Association of patient-to-emergency department staff ratio with the incidence of cardiac arrest: A retrospective cohort study

1. Introduction

Emergency department (ED) overuse and overcrowding are global healthcare problems [1, 2]. In the United States, the total number of ED visits per year increased from 115.3 million (approximately 39.6 visits per 100 persons) to 138.9 million (approximately 43.3 visits per 100 persons) from 2005 [3] to 2017 [4]. In comparison, the total number of ED visits per year increased from 10.2 million (approximately 46.9 visits per 100 persons) to 11.8 million (approximately 50.1 visits per 100 persons) from 1997 to 2018 in Taiwan [5]. Despite a substantial body of literature on ED crowding, there are three measures most frequently linked to the quality of care, namely, the number of patients in the waiting room, ED occupancy rate (EDOR = total number of patients in the ED divided by the total number of licensed beds), and the number of admitted patients in the ED awaiting inpatient beds [6].

ED overcrowding poses a potential threat to patient safety and quality of care and must be addressed urgently. Sun et al. found that ED crowding was associated with increased mortality, length of hospital stay, and costs in a large cohort of admitted patients [7]. Available medical resources in the ED including emergency physicians, nursing staff, medical imaging examinations, blood tests, monitors, and ED beds are limited. ED overcrowding can also slow down systemic operations and delay the process of diagnosis and treatment for ED patients [8]. Chang et al. investigated the relationship between ED crowding and sudden in-hospital cardiac arrest (IHCA) and found that the ED bed occupancy rate (EDBOR = the ratio of the number of beds occupied by patients to the total number of licensed ED beds) and prolonged boarding time to general wards or intensive care units (ICUs) were associated with an increase in the incidence of sudden IHCA in the ED [9]. We consider that the core of primary care, reasonable patient-to-nurse ratio (PTNR) and patient-to-physician ratio (PTPR) should be established to protect patient’s safety. Previous studies have proven the importance of an optimal PTNR to provide the best care possible by maximizing the value of the nursing staff [10, 11]. Jansson MM et al. [11] found that lower nurse staffing and increased nursing workload were associated with a higher incidence of ventilator-associated pneumonia and mortality in the ICU. A systematic review conducted by Andrea Driscoll et al. [12] also showed an association between a higher level of nurse staffing and improved patient outcomes. For every increase of one nurse, patients were 14% less likely to experience in-hospital mortality.

However, to the best of our knowledge, no previous study has explored the association between a reasonable patient-to-ED staff ratio with primary unexpected cardiac arrest in patients in the ED [13, 14]. This is the first study aimed to find the association between unexpected emergency department cardiac arrest (EDCA) and the patient-to-ED staff (physician and nurse) ratio.

2. Methods

2.2 Study population

The conceptual research model is presented in Fig. 1. We analyzed ED visits to the medical center in northern Taiwan from January 2016 to November 2018. Only patients with primary unexpected (non-trauma, adult) ED cardiac arrest events were included. Non-trauma patients with do not resuscitate (DNR) forms signed before a cardiac arrest episode and patients younger than 18 years were excluded. An unexpected primary ED cardiac arrest was defined as a patient without a physician’s order including critical condition notification, and no DNR signed before the cardiac arrest episode (pulseless electrical activity (PEA)/asystole or pulseless ventricular tachycardia (VT)/ventricular fibrillation (Vf)) through chart review.

Fig. 1.Research algorithm.

3. Results

3.2 Main results

Table 2 presents the incidence of EDCA in the context of different ED crowding parameters every 8 hours during the study period. According to the levels of each ED crowding parameter, 3,192 time intervals (1,064 days $\times$ 3 intervals) were classified into three groups to evaluate the association between ED crowding and the incidence of EDCA (Table 2).

