Preparing for Influenza Season

PROJECT BACKGROUND

As analyst, you need technical skills to analyse your data and soft skills to communicate your insights to stakeholders. You’ll start by distilling business requirements and requests into questions you can answer with an analysis. You’ll follow up by sourcing and curating the data to address these questions. After analysing the data and drawing conclusions or formulating recommendations from your results, you’ll present your insights to stakeholders in an easily consumable format.

The United States has an influenza season where more people than usual suffer from the flu. Some people, particularly those in vulnerable populations, develop serious complications and end up in the hospital forcing hospital to overstretch their resources. For this reason hospitals and clinics need additional staff to adequately treat these extra patients. The medical staffing agency provides this temporary staff.

The agency covers all hospitals in each of the 50 states of the United States, and the project will plan for the upcoming influenza season. The main goal is to determine when to send staff, and how many, to each state.

Motivation

Scope and Objectives

What is Influenza? Influenza is a viral infections, often causing fever and aches.

• Data Sourcing , Profiling

• Data Transformation

• Statistical analyses and hypothesis testing

• Data Visualizations:

  - Temporal forecasting

  -Statistical ( histogram, box plot, scatterplot, bubble charts)

  -Comparison (tree map, stacked bar, pie bar) -Spatial analysis

  -Textual Analysis

• Storytelling with data

Techniques

Tools

Dataset

• Excel

• Visualization: Tableau

• Influenza deaths by geography (CDC)

• Population data by geography, time, age, and gender ( Source US census)

• Counts of influenza laboratory test results by state (survey) ( CDC)

• Flu shot rates in children (CDC

Challenges

Influenzas shots data was not related to the general population, therefore measuring the impact of shots as a presentation measure was not possible. Data dated until 2017, lack on more current figures prevail from providing more accurate recommendations.

MARKET ANALYSIS

Research shows that states with a higher proportion of vulnerable population ( people over 65 years old) have higher influenza-related mortality in comparison with states with a lower volume

Hypothesis

Mortality rate for population over 65 years old represents 62.56% ( 411,631) from the total influenza mortality recorded in the Unites States a year

Geographical and seasonal analysis

Data reflects a direct correlation between the states with a higher vulnerable population ( over 65 years old) and the states with a higher mortality rate in the U.S (2009-2017). A monthly analysis can show the fluctuations that reflect seasonality starting in November and significantly dropping in March.

Access to the interactive Dashboard in Tableau the the full list of the top 10 States with higher mortality rate

Conclusions

During this analysis we had studied the correlation between the identified vulnerable population group ( people over 65 years old) and influenza mortality. Confirming a direct correlation between age and influenza mortality.

Even though there is influenza during the entire year 6 and 12 months data shows that there is a seasonality that starts in November spiking in January and ending in March.

Recommendations

Next steps

Based on the size of the country the U.S requires prioritization on the medical staffing based on a series of criteria.

On one side, upon the state mortality rate these are the most impacted states:

1. California and New York ( High mortality)

2. Texas, Illinois, Ohio, Pensilvania ( Medium mortality)

However, to the mortality rate we need to include another factor that is the state´s population size:

3. Florida (high population)

Further analysis is needed to understand the correlation between the flu shots and their impact as a preventive tool.

Additionally, an analysis on further social economical elements might be affecting the country having into consideration that the US lacks on universal health care and private medical attention could influence data.