DDHA 8703 Week 5 Descriptive Analysis Example

More Low-Acuity Arrivals on Weekend Late Mornings, Described and Not Explained: A Descriptive Analysis of a Year of Emergency Department Arrivals Across a Four-Hospital Network

Student Name

Walden University

DDHA 8703: Advanced Health Analytics and Data-Driven Decision Making

Instructor Name

Month Day, Year

More Low-Acuity Arrivals on Weekend Late Mornings, Described and Not Explained: A Descriptive Analysis of a Year of Emergency Department Arrivals Across a Four-Hospital Network

Data and Approach

This analysis describes emergency department arrivals across the four hospitals of Northgate Health, a composite health system, from January through December 2025. It uses the encounter dataset audited in the previous week, after removing 1,140 duplicate encounters at one hospital and 37 records with impossible times, leaving 181,223 encounters. Patients who left before triage are kept in arrival counts and shown separately in acuity tables, since they have no acuity level by design. Acuity follows the five-level triage scale used in all four departments, grouped here as high (levels 1 and 2), middle (level 3) and low (levels 4 and 5). The analysis reports what the tables show and does not argue for any explanation.

Arrivals by Day of the Week

Monday has the highest median and the widest range of any day. Weekend days have the lowest medians, but their spread is similar to midweek days, and the busiest Sundays exceed the quietest Tuesdays. Reported as averages alone, the weekly pattern would look like a smooth decline from Monday; the percentiles show that individual weekend days are less predictable than that suggests. McCarthy et al. (2008), modeling emergency department demand at one academic center, found that arrivals followed strong hour-of-day and day-of-week patterns but that individual days still varied considerably, which is consistent with what appears here.

Table 1

Daily Arrivals Across the Network, by Day of the Week, 2025

Counts combine all four departments; percentiles describe the spread of the 52 or 53 days of each weekday in the year.

Acuity Mix on Weekdays and Weekends

Low-acuity arrivals make up about a third of weekend arrivals, compared with just over a quarter on weekdays. High-acuity arrivals are a slightly smaller share on weekends. The share of patients who left before triage is the same in both. In absolute terms, weekend days bring about 160 low-acuity arrivals across the network, against about 138 on weekdays, so weekends bring more low-acuity patients in absolute terms as well as in share.

Table 2

Share of Arrivals by Acuity Group, Weekdays and Weekends, 2025

Shares are of all arrivals in each group of days; the left-before-triage category is shown because those patients have no acuity level by design.

Arrivals by Hospital

The three larger departments show the same weekly shape at different scales. The flagship department had a median of 186 arrivals a day, the second 111 and the hospital acquired in 2023 a median of 169 after duplicates were removed. Each had its highest median on Mondays and its lowest on Sundays. The low-acuity share on weekends was highest at the second department, at about 37%, and lowest at the flagship, at about 31%. The acquired hospital's daily counts were the most variable of the three, with a wider interquartile range than its size alone would predict, and the audit's finding that interface failures cluster on particular days may explain some of that spread even after duplicate removal.

Low-Acuity Arrivals Across the Day

On weekend days, low-acuity arrivals rise steadily from 08:00 and peak between 10:00 and 14:00, when they account for about 44% of all arrivals in those hours. On weekdays, low-acuity arrivals peak later, in the early evening, at about 33% of arrivals. The weekend late-morning concentration appears at all three larger hospitals. Overnight, low-acuity shares are similar on weekdays and weekends.

Subgroups Reported and Omitted

Patterns by hospital are reported for the three larger departments. The fourth, a rural department averaging about 30 arrivals a day, is included in network totals but not reported separately by hour and acuity, because many hour-and-acuity cells contain fewer than ten patients across the year and would not support stable shares. For the same reason, pediatric arrivals are not broken down by hour at any single hospital. Arrival mode is not analyzed, because the audit found it missing for about 9% of encounters, concentrated at one hospital during busy hours.

What This Analysis Does Not Show

The tables show when patients arrive and how acute they are on arrival. They do not show why. It would be easy to attribute the weekend late-morning rise in low-acuity arrivals to primary care offices being closed, to urgent care hours, to patients' work schedules or to something else, and any of these may be true. None can be tested with these data, which contain no information about patients' alternatives, reasons for coming or usual sources of care. Hoot and Aronsky (2008), reviewing the causes of emergency department crowding, describe input factors such as non-urgent visits alongside throughput and output factors, and emphasize that causes differ across settings. This analysis identifies a pattern that later work can try to explain with data suited to the purpose.

Limits

The data cover one year, so seasonal effects cannot be separated from the weekly pattern. Acuity is assigned at triage and reflects triage practice as well as patient condition; Welch et al. (2011) note the importance of standard definitions for operational measures, and the four departments use the same scale but may apply it somewhat differently. Arrival counts at the hospital with the duplicate problem depend on the matching rule used to remove them. Finally, the analysis describes arrivals, not demand: patients who were diverted or who chose not to come are invisible here.

References

Hoot, N. R., & Aronsky, D. (2008). Systematic review of emergency department crowding: Causes, effects, and solutions. Annals of Emergency Medicine, 52(2), 126-136. https://doi.org/10.1016/j.annemergmed.2008.03.014

McCarthy, M. L., Zeger, S. L., Ding, R., Aronsky, D., Hoot, N. R., & Kelen, G. D. (2008). The challenge of predicting demand for emergency department services. Academic Emergency Medicine, 15(4), 337-346. https://doi.org/10.1111/j.1553-2712.2008.00083.x

Welch, S. J., Asplin, B. R., Stone-Griffith, S., Davidson, S. J., Augustine, J., & Schuur, J. (2011). Emergency department operational metrics, measures and definitions: Results of the Second Performance Measures and Benchmarking Summit. Annals of Emergency Medicine, 58(1), 33-40. https://doi.org/10.1016/j.annemergmed.2010.08.040