Our survey has delivered valuable insights into the state of implementation of key IPC structures and processes in Austria. Overall, the data gathered demonstrated that IPC is at a high, yet in some respects diverse, level in Austria. This generally high level of IPC implementation was expected, as Austria is classified as a high-income country according to the World Bank classification . Noticeably though, a considerable proportion of participating hospitals (42%) were allocated to merely an “intermediate” IPC level. This rather surprising finding could either be explained by a very strict interpretation of the IPCAF by Austrian hospitals, or be an indicator that IPC is not yet adequately addressed by some hospitals in Austria.
Besides the differences observed between the different participating Austrian hospitals with regard to the overall IPCAF score, we noticed pronounced differences between aggregated scores of the respective IPCAF sections. Scores regarding CC1 (IPC Program) were generally high. However, specific questions focusing on IPC staffing and organization of an IPC committee revealed mixed results. Only 26% of hospitals reported employing at least one IPC professional per ≤250 beds, which has to be viewed as a clear target for improvement. Moreover, more than one fourth of hospitals from Austria reported not having an IPC committee actively supporting the IPC team. Where IPC committees existed, crucial professional groups (e.g. senior facility leadership, senior clinical staff, and facility management) were underrepresented in many Austrian hospitals. In CC7 (Workload, Staffing and Bed Occupancy), only 60% of hospitals in Austria reported that an agreed healthcare worker to patient ratio was maintained at all times within the entire facility. Given the evidence, that understaffing is a demonstrated risk factor HAI occurrence , this finding gains relevance as an action point for future measures to improve the quality of care in Austrian hospitals.
A relatively high proportion (26%) of Austrian hospitals reported not conducting surveillance of multidrug-resistant pathogens according to the local epidemiological situation. Given the increase in awareness for the subject, due to recent publications on the burden of antimicrobial resistance in patients with HAI , we found this proportion to be surprisingly high. A crucial element of surveillance is feedback to relevant stakeholders, such as frontline healthcare workers and clinical leadership . The majority of Austrian hospitals reported performing regular feedback of surveillance data. However, regarding the methods employed for this feedback, less than a third of participating hospitals reported performing annual feedback via presentation and interactive problem-solution finding. It is important to acknowledge this result as a target for improvement, since the effectiveness of interactive feedback of surveillance data has been demonstrated in previous publications [15, 16].
The concept of multimodal strategies has become increasingly prevalent in medicine, and in the practice of IPC [17,18,19,20,21,22]. However, given the rather low scores achieved by Austrian hospitals for CC5 (Multimodal Strategies for Implementation of IPC Interventions), it appears that awareness for and implementation of multimodal strategies are not yet fully accomplished. A similar deficit was illustrated in an analysis of the German IPCAF data . A conclusion, which could be derived from these data, is that any efforts to strengthen IPC in high-income countries, should place an emphasis on multimodal strategies.
According to data from this survey, Austrian hospitals do not lack on-site expertise for conducting IPC training, as almost all hospitals reported having staff capable of performing IPC training. However, questions focusing on the execution of IPC training revealed deficits, especially with regard to the frequency of training and the methods employed. For instance, the percentage of hospitals that reported conducting mandatory annual training was lower than in the survey in Germany , where such activities are mandatory in numerous federal states. The importance of regular IPC education and training has been demonstrated in various publications , and given the presence of capable staff on-site, appears feasible in Austrian hospitals. Regarding methods employed for IPC training, similar shortcomings were demonstrated as in the German survey . The lack of interactive (e.g. bedside) training has to be considered as a deficit and potential for improvement, as this form of education has been proven to be efficacious in numerous studies [20, 24].
A secondary objective of our survey was to increase the awareness of and knowledge about WHO tools in Austria. Our data suggests that these are not yet fully implemented in the routine IPC work at Austrian hospitals. For instance, only a minority of hospitals reported regularly conducting the WHO Hand Hygiene Self-Assessment Framework, despite the tool already having been released by the WHO in 2010 , and having been promoted in two global surveys (2011 and 2015) . This finding may be attributable to the fact that the Austrian “Clean Care is Safer Care” campaign has been initiated only in recent years.
The results of our survey gain additional relevance when comparing the Austrian data to data from the IPCAF survey conducted in German hospitals. In Germany, the proportion of hospitals not achieving an advanced IPC level was only 15% (versus 42% in Austria). Moreover, the median aggregated IPCAF score was higher in Germany (690 in Germany vs. 620 in Austria) . This discrepancy between the two economically and culturally similar countries, may partially be attributable to measures strengthening IPC in German hospitals which have been undertaken earlier in Germany than in Austria. The revision of the German Protection against Infection Act in 2011 required hospitals to reinforce existing IPC structures , particularly regarding IPC staffing and surveillance of HAI. It is conceivable that this has led to a more equal state of implementation of IPC in Germany. As of now, Austria does not have a similar legislation in place with regard to infection prevention in order to promote adequate staffing for IPC. Additionally, the long history of surveillance in Germany [27, 28], has to be recognized as factor contributing to high awareness of IPC in Germany.
Adequate interpretation of the data presented in this publication requires acknowledgement of various limitations. The primary purpose of the IPCAF is to enable healthcare facilities to perform a self-assessment of the local situation regarding IPC, not to generate national reference data. Comparisons between countries and healthcare systems are therefore only possible with significant restrictions. The NRZ is responsible for organizing surveillance in Germany, and the KISS network is a German institution. Therefore, collection of data from other countries is not the primary purpose of a survey issued by the NRZ. However, due to the large number of hospitals from Austria participating in the KISS network (approximately 40% of all acute care hospitals in Austria) and in this survey (around 25% of all acute care hospitals in Austria) , careful extrapolations to the overall situation in Austrian acute care hospitals appear justified. Nevertheless, it has to be noted that survey participation was on voluntary basis, therefore, facilities with a high interest in IPC may be overrepresented. This could lead to an overestimation of the state of IPC implementation in Austrian hospitals.
Proper utilization of the IPCAF required a deep understanding of the WHO terminology and underlying concepts. Despite providing footnotes and online support, concepts such as multimodal strategies, may have been misunderstood, and therefore certain data not reported correctly. Furthermore, selected questions of the IPCAF may have been deemed as compromising by some hospitals, which could lead to biased reporting in some cases. However, to decrease the incentive for wishful reporting and to comply with data protection restrictions, the IPCAF survey did not collect structural hospital information (e.g. number of beds, hospital type, and hospital ownership). Regarding data collection, it has to be acknowledged that the web-based mode of data entry did not allow for retroactively correcting entered data after completion of the survey. This may explain some surprising results.