The Anxiety Tax: What Physicians Actually Control, and What’s Just Burning Your Brain’s Battery

The Problem With How Anxious Clinicians Spend Their Mental Energy

There is a particular kind of exhaustion that physicians describe but rarely name precisely. It is not the tiredness that follows a long shift, or the grief that follows a bad outcome, or the frustration of a system that doesn’t work. It is something more diffuse — a background hum of worry that runs continuously, consumes enormous cognitive resources, and produces almost nothing clinically useful. It is the expenditure of mental energy on things that will not change regardless of how much attention is directed at them.

Anxiety in medicine is not rare and not soft. National surveys consistently find that 30–40% of physicians screen positive for significant anxiety symptoms — rates substantially higher than the general population, and higher than most other high-skill professions. The clinical literature has documented the downstream consequences in detail: impaired working memory, reduced decision accuracy under uncertainty, increased diagnostic anchoring, shortened patient interactions, sleep disruption, and, over time, burnout and departure from practice.

What the literature discusses less is the cognitive economics of clinical anxiety — specifically, the distinction between worry that is useful because it is directed at something actionable, and worry that is metabolically expensive but causally inert. This distinction matters practically, because the treatment for the former is problem-solving, while the treatment for the latter is something closer to radical acceptance. Conflating the two — treating genuinely uncontrollable variables as if more analysis will yield control — is one of the most common and costly cognitive errors anxious physicians make.

This article is an attempt to draw that line more clearly, and to offer a framework physicians can actually use.

Part I: The Neuroscience of Why Anxious Brains Don’t Distinguish Well Between Controllable and Uncontrollable

The prefrontal cortex — the seat of executive function, planning, and deliberate reasoning — operates on a limited energy budget. Under elevated stress, the amygdala’s threat-detection circuitry competes with prefrontal resources and, in high-anxiety states, partially wins. The neurological result is a well-documented phenomenon: the brain under anxiety treats ambiguity as threat and responds to uncertainty by generating more cognitive activity directed at the uncertainty, regardless of whether that activity can produce a useful output.

This made evolutionary sense in environments where ambiguous threats were usually physical and usually responsive to attention — a predator in the grass benefits from sustained monitoring. It makes much less sense in a clinical environment where many of the genuine threats to a physician’s career, income, autonomy, and patient outcomes are structural, systemic, and genuinely outside any individual’s control regardless of how carefully they are watched.

The result is what cognitive-behavioral researchers call “worry as a coping strategy” — the use of sustained mental activity directed at problems as a way of managing the anxiety generated by those problems, independent of whether that activity actually reduces the probability of the feared outcome. The physician who lies awake reviewing a clinical decision that was reasonable given the available information, or who spends 40 minutes a day reading healthcare policy news without any clear action to take based on it, or who rehearses a difficult conversation with a hospital administrator dozens of times in advance — these are not failures of intelligence or discipline. They are the predictable output of a threat-detection system being applied to a problem domain it was not designed to solve.

The practical question is not whether this happens — it does, in virtually every anxious physician — but which specific categories of worry fall into the “cognitively costly but causally inert” bucket, and which don’t.

Part II: A Framework — The Two-Axis Test

Before spending mental energy on a concern, apply two questions:

Question 1: Is there a specific action I could take in the next 72 hours that would meaningfully change this outcome?

Not “is there something in principle someone could do” — but specifically, you, in the near term. If the answer is no, continued cognitive engagement with the problem is not problem-solving. It is rumination wearing problem-solving’s clothes.

Question 2: If I took that action, would it actually be likely to change the outcome — or would the outcome be determined primarily by forces outside my influence?

Many anxious physicians identify actionable responses to concerns and pursue them with genuine effort, only to find the outcome was determined upstream of anything they could control. The action was real; the causal influence was not. Recognizing this pattern in advance — rather than after the fact — is the cognitive work that saves energy.

Together, these questions divide a physician’s worry landscape into four quadrants:

Quadrant I — Actionable and Influential. You can do something specific, and doing it meaningfully changes the probability of a good outcome. This is where cognitive energy belongs. Clinical decisions in progress. Conversations that haven’t happened yet. Referral patterns you control. Team dynamics you can affect. Your own training and skills. Your documentation practices. How you talk to patients.

Quadrant II — Actionable but Low-Influence. You can do something, but the outcome is largely determined by factors outside your control. Responding thoughtfully to a prior authorization denial falls here — you can write the appeal, but approval rates for your payer are what they are. Energy: bounded. Do the thing, then release the outcome.

Quadrant III — Not Currently Actionable but Potentially Influenceable. Policy-level concerns, systemic issues, professional advocacy. You cannot act today, but sustained engagement over time may matter. Energy: invest selectively, through organized channels, not through continuous individual rumination.

