An assistant professor of internal medicine is sitting with her CV open in one window and her medical school's promotion criteria open in another, trying to figure out if she is ready.
The criteria say she needs "national reputation" and "sustained scholarly excellence." She has forty publications, two national committee appointments, and a teaching award. She does not know if that is enough, because nobody has told her what "enough" actually looked like for the last three people in her department who went up for promotion and succeeded. Her chair is supportive in the vague way chairs are supportive, and also happens to be the person who will vote on her case. Asking him to be specific feels like asking him to grade his own fairness in advance.
She emails a friend from residency who was promoted last year at a different institution, and the friend is generous but honest: her own department's bar was different, shaped by a different dean, a different specialty mix, a different committee culture. The comparison helps a little and resolves almost nothing.
So she does what almost every faculty member in this position does. She assembles the strongest dossier she can, submits it into a process she cannot see the inside of, and finds out in eight months whether her guess about the bar was close enough.
She is calibrating against a threshold that exists, that has been cleared by real people whose actual dossiers could tell her exactly what it takes, and that no institution will publish.
The gap is not evenly distributed, and the data is specific about who it hits
This would be a frustrating but survivable ambiguity if everyone navigated it with roughly equal access to informal calibration. The evidence says they do not, and the disparity is large enough to be its own finding, not a footnote.
A 2024 study in JAMA Network Open, examining 673,573 medical school graduates, found Black women physicians were 55 percent less likely than white men to be promoted to associate professor. That figure survives standard adjustment for the productivity measures institutions typically point to when explaining promotion gaps, which is precisely why the promotion-criteria literature increasingly frames this as an information and sponsorship gap rather than a productivity gap.
A separate 2024 study in JAMA Surgery, following 31,045 academic surgical faculty from 2005 to 2020, found promotion hazard ratios of 0.68 for Hispanic women and 0.75 for white women, relative to white men. Ten-year retention in the same cohort was 79 percent for non-underrepresented men versus 63 percent for underrepresented women. Faculty who cannot calibrate against the real bar do not just get promoted more slowly. A meaningful share of them leave first.
The bar is real, department-specific, and never written down
Every medical school publishes formal promotion and tenure criteria. Read them closely and the actual operative language is calibrated to be interpreted rather than applied mechanically: "national reputation," "substantial scholarly contribution," "sustained excellence in teaching." These phrases are not vague by accident. They are vague because promotion committees need discretion to weigh genuinely different kinds of academic careers against each other.
That discretion has to be exercised somehow, and it is: against precedent. What actually got the last several comparable candidates in a given department promoted is a real, specific, knowable thing. It is also never published. It exists only in the memory of the department chair and the committee members who evaluated those cases, transmitted person to person, informally, inside the committee room and in the hallway conversations that surround it.
That transmission channel structurally advantages whoever is already socially proximate to the people who hold the knowledge. A candidate whose mentor sits on, or regularly interacts with, the promotion committee gets a far more accurate read on the real bar than a candidate whose mentor does not, regardless of how strong either candidate's actual dossier is. This is precisely the mechanism the disparity literature keeps finding: not a difference in scholarly output, but a difference in who gets told, informally and accurately, what the bar actually requires before they build their case against it.
Institutions cannot publish their own answer, and know it
It would be reasonable to ask why a department simply does not publish its real precedent, the actual comparables that succeeded, stripped of names. The honest answer is that doing so would expose the institution to exactly the kind of scrutiny the disparity data above documents.
If a department published, honestly, what its last five successful associate-professor cases actually looked like, and those five cases were disproportionately drawn from faculty with strong informal mentorship access, the pattern the JAMA Network Open and JAMA Surgery studies quantify would become visible at the department level, attached to specific, identifiable committee decisions. No general counsel's office is going to recommend that. The incentive runs directly against transparency, even for institutions that would, in principle, like to fix the underlying disparity.
This is the same shape of problem this series has documented in other contexts: information that would help the person navigating a system is precisely the information the institution holding it has a structural reason never to make legible.
The institutional owner of this problem is being defunded
There is a further complication, and it is a recent, sharp one. A 2025 systematic review in Family Medicine, examining 33 studies drawn from an initial screen of 2,351, identifies "lack of strong research mentorship and sponsorship" and "unsupportive institutional cultures that lack resources or infrastructure" as the dominant drivers of what the literature calls scholarship delay among early-career underrepresented faculty.
