Satisfaction of nurses and physicians with the introduction of the Rapid Response System in Dutch hospitals

Background: Rapid Response Systems (RRSs) have been introduced in hospitals to improve recognition of and response to deteriorating hospital ward patients. The value of an RRS depends not only on relevant patient outcomes but also on how satisfied nurses and physicians are with the system. The aim of the study was to measure the degree of satisfaction with an RRS and analyse factors influencing the degree of implementation. Methods: Questionnaires were distributed among physicians and nurses on medical and surgical wards participating in the COMET study at 7 and 14 months after introduction of a Rapid Response Team (RRT). The questionnaires included 24 questions regarding the use and the degree of satisfaction with the Modified Early Warning Score MEWS/SBAR tool and the RRT. Results: The response rate was 1005/1920 (52%). Satisfaction with implementation of the RRS was generally higher at t=14 compared with t=7 months and in respondents working on surgical versus medical wards. In a multivariate analysis, independent predictors of high satisfaction were timing of the questionnaire (14 months versus 7 months after the start of an RRT), the support of the RRT system by local ward management, and having an RRT that was considered to be open and approachable. Conclusions: Our findings show that healthcare workers on hospital wards are generally very satisfied with the services offered by the RRT, the use of the MEWS instrument to recognise deteriorating patients and the SBAR communication tool to improve communication between nurses and doctors. Satisfaction with the RRT was higher at 14 months compared with 7 months. Introduction Rapid Response Systems have been introduced in hospitals to improve recognition of and response to deteriorating hospital ward patients.[1] An RRS can be seen as an intensive care-based, organisation-wide preventive approach to the management of deteriorating patients, and implementing the RRS requires more than just standardisation of ‘calling criteria’ and the rapid response of a dedicated acute care team. The RRS consists of three important components. The afferent limb is designed to identify the deteriorating patient by using calling criteria such as the Modified Early Warning Score (MEWS) card and to trigger a response. The efferent limb involves directed action of the Rapid Response Team (RRT) and the third component includes measures to improve the quality of care on the ward, training and feedback.[1,2] An optimal RRS should ensure 1) the support of all physicians and nurses, 2) leadership and support from senior hospital executives, 3) 24/7 response by staff with appropriate skills, knowledge and experience, and 4) the promotion of hospitalwide awareness of the system.[3] The effectiveness of RRSs has not yet been proven conclusively. So far, the effectiveness of the introduction of RRSs in hospitals was shown in only two studies. The study by Priestly[4] showed a reduction in hospital mortality, while the study by Ludikhuize et al.[5] showed a reduction of the composite endpoint including cardiac arrest, death and unplanned ICU admission. Another multicentre randomised study conducted by Hillman[6] in Australia could not demonstrate a benefit of the introduction of a medical emergency team based RRS. Besides effects on relevant patient outcomes, the value of an RRS also depends on how satisfied nurses and physicians are with the system. Satisfaction of healthcare workers with the RRSs is not only a subjective measure of contentment with the support the RRS offers to the care of their patients, it also is a prerequisite for a good implementation and performance of the RRS. Nurses will only call an RRT if they expect to be supported by it. Fear of being criticised by members of an RRT for their care of deteriorating patients was reported to be a barrier for implementing an RRS.[7-9] In the Netherlands, we recently implemented an RRS in 12 hospitals. The aim of this study was to measure the degree of Netherlands Journal of Critical Care 166 NETH J CRIT CARE VOLUME 25 NO 5 SEPTEMBER 2017 satisfaction of nurses and physicians with the implementation of an RRS and the perceived benefit of the system. Material and methods Design, setting, participants This study is part of the Cost and Outcome Medical Emergency Team (COMET) study which was conducted in the Netherlands from 2009 to 2011. The COMET study was a pragmatic prospective before-after multicentre study in which 12 Dutch hospitals participated. The before period in which baseline characteristics were collected lasted five months. Subsequently, the RRS was introduced in a two-step fashion. First, in the MEWS/SBAR phase, which lasted 7 months, the Modified Early Warning Score (MEWS) card and the Situation Background Assessment Recommendation (SBAR) communication tool were introduced to identify patients at risk and to facilitate communication between nurses and physicians. Secondly, the RRT was implemented and this phase lasted 17 months; it was divided into two periods, namely RRT implementation and the Final RRT phase. In each participating hospital, patients of 18 years and older who were admitted to two surgical and two medical wards, the so-called COMET wards, were included. A full description of the study design (figure 1) has been published previously.[5,10] During the second phase of the COMET study, questionnaires were distributed to nurses and physicians in all 12 participating hospitals to measure the satisfaction with the RRS on two different time points: 7 and 14 months after introduction of the RRT. On each occasion, participating hospitals distributed 80 questionnaires on the four COMET wards to nurses and physicians. The questionnaires were completed anonymously. Intervention The questionnaires included 24 questions covering three aspects: 1) questions on how respondents used the MEWS/SBAR tools and RRT, 2) level of satisfaction with MEWS/SBAR and RRT, and 3) characteristics of the respondents (physician/nurse, working on medical/surgical ward, gender, age, experience since graduation (years), employment in the hospital and current ward (years)). Responses to the questions were scored on a scale from 0 -10 (0 = totally disagree or never, 10 = totally agree or always). Ethical consideration The medical ethics committee of the Academic Medical Centre in Amsterdam waived the need for formal evaluation of the study due to the observational nature of the study. Consequently, the need for informed consent was not applicable. Statistical analysis Descriptive analyses are presented as raw numbers and percentages. Continuous data were presented as medians with interquartile range (IQR) due to non-normally distributed data. A bootstrap independent t-test was used for comparison of the time points, drawing 1000 samples of the same size as the original samples and with replacement, stratified by the timing of questionnaire. The generalised estimating equation (GEE) was applied to estimate the univariable association between predictors as measured by the questionnaire and satisfaction. The predictors used in GEE were 1) timing of questionnaire (7 and 14 months), 2) gender of respondent, 3) surgical/medical ward, 4) number of patients with MEWS ≥3 assessed by nurse or physician in the last 2 weeks, 5) age (years) of respondent, and 6) work experience (years) of respondent. In the GEE, a binomial distribution was assumed after recoding the questions scored on a scale from 0 to 10 into a dichotomous Table 1. Demographics RRT implementation phase Questionnaire 7 months 14 months Respondent, n (% of total ) 492 (51) 513 (53) Gender, male, n (%) 55 (11) 73 (14) Age, mean ± SD 32.8 ± 10.5 32.6 ± 10.5