excluding questionnaires with missing, incomplete, or inconsistent responses, data from 605 participants were retained for analysis. The sample included nurses, physicians, health technicians and technologists, and administrative staff, with 53.4 percent of participants female and a mean age of 35.02 years. Professional experience ranged from one to 35 years, with a mean of 7.13 years.

The Digital Stressors Scale, originally developed by Fischer, Reuter, and Mander, was derived from an initial pool of 138 items and consists of 50 items grouped into ten subdimensions, covering complexity, role conflicts, job insecurity, privacy invasion, work overload, security concerns, social environment pressure, lack of usability, insufficient technical support, and distrust. All items are rated on a seven-point Likert scale. In the original validation study, internal consistency coefficients for the subdimensions ranged from 0.84 to 0.92, and the scale showed strong predictive validity for outcomes such as job satisfaction, burnout, and user satisfaction. Previous adaptations into Arabic and Norwegian largely retained the original factor structure, with internal consistency coefficients between 0.81 and 0.91.

Adapting the scale into Turkish followed a multistage procedure. Five academics independently translated the English version forward, after which a committee of four experts reviewed the draft for linguistic clarity and cultural appropriateness. Three independent back-translations were then compared with the original to verify conceptual equivalence, and a panel of 11 experts rated each item on a four-point relevance scale, yielding a scale-level content validity index of 0.86. A pilot study with 40 healthcare professionals assessed the clarity and applicability of the items before the main data collection began.

The statistical results showed that the original ten-factor structure was largely preserved but not untouched. Exploratory factor analysis, conducted with principal component extraction and Direct Oblimin rotation to account for correlations among factors, extracted ten factors explaining 64.01 percent of the total variance. The Kaiser–Meyer–Olkin measure of sampling adequacy was 0.907 and Bartlett’s test of sphericity was significant, indicating the data were highly suitable for factor analysis. Nine items were removed in three sequential stages because of substantial cross-loadings, unstable factor placement, or inconsistency with their original subdimensions, reducing the scale from 50 to 41 items. No original subdimension was eliminated entirely, although the removal of two items each from the insecurity, social environment, and technical support subdimensions narrowed the coverage of some peripheral aspects of those constructs.

Confirmatory factor analysis of the restructured 41-item, ten-factor model produced an acceptable overall fit profile: a chi-square to degrees of freedom ratio of 3.04, a root mean square error of approximation of 0.058, and a standardized root mean square residual of 0.051. However, several incremental fit indices, including a comparative fit index of 0.869 and a Tucker–Lewis index of 0.853, fell below commonly recommended thresholds. The authors attribute this to the complexity of a ten-factor model with 41 items and note that the confirmatory findings should be regarded as preliminary structural evidence rather than independent confirmation, because the exploratory and confirmatory analyses were conducted on the same sample.

Convergent and discriminant validity were assessed using the Fornell–Larcker approach. Discriminant validity was supported across all ten factors: for every factor, the square root of the average variance extracted exceeded its correlations with the other factors, and average variance extracted values were higher than the corresponding maximum shared variance and average shared variance values. Convergent validity, however, showed a mixed picture. Composite reliability coefficients were within acceptable ranges, but the average variance extracted values for five factors remained below the recommended threshold of 0.50, indicating limited convergent validity for those dimensions. The authors caution that convergent validity cannot be considered fully established for the affected subdimensions and recommend reassessment in independent samples.

Reliability results were strong. Cronbach’s alpha for the total scale reached 0.938, with subscale coefficients ranging from 0.712 to 0.852, indicating acceptable to high internal consistency across all ten dimensions. Split-half reliability provided additional support, with correlation between the two halves at 0.755 and both the Spearman–Brown and Guttman coefficients at 0.861. Item-level analyses also supported the instrument’s discriminative capacity: mean scores of the upper 27 percent group were markedly higher than those of the lower 27 percent group for all items, and corrected item–total correlations were generally within acceptable ranges. A Hotelling’s T2 test yielded a significant result, indicating that participants differentiated among the item statements, though the authors note this does not directly demonstrate the presence or absence of response bias.

The study’s authors frame the resulting instrument as a restructured Turkish version grounded in the original Digital Stressors Scale framework rather than a definitively validated shortened form. They emphasize that the high total-scale alpha should be interpreted alongside the large number of items, since alpha can increase with scale length and does not by itself demonstrate unidimensionality. The scope of the study was restricted to structural validation; relationships between digital stress and constructs such as burnout, job satisfaction, turnover intention, and psychological well-being were not examined. As with most questionnaire-based research, response styles, social desirability, and common method variance may have influenced responses to some extent.

For healthcare organizations confronting the psychosocial consequences of digital transformation, the availability of a locally validated measurement tool may prove useful. Growing evidence in the literature associates digital and technostress with burnout, emotional exhaustion, reduced job satisfaction, and impaired well-being among healthcare professionals, and the restructured Turkish scale may support future investigations of these relationships and help organizations identify prominent digital stressors. The authors recommend that future research test the 41-item version in independent and more diverse samples, examine measurement invariance across occupational and demographic groups, and reassess the convergent validity of the affected subdimensions before the instrument is used to inform organizational or individual decisions.

Subject of Research: Cross-cultural adaptation and psychometric validation of the Turkish Digital Stressors Scale for healthcare professionals

Article Title: Psychometric validity and reliability of the Turkish version of the Digital Stressors Scale: a sample of healthcare professionals

Article References: Karabulut Meral, K., Filiz, E., & Kıraç, R. (2026). Psychometric validity and reliability of the Turkish version of the Digital Stressors Scale: a sample of healthcare professionals. BMC Digital Health, 4(1), Article 76. https://doi.org/10.1186/s44247-026-00307-3

Image Credits: AI Generated

DOI: 10.1186/s44247-026-00307-3

Keywords: digital stress, Digital Stressors Scale, psychometric validation, healthcare professionals, Turkish adaptation, factor analysis, reliability, cross-cultural adaptation, technostress, occupational stress, scale validation, Türkiye