Nurse staffing and care process factors in pediatric emergency department - An administrative data study

Kokoteksti

(1)UEF//eRepository DSpace Rinnakkaistallenteet. https://erepo.uef.fi Terveystieteiden tiedekunta. 2020. Nurse staffing and care process factors in pediatric emergency department - An administrative data study Janhunen, Katja Wiley Tieteelliset aikakauslehtiartikkelit © 2020 John Wiley & Sons Ltd All rights reserved http://dx.doi.org/10.1111/jocn.15482 https://erepo.uef.fi/handle/123456789/24338 Downloaded from University of Eastern Finland's eRepository.

(2) Accepted Article. Nurse staffing and care process factors in pediatric emergency department – An administrative data study. Authors Katja Janhunen MSc., RN (Corresponding author) Ph.D-student, University of Eastern Finland, Department of nursing science. P.O. Box 1627, 70211 Kuopio, Finland. Katjajan@student.uef.fi Tel: +358445526182 Orcid: https://orcid.org/0000-0002-7286-0208. Päivi Kankkunen Ph.D., Docent University Lecturer Department of Nursing Science University of Eastern Finland P.O. Box 1627, 70211 Kuopio, Finland paivi.kankkunen@uef.fi. Tarja Kvist Ph.D., RN, Docent Associate Professor Department of Nursing Science, University of Eastern Finland P.O. Box 1627, 70211 Kuopio, Finland tarja.kvist@uef.fi. This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/JOCN.15482 This article is protected by copyright. All rights reserved.

(3) Accepted Article. Conflict of Interest All the authors declare no conflict of interest Funding Source(s) This work was supported by the Finnish Nursing Association; the Helsinki University Hospital; and Finnish Foundation of Nurse Education.. This article is protected by copyright. All rights reserved.

(4) Accepted Article. MRS. KATJA JANHUNEN (Orcid ID : 0000-0002-7286-0208). Article type. : Original Article. Nurse staffing and care process factors in pediatric emergency department – An administrative data study . To assess the quality of care, it is important to consider not only its components but also the relationships between them. However, relationships between the structural factors of care, such as nurse staffing, and the pediatric emergency care process have been little studied.. . Nurse staffing is linked to care process variables, length of stay and factors associated with numbers of patients who leave before treatment is completed in pediatric emergency care. settings.. . Because the nurse-patient ratio affects the care process, it should be used together with data on other process variables when assessing care quality in pediatric emergency departments.. Abstract Aim: To describe the nurse-patient ratio in the pediatric emergency department and whether it is related to emergency care process measurements: length of stay and the number of patients who. leave before treatment is completed. Background: Despite abundant data on nurse staffing, little is known about its relationship with process variables in pediatric emergency departments. Design: This was a cross-sectional study. Administrative data regarding 21,956 patients and nurse staffing (N=49) were collected from a university hospital’s pediatric emergency. department between 1 January and 30 June 2019. Summary statistics were calculated, differences. in the variables were assessed by Kruskal-Wallis and chi-square tests, and relations between This article is protected by copyright. All rights reserved.

