New Scale Measures How Daily Time Rhythms Shape Nurses’ Mental Health

Nurses around the world work against the clock. Rotating shifts, night duty, and unpredictable patient demands fracture any semblance of a stable daily rhythm, and the psychological toll of that temporal chaos has long been measured almost exclusively through the lens of symptoms: stress scales, burnout inventories, depression checklists. A team of researchers in Hangzhou, China, argues that this symptom-first approach misses something fundamental about how nurses actually experience their lives. Their new study, published in BMC Nursing, reports the psychometric validation of the Seven-Time Dimensions Scale (STDS) in a large sample of Chinese nurses, offering evidence that a single, coherent measure of daily time experience quality could become a practical tool for monitoring and protecting the mental health of one of the most chronically time-pressured professions in medicine.

The Seven-Time Dimensions Scale is designed to assess daily life rhythms together with the psychosocial resources that people perceive around them. Rather than cataloguing pathology, it asks how well an individual’s experience of time is functioning: whether routines hold together, whether the day feels structured rather than fragmented, and whether the person has the psychological resources to inhabit that structure. The instrument’s developers positioned it as a response to the limitations of traditional symptom-focused tools, which capture distress only after it has accumulated. A time-experience measure, in principle, could flag deterioration earlier, because the erosion of daily rhythm often precedes the emergence of diagnosable symptoms. Until now, however, the scale had never been validated in Chinese nurses, a population whose work schedules make time experience unusually salient and unusually strained.

The validation study, led by Jingxi Xu of the School of Public Health and Nursing at Hangzhou Normal University, with Hong Luo of the Affiliated Mental Health Centre and Hangzhou Seventh People’s Hospital, Zhejiang University School of Medicine, and colleagues including Xiang Li and Ke’er Mou, recruited 1,301 nurses through convenience sampling. This substantial sample size gave the team the statistical power needed to run a rigorous battery of psychometric tests, the kind of analyses that determine whether a questionnaire actually measures what it claims to measure, consistently, across different kinds of respondents.

The team began with the basics of reliability. Internal consistency, assessed with Cronbach’s alpha, came in at 0.900, and split-half reliability reached 0.893, both figures comfortably above conventional thresholds for a well-functioning scale. Item-total correlation coefficients ranged from 0.764 to 0.890, with all values significant at p < 0.001, indicating that every item pulled strongly in the same direction as the scale overall. In practical terms, nurses who scored high on one aspect of time experience tended to score high across the board, which is exactly the pattern expected if the items all tap a common underlying quality.

Structure was the next question, and here the researchers deployed a split-sample strategy. The 1,301 participants were randomly divided into two independent subsamples: one of 650 nurses for exploratory factor analysis and one of 651 for confirmatory factor analysis. Exploratory factor analysis on the first subsample extracted a single factor accounting for 61.049 percent of the variance, a strikingly large share that suggests the scale behaves as one unified construct rather than a loose collection of separate facets. Confirmatory factor analysis on the second subsample then tested that unidimensional structure formally, and the fit statistics were excellent: a chi-square to degrees-of-freedom ratio of 2.214, an RMSEA of 0.043, a CFI of 0.993, a TLI of 0.990, and an SRMR of just 0.016. Values of CFI and TLI above 0.95 and RMSEA below 0.05 are generally considered hallmarks of a model that fits the data closely, and all of these benchmarks were met or exceeded.

Because modern psychometrics has grown rightly suspicious of easy unidimensionality claims, the team went further. They applied exploratory structural equation modeling (ESEM) on the confirmatory subsample, specifying a one-factor solution, and the results cross-validated the single-factor reading with nearly identical fit indices: chi-square to df of 2.214, CFI of 0.993, TLI of 0.990, RMSEA of 0.043, and SRMR of 0.014. They then ran a bifactor confirmatory factor analysis, a more stringent test that models both a general factor and potential group factors to determine whether the total score genuinely reflects a dominant common dimension. The bifactor model fit excellently, with chi-square to df of 1.800, CFI of 0.996, TLI of 0.989, RMSEA of 0.035, and SRMR of 0.011. Two derived statistics sealed the case. The hierarchical omega coefficient (omega-h) was 0.896, and the explained common variance (ECV) reached 0.953, meaning the general factor accounted for more than 95 percent of the covariation among items. When a general factor swallows that much of the shared variance, summing the items into a single total score is statistically justified, not merely convenient.

