Melanie Blanchette in Science:
When I was a Ph.D. student, a respected professor at our school had a heart attack in his office and died. As he was whisked away to the ambulance, I numbly watched familiar faces in the department succumb to shock. I didn't know it at the time, but this deeply troubling experience would shape my thinking about how to craft my academic career after I faced my own life-changing illness. While I was a postdoc, a sudden neurological disorder left me unable to walk, took my vision, and held me in the grip of vertigo and crushing migraines. With the help of a small army of health professionals, I began to improve. My brain started compensating for the lost neurons, and my muscles learned to fire again, but I don't know whether I will ever recover completely. This harsh reality check has made me think seriously about why academia promotes unhealthy work habits and how I can pursue the research I love while also taking care of myself.
Prior to my illness, I worked extremely long hours, sometimes even sleeping in my office if I faced a deadline. I hoped that my hard-won achievements would eventually be judged worthy of tenure. When I returned to work after my illness—despite its severity, I took just 2 months of because of dwindling sick leave, increasing medical bills, and no certainty of ongoing employment—I fell back into the academic achievement trap. I spent all my time working and worrying, and my health began to decline again. But fear of a relapse made me question my actions and, ultimately, the trajectory of my career. I thought about the professor who died. I thought about a friend who left academia because the pace and environment had negative mental and physical effects. I realized that my years in academia had eroded my mental health. I didn't want to hurt myself permanently by pursuing career advancement at all costs, but I didn't want to leave either. So I decided to accept my physical limitations and—an even more diffcult task—shed my prejudices about what a successful career looks like.
More here.