(5) Accepted Article. them were explored by linear regression analysis. This study is reported in accordance with the STROBE guidelines. Results: Nurse-patient ratios varied between shifts, and were highest at night (mean 0.75; range 0.3-5.3) and the lowest in the evenings (mean 0.17; range 0.1-0.8). Increases in numbers of nurses in the pediatric emergency department reduced the length of stay by 2% per additional nurse on average, and nurse-patient ratios were negatively related to frequencies of patients leaving before treatment completion. Conclusion: The results indicate that nurse-patient ratios affect pediatric patient care processes.. Staffing levels are negatively related to emergency department length of stay and influence factors that could reduce numbers of patients who leave before treatment completion. Relevance to clinical practice: Because the nurse-patient ratio affects the care process, it should be used together with other process measurements when assessing care quality in pediatric emergency departments. Keywords: pediatric nursing, nurse staffing, emergency departments, management. Introduction Nurse staffing is a key structural element of care quality. To provide high-quality care, a holistic approach is required, addressing all the care quality factors in an integrated manner, and deploying both the material and human resources effectively to optimize care processes and outcomes (Donabedian, 1988). To assess the quality of care, it is important to consider not only its components but also the relationships between them. However, relationships between the structural factors of care, such as nurse staffing, and the pediatric emergency care process have. been little studied (Alessandrini et al., 2011). Children are a significant group of emergency. patients, and clearly intense efforts should be made to improve their care (American Academy of Pediatrics et al. 2019). Thus, it is important to obtain information about unit-level nurse staffing and its relationship with process variables, such as emergency department (ED) length of stay. (LOS), and the proportion of patients who leave before treatment is completed (LBTC) in the pediatric emergency setting. Background. This article is protected by copyright. All rights reserved.

(6) Accepted Article. Nurse staffing plays a significant role in the service structure of healthcare organizations and there is a need to define an adequate level of staffing. However, Aiken et al. (2012) found that nurse staffing levels were higher, on average, in the USA than in all of 12 surveyed European countries except Norway. Moreover, in California, strict minimum nurse‐ to‐ patient ratios were set by law in 2004, including one nurse per four patients in pediatric care units, and one per two patients in neonatal or pediatric intensive care units. Such measures and variations are potentially important, because increases in nurse staffing ratios are associated with reductions in patient mortality, and improvements in the nurses’ perceptions of the quality they can provide (Aiken et al., 2010). At the hospital level, nurse staffing varies between types of unit, with a low staffing level being related to increases in patient falls and medication errors (Duffield et al., 2011). In addition, a. relationship has been identified between nurse staffing and patients’ LOS in a Finnish hospital’s acute‐ care units (Pitkäaho et al. 2016). To assess staffing levels in hospital settings, commonly used variables include numbers of nursing hours per patient day and nurse-patient ratios (Min & Scott, 2015). Another variable used in emergency care settings is nursing hours per patient (Ramsey et al., 2018), and others used in emergency care nurse staffing models include nursing hours, staffing levels, and patients’ acuity levels (Lordache et al. 2020; Otegbeye et al. 2015). According to interviewed emergency nurses, safe staffing determinations should not be based solely on numbers of patient beds, but also on information regarding patient acuity and volume as well as experience mix of the nursing staff (Wolf et al. 2017). Clearly, estimating or calculating adequate nurse staffing levels is not straightforward, and particularly challenging in pediatric emergency departments, where patient arrival times vary diurnally, with peaks in both the evenings and weekends (Barata et al., 2015). It is also clearly crucial for nurses to have the necessary skill, knowledge, and training to provide emergency care for children of all ages (American Academy of Pediatrics et al. 2009). One solution for. estimating adequate nurse staffing is to use computer modeling to optimize it in relation to diurnal patient flows (and other relevant temporal variations) in pediatric emergency departments (Michelson et al. 2016).. This article is protected by copyright. All rights reserved.

(7) Accepted Article. Two variables commonly used to measure nurse staffing in relation to care processes are the number of patients who LBTC and the ED LOS. Relationships between nurse staffing and. process variables in adult or general EDs have been addressed in several studies. These studies have found that nurse to patient ratios are negatively related to both waiting times until diagnostic evaluation (Shindul-Rothschild et al., 2017) and ED LOS (Ramsey 2018), and positively related to patients’ care experience (Nelson et al., 2018). Hofer and Saurenmann (2017) have also found that prolonged LOS (more than four hours) is related to patients' nonacuity triage classification, emergency visits during winter, and patients arriving in mornings in a pediatric ED. However, relationships between nurse staffing and patient process variables have received little attention in pediatric emergency care, and there are clear needs to obtain more information about them. Aims Aims of this study were to describe the nurse-patient ratio in the pediatric emergency department (PED) and whether it is related to emergency care process measurements: length of stay and the. number of patients who leave before treatment is completed. The following research questions were specifically addressed: 1. What is the shift-related pattern of the nurse-patient ratio in the PED?. 2. How is the nurse-patient ratio related to patients’ emergency department length of stay? 3. Is the nurse-patient ratio related to patients leaving before treatment completion?. Methods. Data collection This was a cross-sectional study, based on data collected from a level 1 trauma center, with over 40,000 annual ED visits, in a university hospital in Finland. The PED offers tertiary and primary care for the population under 16-years-old in its catchment area. The data used in this study were. administrative data and information on PED nurse staffing levels between 1 January and 30 June 2019. As in Finland generally, the nurse staffing structure in the ED is two-tiered (Rafferty et al. 2019), and the nursing staff including 45 registered nurses and four practical nurses. Durations of the nursing staff’s shifts varied between 8 and 12 hours.. This article is protected by copyright. All rights reserved.