A questionnaire is only useful, however, if it means the same thing to different people, and this is where the multi-group confirmatory factor analyses (MG-CFA) came in. The team tested measurement invariance across demographic subgroups of nurses, and the results supported full scalar invariance for nearly every demographic variable examined, with changes in CFI no greater than 0.003 and changes in RMSEA no greater than 0.007, thresholds far inside the conventional tolerances for claiming invariance. The exceptions were sex and ethnicity, where full invariance did not hold across the full scale, a nuance that matters for anyone planning cross-group comparisons. For the remaining demographic splits, though, the finding means that a given STDS score carries the same psychological meaning whether the respondent is a young nurse or an older one, newly qualified or a veteran, so researchers can legitimately compare average scores between such groups without worrying that the instrument itself has shifted meaning.

Validity is ultimately a question of relationships, and the STDS passed its relational tests in both directions. Convergent and criterion-related validity were evaluated with Pearson correlations against four well-established instruments. STDS scores correlated positively with the WHO-5 Well-being Index (r = 0.650), the Perceived Social Support Scale (r = 0.492), and the 10-item Connor-Davidson Resilience Scale (r = 0.555), all at p < 0.001. Nurses whose daily time experience was of higher quality reported greater well-being, stronger perceived social support, and higher resilience. The relationship ran the other way, as predicted, with the Perceived Stress Scale: the correlation was negative at r = -0.526, also p < 0.001. Beyond simple correlation, hierarchical regression analyses showed that the STDS explained significant incremental variance in the criterion variables even after demographic covariates were taken into account, demonstrating that the scale adds explanatory power that demographic variables alone cannot supply.

For hospital administrators and occupational health researchers, the practical implications are considerable. Nurse burnout and attrition have become acute concerns for health systems worldwide, and the tools available to detect psychological risk in nursing staff have largely been borrowed from clinical populations or focused narrowly on stress and exhaustion. A validated measure of daily time experience quality offers a different entry point: rather than waiting for burnout symptoms to surface, health services could track how nurses’ temporal lives are holding up, potentially designing scheduling interventions, rest protocols, or peer-support structures that protect rhythm before distress crystallizes. The authors conclude that the STDS demonstrates acceptable reliability and validity as a preliminary measurement tool for assessing daily time experience quality associated with mental-health-related outcomes among Chinese nurses, and they propose that it can be used in future research and practice to develop mental health regulation strategies targeting this population.

Caveats remain, as they always do in psychometric validation. The study relied on convenience sampling within one country, and the authors themselves frame the STDS as a preliminary tool, with the validation reported here being a first step rather than a final verdict. The lack of full scalar invariance across sex and ethnicity signals that subgroup comparisons on those dimensions should be handled with care until further work clarifies the pattern. Nevertheless, the statistical case presented is unusually thorough for a scale-validation study: two independent subsamples, exploratory and confirmatory factor analyses, ESEM cross-validation, a bifactor model, multi-group invariance testing, and hierarchical regression all converging on the same conclusion. If the STDS travels well to other populations and settings, the humble act of asking people how their days hold together may become a frontline instrument in a field that has, for too long, measured the damage only after it was done. The study received no dedicated funding, and the authors declare no competing interests.

Subject of Research: Psychometric validation of a daily time experience quality scale for nurses' mental health

Article Title: Psychometric validation of the seven-time dimensions scale: a measure of daily time experience quality for nurses’ mental health

Article References: Xu, J., Luo, H., Li, X., & Mou, K. (2026). Psychometric validation of the seven-time dimensions scale: a measure of daily time experience quality for nurses’ mental health. BMC Nursing, 25(1), Article 928. https://doi.org/10.1186/s12912-026-05362-9

Image Credits: AI Generated

DOI: 10.1186/s12912-026-05362-9

Keywords: psychometrics, nurses, mental health, Seven-Time Dimensions Scale, scale validation, confirmatory factor analysis, bifactor model, measurement invariance, well-being, perceived stress, occupational health, China