Quadrant IV — Not Actionable, Not Influenceable. This is where most clinical anxiety’s energy actually goes, and where it produces nothing. The outcome of a malpractice case already in litigation. Whether a patient followed your discharge instructions after leaving. Whether your hospital system will be acquired. Whether the reimbursement environment will improve. Whether a patient’s cancer will respond to treatment. These are real concerns. They are not cognitive problems you can solve.

Part III: The Specific Categories Where Physician Anxiety Spends Its Budget

The following are common anxiety drains in clinical practice — mapped to their actual controllability.

Other physicians’ opinions of you. Moderate to low controllability. You can control how you present your reasoning, how you communicate with colleagues, and whether you are technically excellent. You cannot control whether a specific colleague respects you, or whether your judgment in a particular case will be retrospectively vindicated. Anxious physicians often spend enormous energy on this category, particularly early in their careers and after any critical incident. The cognitive effort does not improve outcomes.

Whether a difficult patient will be satisfied. Low controllability. Patient satisfaction correlates with how well you communicate, how much time you spend, and how thoroughly you address their concerns — all of which you influence. It also correlates heavily with factors you do not control: their prior experiences with the healthcare system, their pain levels, their expectations shaped before they walked in, their personality and attachment style, and whether the outcome of their care matches their hopes. Adjusting your behavior to improve patient experience is worthwhile. Ruminating after an unsatisfying interaction about what you could have done differently is, beyond a brief honest reflection, not.

The healthcare system’s dysfunction. Near-zero individual controllability. Prior authorization burdens, EHR usability, staffing ratios, reimbursement rates, hospital administrative decisions — these are real and legitimate sources of moral distress for physicians. They are also determined by institutional, political, and economic forces that no individual clinician can materially alter through private worry. The physicians who do influence these systems do so through organized advocacy, published research, leadership roles, and collective action — not through solo rumination that consumes clinical cognitive bandwidth without producing systemic change.

Whether your clinical decision was correct in retrospect. Very limited, and time-bound, controllability. Before a decision is made, you can improve your input by gathering more information, consulting colleagues, applying guidelines rigorously, and acknowledging uncertainty explicitly. After the decision is made and the clinical situation has evolved, the decision itself is fixed. You can — and should — learn from outcomes. That is professional development. What you cannot do is retroactively improve a decision by continuing to review it. The purpose of extended post-hoc rumination is not learning; it is anxiety management through the illusion of continued engagement with a controllable variable.

Your patients’ adherence. Low controllability. There is a significant body of evidence on what improves adherence: clear explanations, patient-centered communication, reducing medication complexity, addressing practical barriers, involving caregivers. You can do all of those things and still have a substantial proportion of patients not follow the plan. Investing heavily in patient education and communication is appropriate. Carrying the weight of each patient’s adherence decisions as a personal failure is a cognitive budget drain with no clinical return.

Whether you will face a malpractice claim. Moderate controllability for the underlying risk; near-zero controllability for the occurrence of a claim given that risk. Good documentation, sound clinical reasoning, appropriate communication with patients about uncertainty and risk, and consistent application of evidence-based practice all reduce malpractice risk at the margin. None of them approach zero risk. And the relationship between actual clinical negligence and malpractice claim occurrence is weaker than most physicians believe — many legitimate claims are never filed; many filed claims involve defensible care. Sustained anxiety about malpractice beyond what motivates excellent practice and appropriate documentation is functionally inert.

Your specialty’s reimbursement trajectory. Near-zero individual controllability. Fee schedule changes, CMS payment rules, payer contracting — these are the output of political and institutional processes in which individual physicians have very limited influence. Staying informed is appropriate. Collective professional advocacy through specialty societies is the appropriate channel for influence. Daily monitoring of proposed rule changes while unable to act on that monitoring is energy spent in Quadrant IV.

Part IV: What Physicians Actually Can Change — and Often Underestimate

The framework above risks reading as a counsel of learned helplessness — that physicians are at the mercy of uncontrollable systems and should simply accept it. That is not the argument. The argument is more specific: anxiety is most costly and least useful when directed at variables that are genuinely uncontrollable. It is appropriate and productive when directed at variables that are within reach.

Physicians systematically underestimate their influence in the following areas:

Their own communication quality. How a physician delivers a diagnosis, discusses uncertainty with a patient, explains a treatment plan, or responds to a patient who is frightened — these interactions are highly controllable and have documented effects on patient satisfaction, adherence, and outcomes. Communication skill is learnable, improvable, and its effects are immediate and visible. Anxious physicians often ruminate about how conversations will go or went rather than investing in the skills that make conversations go better. The latter is a much better use of the same cognitive energy.

Their own physical and cognitive functioning. Sleep, nutrition, physical activity, and the management of their own mental health are variables physicians control directly and that have outsized effects on clinical performance. The neuroscience here is unambiguous: sleep deprivation degrades decision-making accuracy in ways that clinicians themselves are poor at detecting. A physician who spends 45 minutes lying awake worrying about an uncontrollable systemic concern and loses cognitive performance the next day has paid a real clinical cost for a zero-return cognitive expenditure.

