“Zeroes and ones will take us there!”
– Jesus Jones (1993)
One of the reasons I love strategy
Is you can use anything to determine what is possible.
Data is one of the most available and,
Especially if you have a lot,
It can make for one part of a great story.
But data is not without flaws.
In clinical research, we use data to assess where patients are
And how they’ve been recruited in the past.
This previous data is a part of what helps us assume
How recruitment will happen in the future.
Some people like to use the mean or average for this data,
Others the median or midpoint.
[And there’s still the mode if anyone wants to sing along]
When it comes to the “right” way,
Every scenario is unique,
But regardless of which you prefer,
They can both be useful!
For example, a previous trial had 15 recruiting sites,
10-12 of them we were hoping to get for a similar* trial.
The mean/average showed patients had enrolled quickly,
Meaning the rate for a new trial might be quite high.
While this was exciting,
Other trials had enrolled very slowly…
And I remembered thinking “something is off.”
Sure enough, the median rate showed it was much lower.
The difference between one patient enrolling per month
And one patient every 3 – 4 months.
The difference?
One investigator enrolled 7 times the number of patients.
15 sites, one outlier,
[A terrific outlier, but outlier nonetheless]
Creating a much higher mean/average than median,
And may have been an overestimate for how the new trial would enroll.
The takeaway?
Both rates are valuable, in context.
The average often shows potential for higher enrollers,
[If you can get those enrollers involved, of course]
The median often more accurate to how sites will perform overall.
One is not necessarily better than the other,
But each has their place,
And adds an extra storyline to your strategies!
*Similar has many layers of course. Several other factors are involved like study design, the drug itself, ongoing competition, excitement etc., but that’s for another post.