A companion 2025 Family Medicine review goes further, and its finding is blunt: DEI offices, which have functioned in many institutions as the informal owner of exactly this equity gap, absorbing some of the calibration and sponsorship work that would otherwise fall entirely on chance mentorship access, are, in the paper's own words, "actively being dismantled."
That timing matters. The literature quantifying this gap has matured in roughly the same window that the institutional infrastructure built to partially compensate for it is being cut. A documented problem now has a shrinking, not growing, institutional owner.
Why AAMC, individual schools, and consultants each fall short
AAMC operates at the aggregate, policy level. Its faculty development content and workshop materials describe the promotion process in general terms, which is valuable for orientation and useless for the specific question a candidate actually has: what did it take in my department, in my specialty, at an institution like mine.
Individual schools will not, and structurally cannot, publish their own precedent set. Beyond the liability exposure described above, most departments simply do not have enough promotion cases in a given year, sometimes a handful across an entire specialty, to constitute a useful comparable set even if they wanted to share it. The sample size problem is real independent of the willingness problem.
Commercial promotion consultants sell generic advice, not verified comparables. A consultant can help a candidate write a stronger CV narrative. They cannot tell a candidate what the last three people who cleared this specific department's bar for this specific rank actually had, because they do not have access to that information any more than the candidate does.
What would actually work
A cross-institution, verified comparables set. Anonymized, aggregated dossiers of what actually succeeded, organized by specialty, target rank, and institution type (R1, community-based, safety-net-affiliated), assembled from enough institutions that no single department's participation is individually identifiable.
A route to a person, not just a data point. A candidate matched to a verified peer who was promoted recently in a comparable department, specialty, and rank, for a structured review of their actual dossier against what that peer's committee required, replicates the informal-network advantage that currently depends entirely on luck.
An explicit anonymization standard built before the first comparable is collected. Because the underlying disparity data shows exactly how this information could be misused or could expose institutions unfairly, the aggregation threshold, how many cases must be pooled before a pattern is shared, has to be decided deliberately and conservatively, not improvised after the fact.
Framed as calibration, not as a leak of institutional secrets. No department is being asked to publish its committee's deliberations. The service is a cross-institution benchmark built from what candidates themselves already know about their own successful cases, aggregated by a neutral party none of them individually controls.
A feedback channel to faculty affairs offices, not a bypass of them. The same aggregated data that helps an individual candidate calibrate is, over time, exactly the benchmark a faculty affairs dean needs to know whether their own department's informal bar is drifting from comparable institutions, or from written criteria.
Legal and equity review built in from the start. Any product resembling this has to be built with explicit attention to how comparables could be read under Title VII and EEOC frameworks, since the underlying disparity is itself protected-class-relevant, and a poorly designed tool could reproduce, rather than correct, the exact unevenness it targets.
What you can do now
If you are preparing a promotion case
Ask directly, in writing if needed, for the specific comparables your chair has in mind. Vague reassurance is not calibration. Ask what the last two or three successful cases in your department actually included, not what the written criteria say.
Find a peer at a different institution, in your specialty, at your target rank, who was recently promoted, deliberately, not by chance. The single highest-value thing you can do without waiting for any new infrastructure is to build this comparison yourself, now, rather than assuming your mentor's advice is calibrated to the actual bar.
Keep your own dossier, and eventually your own successful case, in a form you would be willing to anonymize and share with someone in your position five years from now. The precedent that would have helped you does not exist because nobody before you wrote it down in a shareable form either.
If you chair a department or sit on a promotion committee
Say the disparity numbers out loud to your committee before the next cycle starts. A 55 percent promotion gap for Black women physicians relative to white men, and a 63 percent versus 79 percent ten-year retention gap for underrepresented women in academic surgery, are not abstractions; naming them at the start of a cycle changes how a committee reads a marginal case.
Offer explicit calibration conversations to every candidate, not just the ones who happen to ask. The candidates most likely to lack informal access to this information are, by definition, the ones least likely to know to ask for it.
If you lead a faculty affairs office or specialty society
Treat the loss of DEI-office capacity as a gap you now need to fill by other means. If that informal calibration function is being dismantled at your institution, name what replaces it explicitly rather than assuming the gap closes itself.
Consider participating in a cross-institution comparables benchmark. A properly anonymized, aggregated version of what this article describes protects your institution from the liability of publishing its own precedent alone while giving your faculty the calibration they currently have to find informally or not at all.
