How to Leave Medicine Without Feeling Like a Failure

WOMEN IN MEDICINE • March 19, 2026

Forty Minutes in the Parking Garage

It’s 6:40 on a Wednesday morning and Mira is sitting in her car on level three of the hospital parking garage, engine off, badge already clipped to her white coat. She’s 38, a pediatric hospitalist at a large academic medical center, the attending everyone wants running a code because she doesn’t rattle. Her travel mug of coffee has gone lukewarm in the cupholder. On the passenger seat sits a resignation letter she has now drafted four times, printed once, and never given to anyone. She isn’t late. She just can’t make her hand find the door handle yet.

In my work with physicians navigating whether to leave clinical medicine, I notice the same paralysis almost every time, regardless of specialty, and it rarely looks like burnout in the way the word gets used casually. It looks like this: forty minutes in a parking garage, a letter that keeps getting written and never delivered, a woman who is, by every external measure, excellent at her job and privately terrified that leaving it will prove she never deserved it.

Mira isn’t stalling because she’s unsure whether she wants to leave. Somewhere underneath the paralysis, she already knows. She’s stalling because no one in her training, her family, or her professional culture ever gave her language for what she’s feeling, and without language, the feeling just circles. “I used to love this,” she told me, turning her cold coffee cup in slow half-turns on the table between us. “Now I walk in and I feel like I’m performing a version of myself from four years ago. I don’t know when she left. I just know I can’t find her anymore, and I don’t know if that means I’m burned out or if it means I’m supposed to leave, and not knowing which one it is has kept me stuck for a year.”

Sitting with Mira that first hour, I felt the particular weight I’ve come to recognize in physicians at this exact juncture. Not the weight of a woman who hates her work. The weight of a woman grieving the loss of her relationship to work she once loved, while still performing that work at a level nobody would guess was costing this much. What Mira is describing isn’t one thing with one name. It’s several overlapping experiences, burnout, moral injury, identity disruption, and ordinary uncertainty about the next chapter, and one of the most useful things I can offer is help telling those apart, because each one calls for a different kind of care.

Is It Normal to Grieve a Career You Chose to Leave?

Yes, and I want to say that plainly before anything else in this guide, because the guilt of grieving something you’re choosing to walk away from is often heavier than the decision itself. Grief doesn’t check whether your departure was voluntary before it shows up. It only registers that something real is ending: a role, a community, a version of the future you’d planned around, a self-concept assembled over a decade or more of training that cost you your twenties, your sleep, and in many cases your health.

Here’s the clinical mechanism underneath that gap. When a loss is socially recognized, other people’s acknowledgment does regulatory work your nervous system can’t fully do alone: a friend saying “I’m so sorry, that must be devastating” actually helps metabolize the loss. Which means in practice that when the loss goes unacknowledged, because you chose it, because it looks like success from the outside, the metabolizing doesn’t happen the same way. The grief doesn’t disappear. It just has nowhere to go, so it tends to surface sideways: as irritability, as insomnia, as a strange flatness during what should be a celebratory milestone.

Deborah Lathrop’s 2017 analysis in the Annals of Family Medicine names something I see constantly in my own practice: physicians accumulate ongoing, often unacknowledged professional losses over the course of a career, the erosion of autonomy, the changing shape of the doctor-patient relationship under electronic health record demands, the slow loss of the version of medicine they trained for, and those losses function as a form of grief even when no single dramatic event triggered them (Lathrop, 2017). Lathrop’s piece is a conceptual argument rather than a data study, and she’s careful about that herself, but the framing has held up in every physician-transition case I’ve worked since I first read it. She also raises something I think about often with clients: resilience-training and individual-focused coping programs can quietly imply that burnout is a personal failing to be trained out of, rather than a legitimate response to systemic and identity-level loss that deserves acknowledgment, not correction.

Of course you’re tired. You’ve been carrying a loss that had no name and no witness, on top of the demands of a job you were still, until very recently, showing up for at full capacity.

Why Does Leaving Medicine Feel Like Failure When It Isn’t?

