>The agglomeration method that we adopt is Ward’s method, which aims to min- imize the total within-cluster variance. At each step in Ward’s method, the algorithm finds two clusters that will result in minimum increase in total within-cluster variance after fusing. The number of clusters was narrowed by cluster size and similarity to 9 total groups through review of dendrograms and heat maps by non-clinician investigators (A.G. and N.A.).

1. Pick an algorithm.

2. Get a number of clusters. Preferably below 10, because otherwise it gets too messy for publication.

3. Justify algorithm based on its first paragraph on Wikipedia.

Yes, this!

I am by no means an expert, but I have spent time and experience enough with chronic pain to state that this rather one dimensional. Meaning their group although big was narrow in scope. For example, no chronic head pain? We have been working on a VR body mapping app(https://www.hatsumivr.com/), I'm the art lead on it. And you quickly find that there is an ocean if different experiences out there, and thats probably the biggest obstacle to helping people with chronic pain. Its is often treated (at the policy level) as one condition with a select few options for treatment. Studies with conclusion like this reinforce a reductionism in medical spaces and leave the people that don't fit into their pidgin holes out in the cold, alone, and in pain.

As I suffer from relentless chronic pain, I was hoping to read something insightful. What I saw might as well been on reddit/r/coolguides. Your post gave me a chuckle.

I'm excited to adopt these new categories.

Discussing pain with my care providers is very suboptimal.

The 9 categories from this OC are far superior to our respective current vocabularies.