(8) Accepted Article. Nurse staffing hours were obtained from the PED daily payroll documents, which include data on individual nurses’ work shifts, with their exact working time and a code indicating the type of employment for which they are being paid (e.g., direct care, sick leave, or education).. Administrative patient data were extracted from the hospital’s electronic health records. These include demographic information, such as each patient’s gender and age, the day of the week and time of day when the patient presented to the PED, and the patient’s time of discharge from the. PED. In addition, information on each patient’s triage classification and PED LOS were. gathered. Both the nurse staffing and patient datasets were complete. Sample size The data covered all (22,517) visits to the Children Hospital’s emergency department between 1 January and 30 June 2019, and numbers of the 49 nurses working each of the shifts (543 in total), during that time. Patients in the ED’s observation status bed in each shift were excluded because their care process differs from that of other PED patients (inter alia, they had a longer LOS: mean 399 minutes), which may have biased the results. Following their exclusion, the patient data covered 21,956 visits. Calculations made before statistical analysis Daily nursing hours were determined from nursing staff records for registered nurses and licensed practical nurses then summed for each shift (0.00-7.59 am, 8 am-3:59 pm, and 4 pm11.59 pm). The total number of shifts during the 181-day study period was 543. When calculating nursing hours, only actual bedside hours were counted, and education time, vacation time, and times covered by codes describing anything other than direct patient care were excluded (Emergency Nurses Association 2018). The hours of the triage nurses and charge nurse were included in the count because they are involved in the patients’ direct care. The nursepatient ratio was calculated for every shift by dividing the patient volume by the reported nurse staffing, and for further analysis a nursing ratio was assigned to each patient's data. As the. proportion of licensed practical nurses in the total workforce was small, and usually either none or one was working during a shift, hours worked by all nurses were included when calculating nurse-patient ratios. Patients’ PED LOS was calculated in minutes between the time of presenting to the PED and the time of either their discharge or admittance to a ward. Next,. This article is protected by copyright. All rights reserved.

(9) Accepted Article. descriptive statistics were calculated, including daily staffing numbers per shift, mean numbers of arriving patients, triage (acuity) classification, and admitted patients per shift. Data analysis As the data did not meet normality of distribution criteria for parametric tests, the significance of differences between shifts in numbers of patients’ acuity and numbers of admissions were analyzed using the chi-square test. The significance of differences between shifts in nurse-patient ratio was gauged using the Kruskall Wallis test. Predictors of patients’ LOS were then explored. by linear regression analysis. For this, the triage classification variable was recoded into a twoclass variable: emergency severity index (ESI) triage levels 1–3 were re-coded as ‘high acuity’, and levels 4–5 as ‘low acuity’ (Green et al., 2012). In addition, variable patient arrival time in shifts (3 groups variable) was re-coded into two binary-class variables; variable day arriving (1= day, 0=other) and variable night arriving (1=night, 0=other), evening arriving was used as a. reference group. The non-parametric Mann Whitney U test and chi-square test was applied to. assess differences between patients who LBTC and those who completed their care in terms of: nurse-patient ratio during their visit, LOS and acuity level, and arrival time at ED. The. significance threshold was set at p < 0.05 for all tests, and statistical analyses were performed with IBM SPSS Statistics 23 software for Windows (IBM Corp., Armonk, N.Y., USA).. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed throughout the study (Supplementary File 1). Ethical considerations Ethical approval for the study was granted by the Research Ethics Committee of the Hospital District of Northern Savo (permission number 205/2015, supplemented 6/2019). The participating hospital also granted permission for this study.. Results Characteristics of children’s ED visits Between 1 January and 30 June 2019, number of children who visited the PED per day varied between 40 and 181 (mean, 124; 21,956 children in total). The patients’ mean age was 5.5 (SD. This article is protected by copyright. All rights reserved.

