A multinational survey on the infrastructural quality of paediatric intensive care units

Deficits regarding structural quality were found in all three studied countries but especially in Germany (compare Fig. 1). The most plausible reason seems to be the more centralized organization of paediatric intensive care in Switzerland when compared to the other two countries. Despite the fact that is been shown that the use of single-patient rather than multi-patient rooms improves nurses’ perceptions of the PICU environment for promoting patients’ sleep and the nurses’ own work experience [22], the recommended single and double bedrooms were available in only a low number of PICUs. A possible underlying reason may be the fact that most hospitals in the studied countries are based on older buildings with architectural limitations. Since isolation is a recognized preventive measure in infectious patients [23], this insufficiency needs to be addressed in the near future. PICUs were centrally located in about three quarters of hospitals. In our questionnaire, a 3-min walking distance was defined as proximity, because this seemed to be a reasonable distance in case of life-threatening emergencies to the authors. Nevertheless, there are no clear guidelines in this regard and the cut-off was arbitrary. In future, on-site visits could lead to an optimization in the evaluation of the wards’ location [24].

Comparing the three countries, Switzerland had the highest mean number of patients per year on PICUs. Astonishingly, the median stay on the PICU in Switzerland was about half that of Austria and Germany. This may be due to differences regarding the availability of alternatives such as IMCUs for less severe cases (who occupy PICU capacity). Additionally, Swiss hospitals provided the highest number of SOPs also including admission, triage and discharge (compare Fig. 4), which may have impact on the length of stay. On the other hand, there may be differences in the quality of aftercare (allowing earlier transfer to the normal ward to free PICU capacity) which is difficult to assess and was thus not evaluated in this survey.

The diagnostic and therapeutic equipment was on a very high technical level in the majority of PICUs in all three studied countries (compare Fig. 2). However, our survey has revealed two obvious deficits: the availability of respirators and the implementation of a PDMS. Since the ESICM does not specify this item the requirement of one respirator per available bed was based on DIVI guidelines [12]. The reasonability of one respirator per bed on PICUs, however, can be discussed. The low number of PICUs using PDMS—especially in Germany—has to be criticized because PDMS has been shown to play a pivotal role in QM [15, 25].

Deficits regarding the personnel were encountered in all three surveyed countries (see Fig. 3). The training, qualification and presence of the medical head were insufficient in many German PICUs. With regard to patient safety and QM, a strict adherence to the guidelines has to be demanded [11‐13, 26]. The low number of medical doctors (MDs) on the PICUs and their qualification mirrors tight resources in all countries. This fact cannot be overstressed since it has been shown that paediatric critical care provided in the PICUs staffed with a 24/7 intensivist presence is associated with improved overall patient survival [27]. Therefore, future training programmes should aim to raise the qualification of MDs on PICUs in order to further improve patient care [17].

The qualification of the nursing staff and the head nurse was very high, and a large proportion of PICUs met the requirements. However, the head nurse was also occupied with patient care in many wards, which contradicts the recommendations [11, 12]. The ratio between beds and nurses was tremendously different between the countries. While the requirements were met by the majority of PICUs in Austria and Switzerland, grave deficits were encountered in Germany where one nurse had to care for 3–5 critically ill children during the night shifts. Therefore, prompt measures have to be taken to improve this deficit [7, 11, 12, 23, 28‐30].

Concerning organization and quality improvement, Switzerland took a clear leading position in this survey (Fig. 4). In future, patient care and management could be improved by the increasing the number of available SOPs. In particular, the important role of a weaning protocol is well known [32] and the request for improved patient management warrants improvements in SOPs in all countries in this regard.

Moreover, it has previously been shown that both benchmarking and internal quality control improve patient safety and management [24, 31]. Although systems of quality comparison are available in Europe (i.e. Vermont Oxford Neonatal Networks, Extracorporeal Life Support Organization Registry), only half of the PICUs in Austria and Germany participate in these tools.

The number of returned questionnaires was 100% for Austria and Switzerland. Thus, the results demonstrated above are representative for these countries. In contrary, the response rate for Germany was lower (61%). Nevertheless, the response rate of this survey is higher than that reported in previous smaller studies [19, 33].

A possible limitation of the present study is the fact that a survey is always based on the subjective opinion of the persons reporting for the centre. However, this “self-reporting” bias is the nature of a questionnaire-based survey. This could be mitigated by on-site visits from an independent committee. However, this was beyond the scope of this study.

Inclusion of a case mix index (CMI) would have further enhanced this study. The CMI was part of the questionnaire. However, response rates were very low (< 10%) and therefore not representative. All participating hospitals and the national societies were contacted a second time, but the response rate could not be increased. Therefore, this item had to be excluded in the final analysis. Additionally, the calculation of the CMI differs between European countries making a direct comparison difficult.

Another limitation lies in the interpretation of the guidelines. First of all, there are no available European guidelines specifically for paediatric patients. For a lack of alternatives, we had to rely on adult guidelines. Additionally, a similar approach has been chosen in a previous smaller survey [33]. The major difference between PICUs and adults ICUs can be found in the demands and (patho)physiology of the patients treated. Besides different requirements concerning medical equipment, the heterogenic spectrum within paediatrics causes difficulties in creating SOPs as well as guidelines. Thus, certain adult-based items may be the difficult to meet for PICUs.

Moreover, ESICM guidelines are based on treatment levels to a certain amount [10, 11]. The definition of these levels either is absent or differs between the countries making their comparability difficult. Additionally, ESICM guidelines are unspecific in certain items and had to be supplemented by DIVI recommendations [12] in certain questions to obtain a more complete image of the infrastructure of PICUs addressed in this study. DIVI guidelines were chosen because all countries were German speaking and the national societies of Austria and Switzerland have not published separate recommendations.