HCS 493 Week 2 Data Tables and Graphic Displays Example

Four Percent Walked Out: Building the Tables and Charts That Show a Hospital Board When and Why Emergency Patients Leave Before Being Seen

[Student Name]

University of Phoenix

HCS/493: Data Analytics for Health Care Managers

Week 2 Assignment

[Instructor Name]

[Date]

The hospital and its visit counts are composites written for a model paper; the national rate comes from the public CMS data set listed, and display principles come from the sources cited.

After a patient wrote to the local newspaper that she had waited four hours with a broken wrist and gone home untreated, the board of a composite 210-bed community hospital asked a simple question: how many people leave our emergency department without being seen, and when does it happen? The analytics team had the raw data, a count of visits and walkouts for each month and each hour of arrival, but raw counts in a spreadsheet would not answer the question in a board meeting. This paper builds the tables and charts, explains each design choice and answers the board's interpretation questions.

The Data

Over 12 months the department recorded 41,600 visits. Of these, 1,712 patients registered and left before a physician, nurse practitioner or physician assistant examined them, a group the hospital's measure calls left without being seen. For each visit the data include the month and hour of arrival and the triage level. Month and hour are the grouping variables; the outcome is whether the patient left.

Why Rates, Not Counts

Monthly visit volume ranged from 3,080 in June to 3,910 in January. A month with more walkouts may simply have had more patients. Dividing walkouts by visits and multiplying by 100 gives a rate per 100 visits, which can be compared across months of different sizes. The overall rate was 4.1 per 100 visits, or 4.1%.

Table 1: Walkouts by Month

Table 1 lists, for each month, visits, walkouts and the rate. Selected rows show the pattern:

January: 3,910 visits, 219 walkouts, 5.6%.

February: 3,720 visits, 190 walkouts, 5.1%.

May: 3,310 visits, 96 walkouts, 2.9%.

June: 3,080 visits, 83 walkouts, 2.7%.

October: 3,420 visits, 117 walkouts, 3.4%.

December: 3,780 visits, 181 walkouts, 4.8%.

The table follows APA style: a number and title above, clear column heads with units and one decimal place for every rate.

Table 2: Walkout Rate by Hour of Arrival

Table 2 groups visits into four-hour blocks. Arrivals from 7:00 to 10:59 had a rate of 1.9%. From 11:00 to 14:59, 3.8%. From 15:00 to 18:59, 6.2%. From 19:00 to 22:59, 7.4%. From 23:00 to 2:59, 3.1%, and from 3:00 to 6:59, 1.2%. Four-hour blocks were chosen because hourly rows produced a table too long to read in a meeting, while two blocks per day hid the evening peak.

Figure 1: Line Chart of the Monthly Rate

Figure 1 plots the monthly rate as a line, with months on the horizontal axis and the rate per 100 visits on the vertical axis, starting at zero. A dashed horizontal line marks the national rate of 2% reported by the Medicare agency for 2024 (Centers for Medicare & Medicaid Services, 2025). A line chart suits this display because the months are ordered in time and the reader should see the rise into winter and the fall into summer as a shape.

Figure 2: Bar Chart by Hour of Arrival

Figure 2 shows the six arrival blocks as vertical bars in time order, with the rate on the vertical axis starting at zero. Bars suit separate categories, and starting at zero keeps the evening bars from looking more extreme than they are. Schwabish (2014) recommended showing the data clearly, reducing clutter and integrating text with the graphic, so each bar carries its value on top and the chart title states the finding: evening arrivals are most likely to leave.

Why Not a Pie Chart or a Table Alone

A draft used a pie chart of walkouts by arrival block. It was dropped because a pie shows each block's share of all walkouts, not the chance that a patient arriving in that block will leave, and readers struggle to compare slices of similar size. Gelman et al. (2002) argued that many tables in published research would communicate better as graphs, because patterns and comparisons are easier to see. The tables remain in the appendix for board members who want exact values, while the charts carry the message.

Interpretation: What Do the Displays Show?

Three findings stand out. First, the department's rate of 4.1% is about double the national figure of 2%. Second, the rate is seasonal, peaking at 5.6% in January and falling to 2.7% in June, in line with respiratory illness and visit volume. Third, the rate is highest for patients arriving between 15:00 and 23:00, reaching 7.4% in the evening block, roughly six times the rate for early-morning arrivals. The problem is concentrated in winter evenings, which is a staffing and flow question, not a general failure of the department.

Interpretation: What Might Explain the Pattern?

The displays show when walkouts happen, not why. Possible explanations include provider staffing that drops after 19:00 while arrivals remain high, boarded inpatients occupying treatment rooms in the evening and winter volume. A further table comparing provider hours with arrivals by hour would test the first explanation.

Interpretation: How Would a Manager Use This?

The emergency department manager will use Figure 2 in a request to extend a midlevel provider shift from 19:00 to 23:00 during December through March and will add a nurse-initiated protocol for common tests in the waiting room. The monthly line chart will be updated each month on the board's dashboard with the national line, so progress is visible.

Accuracy Checks

Before the displays went to the board, the analyst confirmed that monthly walkouts summed to the annual total, that every rate used the same definition and that the national figure used the same measure. Two months had been entered with walkouts and visits reversed in the original spreadsheet, an error the rate column exposed because it produced impossible values above 100.

Conclusion

The same 1,712 walkouts looked like an unexplained total in a spreadsheet. Converted into rates, organized into two short tables and displayed as a line chart against the national figure and a bar chart by hour, they showed a department with a seasonal, evening-concentrated problem about twice the national rate. That picture gave the board a question it could act on and the manager a targeted request.

References

Centers for Medicare & Medicaid Services. (2025). Timely and effective care: National [Data set]. https://data.cms.gov/provider-data/dataset/isrn-hqyy

Gelman, A., Pasarica, C., & Dodhia, R. (2002). Let's practice what we preach: Turning tables into graphs. The American Statistician, 56(2), 121-130. https://doi.org/10.1198/000313002317572790

Schwabish, J. A. (2014). An economist's guide to visualizing data. Journal of Economic Perspectives, 28(1), 209-234. https://doi.org/10.1257/jep.28.1.209