Relationship between hospital ward design and healthcare-associated infection rates: a systematic review and meta-analysis

The purpose of this review was to systematically identify and analyze primary research studies, wherein infrastructural measures were examined as determining factors for infection control. Our research reveals a strong correlation between hospital ward design and healthcare-associated colonization and infection rates. According to our analysis, the implementation of single-patient rooms and the installation of easily accessible antiseptic hand rub dispensers near patient beds are two important facilitators for infection control. Research data about the relationship between the patient room size or the proximity between patients in adjacent beds and the colonization or rates of infection is scarce. We identified three studies, which had entirely different study environments and outcomes. Jones et al. investigated the space per cot in a neonatal intensive care unit. They concluded that a significant association exists between a higher square footage per cot and lower late-onset sepsis rates [27]. Jou et al. determined an increased risk of nosocomial C. difficile infection in patient rooms with larger square footage [26]. Due to the characteristics of the evaluated pathogen C. difficile, it is likely that spores contaminated the surface. This is attributable to the fact that a larger room allows more surface to be contaminated, which leads to an increased transmission risk as cleaning in a larger room could be performed rather inadequately [29]. However, transmission seems to be a minor issue for infection with C. difficile. Widmer et al. presented a very low rate of transmission in their prospective observational study during an 11-year study period: transmission was detected in 1.3% (6/472) of all contact patients [30]. Another structural aspect was investigated by Yu et al., who investigated the association between the distance between beds and the outcome severe acute respiratory syndrome [28]. They concluded that a minimum of ?1m between beds is needed to reduce the risk of transmission and thus infection. As this outcome describes a pathogen, which is transmitted via droplet infection, it is questionable to transfer their results to other pathogens. More research is needed on this specific topic to further analyze the implications for infection control measures.

Proper hand disinfection has been proven to be one of the most effective infection control measures. It is quite conceivable that factors improving the compliance rate support the barrier against pathogen transmission [7, 8]. We did not identify any studies investigating on the impact of the location of hand-rub dispensers on healthcare-associated infection rates. However, the results of this review indicate that sustainable improvement of hand hygiene compliance can be supported by locating the hand rub dispenser in the point of care and facilitate its accessibility for healthcare workers [31–33]. Therefore, this review confirms the conclusions made by Kendall et al. who suggest to ensure the availability of the hand rub dispenser in the point of care [33]. Likewise, Zingg et al. concluded that a hand-rub dispenser directly in sight of healthcare workers and facilities for hand hygiene at the point of care both improved hand hygiene performance in their systematic review about hospital organization, management and structure for the prevention of HAIs [34] However, as Giannitsioti et al. found out in their follow-up study, a directly accessible dispenser alone may not lead to a sustained compliance improvement [14]. We suggest that easily accessible hand rub dispensers be placed near the patient’s bed at the point of care. This should be combined with other useful compliance improvement measures such as regular staff training and feedback on compliance rates to ensure improved hand hygiene.

The review shows that single-patient rooms are a significant infection control measure in preventing transmission of pathogens from one patient to another due to the fact that no contact transmission can occur either directly from a roommate or indirectly from a healthcare worker taking care of a roommate. Moreover, boundaries that enhance the health care workers’ hand hygiene compliance rate are more firmly established [35]. Conversely, infections can also be caused by the acquisition of pathogens from a prior room occupant [36]. However, a single patient room is considerably easier to clean after the discharge of a patient. Therefore, the risk of environmental contamination could be reduced in comparison to larger and more heavily equipped multi-patient bedrooms.

This review has several limitations. It cannot be ruled out that due to the before-/after-intervention concept the general improvement of medical care over time might have biased the results of some of the studies and consequently biased the results of our meta-analysis (see Figs. 4, 5 and 6). This does not affect three studies [16–18]. Two of these three studies comparing the intervention and control group in the same time period also revealed a benefit in single patient rooms [17, 18]. The study conducted by Ellison et al. is found to be the only statistical outlier [16]. Confirmatively, the authors describe what may have compromised the intervention’s potential: shortly after the study began, three single-patient rooms were converted to multi-patient rooms with proximity of 1m between beds. Approximately 50% of the intervention group’s patients stayed in these converted rooms. The considerable heterogeneity in the meta-analysis of studies comparing single vs. multi-bedrooms with the outcome of healthcare-associated colonization/infection could be partially explained by the study design. Hagel et al. considered a strong Hawthorne effect on hand hygiene performance, which might attenuate the reported results regarding hand hygiene compliance rate [37].