Limits...
Skåne Emergency Department Assessment of Patient Load (SEAL)-A Model to Estimate Crowding Based on Workload in Swedish Emergency Departments.

Wretborn J, Khoshnood A, Wieloch M, Ekelund U - PLoS ONE (2015)

Bottom Line: The variables Patient hours, Occupancy, Time waiting for the physician and Fraction of high priority (acuity) patients all correlated significantly with the workload assessments.Our model may be applicable to EDs with different sizes and characteristics, and may be used for continuous monitoring of ED workload.Before widespread use, additional validation of the model is needed.

View Article: PubMed Central - PubMed

Affiliation: Department of Emergency Medicine, Skåne University Hospital, Lund, Sweden.

ABSTRACT

Objectives: Emergency department (ED) crowding is an increasing problem in many countries. The purpose of this study was to develop a quantitative model that estimates the degree of crowding based on workload in Swedish EDs.

Methods: At five different EDs, the head nurse and physician assessed the workload on a scale from 1 to 6 at randomized time points during a three week period in 2013. Based on these assessments, a regression model was created using data from the computerized patient log system to estimate the level of crowding based on workload. The final model was prospectively validated at the two EDs with the largest census.

Results: Workload assessments and data on 14 variables in the patient log system were collected at 233 time points. The variables Patient hours, Occupancy, Time waiting for the physician and Fraction of high priority (acuity) patients all correlated significantly with the workload assessments. A regression model based on these four variables correlated well with the assessed workload in the initial dataset (r2 = 0.509, p < 0.001) and with the assessments in both EDs during validation (r2 = 0.641; p < 0.001 and r2 = 0.624; p < 0.001).

Conclusions: It is possible to estimate the level of crowding based on workload in Swedish EDs using data from the patient log system. Our model may be applicable to EDs with different sizes and characteristics, and may be used for continuous monitoring of ED workload. Before widespread use, additional validation of the model is needed.

No MeSH data available.


Related in: MedlinePlus

Correlation between the SEAL model score and workload assessments.Variables in bold are included in the final model.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4470939&req=5

pone.0130020.g001: Correlation between the SEAL model score and workload assessments.Variables in bold are included in the final model.

Mentions: Four of the 14 variables: Patient hours, High priority, Awaiting MD and Occupancy each demonstrated a significant correlation to the assessed workload (Table 2). When analyzed together in a reduced model, these four variables explained 96.4% of the full model based on the r² value, and correlated well with the workload assessments (r² = 0.509, p < 0.001, Fig 1). The correlation between the assessments and Occupancy alone was r2 = 0.334, p < 0.001. A retrospective calculation of the reduced model score (1–6) over the initial collection period (503 h) gave an average score of 3.5, 3.4, 3.3, 2.9 and 3.0 for EDs A-E respectively. The difference in mean score between the validation EDs, Lund (A) and Malmö (B), was not statistically significant (p = 0.084). The SEAL score is calculated by adding 1.589 to the sum of the four variables highlighted (bold) in Table 2, each variable multiplied by its coefficient. Its value varies between 1 and 6, 6 representing the highest workload.


Skåne Emergency Department Assessment of Patient Load (SEAL)-A Model to Estimate Crowding Based on Workload in Swedish Emergency Departments.

Wretborn J, Khoshnood A, Wieloch M, Ekelund U - PLoS ONE (2015)

Correlation between the SEAL model score and workload assessments.Variables in bold are included in the final model.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4470939&req=5

pone.0130020.g001: Correlation between the SEAL model score and workload assessments.Variables in bold are included in the final model.
Mentions: Four of the 14 variables: Patient hours, High priority, Awaiting MD and Occupancy each demonstrated a significant correlation to the assessed workload (Table 2). When analyzed together in a reduced model, these four variables explained 96.4% of the full model based on the r² value, and correlated well with the workload assessments (r² = 0.509, p < 0.001, Fig 1). The correlation between the assessments and Occupancy alone was r2 = 0.334, p < 0.001. A retrospective calculation of the reduced model score (1–6) over the initial collection period (503 h) gave an average score of 3.5, 3.4, 3.3, 2.9 and 3.0 for EDs A-E respectively. The difference in mean score between the validation EDs, Lund (A) and Malmö (B), was not statistically significant (p = 0.084). The SEAL score is calculated by adding 1.589 to the sum of the four variables highlighted (bold) in Table 2, each variable multiplied by its coefficient. Its value varies between 1 and 6, 6 representing the highest workload.

Bottom Line: The variables Patient hours, Occupancy, Time waiting for the physician and Fraction of high priority (acuity) patients all correlated significantly with the workload assessments.Our model may be applicable to EDs with different sizes and characteristics, and may be used for continuous monitoring of ED workload.Before widespread use, additional validation of the model is needed.

View Article: PubMed Central - PubMed

Affiliation: Department of Emergency Medicine, Skåne University Hospital, Lund, Sweden.

ABSTRACT

Objectives: Emergency department (ED) crowding is an increasing problem in many countries. The purpose of this study was to develop a quantitative model that estimates the degree of crowding based on workload in Swedish EDs.

Methods: At five different EDs, the head nurse and physician assessed the workload on a scale from 1 to 6 at randomized time points during a three week period in 2013. Based on these assessments, a regression model was created using data from the computerized patient log system to estimate the level of crowding based on workload. The final model was prospectively validated at the two EDs with the largest census.

Results: Workload assessments and data on 14 variables in the patient log system were collected at 233 time points. The variables Patient hours, Occupancy, Time waiting for the physician and Fraction of high priority (acuity) patients all correlated significantly with the workload assessments. A regression model based on these four variables correlated well with the assessed workload in the initial dataset (r2 = 0.509, p < 0.001) and with the assessments in both EDs during validation (r2 = 0.641; p < 0.001 and r2 = 0.624; p < 0.001).

Conclusions: It is possible to estimate the level of crowding based on workload in Swedish EDs using data from the patient log system. Our model may be applicable to EDs with different sizes and characteristics, and may be used for continuous monitoring of ED workload. Before widespread use, additional validation of the model is needed.

No MeSH data available.


Related in: MedlinePlus