Visualising the prevalence of the top 5 diseases in the US


#1

Tool: Tableau
Country: US
Diseases:

  • Gonorrhoea
  • Influenza
  • Measles
  • Scarlet Fever
  • Whopping Cough

Year: 1892 - 1974
Visual Mappings:

  • x-axis: Date
  • y-axis: Count of disease incidents
  • colour: Disease type (only top 5 shown)

Data Preparation:

  • Filter the epiweek between the two year points to look specifically at the
    spike in disease occurrences in early 1900s.
  • Aggregate the following data sets together, omitting blank or non-useful fields.

Source: Project Tycho
DOIs:

  • Gonorrhoea 10.25337/T7/ptycho.v2.0/US.15628003
  • Influenza 10.25337/T7/ptycho.v2.0/US.6142004
  • Measles 10.25337/T7/ptycho.v2.0/US.14189004
  • Scarlet Fever 10.25337/T7/ptycho.v2.0/US.30242009
  • Whopping Cough 10.25337/T7/ptycho.v2.0/US.27836007

My question:

  • I notice that pertussis or whopping cough has no values between 1910 and 1923 roughly, what could be the reason for that?
  • Are the choice of colours particularly good? If not which colour mapping would you suggest?
  • Could this visualisation be misleading? If so what could be done to limit this?

#2

Hi Alex,

There are many reasons why you are finding no values for whopping cough between 1910-1923. Wikipedia tells me that it was first discovered in 1906 which might suggest that knowledge of it might not yet have circulated. Records may also be missing due to physical reasons such as poor record keeping, fires, etc… The lesson here is that you should be skeptical of the accuracy of historical data.

For your choice of colours, I would recommend looking at the qualitative selection on ColorBrewer, your choice of colours should be more distinct.