Editor’s Note: Matthew Goldstein is a research fellow at Doximity. He is currently completing his MD/PhD at Stanford University.

We all know the stereotypes. Pediatricians love their patients, love their colleagues, and well, love everyone. Surgeons throw scalpels, work long hours, and don’t have time for chit-chat. Recently, we came up with what seemed like an intriguing way to test how well those images held up in data form: Doximity colleaguing patterns as they occur within physicians’ own specialties. These connections would seem to be less strongly tied to the financial incentives of referrals (a cardiologist and a GP can treat the same patient for different things, less so two cardiologists). In other words, such numbers could, perhaps, show us a little about which doctors tend to connect the most for other reasons.

When we ran the data, at the top were the general specialties like pediatrics, family medicine, internal medicine, and anesthesia. All had more than 30% of colleagues coming from the same field. At the bottom–with less than 10% of colleagues coming from the same field–were plastic surgery, thoracic surgery, allergy, immunology, radiation oncology, and oncology.

So maybe pediatricians are, in fact, friendlier and more collaborative. But, maybe, too, these trends reflect something of the economics and other forces underlying medicine today.

Here, three possibilities:

1) We’re moving to more of a group-practice model where general physicians are working in shifts. The hospitalist movement has almost exclusively taken over the management of hospitalized patients. The increased collegiality within general specialities may reflect this trend.

2) Specialty practices such as surgery and oncology are so profoundly driven by an outside referral base that, even if individuals within those groups are very engaged with others in their field, those outside percentages will always be higher.

3) Physicians are putting the power of professional networks to work. When referrals can be done online, connecting with a colleague in the same specialty who lives in a different state might be starting to make more sense than it has historically. If a patient ever moves cities or states, this is an easy way to be at the click of a search button, as opposed to a number in the phone book or a business card at the bottom of a drawer.

It’s not surprising to think that practice dynamics and economics might be driving online colleaguing behavior in these kinds of ways, and it’s certainly something worth considering when designing and optimizing our own professional networks. How does your own colleague network compare to these specialities?