Regarding the crowding parameter, significant increments in EDCA incidence were noted when the total patient number in the ED was $>$ 361 patients (RR: 1.54, 95% CI: 1.239, 1.917), EDOR $>$ 280 (RR: 1.54, 95% CI: 1.245, 1.898) and EDBORs $>$ 184 (RR: 1.63, 95% CI: 1.308, 2.034). For the efficiency parameter, higher EDCA incidences were noted when the total number of patients waiting for boarding to the general ward and ICU were $>$ 134 (RR: 1.45, 95% CI: 1.164, 1.805) and total patients in ED $>$ 48 hours were $>$ 36 (RR: 1.27, 95% CI: 1.029-1.558). For the loading parameter (Table 3), it was observed that increasing both PTAPR and PTNPR led to no significant increments in EDCA incidence (Table 3). For the PTNR categories of $>$ 9.5, 8.5-9.5, and $<$ 8.5, higher EDCA incidence were noted for PTNRs of 8.5-9.5 (RR: 1.33, 95% CI: 1.054, 1.672) as well as those $>$ 9.5 (RR: 1.54, 95% CI: 1.187, 1.994).

Additionally, the incidence of EDCA significantly increased for PTNR $>$ 9 compared to the average incidence of EDCA (0.159), as shown in Fig. 2.

Fig. 2.Association between probability of cardiac arrest and patient-to-nurse ratio. PTNR, patient-to-nurse ratio.

4. Discussion

Previous studies already showed that many crowding parameters [9, 17, 19] are associated with the incidence of EDCA, including the total number of patients in the ED, EDOR, EDBOR, number of patients waiting for boarding and the number of patients in the ED with a prolonged waiting time. However, this is the first study to investigate the association between the incidence of EDCA and the patient-to-ED staff (nurse or physician) ratio.

McCarthy et al. [17] first introduced the EDOR for evaluating ED crowding in 2008. Other studies use EDOR to survey the association of ED crowding and its clinical impact, including overall ED patient’s mortality and poor outcome in ED patients with acute stroke and severe sepsis [20, 21]. Likewise, YH Chang et al. [9] first introduced the EDBOR for assessing the association of sudden IHCA with ED crowding in 2019. Bed-occupying patients were immobilized and severely sick, with higher demand for nurse and physician care. As Chang et al. [9] proved, our analysis also showed that EDBOR is a more representative quality parameter compared to the EDOR. This might be because EDBOR, considering the number of beds occupied by patients, which would highlight the number of ill patients more efficiently than the total number of patients in the ED.

Our findings showed that a higher number of patients waiting for boarding in the ED was associated with a higher EDCA incidence compared to Chang et al.’s study, which only showed the trend [9]. This could be due to the ED observation room having more beds (128 beds) than other facilities, leading to a greater number of patients with urgent and relatively unstable conditions who needed frequent examination and evaluation. Kang et al. [22] have shown that ED overcrowding was not associated with the return of spontaneous circulation rate, survival, and neurological outcomes of out-of-hospital cardiac arrest. Further studies to assess the association between ED overcrowding and the survival outcome of EDCA would be of interest.

Andrea Driscoll et al. [12] reported an association between a higher level of nurse staffing and improved patient outcomes. For every increase of one nurse, patients were 14% less likely to experience in-hospital mortality. In this study, we also found that higher PTNR was associated with an increment of the incidence of EDCA. Considering that both senior residents and nurse practitioners would approach and manage ED patients directly, two kinds of PTPR (PTAPR (attending physician only) and PTNPR (nurse practitioner + attending physician + resident)) were calculated in Table 2. Both PTAPR and PTNPR were not significantly associated with the incidence of EDCA. This can be explained by the fact that emergency physicians simultaneously care for multiple patients due to the multi-tasking nature of emergency management [23]. Besides, unlike nursing staff who routinely provide primary care for patients (vital sign check, IVF adjustment, medication administration, interaction with patients and their families), ED physicians spent more time on indirect care activities (charting, review of medical records, interacting with consultants, and interpretation of diagnostic tests) [23]. Moreover, emergency physicians focus more on patients’ diseases with higher complexity, patients needing frequent resuscitation, and those in worse conditions without any definite diagnosis or disposition through any kinds of diagnostic tests and specialist’s consultation. Hence, the higher PTPR might not show the real workload of the emergency physician compared to the higher PTNR that directly shows the workload of primary care on nursing staff. Another possible explanation for the association between a higher PTNR and a higher EDCA incidence could be that nurses function as the primary surveillance system in the complex ED environment. Nurses provide bedside patient care 24 hours a day and are responsible for early warning system monitoring in addition to having direct knowledge of a patient’s condition and any associated changes [24]. Previous research has shown that sufficient nursing staff is important for patient’s surveillance because nurses manage various patient care needs, identify and intervene rapidly when something has gone wrong [25, 26]. Our study indicates that the hospital should adjust ED staff with more elasticity in the context of ED overcrowding, especially considering nursing staff [18, 27]. In daily practice, a higher PTNR would postpone performing the physician’s order or provide primary patient care and even worse, increase the possibility of burnout of nursing staff due to overloading [25, 28].