One of the most persistent myths in medical culture is that leaving equals failing. It’s rarely stated outright, but it’s embedded everywhere: in the language of “attrition,” in the way departure gets discussed in hallway conversation as something that happened to a colleague rather than something she chose, in the subtle recalibration of how people talk to you once you’ve said the word “leaving” out loud.

Failure, properly defined, means not meeting a standard you set out to meet, or lacking the capacity to continue. Neither describes what’s happening when a driven, capable physician decides to leave clinical practice. She met the standard. She exceeded it, often for years, at a cost nobody ever asked her to calculate honestly. Continuing to perform at a high level while your health erodes underneath you isn’t success being interrupted by weakness. It’s a different kind of failure already in progress, a failure to honor what your own body and mind have been signaling.

Medical training instills endurance as a core virtue, and there’s something genuinely admirable in that. It’s also, without a corresponding permission to stop, a setup. Physicians are trained to read a chart, notice small deviations from baseline, and act before a crisis. The same physicians are rarely trained to apply that attention to their own baseline. Mira could tell you the early warning signs of a decompensating patient in her sleep. It took her a full year sitting in that parking garage before she could name her own.

What I want to say directly to the physician reading this: your worth was never actually tied to your job title, even though your training environment made it feel that way for years. It’s inherent, and it survives the decision to leave completely intact.

Who Are You When the White Coat Comes Off?

This is the question that keeps physicians circling in the parking garage long after they already know the practical answer. Your identity as a doctor can feel completely fused with your identity as a person, not metaphorically fused but structurally fused, built during the exact years, residency through early attending life, when most people are still forming a stable adult sense of self. When training coincides with identity formation itself, the two get soldered together in a way that makes “who am I without this” feel less like a question and more like a threat.

Hana is a 47-year-old anesthesiologist who came to see me eight months after she’d already left her hospital position, not before, which is its own common pattern I want to name. She’d handled the operational side of leaving with the same competence she brought to every case: she’d negotiated her exit, lined up part-time locum work, built a spreadsheet of options. What she hadn’t handled, what she didn’t know how to handle, was that she’d wake up at 4:12 most mornings, the exact time her alarm used to go off for early cases, and lie there with what she described as “a low hum of nothing, like static where a person used to be.” She said, “I did everything right. I planned this for two years. Why do I feel like I disappeared.”

Hana hadn’t disappeared. What she’d lost wasn’t herself. It was the external scaffolding that had been organizing her identity for twenty-two years: the badge, the OR schedule, the specific way strangers’ faces changed when she said what she did for a living. Removing scaffolding doesn’t remove the building. It does, for a while, make the building feel unfamiliar and slightly exposed, and that disorientation isn’t evidence that something went wrong. It’s evidence that something enormous just changed structurally, and structural change takes longer to feel normal than most people expect going in.

The work in this phase isn’t to rush toward a replacement identity, a new label to fill the space the old one occupied. It’s to tolerate the in-between long enough to notice what’s actually still there underneath the white coat: the curiosity that made you good at diagnosis before it made you exhausted by documentation, the steadiness under pressure, the specific way you’ve always paid attention to people in distress. Those capacities don’t require a medical license to exist. They just temporarily lost their container.

I find myself returning to that question often with physicians in this exact in-between, not as a prompt to answer immediately, but as permission to sit with it honestly. Hana didn’t have a tidy answer the first time I asked what she wanted her one wild and precious life to hold. Neither did Mira. Neither, most weeks, do I, about my own. The question doesn’t require an instant answer. It requires that you stop letting a job title answer it for you by default.

What Does the Research Actually Show About Physician Burnout and Moral Injury?

I want to slow down here, because most articles about physicians leaving medicine either skip the research entirely or cite a single alarming statistic without the context that makes it useful. Here’s what the actual literature shows, and where it’s more limited than the headlines suggest.

