Influenza Forecasting

A medical staffing agency, responsible for providing temporary workers to hospitals on an as-needed basis, requires assistance in planning for influenza season.

View Full Project on GitHub

  • As a data analyst, my goal was to analyse influenza trends focusing on vulnerable populations, especially those over 65 years old, to proactively plan for staffing needs across the U.S.

  • This project used 2 datasets:

  • Excel was used for a large portion of this project, from data cleaning, integration, transformation, and statistical hypothesis testing.

    Tableau was used for further visual analyses as well as constructing a final presentation using dashboards.

  • This is a project I completed as part of the data analytics course at CareerFoundry.

Approach and Process

Data Preparation

  • Cleansed both datasets of inconsistencies, duplicate values, and addressed missing values.

  • Integrated datasets using the VLOOKUP function in Excel.

  • Grouped age ranges and normalised influenza mortality rates for data analysis.

Further details of the data transformations and merging can be looked at in this Excel file.

Final integrated dataset, fully prepared for statistical analysis and Tableau analysis.

Data Analysis

  • Hypothesis testing - determined whether the mortality rate for older individuals (65+) was higher than other ages. This helped direct the focus of the analysis on this vulnerable population.

  • Visual analysis - used Tableau Public to determine the seasonality of influenza cases and deaths. Maps were also used to show the distribution of influenza cases and vulnerable populations across the United States, thus leading to recommendations for where to send additional staff.

    • Filters and highlighters were also added to various charts to create a more dynamic final presentation.

Map of the United States highlighting states with the largest 65+ populations and flu deaths among that age group, with a bar graph showing top states by senior population, and a small graph showing total senior population and flu deaths. The US map uses colored circles to indicate the number of flu deaths in each state, and states are highlighted in various shades of blue and red.

Dashboard with map and bar chart comparing vulnerable population totals and deaths across each state.

A web page showing influenza season data with two charts. The first chart displays historical flu death counts in the U.S. from 2009 to 2018, with peaks around 2012 to 2015. The second chart shows total influenza deaths by month, with the highest in January and December, and a death count bar at the bottom indicating over 55,000 overall deaths.

Dashboard showing seasonality of influenza. The year filter allows further analysis into changes overtime.

Presentation and Recommendations

  • Tableau Dashboards were refined and compiled into a storyboard. This storyboard was used in a video presentation for [hypothetical] stateholders at the medical staffing agency.

  • The final dashboard shows a map categorising each states’ staffing priority level assigned by total deaths amongst the vulnerable population in each state.

  • This dashboard also highlights how staffing prioritisation could be aided by providing hospitals’ and clinics’ staff-patient ratios, thus understanding where shortages are truly affecting the quality of care to patients.

A map of the United States showing states color-coded by staffing priority levels for flu-related hospital staffing. States are marked in shades from white to dark red indicating priority from low to high, with a legend at the bottom indicating the number of flu deaths for each priority level.

Final dashboard highlighting findings, recommendations, and ways to improve the analysis in future.

Results and Deliverables

Summary of Results

  • Categorised states into 3 tiers of staffing priority.

  • Identified peak staffing needs will most likely occur from December to March (influenza season).

  • Recommendation for further analysis regarding hospital and clinic staff-patient ratios made.

Deliverables

  • Interim Report: documentation of project progress and findings, including data limitations, descriptive and statisitical analysis, and next steps.

  • Tableau Storyboard: interactive storyboard containing visualisations and recommendations.

  • Video Presentation: Screecast walkthrough of storyboard with explanations of project analysis and highlights.