top of page

Emergency Service Triage South Korea

Data Analysis to identify emergency department triage accuracy using the Korean Triage and Acuity Scale (KTAS) and evaluate the causes of mistriage.

Business question: To improve the triage system, how do mistriage cases impact the emergency service?

Tools used: Tableau, SQL Server.

Triage-1.jpg
Tableau_02.jpg
Key Insights

​

  • Mistriage accounts for 14% (186 cases) of the total number of cases, and under triage (10%) is higher than over triage (4%) error. This implies nurses may need better and clearer guidelines when classifying patient's symptoms.

  • Other examples that show the importance for clearer guidelines are:

    • Under triage includes the most serious disposition types, such as death, surgery and intensive care unit (ICU) admission.

    • In the ressuscitation level, apart from the 18 cases classified by nurses, experts classified an additional 11 cases as ressuscitation that nurses classified as emergent, those being under triage.

  • The age groups 55-60 and 75-80 account for highest percentage of people in the emergency department. The first group 55-60 may represent the key age were health problems become more symptomatic and extra care / vigilance are needed.

  • Regardless of the main complaint of each patient, the majority has high or elevated blood pressure, summing 72% of cases. This is a relevant health issue, due to the risks that entail, putting further pressure into the healthcare system.

1. Data Source

​

  • Data source from Kaggle, then linked to the original data set in PLOS Journals (Kaggle Link).

  • This cross-sectional retrospective study was based on 1267 systematically selected records of adult patients admitted to two emergency departments between October 2016 and September 2017.

  • Twenty-four variables were assessed, including chief complaints, vital signs according to the initial nursing records, and clinical outcomes.

  • Three triage experts, a certified emergency nurse, a KTAS provider and instructor, and a nurse recommended based on excellent emergency department experience and competence determined the true KTAS.

  • Triage accuracy was evaluated by inter-rater agreement between the expert and emergency nurse KTAS scores.

  • CSV file with 1267 rows and 24 columns.

2. Data Exploration
 

  • In SQL, the main skills used were Joins, CTE, Windows Functions, Aggregate Functions, Creating Views, Converting Data Types.

  • Get aquainted with the data and performed the main tasks:

    • Check for null values in KTAS_RN and KTAS_expert.

    • Cast age from double to integer.

    • Convert 1 and 2 to female and male.

    • Calculate mistriage and blood pressure calculation.

    • Minimum, average, maximum length of stay in minutes, group and order by disposition.

    • Search for acute diagnosis and compare patients complaint with nurse diagnosis.

  • Creating views to store data for later visualization:

    • All records partition by mistriage disposition.

    • Records under and over triage.

3. Data Visualization
 

  • In Tableau, created parameters, bins and a slider for age variable.

  • The graphs used were: 

    • Pie chart  for showcase gender, blood pressure.

    • Vertical bar chart for age distribution.

    • Horizontal bar chart for disposition cases by over / under triage and lenght of stay.

    • Scatter plot for comparing triage levels given by experts versus nurses.

    • Highlight table with the number of cases by triage level.

    • Heatmap with the distribution and percentage of correct, under and over triage cases.

    • Each vizualization has its specific tooltip.

Other projects

bottom of page