Frequently asked questions
What are the criteria for promotion to associate professor in academic medicine? Written criteria typically require sustained scholarly contribution, evidence of a developing or established national reputation, and demonstrated excellence in teaching or clinical care, but these phrases are deliberately open to interpretation. The real, applied bar is set by department-specific precedent, what actually got the last several comparable candidates promoted, which is transmitted informally and is not published by any medical school.
Why are women and underrepresented faculty promoted more slowly in academic medicine? A 2024 JAMA Network Open study of 673,573 medical school graduates found Black women physicians 55 percent less likely than white men to be promoted to associate professor. A 2024 JAMA Surgery study of 31,045 academic surgical faculty found promotion hazard ratios of 0.68 for Hispanic women and 0.75 for white women relative to white men. A 2025 Family Medicine systematic review of 33 studies attributes the pattern primarily to gaps in mentorship, sponsorship, and institutional support rather than to productivity differences.
What is scholarship delay in academic medicine? Scholarship delay refers to slower research productivity and career advancement among early-career, often underrepresented, faculty, attributed by a 2025 Family Medicine systematic review to lack of strong research mentorship and sponsorship and to unsupportive institutional cultures lacking dedicated resources or infrastructure.
How do I find out what a promotion committee actually expects? Currently, mostly through informal access: a supportive mentor already close to or on the committee, chance conversations with recently promoted colleagues, or hired promotion consultants who can improve dossier presentation but cannot supply department-specific comparables, since they do not have access to that information either. No cross-institution, verified comparables resource currently exists.
Why won't medical schools just publish their real promotion bar? Publishing actual comparables risks exposing the institution to the exact disparity claims the JAMA Network Open and JAMA Surgery data document, since a department's real precedent set is often small enough that patterns tied to specific committee decisions could become identifiable. Legal exposure, not indifference, is the primary structural reason institutions have not done this.
Are DEI offices still helping faculty navigate promotion? Increasingly, no. A 2025 Family Medicine review notes that DEI offices, which have functioned informally as a partial calibration and sponsorship resource for underrepresented faculty in many institutions, are being actively dismantled, removing institutional capacity at the same time the disparity literature quantifying the need for it has matured.
The bottom line
The bar for promotion in academic medicine is real, specific, and knowable. It has already been cleared by real people whose actual dossiers, stripped of names, would tell any candidate exactly what it took. No institution publishes it, for reasons that are legally rational even when they leave every candidate without a senior sponsor guessing.
The evidence on who that guessing hurts most is not ambiguous. Black women physicians are 55 percent less likely than white men to reach associate professor, across more than six hundred thousand graduates. Underrepresented women in academic surgery are promoted at a fraction of the rate of white men and leave at nearly double the rate over a decade. The informal channel that currently substitutes for a published bar runs, predictably, through existing proximity and existing sponsorship, which means it reproduces exactly the disparity the data documents.
The institutional infrastructure that used to partially offset this, DEI and faculty-affairs equity offices, is shrinking at precisely the moment the evidence quantifying the gap has become undeniable. What remains is a candidate alone with a CV, a vague written standard, and a chair who cannot ethically grade his own fairness in advance.
She will submit her dossier in a few months, into a process she still cannot see the inside of, and find out whether her guess about the bar was close enough.
Part of a series on the missing professional infrastructure of healthcare. Previously: Two-Sided Investigator Opacity: Sites Can't See Sponsors, Sponsors Can't See Sites
Evidence note: promotion-disparity figures come from a 2024 JAMA Network Open study of 673,573 medical school graduates and a 2024 JAMA Surgery study of 31,045 academic surgical faculty tracked from 2005 to 2020; both are large, adjusted cohort analyses, though the surgery-specific findings should not be assumed to generalize precisely to every specialty without further replication. The scholarship-delay and DEI-office-dismantling findings come from two related 2025 Family Medicine systematic reviews (33 studies from an initial pool of 2,351), which describe the driving mechanisms qualitatively; the claim that DEI offices are being actively dismantled reflects that review's own characterization of the current environment and should be understood as describing a trend rather than a precisely quantified rate of closure. The approximately 195 LCME-accredited medical school figure reflects current accreditation counts. Nothing in this article constitutes legal advice regarding Title VII, EEOC, or institutional liability; any product addressing this gap would require dedicated legal review beyond what is summarized here.