(10) Accepted Article. 4.6) years. The most of the patients (52%) attended the PED during the evening (4-11:59 pm), 41% arrived during the daytime (8-3:59 pm), and 7% during the night (12-7:59 am). In total,. 70% of the patients, (n = 15,278) were classified as low acuity (ESI levels 4–5), and the rest (n = 6678) as high acuity (ESI levels 1–3). The proportion of acute patients was higher at night (n =553, 36%) than during the day (n=3116, 34%) and evening (n=3009, 26%), (𝑥 2 = 173.8, df 2, p <.001). Nearly 10% of the patients (n = 2081) were admitted to the hospital. The proportion of. patients admitted to the hospital was highest (n = 192, 12%) between midnight and eight in the morning, (𝑥 2 = 20.9, df 2, p < .001; Table 1). Nurse-patient ratios The nurse-patient ratio varied between the daily shifts (Figure 1). The ratio were both highest and most variable in night shifts (𝑥 2 = 5615.19, df 2, p < .001; Table 2), during which there were two peaks with more than five nurses per patient, one in February and one in June, and on average there were three nurses per four patients (mean ratio, 0.75; range, 0.3-5.3). During the day and evening shifts there was one nurse per five and six patients, respectively, on average (mean ratios, 0.2 and 0.17, with ranges of 0.1-0.6 and 0.1–0.8, respectively). Nurse-patient ratios relation to length of stay Patients’ mean ED LOS was 135 minutes (range 1-1699 minutes). Linear regression analysis. clearly showed (𝑅 2 =.214, DF = 5, F = 1225.3, p < .001; Table 3) that high acuity (relative to low acuity), hospital admission, and increases in patients’ age were all associated with increases in patients’ LOS (β = .384, .133 and. 0.30, respectively; p < .001 in each case). In addition, the nurse-patient ratio was negatively associated with PED LOS (β = -.020, p = .005), indicating that a unit increase in the number of nurses decreased the patients’ ED LOS by 2% on average. Relation between nurse-patient ratio and patients leaving before treatment completion Of all patients, 0.6% left before treatment was completed (n = 129). The mean LOS of patients. who did not complete their treatment was 215 minutes (SD 224 minutes). During the work shifts coinciding with PED visits of these patients, there was a significantly lower (Z=-3.318, p<.001) nurse-patient ratio, one nurse per six patients (mean .16, SD=.12) on average, than in shifts. coinciding with visits of the other patients, one nurse per five patients (mean .20, SD=.16), as shown in Table 4. They also had a significantly longer LOS (Z=-3.731, p < .001), on average,. This article is protected by copyright. All rights reserved.

