Hospital environment health monitoring techniques for identifying pathogens are widely used in the environmental epidemiological study of nosocomial infections to identify the emergence of hazardous environmental conditions over time, accurately evaluate the effect of disinfectants on pathogens, and to identify transmission media and methods of nosocomial infection[13, 14]. Hospital environment health monitoring techniques is an essential means to control nosocomial infections.
High-touch objects such as partition curtains, which are used widely in diagnosis and treatment areas, are sources of nosocomial infection. Appropriate techniques for monitoring such high-touch objects play a vital role in the prevention of cross-infection, thereby ensuring medical quality and patient safety.
The results of our study are consistent with those of Lemmen et al., who reported that Gram-positive bacteria are more common than Gram-negative bacteria in hospital wards [11]. Gram-positive bacteria such as Staphylococcus and Enterococcus are major contributors to cross-infection incidences in hospitals, as they can survive for days or even months on dried medical textiles and plastics [15]. Gram-negative bacteria lose activity faster than Gram-positive bacteria in dry conditions. Hence, with Gram-positive bacteria there will be a wider contamination area on the surface of an object [16]. Further studies on a more sensitive detection method for Gram-negative bacteria will be needed.
Obee et al. found that the impact of the sampling method on the detection rate of the microorganisms on the environmental surface is greater than that of the used culture medium[17]. However, Niskanen and Pohja reported a different result in their comparison between the RODAC agar plate and swabbing technique that sampling methods should be considered on the basis of monitoring purposes[18]. Lemmen et al. found that RODAC agar plate was more sensitive to Gram-positive bacteria, with a detection rate of 69.5%, while swabbing technique was more sensitive to Gram-negative bacteria, with a detection rate of 74.2%[11]. In an environmental surface test for carbapenem-resistant enterobacteriaceae, Lerner et al. reported the recovery rates of the three sampling methods were 32%, 24%, and 16% for contact plates, contact plates, eSwab sampling followed by broth enrichment prior to plating, and eSwab sampling with direct plating onto the plates, respectively [19]. However, in our study the detection of Gram-positive bacteria using the swabbing technique was superior to that of the RODAC agar plate method. We found no significant difference between the two methods in the detection of Gram-negative bacteria.
Rabuza et al.reported that the hygroscopicity and porosity of the object surface can affect the efficiency of the swabbing technique [20]. Among the methods tested, they found the detection rate of the swabbing technique to be the lowest [20]. Furthermore, Dalmaso et al. and Hedin et al. found that the results of sampling techniques could be affected by the materials contained in the swabs [21, 22]. The sampling efficiency of the RODAC agar plate can also be affected by contact time and pressure. Foschino et al. found that the bacterial detection rate in the contact plate was 35% after a 1-hour contact period, which was much higher than that after a 1-minute contact period; they also reported that the recovery rate of Escherichia coli reached 100% after 4 hours’ contact [23]. However, to date there is no consensus as regards optimal time or pressure for contact. Several studies suggest that the contact time can be up to 30 minutes, while others suggest a more realistic contact time of 5–15 seconds [11, 23–27]. Kusumaningrum recommended a range of 20-25g/cm2 as the optimal contact pressure between the contact plate and the object surface [28].
Rabuza et al. [20] compared the sampling efficiency of different sampling methods, and found that the number of colonies varied with different concentrations of swab sampling and contact plate culture. No microorganisms were found on the swabs at low concentrations. Our mixed model has shown that there are interaction effects between sampling method and sampling time. Our independent effect analysis of sampling methods has shown that none of the sampling methods had significant impact on the amount of bacteria captured from the curtains. Our study employs the mixed linear model based on the characteristics of the monitoring data, which permits scope for some correlation and diversity within the data of the covariance matrix so as to better adapt to the characteristics of repeatedly measured data.
The mixed linear model is a useful tool for the study of complex internal relationships across time [29]. The results of the comparative analysis of time trends is consistent with the change of time effect shown in the mixed model, which makes the results more reliable. Our study indicates that the amount of bacteria in the curtain initially increased, then peaked and decreased over time; this pattern contrasts with the research results of Kevin et al. [2]. This dissimilarity might be ascribed to different monitoring intervals and cycles.