How are price-demand curves for pharmaceuticals determined?
By Gary Johnson, expert-trainer of the 2-day course Value Pricing for Market Access - The Fundamentals.
The pharmaceutical industry uses several decision-support tools and techniques in the process of determining the optimum price range for a pharmaceutical. These include conjoint analysis, the Gabor and Granger technique, the van Westendorp Price Sensitivity Meter and Monadic Price Testing.
They have all been shown to have the potential to produce misleading and biased results.
For example, take the Monadic Price Test – which is often claimed to be the “Rolls Royce” of pricing research techniques. Respondents are shown a product profile. All respondents are shown exactly the same profile except that different respondents are shown different prices. So, for example:
- 20 respondents are shown product X with a price of €5
- 20 respondents are shown product X with a price of €10
- 20 respondents are shown product X with a price of €15
- 20 respondents are shown product X with a price of €20
Each respondent is asked whether they would reimburse/use (as appropriate to the survey) the product. We then plot how many respondents would reimburse/use the product at each price. This is supposed to represent the price-demand function for the product.
The problem is that the questions we ask are hypothetical questions. And, one of the most studied and certain effects in survey research is the so-called “hypothetical bias”: the responses to hypothetical questions are systematically biased. This means that the price-demand curve is systematically biased.
Gary Johnson's value-pricing course explains what this bias is and why it occurs.
Last update: March 2017