The environment they create for their immediate team. Ward culture, the psychological safety of a team to raise concerns, the quality of handoffs, the tone set in teaching interactions — these are highly proximate to the individual physician and respond to individual behavior. This is a high-leverage area where physician effort consistently produces measurable results, yet anxious physicians often feel too depleted to invest here, having spent their energy on less tractable concerns.

Whether they seek help. The decision to engage with a therapist, a physician health program, a peer support resource, or even a trusted colleague who can offer perspective is entirely within a physician’s control and has the highest return on investment of any item on this list. The primary barrier is not access — it is the stigma that remains embedded in medical culture, compounded by the erroneous belief that seeking help signals unfitness to practice. The evidence runs in precisely the opposite direction: physicians who engage with mental health support maintain clinical performance better over time than those who manage in isolation.

Part V: A Practical Protocol

For physicians who recognize themselves in the anxiety patterns described above, three evidence-consistent practices are worth integrating:

The Worry Triage. When a concern surfaces, spend no more than 90 seconds applying the two-axis test. If there is an action to take, note it and take it at the appropriate time. If there is not, explicitly label the concern as “not currently actionable” and practice deliberate disengagement. This is not denial — the concern may be real. It is cognitive hygiene: preserving mental resources for problems that respond to them.

Scheduled concern time. Counterintuitively, designating a specific, time-bounded period for processing legitimate professional concerns — typically 15–20 minutes in the late afternoon, away from clinical time — reduces the intrusion of those concerns at other times. The brain’s threat-detection system is partly appeased by the promise of attention at a designated time. This technique has solid empirical support in CBT for generalized anxiety and translates well to professional rumination.

Distinguishing post-case reflection from rumination. After a difficult case, a brief structured review — what information was available, what guidelines applied, what decision was made, what you would do similarly or differently — is professional learning. It has a beginning and an end. Rumination is the same material cycling without resolution, typically in the absence of new information. When you notice you are reviewing a case for the third time without accessing any new facts or perspectives, you have moved from reflection to rumination. The appropriate intervention is explicit transition: write down what you’ve concluded, close that cognitive file, and redirect.

Anxiety in medicine is not a personal failing. It is a predictable response to a genuinely difficult environment: high stakes, genuine uncertainty, complex systems, significant autonomy combined with significant dependency on others, and a professional culture that has historically discouraged vulnerability. The physicians who experience it are not weaker than those who report they don’t — they are, in many cases, the physicians most attuned to the genuine risks and ethical weight of the work.

The problem is not the anxiety. The problem is where it spends its energy.

A physician’s cognitive budget is finite and directly linked to clinical performance. Depleting it on variables that will not change regardless of attention directed at them — the healthcare system’s structural dysfunction, retrospective case reviews that yield no new information, other people’s judgments that are already formed, outcomes that are already determined — is not just personally costly. It is clinically costly. The patient seen by an exhausted, rumination-depleted physician at 4pm is receiving a materially different quality of cognitive engagement than they would from a physician who has conserved mental resources by applying them more selectively.

The anxious brain is not malfunctioning. It is running a program that evolved for a different environment. Learning to redirect it — toward the substantial subset of clinical and professional life that genuinely responds to effort, and away from the larger subset that does not — is not a luxury skill. For physicians practicing in 2026, it is a clinical competency.

Selected References

1. Shanafelt TD, et al. “Burnout and satisfaction with work-life integration among US physicians relative to the general US working population.” Mayo Clinic Proceedings, 2012 (updated surveys through 2023).
2. West CP, Dyrbye LN, Shanafelt TD. “Physician burnout: contributors, consequences and solutions.” Journal of Internal Medicine, 2018.
3. Brosschot JF, Gerin W, Thayer JF. “The perseverative cognition hypothesis: a review of worry, prolonged stress-related physiological activation, and health.” Journal of Psychosomatic Research, 2006.
4. Arnsten AFT. “Stress signalling pathways that impair prefrontal cortex structure and function.” Nature Reviews Neuroscience, 2009.
5. Ely EW, et al. “Inattention, disorganization, and fluctuation in the neurological intensive care unit.” JAMA, 2001. (Sleep deprivation and cognitive performance in clinical settings.)
6. Balch CM, et al. “Personal consequences of malpractice lawsuits on American surgeons.” Journal of the American College of Surgeons, 2011.
7. Clark DA, Beck AT. Cognitive Therapy of Anxiety Disorders: Science and Practice. Guilford Press, 2010. (Foundational text on worry as a coping mechanism.)
8. AMA: “Physician Health and Well-Being” survey data, 2025. ama-assn.org
9. Oken BS, et al. “Vigilance, alertness, or sustained attention: physiological basis and measurement.” Clinical Neurophysiology, 2006.