(11) Accepted Article. were fewer classified in triage as high acuity patients (𝑥 2 =18.217, df 1, p < .001), and visited the PED in the evening more often than those who completed their care (𝑥 2 = 11.760, df 2, p < .003; Table 4). Discussion The results of this study on nurse staffing in a PED provide new insights into the relationships between a key structural variable (nurse staffing) and quality related process parameters in pediatric emergency care In the evening, over half of the patients attended to PED. A high frequency of evening arrivals,. such as this, is typical in PEDs (Barata 2015). However, we also found that just 26% of patients were classified as high acuity during the evenings, compared to more than 30 percent during evening and day shifts. The large numbers of patients in the evenings were also reflected in the nurse-patient ratios, which were lowest in the evening (approximately one nurse per six patients), and highest at night (almost one to one). Thus, although there were more nurses, there were still. deficiencies in staffing in the evening, in relation to both average diurnal levels in the studied PED, and the minimum Californian level of one nurse per four patients (Aiken et al., 2010). A nurse increase in the nurse-patient ratio was associated with a two percent decrease in patients' LOS in PED. This is a novel finding in pediatric emergency care. Even though the two percent decrease is rather small, this result reflects the significance of nurse staffing on the patients' flow. However, the findings indicate that increasing the number of nursing staff is not the only solution. for balancing the daily workload. Another is to allocate nursing staff to shifts based on patient flow and acuity, with a regular (perhaps quarterly) review of the success of the approach. The results show that it is not advisable to determine nurse-patient ratio only once a day in pediatric emergency care settings, as in a previous study of an adult ED (Ramsey, 2018). As patient flows often peak in the evenings, 24-hour counts may lead to incorrect conclusions, and shift-based counts would provide more accurate indications of optimal staff allocation patterns. There was also substantial daily variation in nurse-patient ratios in all shifts. However, the peaks were highest in the night shifts, especially in February and May, probably due to low patient flows and because EDs must be adequately prepared to care for critical patients at all times of the day. Thus, sufficient nursing levels must be maintained, but EDs should be prepared for changes in. This article is protected by copyright. All rights reserved.

(12) Accepted Article. patient flows, for example, through maintaining some form of buffer resources or equalizing inhospital nurse resources. The study confirmed the relation between nurse staffing and the number of patients who LBTC (Andersson et al., 2015; Ramsey et al., 2018). More precisely, there were six patients per nurse,. on average, during visits of these patients and five per nurse during other patients’ visits. Nurses having a greater number of patients to care for will inevitably lead to shorter care contacts with the patients and their families (in the absence of mitigating factors), as well as longer waiting times for patients in the care process. This may motivate the patients and their families to leave the PED before treatment is completed. In addition, the results confirmed earlier findings (Tropea et al., 2012) of associations between LBTC and: evening arrival, triage classification as a low-acuity patient, and a relatively long LOS. Consideration of these factors could facilitate identification of effective strategies to reduce numbers of patients who LBTC. For example, an electronic health record system that provides automated reminders of the extended LOS of lowacuity patients in the evening could enhance staff’s awareness, of them and efforts to streamline the process for them. Conclusions Nurse staffing is a key structural quality of care-related parameter as it is linked to important process variables, including LOS and factors associated with numbers of patients who LBTC in pediatric emergency care settings. In conjunction with administrative data, unit-level shift-based nurse-patient ratios can provide valuable insights into nurse staffing requirements and important indications of optimal staff allocation patterns. Limitations This cross-sectional study considered emergency care data covering just six months from one. PED. A longitudinal, multicenter study might have detected patterns that were not apparent in our dataset, such as seasonal variations in patient flow, and yielded more generalizable results. It should also be noted that there may be variations in the accuracy of triage, especially for pediatric patients, and thus affect the results obtained. In addition, this study considered only. nurse staffing, rather than levels of engagement of all the multiple actors with different. This article is protected by copyright. All rights reserved.