Physician burnout prevalence varies enormously by specialty, geography, and the measurement tool used, an important caveat before quoting any single number as universal. A systematic review and meta-analysis of 37 studies covering 15,183 French physicians found a pooled burnout prevalence of 49%, with 5% meeting criteria for severe burnout, and identified emergency physicians and junior residents as carrying the highest risk (Kansoun et al., 2019). That figure describes French physicians during the studies’ sampling window, not a universal global rate, and the high heterogeneity typical of burnout meta-analyses means it should be read as a range, not a precise parameter.

Burnout isn’t only a wellbeing concern. It has measurable downstream consequences for patient care. A meta-analysis of 13 studies covering 20,643 physicians and residents found that overall burnout was significantly associated with increased self-reported medical errors, with an odds ratio of 2.72 (Owoc et al., 2022). The outcome measured was self-reported errors, not independently audited errors, a real limitation. It still tells you something clinically important: burnout doesn’t stay contained to the person experiencing it. It reaches the patients in front of her too.

Subspecialty matters too, and it matters a great deal. A 2025 systematic review and meta-analysis of 19 studies covering 4,634 trauma surgeons found a pooled burnout prevalence of 60.0%, among the highest reported of any surgical subspecialty, with mentorship and protected non-clinical time identified as the strongest protective factors (Kirdar-Smith et al., 2025). A companion review by the same research team, covering 15 studies and 2,757 pediatric surgeons, found a notably lower pooled prevalence of 29.4%, with personal accomplishment, family-centered care, and structured mentorship named as protective (Kirdar-Smith et al., 2025). Because both reviews share a first-author team, I’d caution against treating them as fully independent replications. What they show, read together, is that burnout prevalence isn’t a fixed physician-wide constant. It moves depending on specialty culture, autonomy, and the presence or absence of mentorship.

Burnout and moral injury are related but not identical, and the distinction matters clinically because it changes what actually helps. A 2023 systematic review synthesizing 18 empirical studies on moral injury among healthcare clinicians found it consistently associated with reduced professional wellbeing and worse mental health outcomes, while explicitly noting the construct still needs further theoretical development (Thibodeau et al., 2023). I’ve come to think of the difference this way in session: burnout is what happens when the tank runs empty from sustained overwork. Moral injury is what happens when you’re required, repeatedly, to act against your own values, discharging a patient before she’s stable because the bed needs to turn over, and the injury isn’t exhaustion. It’s the specific wound of having participated, against your will, in something you believe was wrong. A physician can be exhausted without being morally injured, and she can be morally injured while still, on paper, managing her hours reasonably well.

Systemic framing matters as much as any individual statistic. A widely cited 2019 BMJ analysis, co-authored by Christina Maslach, PhD, the psychologist who developed the Maslach Burnout Inventory used in most of the studies above, argues that burnout is primarily driven by organizational factors, workload, control, reward, community, fairness, and values alignment, rather than by deficits in individual resilience (Montgomery et al., 2019). This is an expert synthesis piece rather than a data study, but it’s the one I return to most often with clients who’ve been told to simply meditate their way out of a structural problem.

Gendered workload disparities compound all of the above. A 2023 narrative review found that women physicians carry measurably higher workload per patient, roughly 10% more time per visit and 15.7% more direct patient-care time in the studies reviewed, alongside more after-hours electronic-record work, less schedule control, and lower compensation than male peers in comparable roles (Lyubarova et al., 2023). This is a narrative report drawing on surveys of varying rigor, not a single controlled study, so I treat its numbers as directional evidence of a real pattern rather than a precise, citable statistic on their own.

Taken together, here’s the honest epistemic line I hold with physicians in my office. The research clearly supports that burnout is common, specialty-dependent, and driven substantially by organizational conditions rather than individual failing. It’s less settled on exactly how moral injury should be measured as distinct from burnout, and on precisely how large the gendered workload gap is across every specialty. Where the research is thinner, I say so. Where it’s solid, I lean on it.

Both/And: Can Medicine Have Mattered and Still Be the Thing You Need to Leave?

Physicians leaving medicine often feel pressured into a binary they didn’t choose: either medicine was your calling and leaving is a betrayal of that calling, or medicine was a mistake and you should feel relief, not grief, about walking away. Neither framing survives contact with an actual physician’s actual history. Both/And means refusing that binary entirely.

