Preparing epidemiology data for use in a forecast model - do you know your target population?

Use our guide to check your thinking.

Identifying patient segment(s):

  • Do you know the indication for the product in sufficient detail for a forecast?
  • Are there specific inclusion or exclusion criteria mentioned in the label?
  • Are there specific inclusion or exclusion criteria mentioned in the label?
  • Are these mentioning specific comorbidities (e.g. diabetes, atrial fibrillation, COPD etc)
  • Are these mentioning specific clinical parameters? (e.g. blood pressure, kidney function, ejection fraction?)
  • Are these mentioning specific drug interactions / contra-indications with other medications? (e.g. anti-fungals, statins, ace inhibitors etc)
  • Are there any age or gender exclusions mentioned? (e.g. not suitable for women of child bearing age or under 12 year so of age?)
  • Does the label mention restriction to populations treated with specific prior drug treatments or specific line of therapy, or those refractory to certain treatment options?
  • Do you have a suitable source to go to for epidemiology data?
    Does the source have data for the head-line epidemiology in a suitable format?
  • Does the epidemiology data set extend across the time period needed for the forecast model?
  • Does the data set include parameters / subpopulations that allow suitable identification of patients to align with the label indication?
  • Has the data been based on country specific sources or extrapolated from another country?
  • What methodology was used to construct the dataset?
  • How was the data projected for the forecast period?
    Have impacts of medications / vaccinations / diagnostics been included in the data set? (where relevant)