(13) Accepted Article. professional backgrounds that are involved in the care process, and thus warrant detailed attention in future analyses. Relevance to clinical practice Findings from this study suggest that a lower nurse-patient ratio could reduce LOS and LBTC. Therefore, PED nurse managers should focus on the nurse-patient ratio, and particularly its daily fluctuations. Instead of simply increasing numbers of nursing staff to target levels based on patient acuity and flows, it may be more effective in some cases to reallocate nursing staff to other shifts. Because the nurse-patient ratio affects the care process, it should be used together. with data on other process variables when assessing care quality in pediatric emergency departments.. References Aiken, L.H., Sloane, D.M., Cimiotti, J.P., Clarke, S.P., Flynn, L., Seago, J.A., Spetz, J. and Smith, H.L. (2010), Implications of the California Nurse Staffing Mandate for other States. Health Services Research, 45, 904-921. doi:10.1111/j.1475-6773.2010.01114.x. Aiken, L. H., Sermeus, W., Van den Heede, K., Sloane, D. M., Busse, R., McKee, M., Bruyneel, L., Rafferty, A. M., Griffiths, P., Moreno-Casbas, M. T., Tishelman, C., Scott, A., Brzostek, T., Kinnunen, J., Schwendimann, R., Heinen, M., Zikos, D., Sjetne, I. S., Smith, H. L., & Kutney-Lee, A. (2012). Patient safety, satisfaction, and quality of hospital care: cross sectional surveys of nurses and patients in 12 countries in Europe and the United States. BMJ, 344, e1717. https://doi.org/10.1136/bmj.e1717. American Academy of Pediatrics, Committee on Pediatric Emergency Medicine, American College of Emergency Physicians, Pediatric Committee, Emergency Nurses Association Pediatric Committee. (2009). Joint Policy Statement—Guidelines for Care of Children in the Emergency Department. Pediatrics, 124 (4) 1233-1243. doi: 10.1542/peds.2009-1807. Alessandrini, E., Varadarajan, K., Alpern, E.R., Gorelick, M.H., Shaw, K., Ruddy, R.M., Chamberlain, J.M. (2011), Emergency Department quality: An analysis of existing. This article is protected by copyright. All rights reserved.

(14) Accepted Article. pediatric measures. Academic Emergency Medicine, 18, 519-526. doi:10.1111/j.15532712.2011.01057.x. Anderson, D., Pimentel, L., Golden, B., Wasil, E., & Hirshon, J. M. (2015). Drivers of ED efficiency: A statistical and cluster analysis of volume, staffing, and operations. American Journal of Emergency Medicine, 34(2), 155–161. doi:10.1016/j.ajem.2015.09.034. Barata, I., Brown, K. M., Fitzmaurice, L., Griffin, E. S., Snow, S. K., . . . Emergency Nurses Association Pediatric Committee. (2015). Best practices for improving flow and care of pediatric patients in the emergency department. Pediatrics, 135(1), e273–e283. doi:10.1542/peds.2014-3425. Donabedian, A. (1988). The quality of care: How can it be assessed? JAMA, 260(12), 1743– 1748. doi:10.1001/jama.1988.03410120089033. Duffield, C., Diers, D., O'Brien-Pallas, L., Aisbett, C., Roche, M., King, M., & Aisbett, K. (2011). Nursing staffing, nursing workload, the work environment and patient outcomes. Applied Nursing Research, 24(4), 244–255. doi://doi.org/10.1016/j.apnr.2009.12.004. Emergency Nurses Association (2018). Position statement: Staffing and productivity in the emergency department. Retrieved from https://www.ena.org/docs/defaultsource/resource-library/practice-resources/positionstatements/staffingandproductivityemergencydepartment.pdf?sfvrsn=c57dcf13_6 26.2.2020. Green, N. A., Durani, Y., Brecher, D., DePiero, A., Loiselle, J., & Attia, M. (2012). Emergency Severity Index version 4: A valid and reliable tool in pediatric emergency department triage. Pediatric Emergency Care, 28(8), 753–757. doi:10.1097/PEC.0b013e3182621813. Hofer, K. D., & Saurenmann, R. K. (2017). Parameters affecting length of stay in a pediatric emergency department: A retrospective observational study. European Journal of Pediatrics, 176(5), 591–598. doi:10.1007/s00431-017-2879-y. This article is protected by copyright. All rights reserved.