Mira, six months into our work, put it more precisely than I could have. “I saved a kid’s life in the PICU two weeks before I wrote that resignation letter,” she said. “I mean that literally. He coded and we got him back and his mother hugged me so hard I couldn’t breathe for a second. That happened. It was real. It mattered more than almost anything I’ve ever done. And also, that same week, I sat in my car for forty minutes because I couldn’t make myself walk in. Both of those are true about the same job in the same month, and I don’t know how to hold both without one canceling the other out.” I told her she didn’t have to make one cancel the other. She could let both stay exactly as true as they were.

Both/And means medicine can have been the site of some of the most meaningful work Mira will ever do, work that used every capacity she has, her attention, her steadiness, her years of training, and it can simultaneously be the environment that’s costing her more than it’s giving back right now. It means Hana’s twenty-two years in the operating room weren’t a waste, and her decision to stop isn’t a repudiation of those years. Gratitude for what a career gave you and clarity that it’s time to leave aren’t in tension. They’re both accurate descriptions of the same set of facts, and a physician doesn’t have to choose which one is more true.

This also means the decision to leave doesn’t require reinvention on a grand scale to be legitimate. You don’t owe the world a pivot to health policy or a wellness startup to justify walking away from a career that was hurting you. Sometimes the honest story is smaller and more private than that: I gave medicine everything I had for as long as I could sustain it, and now I’m choosing myself. That sentence is complete on its own. It doesn’t need a sequel to prove it was the right call.

The Systemic Lens: Why Does the System Make Leaving So Hard?

What Mira and Hana are each navigating isn’t a personal deficiency in resilience. It’s patterned, and the pattern has identifiable structural sources that show up across specialty, geography, and career stage.

Medical training is organizationally designed around workload, control, and values alignment in ways that predictably erode wellbeing over time, per the organizational-change framework that Christina Maslach, PhD, and her co-authors have documented (Montgomery et al., 2019). Residents and early attendings have limited control over their schedules during precisely the years identity is still consolidating. Reward structures often prize throughput over the relational, unhurried care most physicians entered medicine to provide, creating a chronic values conflict between what the system rewards and what the physician believes good care looks like. That mismatch, sustained for years, is a documented mechanism of burnout and moral injury, not a character flaw in the physician experiencing it.

Gendered structural pressures compound this further for women physicians. Measurably higher per-patient workload, more uncompensated after-hours documentation time, less schedule autonomy, and lower relative compensation aren’t incidental frictions. They’re structural features of how many healthcare systems are organized, and they land disproportionately on women, who are also more likely to carry a disproportionate share of caregiving responsibilities outside of work (Lyubarova et al., 2023).

Here’s how that inheritance shows up on an ordinary Tuesday afternoon. It’s Mira finishing rounds twenty minutes late because she stayed to actually talk with a frightened parent, then eating lunch at her desk while finishing notes the system counts as “after hours” even though the caring itself happened during her shift. It’s Hana calculating, at 4:12 in the morning eight months after leaving, that she gave twenty-two years to an institution that never once asked whether the pace was sustainable. It’s every physician doing the mental arithmetic of what she sacrificed against what the system gave back, discovering the math doesn’t balance, and then, often, blaming herself for noticing.

You are not broken for finding this unsustainable. A system that demands relentless output while treating burnout as a personal wellness gap, rather than a structural design flaw, has no interest in adjusting its own expectations. Naming that clearly doesn’t fix the system. It does stop a driven, capable physician from internalizing a structural failure as evidence of her own inadequacy, which is often the first real relief available to her.

How Do You Actually Leave Medicine Without Breaking Yourself?

I want to say something directly before anything else in this section: nothing here is a substitute for personalized financial, legal, or clinical advice, and I’m not suggesting anyone resign abruptly based on an article. Leaving medicine safely and sustainably almost always benefits from working with a financial planner who understands physician debt loads, an employment attorney if contract or licensure questions are involved, and a therapist or coach who can help you pace the emotional and identity work alongside the practical steps. What follows are the patterns I’ve seen actually help, not a script to execute on your own timeline without support.

