Healthcare Staffing Distribution
Influenza in the United States
Completed April - May 2023 (CareerFoundry)
-
This project features a diverse set of stakeholders, including:
Medical agency frontline staff, hospital and clinic leadership using the agency’s services, influenza patients, and staffing agency administrators.
-
I was provided with various data sets to profile and select as most relevant. Ultimately, the main datasets were:
Influenza Deaths by Geography (CDC)
Population Data by Geography, Time, Age, and Gender (US Census Bureau)
-
Outlining data limitations tied to collection methods and patient demographics.
Performing analysis to extend impact of limited resources
-
Data Profiling
Healthcare Industry Knowledge
Social Equity within Health Outcomes
Domain-Specific Research
Descriptive Analysis
Hypothesis Testing
Data Transformation
Data Privacy
Dashboard design
Geospatial analysis and visualizations
Forecasting
Presenting results to non-technical audiences
Synthesizing analysis into digestible Tableau storyboard
-
Excel, Microsoft Word, Tableau
Context and Objective:
A medical agency that provides temporary staffing across the United States wants to create a staffing plan to prepare for the upcoming (2018) flu season. The goal is to use a descriptive analysis, with a particular emphasis on vulnerable populations. Hospitals need additional staffing during flu season to lower death rates. Insights from the analysis will help determine when to send staff, and how many, to each state.
Medical analyses often balance data ethics, limited resources, and a desire to improve patient outcomes.
Vulnerable populations suffer from the most severe cases of influenza. The agency seeks to meet patient needs based on financial constraints with a limited number of nurses, physician assistants, and doctors on staff. Success factors include:
A staffing plan that utilizes available agency staff per state need, without additional resources.
Minimal instances of understaffing and overstaffing across states.
To meet these goals, stakeholders requested an analysis to:
Provide Information to support a staffing plan, detailing what data can help inform the timing and spatial distribution of medical personnel throughout the United States.
Determine whether influenza occurs seasonally or throughout the entire year. If seasonal, does it start and end at the same time (month) in every state?
Prioritize states with large vulnerable populations. Consider categorizing each state as low-, medium-, or high-need based on its vulnerable population count.
Assess data limitations that may prevent you from conducting your desired analysis
Designing A Research Project
Choosing the right sources and capturing data limitations, particularly related to which demographics of people are counted during government data collection, are essential aspects of data ethics.
I conducted exploratory analysis with large data sets, cleaned data, transformed and integrated data to be usable, then completed statistical analysis. Statistical interpretations led to my central hypothesis, "If a state has a higher proportion of the population of people 65 years and older than people 5 to 64 years old, then the influenza death rate in that state will increase." I summarized these steps as an interim report.
I utilized Tableau and Excel tools to visualize variables in composition and comparison charts to begin to translate statistical insights.
Tableau scatterplots provided interactivity and an accessible version of hypothesis testing.
Geospatial analysis and treemaps allowed me to demonstrate stand out states, both for high performing patient outcomes and worrisome areas.
Although the analysis involved a complex critical thinking and produced extensive content, it is my job as an analyst to present stakeholders with simplified, relevant, engaging insights. I was able to answer key questions and provide practical recommendations to the staffing agency. I created an interactive storyboard outlining the key insights as well as asking important follow up questions related to equity and patient care.