(15) Accepted Article. Lordache S., Elseviers M., De Cock R., van Rompaey B.(2020). Development and validation of an assessment tool for nursing workload in emergency departments. Journal of Clinical Nursing, 29, 794 – 809.. Michelson, K. A., Stack, A. M., & Bachur, R. G. (2016). Development of a model to measure emergency department staffing limitations. Pediatric Emergency Care, 32(9), 599–602. doi:10.1097/PEC.0000000000000892. Min, A. & Scott, L. D. (2016). Evaluating nursing hours per patient day as a nurse staffing measure. Journal of Nursing Management, 24(4), 439–448. doi:10.1111/jonm.12347. Nelson, D., Hearld L .R. & Wein D. (2018). The impact of emergency department RN staffing on ED patient experience. Journal of Emergency Nursing, 44(4), 394–401. doi:S00991767(17)30255-6. Otegbeye, M., Scriber, R., Ducoin, D. & Glasofer, A. (2015). Designing a data-driven decision support tool for nurse scheduling in the emergency department: A case study of a southern New Jersey emergency department. Journal of Emergency Nursing, 41(1), 30– 35. doi:10.1016/j.jen.2014.07.003. Paulsen, R. A. (2018). Taking nurse staffing research to the unit level. Nursing Management, 49(7), 42-48. doi:10.1097/01.NUMA.0000538915.53159.b5. Pitkäaho, T., Partanen, P., Miettinen, M. H., & Vehviläinen-Julkunen, K. (2016). The relationship between nurse staffing and length of stay in acute-care: A one-year timeseries data. Journal of Nursing Management, 24(5), 571–579. doi:10.1111/jonm.12359. Rafferty AM, Busse R, Zander-Jentsch B, et al., editors. (2019). Strengthening health systems through nursing: Evidence from 14 European countries. Copenhagen (Denmark): European Observatory on Health Systems and Policies. Health Policy Series, No. 52. Available from: https://www.ncbi.nlm.nih.gov/books/NBK545724/ 5.5.2020. Ramsey, Z., Palter, J. S., Hardwick, J., Moskoff, J., Christian, E. L., & Bailitz, J. (2018). Decreased nursing staffing adversely affects emergency department throughput metrics. The Western Journal of Emergency Medicine, 19(3), 496–500. doi:10.5811/westjem.2018.1.36327. This article is protected by copyright. All rights reserved.

(16) Accepted Article. Shindul-Rothschild, J., Read, C. Y., Stamp, K. D., & Flanagan, J. (2017). Nurse staffing and hospital characteristics predictive of time to diagnostic evaluation for patients in the emergency department. Journal of Emergency Nursing, 43(2), 138–144. doi://doi.org/10.1016/j.jen.2016.07.003. Tropea, J., Sundararajan, V., Gorelik, A., Kennedy, M., Cameron, P., & Brand, C. A. (2012). Patients who leave without being seen in emergency departments: An analysis of predictive factors and outcomes. Academic Emergency Medicine, 19(4), 439–447. doi:10.1111/j.1553-2712.2012.01327.x. Wolf, L., Perhats, C., Delao, A., Clark, P & Moon, M. (2017). On the threshold of safety: A qualitative exploration of nurses’ perceptions of factors involved in safe staffing levels in emergency departments. Journal of Emergency Nursing, Volume 43 (2), 150 – 157.. Tables and figures. Table 1. Descriptive statistics of visits to the focal PED, by shifts (n, %) (N = 21956). P-values are according to chi-square test. Characteristics Day. Total number of. Evening. Night. 8 am – 3.59. 4 pm – 11.59. 0 am – 7.59. pm (n, %). pm. am. (n, %). (n, %). 11362 (51.7%). 1533 (7.0%). 9061 (41.3%). Total (n). P value. 21956. arrived patients per shift. Patients’ acuity Number of high-. <. 001 3116 (34.1%). 3009 (26.5%). 553 (36.1%). 6678 (30.4%). 5945 (65.9%). 8353 (73.5%). 980 (63.9%). 15278 (69.6%). acuity patients Number of lowacuity patients. This article is protected by copyright. All rights reserved.

