A full-time GI position comes with a set of assumptions nobody questions: one physician, one panel, one endoscopy block, one call schedule. Given how much we talk about burnout and workforce shortages in this specialty, it's strange how rarely we question its day-to-day responsibilities. Primary care has experimented with job sharing for decades. As far as I can tell, gastroenterology has almost entirely ignored it.
I currently practice-share my GI practice, and my colleague and I may be the only gastroenterologists doing so.
It's important to note this wasn't built from scratch. I already had a large, full-time practice, an established full-time NP, and a four-day-a-week schedule. What made the arrangement possible was noticing something structural: procedure time was going unused. Instead of treating that as a scheduling headache, I offered to physically share my own office with an incoming colleague — which removed the objection that kills most of these conversations before they start: Where would they even sit? My department head and administration were supportive early on, and that mattered more than any spreadsheet.
My colleague was leaving a busy private practice, also at four days a week — I recruited her over to make this happen. We restructured into a 0.6 FTE (me) and 0.7 FTE (her) — two separate appointments, not one job split in half. A practice previously staffed four days a week now had six combined days of coverage. That's the part worth considering if you're skeptical: this isn't about doing less, it’s about a practice gaining capacity.
In our arrangement, I work Monday through Wednesday; she works Wednesday through Friday, with Wednesday as a deliberate overlap for handoffs. We keep mostly separate patient panels and cover each other when needed, supported by our NP, whose role predates the arrangement. Call is split into two tiers: routine call we cover together as a single unit, the same load a full-time physician would carry; hospital weeks and weekends we divide, mirroring our outpatient split.
The one real challenge is inbox coverage — labs, messages, EMR tasks don't pause on your days off, so each of us handles our own inbox in short blocks, usually 60 to 90 minutes. We looked at having each other cover instead and rejected it: a physician who doesn't know the patient takes longer to triage safely, every time.
The arrangement has been tested by real absences, not just theory, and coverage has held without disruption every time. On the professional side, I found myself enjoying teaching more, and I kept up with the literature more consistently than I had full-time. My burnout didn't disappear — no schedule change does that — but it improved substantially.
I also have to be honest about what surprised me most. Reception from colleagues wasn't uniformly warm, and some of that hasn't faded with time — mostly as subtle disapproval rather than anything said to my face, especially around how we handle calls. What struck me is that it wasn't limited to any one group; if anything, I expected more solidarity from people under similar pressures than I got. I've come to see it less as personal and more as systemic: a visible, working alternative to the standard full-time model tends to unsettle the status quo others have built their schedules around. In a few cases, it seems to have nudged colleagues to renegotiate their own arrangements — which, discomfort aside, tells me the model is doing something worth paying attention to.
Gastroenterology has a real workforce problem and a real burnout problem, both of which are tied to procedural volume and call intensity. Practice sharing won't fit every physician or every group, but it belongs in the same conversation as part-time work, locum coverage, and early retirement — except unlike those options, it doesn't remove an experienced physician from full engagement. Done carefully, it doesn't shrink a practice. It can grow one.
If you've tried something like this, or you're working out the logistics right now, I'd genuinely like to hear from you. The absence of anyone writing about practice sharing in a procedural specialty isn't proof it doesn't work here. It's just proof nobody's written it down yet.
