Workload Volume and Quality Control Performance in Clinical Molecular Laboratories: Testing the Mediating Role of Staff Fatigue and the Moderating Role of Laboratory Automation

Abstract

Clinical molecular, PCR, and virology laboratories in Punjab, Pakistan have absorbed a sharp rise in diagnostic testing volume over the past several years, raising concern about whether staff can sustain quality control (QC) performance under mounting pressure. This study examined the relationship between workload volume and perceived QC performance among laboratory professionals, and tested whether staff fatigue transmits this relationship and whether laboratory automation buffers it. A cross-sectional survey collected 122 valid responses from technologists, supervisors, and laboratory managers working in molecular diagnostics, PCR/RT-PCR, and virology sections across public, private, hospital-based, and independent laboratories. Workload volume (9 items), staff fatigue (8 items), laboratory automation (11 items), and QC performance (14 items) were each measured on five-point scales and showed strong internal consistency (Cronbach's alpha ranging from .836 to .982). Contrary to the hypothesised direction, workload volume was positively associated with QC performance (r = .384, p < .001), and this positive direct effect persisted in multiple regression after fatigue and automation were entered (β = .206, p < .001). Workload volume was positively associated with fatigue (r = .318, p < .001), but fatigue was not significantly associated with QC performance once workload was controlled (p = .135), and the bootstrapped indirect effect of workload on QC performance through fatigue was not significant (effect = -.050, 95% CI [-.139, .010]). Laboratory automation did not significantly moderate the workload-fatigue pathway (interaction p = .986), and the index of moderated mediation was likewise non-significant (index = .0002, 95% CI [-.034, .033]). Laboratory automation instead emerged as the strongest direct predictor of QC performance in the model (β = .782, p < .001), and the full model explained 73.2% of the variance in QC performance. These findings do not support the hypothesised fatigue-mediation and automation-moderation pathways in this sample, but point instead to automation support as the variable most closely tied to perceived QC performance, with practical implications for laboratory quality management in resource-constrained settings.