(17) Accepted Article. <. 001. Admission Admitted to hospital. 875 (12.5%). 1014 (9.7%). 192 (9.5%). 2081 (9.5%). Discharged from. 8186 (87.5%). 10348 (90.3%). 1341 (90.5%). 19875 (90.5%). PED. This article is protected by copyright. All rights reserved.

(18) Accepted Article. Table 2. Nurse staffing, numbers of patients and nurse-patient ratios by shifts (mean, SD, minmax) (n=543). Shifts. Mean (SD). Min - Max. 8.8 (.7). 7.3 - 11.0. Registered nurses. 7.8 (0.8). 6.3 – 9.5. Licensed practical nurses. 1.0 (0.5). 0.0 – 2.3. Number of patients. 50.1 (15.6). 16.0 - 104.0. Nurse-patient ratio. .2 (.1). .10 - .60. 9.9 (.5). 8.0 - 11.7. Registered nurses. 9.5 (0.6). 7.5 – 10.9. Licensed practical nurses. 0.4 (0.4). 0.0 – 1.7. Number of patients. 62.8 (12.3). 13.0 - 86.0. Nurse-patient ratio. .17 (.09). 0.1 - .8. 5.3 (.09). 4.80 - 5.50. Registered nurses. 5.2 (0.1). 4.2 – 5.3. Licensed practical nurses. 0.1 (0.10). 0.0 – 1.0. Number of patients. 8.5 (3.13). 1.00 - 17.00. Nurse-patient ratio. .75 (.46). .30 - 5.30. Day (8 am – 3:59 pm) Number of nursing staff. Evening (4 pm – 23:59 pm) Number of nursing staff. Night (0 am – 7:59 am) Number of nursing staff. Table 3. Linear regression model of predictors for ED length of stay (R2 = .214, DF=5, F =1225.3, p< .001). Predictor. B (95% Cl). SD. Patient's high acuity triage. 105.813 (102.149,. level (ref. low acuity). 109.477). Admitted (ref. discharged). t. p-value. 1.869 0.384. 56.607. <0.001. 57.535 (51.810, 63.259). 2.921 0.133. 19.698. <0.001. Age (years). 0.795 (.481, 1.108). 0.160 0.030. 4.963. <0.001. Nurse-patient ratio (per shift). 16.504 (28.004, 5.004). 5.867 -0.020. -2.813. 0.005. This article is protected by copyright. All rights reserved. Beta.

(19) Accepted Article. Arrival time (ref. evening 4 pm – 23:59) Day (8 am – 3:59 pm). 2.476 (5.345, .394). 1.464 0.009. 1.431. 0.153. Night (0 am – 7:59 am). 7.558 (-1.762, 16.877). 4.755 0.015. 1.590. 0.112. Table 4. Comparison of variables associated with patients who left before treatment completion (LBTC) (n=129) and patients who completed the care (n= 21 827) P-values are according to Mann-Whitney U test (*) and chi-square test (†) Variable LBTC. Patients who completed the. Mean (SD). p-value. care Mean (SD). Nurse-patient .16 (.12). .20 (.16). .001*. 133.6 (125.9). .001*. Patients who completed the. P-value. ratio. ED length of stay 214.8 (224.5) Variable LBTC (n, %). care (n, %). Patients’ acuity Low acuity 112 (87%) Arriving time Day (shifts) 38 (29%). High acuity. Low acuity. High acuity. 17 (13%). 15166 (69%). 6661 (31%). Evening Night. Day. Evening Night. 86. 5. 9023. 11276. 1528. (67%). (4%). (41%). (52%). (7%). This article is protected by copyright. All rights reserved. .001(†) .003(†).

(20) Accepted Article. Figure 1. Nurse-patient ratios from 1.1.2019 to 30.6.2019 by shifts (n = 543).. This article is protected by copyright. All rights reserved.

(21)