Effectiveness of ultraviolet–C robots in isolation room disinfection: Impact of dose, distance, and direct exposure on environmental surface decontamination

Background To minimize environmental services staff’s exposure to pathogens, ultraviolet-C (UV-C) disinfection robots were applied before terminal manual environmental disinfection. This study aimed to evaluate the disinfection efficacy of the UV-C robot and to assess the effects of ultraviolet-C dose, distance, and direct exposure (line-of-sight irradiation) on disinfection outcomes. Methods This before-after study was conducted in a negative-pressure isolation ward comprising 17 single-patient rooms at a medical center in Taiwan between August 24, 2024, and January 3, 2025, yielding a total of 270 paired samples. An ultraviolet-C-emitting robot with a wavelength of 254 nm was sequentially positioned on both sides of the patient bed and operated for 5 minutes on each side before routine manual environmental disinfection. Total aerobic colony counts (ACC) and bacterial culture positivity rates were obtained from high-touch environmental surfaces (e.g., bed rails, call bells and bedside chairs) before and after UV-C disinfection. The effects of UV-C dose, distance between the robot and surfaces, and the presence of direct exposure (line-of-sight irradiation) versus indirect exposure on disinfection efficacy were analyzed using multivariate binary logistic regression, performed using SPSS version 27.0. The results of the regression analysis were expressed as odds ratios (ORs) with 95% confidence intervals (CIs). Results The median ACC and bacterial culture positivity rates on surfaces significantly decreased after UV-C disinfection (0.76 vs. 0.08 CFU/cm²; 83% vs. 57.4%, P < 0.001). A weak negative correlation was observed between ACC and UV-C doses after disinfection ( r s = −0.281, 95% CI: −0.390 to −0.164, P < 0.001), while no correlation ( r s = 0.040, 95% CI: −0.083 to 0.162, P = 0.510) and weak correlation ( r s = 0.317, 95% CI: 0.202 to 0.424, P < 0.001) were found between ACC at short and long irradiation distances. The ACC on environmental surfaces with direct UV-C exposure were significantly lower than those in shadowed areas [median: 0.00 (IQR: 0.00 to 0.44) vs. median: 0.16 (IQR: 0.00 to 0.80); P < 0.001]. Multivariate logistic regression identified direct exposure (line-of-sight irradiation) as the strongest predictor of negative bacterial culture results, reducing the odds of bacterial culture positivity by 79.5% compared to indirect exposure (OR = 0.205, 95% CI: 0.083 to 0.508, P < 0.001) in the layout of the negative pressure isolation room in this study. Conclusions Ultraviolet-C robots serve as an effective disinfection intervention for hospital environments. Their efficacy is most pronounced when direct exposure (line-of-sight irradiation) is achieved in the isolation rooms. Therefore, aligning movable items toward the UV-C robot prior to disinfection, alongside subsequent manual environmental disinfection and physical debris removal, remains essential, particularly in shadowed areas.