Life and work are full of decisions — many small decisions and a few big decisions. One of the big decisions drug and biotech manufacturers make is selecting analogs (sometimes spelled “analogues”) for forecasting purposes, spanning a potential array of predictions (e.g., provider prescribing, patient requests, or payer and health system formulary management). In turn, these forecasts can impact licensing decisions, decisions related to pricing and contracting, and trial design decisions. Selecting analogs, then, can be thought of as an intermediate step for high-stakes decisions.
Games give us opportunities to practice making strategic and tactical decisions. And unless betting is tied to the outcomes, these are low-risk decisions. Scrabble is a board game with broad appeal, and playing Scrabble can be a useful metaphor for selecting analogs. [Speaking of games, did you spot the anagram in the last sentence?]
The right analogs
The tiles a Scrabble player uses during any given round depend on the tiles available in his or her hand and those already played. The spaces available around words that have been played must also be considered, as should the availability of bonus spaces. Bottom line: Context matters. Similarly, analog selection should consider the context — the variables that not only vary but that impact decisions as well.
Too often, analogs are mistakenly constrained to competing products that are structurally similar or used for the same indications. Another common mistake is to limit analogs to products with shared characteristics, like dosing or mechanism of action, when these characteristics only correlate with, but are not causally related to, stakeholders’ decisions. Perceived clinical and economic benefits and costs are important drivers, but the right analog typically has more to do with the market context (e.g., the order of market entry, the competitiveness of the market, levels of need, demand or evidence) than product characteristics. After all, every market event, including the entry of a new product, changes the market context to some degree.
The right number of analogs
Scrabble players must decide which of their tiles to place on the board. Sometimes, a word with fewer letters maximizes the number of points earned; in other scenarios, a word with more letters is the point-maximizing move. While there is no “right” number of analogs to select, there are disadvantages in selecting too few or too many. Starting an analysis with too few analogs risks yielding a desired prediction but one that may not be representative. Starting an analysis with too many analogs may unnecessarily complicate the analysis and should be a red flag that the selection criteria are not stringent enough. A Scrabble player has only seven tiles to use. If the criteria used to select analogs yield an initial list of more than seven products, then the criteria are likely not right.
As a rule of thumb, ending the selection process with two to three analogs is often practical. This small set of analogs should yield highly aligned predictions. If the predictions are dissimilar, then by definition, one or more of the analogs are not predictive and wrong. It is important to note that the majority does not rule — the outlier(s) may be right. And, of course, none may be right even if the outcomes are highly correlated. A smallish, parsimonious set of analogs is easier to analyze and report and is more likely to define a practical number of segments based on the target population’s behaviors tied to actionable implications. Interestingly, learning two- and three-letter words is an oft-cited recommendation for boosting Scrabble scores.
The ability to pivot
Some Scrabble players spot a nearly complete word in their hand and then stubbornly wait turn after turn to find an opportunity to play it or to pick a missing letter. Similarly, confirmation bias can result in the selection of analogs with desired outcomes, like when a manufacturer cherry picks instances of premium pricing that optimize market access and discounts or ignores instances that trigger “nasty” restrictions.
To avoid the hazards of confirmation bias, analogs should not be selected on the basis of success and failure. Analogs should be selected on the basis of shared characteristics about the product and/or the market that predict outcomes of interest, not a desired outcome. And just as Scrabble players need to be flexible as the tiles in hand and spaces available on the board change, analysts must be willing and able to pivot when an objective analysis yields unexpected evidence that contradicts initial assumptions or when the context changes.
Knowledge is power
Without a doubt, a larger vocabulary is an advantage when playing Scrabble. The ability to recall a greater number of words improves a person’s ability to spot playable combinations. When a player challenges another player’s word, an accepted dictionary can be consulted.
Accessing relevant information about stakeholders’ past decisions is not nearly as easy as looking up a word in a dictionary. Because some of the best analogs may not be current or even recent history, it may take additional time and money to look up the history. It is even more challenging to collect the intelligence that drove past decisions. And for those who get this far, remember that secondary data alone may not predict the future. In fact, overreliance on secondary data can prove deceptive; knowledge of a stakeholder’s readiness to change is best determined through interviews and surveys. The more relevant history a manufacturer knows, the greater its chance of success.
It’s how you play the game
There are many parallels between playing Scrabble and selecting analogs, but their likeness has some limitations. In Scrabble, the player with the most points wins the game. In contrast, the team(s) selecting the right analogs win. Competing manufacturers (each arriving at predictive sets of analogs, even if not identical), as well as payers and health systems (which can use analogs when considering the most prudent formulary and medical policy decisions) can all be winners. Winning can depend on knowing how to play the game, and low-stakes practice is the best preparation for high-stakes decisions. So, let the games begin!
No Comments