Introduction

I received a task asking me to analysis data for inflammatory bowel disease patients.

Dataset

The data contains information about 33K patients The data columns as follow

• pt_ID: unique ID for each patient
• gender: integer value indicates the gender of each patient
  • 1 means male
  • 2 means female
• diagnosis: integer value indicates the type of IBD
  • 1 means Crohn's
  • 2 means UC
  • 3 means Unspecified
  • 4 means Other
• hosp_admin: integer value indicates if the patient recorded in a hospital or not
  • 1 means Yes
  • 2 means No
  • 3 means Unknown
• age_band_jan1: the age value for each patient
• bmi: contains the body mass for each patient
• smoker: is the patient smoker or not
• past_smoker: is the patient was smoker or not
• diet_preference: which diet the patient follows
• cancer: is the patient has cancer or not
• heart_disease: is the patient has heart disease or not

Questions

The task requires me to answer some questions

• What is the percentage of each BMI category in each age group
• Which type of IBD tends to have sufferers of a higher BMI?
• Which type of IBD has more sufferers that smoke or previously smoked?
• Which type of IBD has more cancer sufferers?
• Which type of IBD has more heart disease sufferers?