Give the grief its full address before you rush to reinvent. You’re not only leaving a job. You’re leaving an identity built over a decade or more, a community, a future you’d planned around, and in many cases a version of yourself that sacrificed enormously to arrive here. That’s worth mourning fully, not in the five-minute windows between obligations. Physicians who skip this step tend to carry the unprocessed grief directly into whatever comes next, where it resurfaces later, often at a worse time.

Separate the vocation from the institution. What drew you to medicine, solving complex problems, sitting with people in crisis and knowing what to do, the specific satisfaction of competence applied to something that matters, doesn’t evaporate when your hospital badge does. Many physicians who leave clinical practice find those exact capacities show up in health policy, medical writing, consulting, education, or coaching. The desire to contribute doesn’t die with the resignation letter. It redirects.

Respect the sunk-cost pull without letting it run the decision. The training was real. The debt is often real and substantial. The years were real. None of that obligates you to keep paying with your health indefinitely. The more useful question usually isn’t “what did I sacrifice to get here,” but “what’s staying actually continuing to cost me, and is that cost sustainable for another five years.” A financial advisor who specializes in physician debt can help you model the actual numbers rather than the anxious estimate in your head, which is almost always worse than reality.

Sequence the licensure and logistics deliberately. Whether and when to let a license lapse, how to structure a transition period, and what a part-time or locum bridge might look like are questions worth working through with someone who knows the rules in your jurisdiction, since the details vary and getting them wrong can be costly to reverse.

Find one person who has actually done it. Not to copy her path, but to see that the path exists and that she’s still standing on the other side of it. Isolation is one of the most corrosive parts of this transition, partly because the profession doesn’t celebrate the people who leave. Even one conversation with a physician who built something sustaining afterward can shift the internal frame from failure to possibility.

Hana, when I last saw her, was four months into a part-time medical-legal consulting role she’d taken almost as an experiment, still uncertain whether it would become permanent. “I still wake up at 4:12 some mornings,” she told me. “It doesn’t happen every day anymore. When it does, I don’t spiral the way I used to. I just lie there for a minute and let it be what it is, and then I go back to sleep.” She hasn’t landed anywhere final. She isn’t sure this is the shape her next chapter takes. What she has is a little more room between the waking and the spiral, and on the morning she told me this, that was enough.

Mira is still, as of this writing, working the same hospitalist position, though the resignation letter is no longer being rewritten. She’s given herself a defined six-month window, not a deadline to decide, but a deliberate period of gathering information: informational conversations with physicians who’ve left, a session with a financial planner, an honest look at what parts of the work still feel like hers. “I don’t know yet if I’m staying or going,” she told me recently, coffee this time actually still warm in her hand. “I know I’m not stuck in the parking garage anymore. That’s different. That’s something.” Whichever way Mira’s six months resolve, the paralysis itself has already loosened, and that loosening, more than any specific outcome, is usually where this work actually begins.

Whether you’re a few months into the questioning or years into a decision you haven’t yet acted on, you don’t have to sit with this alone, and you don’t have to figure out the entire next chapter before you’re allowed the first honest step.

You might also find these useful: a look at co-regulation and how it shapes nervous system safety during major life transitions, a guide to inner child work for anyone untangling old beliefs about worth and achievement, an overview of attachment styles for understanding why identity loss can feel so destabilizing, and a companion resource on EMDR therapy for physicians whose bodies are still holding stress their minds have already tried to process.

This article was drafted with AI assistance as part of Annie’s editorial process and reviewed and approved by Annie for clinical accuracy and voice. Composite client stories in this piece, including Mira and Hana, are fictionalized amalgams drawn from patterns Annie has observed across her clinical work; they don’t describe any single real client, and no identifying details are included. Read more in our Editorial Policy. Questions can be sent to support@anniewright.com.

Warmly,
Annie.

Medical Disclaimer