“Pajama Time” isn’t cozy. It’s a symptom.
Since 2020, patient portal message rates have increased by over 200%.(1) On the surface, that seems like progress: patients are more connected to their care and experience support that extends beyond the clinic walls. But for many physicians, like myself, this introduced a different kind of burden—one that’s harder to see and even harder to solve.
We call it “pajama time.”
Every day, physicians log in after hours to respond to patient messages, clarify prescriptions, check labs, and more. Messages that are often best suited to be handled by someone else on the care team oftentimes still land right in the physician’s inbox. Sorting and answering them became part of the job during work (distracting me from the patients I saw in clinic) and after hours.
I experienced this firsthand, especially during the height of the COVID-19 pandemic. A typical day for me during this time consisted of spending long hours treating patients back-to-back, only to be met with a mountain of messages that needed attention, waiting for me at home. I knew that if I waited to respond, the messages would only continue to pile up, and I’d fall further behind. So I felt forced to keep going, even when I was completely spent and just wanted to spend time with my family.
And I knew I wasn’t alone with this.
In 2024, 93% of physicians reported feeling burned out and spending an additional 15 hours of administrative work at home or after hours.(2) Pajama time is, without a doubt, a leading cause of burnout across the board. The most frustrating part is that many of these message loads are fixable; if only their system could triage, route, and respond in smarter ways.
I realized this issue early on, and it’s the reason why my company,Switchboard, MD, exists today.
In 2021, I built an NLP model to intercept inbound messages and correctly classify, triage, and route them to the most appropriate recipient before they even hit the inbox. This essentially eliminates the need for physicians to forward messages to others or try to handle the messages themselves.
Since then, the model evolved into the MDAware model that we know today: with 98% accuracy, 84% faster resolution of patient conversations, and 40% fewer message touches by staff.(3) By easily integrating with the EMR, we’re proving that automating the triage of clinical communications can reduce delays and burnout, without compromising care.
If you’re interested in how stories like mine are reshaping how care is delivered (for both patients and staff), we invite you to join our newsletters. We’ll be sharing more insights on the people, the research, and the innovations that we’re building here at Switchboard to help move the field forward.
All the best,
Blake Anderson, MD
Nath B, Williams B, Jeffery MM, et al. Trends in electronic health record inbox messaging during the COVID-19 pandemic in an ambulatory practice network in New England. JAMA Netw Open. 2021;4(10):e2131490. doi:10.1001/jamanetworkopen.2021.31490
2023 Physician Sentiment Survey, commissioned by athenahealth and fielded by Harris Poll, Jan 2024.
Anderson B, Zia ul Haq M, Zhu Y, Hornback A, Cowan A, Mott M, Gallaher B, Harzand A. Development and Evaluation of a Model to Manage Patient Portal Messages. NEJM AI 2025;2(3) doi: 10.1056/AIoa2400354