Ramsey et al. [29] reported that a lower nurse-to-patient ratio directly influences efficiency, thereby increasing ED length of stay and the percentage of patients who left without being seen. Even if facilities do not possess the capacity to cope with the amount of ED visits, particularly with the availability of hardware devices (e.g., beds, time taken for different diagnostic testing), warning values for the number of patients that can be accommodated according to the capacity of each ED must be set. During various seasons and holidays, or in the case of a mass casualty incident, hospitals should implement dynamic management of their ED workforce to maintain patient safety and quality of care [30]. Fig. 2 shows the association between the incidence of EDCA and PTNR by the restricted cubic spline model. In our study, when PTNR exceeded 9, the incidence of EDCA crossed over the average incidence of EDCA that may be of interest and may serve as a cut-off for the emergency department managers to avoid the increasing incidence of EDCA and potential burnout on nursing staff. The results pointed out the possibly alarming threshold of the ED’s capacity.

In the past decade, there have been many quality parameters with proven associations with poor ED quality and patient’s safety were proven. However, this is the first study to analyze and prove that a higher patient-to-nurse ratio was associated with a higher EDCA incidence.

5. Limitations

This study should be interpreted in the context of the following limitations.

First, the time spent in the observation room waiting for boarding, after initial stabilization and resuscitation, was not analyzed in the current study, as we aimed to focus on different crowding parameters. Second, this study was conducted at a medical center in northern Taiwan within a limited period, which may have restricted the general applicability of our findings. The warning threshold of each parameter from our results might not be suitable for hospitals with different levels. More validation studies conducted in different settings and regions would be of interest. Third, no subgroups based on whether the patient was transferred from other hospitals or had utilized emergency medical services were analyzed. The severity of illness might also contribute to our study’s limitations. However, our study population focused on patients with low potential risk for EDCA and aimed to find the relationship between the workload of emergency physicians and nurses and EDCA. Finally, no analysis of a survival outcome was conducted in our study. Although the incidence of EDCA was associated with many quality parameters, further studies investigating the cerebral performance categorylevel of patients with EDCA episodes after discharge would be of interest.

6. Conclusions

A higher patient-to-nurse ratio ($>$ 8.5) was associated with an increment in the incidence of EDCA. Our findings could provide a basis for setting different thresholds for different ED settings to adjust ED staff and develop individually tailored approaches corresponding to the level of ED overcrowding.

Author contributions

Conceptualization, L-H, Tsai, Y-M, Weng, C-C, Lin, C-J, Ng, C-J, Seak, C-H, Huang and C-Y, Chien; Data curation, S-L, Tsai; Formal analysis, C-B, Chen, C-H, Chaou and C-Y, Chien; Investigation, W-C, Chien, Y-M, Weng; Methodology, L-H, Tsai, W-C, Chien and S-Y, Fang; Resources, C-J, Seak; Software, W-C, Chien; Supervision, S-Y, Chen; Writing-original draft, L-H, Tsai; Writing-review & editing, C-Y, Chien.

Ethics approval and consent to participate

This study was approved by the Hospital Ethics Committee on Human Research of the Chang Gung Medical Foundation (IRB 201900275B0). The study protocol was reviewed, and the need for informed consent was waived.

Acknowledgment

Thanks to all the peer reviewers for their opinions and suggestions.

Funding

Funding was provided by Chang-Gung memorial hospital, Linkou. CMRPG3J1221.

Conflict of interest

The authors declare that there is no conflict of interest regarding the publication of this article.

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.