Is wetness binary? Understanding what matters...

‘Wet’ seems binary, huh? Once you’re not dry, you’re wet…?

Of course, like so much to do with endpoints in pharma (what does ‘efficacy’ mean..?), there is an interesting rabbit hole once you decide to define ‘wet’ or ‘wettest’…

If you?look at a chart like this?one, it defines ‘rainiest’ as ‘how many days did place?x have rain (or snow) of more than 1mm?’

Is that?your?definition? If a place you lived got 1mm of rain at 2am, and then was at a glorious 75 degrees Fahrenheit from 8am to 8pm, would you regard that as rainy??These people?‘found the average rainfall in millimetres and the average number of wet days per month for each UK city with a population of more than 100,000 people. These results have then been divided by 12 in a bid to calculate the average monthly rainfall and the average amount of days of rainfall per month’ for their definition… As if a ‘rainy season’ is comparable to endless days of grey skies (to an ‘average’ they are the same).

Does ‘amount of rain’ as well as ‘number of days with rain’ feature in your definition? How about ‘duration of rain’, or ‘days since the last sunshine’ or ‘predictability’ feature? Given we all use different words for rain (especially us Brits: downpour, drizzle, mizzle, stair rods…), it is clear that anyone claiming ‘wettest’ is using a singular definition that is likely to be arguable. If your definition of ‘wettest’ was as simplistic as ‘which gets the most water from the sky per year?’ you’d have a different table:

Area - Rainfall (mm)

Argyllshire 2274.9

Dunbartonshire 2066.5

Inverness 2034.8

Merionethshire 1914.0

Ross and Cromarty 1858.1

Carnarvonshire 1809.7

Buteshire 1771.2

Kirkcudbrightshire 1696.7

Westmorland 1652.6

Brecknockshire 1643.7

Let’s apply that thinking to ‘productivity’ too: especially at this time of year, many companies’ slide decks talk about their productivity – read further and it might be ‘we kept products alive in early phase’ or ‘we launched products’ or ‘we launched products with commercial potential’, all masquerading as rainfall…

The more obvious overlay is in whether we believe our drugs?work. What does a drug that works on pain ‘do’? Is it number of days or hours with no pain? What does no pain mean? (Answer: it is not an absence of pain, as you might assume…) What does a drug that ‘treats’ schizophrenia do? Can you measure it by numbers of days with ‘relapse’ or is it by the amount of sun that shines?

All learning starts with understanding. The more time spent in looking for alternative definitions, the more opportunity seeking that can take place. Understanding what matters to people about rainfall might mean that we’d look beyond statistical significance to real significance. Understanding what really matters to patients might be the greatest asymmetry that you can bring to bear on a treatment option…