The scale of short-video consumption in China provides the backdrop. By June 2024, the country counted 1.05 billion short-video users, representing 95.5 percent of its internet population, according to the China Internet Network Information Center. For most people, these platforms are harmless entertainment or a convenient way to fill idle minutes. The concern, researchers emphasize, arises when use becomes dysregulated—marked by salience, withdrawal, tolerance, failed attempts to cut back, escapism, deception, loss of interest in other activities, functional impairment, and continued use despite harm. Previous work has tied such patterns to poorer sleep quality, procrastination, and depressive symptoms, but nurses occupy a particularly consequential position: their attention and energy directly affect patient safety.

The theoretical engine behind the study is Conservation of Resources theory, a foundational framework in occupational psychology proposed by Stevan Hobfoll in 1989. The theory holds that people strive to obtain and protect valued resources—time, energy, emotional stability—and that stress and impairment follow when those resources are threatened or inadequately replenished. Nursing, with its shift schedules, night work, emotional labor, and heavy clinical responsibility, is a context of sustained resource expenditure that demands effective off-job recovery. The researchers proposed a specific mechanism: a recovery opportunity cost. Problematic short-video use during nonwork time may displace sleep, relaxation, and psychological detachment—the very activities needed to restore resources spent at work—leaving nurses with less capacity to contain work demands within the work domain.

To test this, the team recruited nurses from three tertiary hospitals in Wuhan between October and December 2025, using convenience sampling and an online questionnaire distributed through WeChat. After screening out duplicates, logically inconsistent responses, and invariant answering patterns, 680 valid questionnaires remained. Participants completed the nine-item Problematic Short-Video Use Scale, which assesses the nine symptoms of dysregulated use on a six-point scale; the six-item Stanford Presenteeism Scale, which measures the tendency to attend work despite conditions warranting rest; and a five-item subscale measuring work interference with personal life—the unidirectional intrusion of work demands into personal time. All three instruments showed strong internal consistency in this sample, with Cronbach’s alpha values of 0.907, 0.965, and 0.943 respectively.

Rather than relying solely on total scores, which describe overall severity but can obscure how symptoms combine, the researchers used latent profile analysis—a person-centred statistical technique that identifies subgroups sharing similar response patterns. They estimated models with one to five classes and evaluated fit using information criteria, bootstrap likelihood ratio tests, entropy, class sizes, and interpretability. The four-class solution emerged as the best balance: it showed significantly better fit than the three-class model, achieved high entropy of 0.904 indicating clear class separation, and no class contained fewer than 5 percent of participants. A five-class model fit better on information criteria but failed the likelihood ratio test and showed reduced entropy.

The four profiles revealed striking heterogeneity. Just over half the nurses (50.9 percent) fell into a low-symptom profile with consistently minimal scores. The largest at-risk group, 30.3 percent, showed a moderate escapism–tolerance pattern—elevated tolerance, failed attempts to reduce use, and escapism, but without generalized impairment. A smaller group of 9.0 percent displayed a high escapism–persistent-use profile, marked by pronounced escapism, tolerance, failed control, and continued use despite harm, yet comparatively low deception and functional impairment. Finally, 9.8 percent showed a generalized high-symptom profile with elevated scores across nearly all nine indicators. Notably, 38.4 percent of the full sample met the screening threshold of 24 or higher on the nine-item scale, a cutoff the original validation study reported with an area under the curve of 0.990.

The correlations among the three core variables told a coherent story. Problematic short-video use was positively associated with work interference with personal life (r = 0.305) and with presenteeism (r = 0.370), while work interference with personal life showed a strong correlation with presenteeism (r = 0.668). In hierarchical regression, sociodemographic and occupational characteristics explained 9.2 percent of the variance in presenteeism; adding work interference with personal life contributed another 38.2 percent, and profile membership added a further 3.2 percent, for a final model explaining 50.6 percent of the variance. Work interference with personal life carried the largest standardized coefficient (β = 0.601). Monthly income was the only demographic characteristic retained: nurses earning more than 5,000 renminbi per month reported lower presenteeism than those earning less, a pattern the authors suggest may reflect differences in financial pressure, though that mechanism was not directly measured.

The mediation analysis sharpened the picture. Using a path model with 5,000 bootstrap samples and bias-corrected confidence intervals, the researchers found that all three non-low-symptom profiles were indirectly associated with higher presenteeism through greater work interference with personal life. The indirect effects were 1.117 for the moderate profile, 3.004 for the high escapism–persistent-use profile, and 2.870 for the generalized high-symptom profile, all statistically significant. For the moderate and generalized profiles, direct associations with presenteeism also remained significant, hinting at additional unmeasured mechanisms such as sleep disruption or emotional exhaustion. For the high escapism–persistent-use profile, however, the direct association vanished once work interference was accounted for—its relationship with presenteeism appeared statistically concentrated in the recovery-displacement pathway. The model explained 49.5 percent of the variance in presenteeism overall.

The authors are careful about what these findings do and do not establish. Because the data are cross-sectional, the proposed sequence—problematic use displacing recovery, recovery deficits amplifying perceived work intrusion, and intrusion fueling presenteeism—is a theory-guided association rather than a confirmed causal chain. Reverse and cyclical dynamics remain entirely plausible: occupational stress or work–life interference may itself drive escapist viewing, potentially creating a resource-loss spiral in which depleted nurses retreat further into their screens. The profiles, moreover, represent relative response patterns within this particular sample of Wuhan tertiary-hospital nurses and should not be treated as validated diagnostic categories. Convenience sampling, self-reported data, the small number of male participants, and the absence of direct measures of sleep, psychological detachment, and exhaustion all temper the conclusions.

Even so, the practical implications are pointed. The authors suggest that universal education could help low-symptom nurses protect recovery time and maintain healthy digital boundaries, while support for those resembling the escapism-oriented profiles might focus on recognizing uncontrolled use, building alternative emotion-regulation strategies, and establishing device-free periods before sleep. Nurses with generalized high symptoms may warrant confidential assessment of sleep, exhaustion, and work functioning, with referral when impairment persists. Crucially, the researchers argue that individual strategies alone are insufficient when organizational demands remain high: predictable scheduling, adequate intershift rest, protected breaks, supervisor training, and accessible occupational-health services all matter. Restricting personal screen time, they caution, does nothing to address the occupational stress and inadequate recovery that make escapist viewing attractive in the first place. Future work, they say, should validate the four-profile solution in multi-centre longitudinal studies and incorporate objective measures of media use and recovery to test the mechanism directly.

Subject of Research: Problematic short-video use profiles and presenteeism among hospital nurses, mediated by work interference with personal life

Article Title: Latent Profiles of Problematic Short‐Video Use and Presenteeism Among Hospital Nurses: The Mediating Role of Work Interference With Personal Life

Article References: Liu, Z., Li, M., Liu, Y., Yang, P., Liu, W., & Hu, B. (2026). Latent Profiles of Problematic Short‐Video Use and Presenteeism Among Hospital Nurses: The Mediating Role of Work Interference With Personal Life. Nursing Open, 13(10), Article e70879. https://doi.org/10.1002/nop2.70879

Image Credits: AI Generated

DOI: 10.1002/nop2.70879

Keywords: problematic short-video use, presenteeism, nurses, work interference with personal life, latent profile analysis, Conservation of Resources theory, occupational health, recovery, work-life conflict, China, cross-sectional study, mediation analysis