Giri v. National Board of Medical Examiners

UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF COLUMBIA

LATIKA GIRI,

Plaintiff,

v. Case No. 24-cv-410 (CRC)

THE NATIONAL BOARD OF MEDICAL

EXAMINERS,

Defendant.

MEMORANDUM OPINION AND ORDER

This case is about the National Board of Medical Examiners’ (“NBME” or “the Board”)

efforts to uphold the integrity of medical licensing examinations in the face of allegations of

organized cheating. In early 2023, NBME received multiple tips concerning a cheating ring

centered in Nepal. After an investigation verified these allegations, NBME performed a

statistical analysis of recent test results from individuals who either graduated from a Nepali

medical school, sat for an examination at a test center in Nepal, or self-reported as a citizen of

Nepal. Based on that analysis, in early 2024, NBME invalidated scores of 832 examinees (or

nearly forty percent of the sample group) whose test results were highly irregular, suggesting the

test-takers may have had prior access to exam questions.

One of the individuals whose exam results were invalidated, Dr. Latika Giri, filed suit

under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., and 42 U.S.C. § 1981,

alleging that NBME unlawfully discriminated against examinees of Nepali ethnicity and national

origin by targeting them for special treatment. With the 2024 National Resident Matching

Program (“the Match”) around the corner, Dr. Giri filed an emergency motion for a preliminary

injunction to reinstate her scores. And not just her own: Dr. Giri seeks to provisionally certify a

class of all 832 test-takers who had their scores invalidated through this “irregular” process that

broke from the Board’s ordinary course of performing an individualized evaluation, including

consideration of any explanation from the test-taker, before invalidating scores.

Weighing the four relevant factors, the Court concludes that such extraordinary relief is

not warranted. Dr. Giri has not shown that she is likely to succeed on the merits of her claims

because the current record demonstrates that NBME took action against the putative class

because of credible reports of cheating, not discriminatory animus against Nepalis. The balance

of equities and public interest also weigh against permitting potentially unqualified doctors from

matriculating to residency programs and administering care to patients. The Court will,

accordingly, deny Dr. Giri’s motion for a preliminary injunction and deny the corresponding

motion for class certification without prejudice to renewal at an appropriate time should the case

proceed in this Court.

I. Background

After a brief overview of the medical-examination and residency-application processes,

the Court turns to the facts at hand. In doing so, it will credit only factual allegations supported

by the record and allegations in Dr. Giri’s verified Complaint that are within her personal

knowledge. See Grimes v. District of Columbia, 794 F.3d 83, 94 n.5 (D.C. Cir. 2015).

A. The USMLE and the Match Program

The United States Medical Licensing Exam (“USMLE”) is a standardized test

administered every year by NBME and the Federation of State Medical Boards to aspiring

doctors who wish to practice medicine in the United States. See Opp’n, Ex. 3 (“Mechaber

Decl.”) ¶¶ 4, 10. The test consists of three steps: Step One is a written exam consisting of

around 280 multiple-choice questions that measure students’ grasp of various scientific concepts.

2

Opp’n, Ex. 1 (“Ward Decl.”), Attach. D (“2023 Bulletin of Information”) at 5. Step Two is

another written exam, with over 300 multiple-choice questions, assessing students’ ability to

apply “medical knowledge, skills, and understanding of clinical science essential for the

provision of patient care under supervision.” Mechaber Decl. ¶ 8; see also 2023 Bulletin of

Information at 5. Step Three “provides a final assessment of physicians assuming independent

responsibility for delivering general medical care.” Mechaber Decl. ¶ 9. One portion consists of

a multiple-choice examination, and the other involves computer-based clinical simulations. See

2023 Bulletin of Information at 5. Although no preset percentage of examinees will pass or fail

any of these steps, examinees typically must answer roughly 60% of questions correctly to pass.

Id. at 17.

Medical school graduates who have completed at least the first two steps of the USMLE

are eligible to apply for residency through the Match program. See Opp’n, Ex. 4 (“Feddock

Decl.”) ¶ 28. Before the Match process begins, applicants first apply to their chosen residency

programs, which will, in turn, invite applicants of interest for interviews. Id. ¶ 23. To participate

in the process, applicants register with the Match program and then rank their target residency

programs in order of preference. Id. ¶¶ 14, 16. Residency programs, meanwhile, rank applicants

of interest (who will, in all likelihood, be a subset of those applicants they chose to interview).

Id. ¶¶ 16, 26. The Match program then uses an algorithm to place applicants with residency

programs and unveils initial results on Match Day. Id. ¶ 16. The deadline this year for both

applicants and residency programs to submit their rank-order lists is February 28. Id. ¶ 36.

Many state medical boards also require passing scores on the USMLE before issuing full

or temporary licenses. See id. ¶ 7; Opp’n, Ex. 5 (“Johnson Decl.”) ¶ 12. And for graduates of

foreign medical schools, passing scores on the USMLE are also a prerequisite to receive final

3

certification by the Educational Commission for Foreign Medical Graduates (“ECFMG”), which

is required to participate in the Match. See Feddock Decl. ¶ 24.

The USMLE is administered multiple times during the year, and some questions are

recycled from one test administration to the next. See Mechaber Decl. ¶ 13. That repetition

creates some risk that examinees may memorize and disseminate such questions to future test-

takers—a risk NBME has foreseen. Its default procedure when the validity of any particular

exam result is called into question is to delay reporting the result (if the score has not yet been

released) or suspend further distribution of the score and inform the examinee (if the score has

already been distributed). Id., Attach. A (“USMLE Score-Validity Policy”) ¶¶ B.3–B.5. The

score is cancelled only if the examinee fails to timely respond to an inquiry by NBME staff or

provides an inadequate explanation. Id. The Board, however, reserves the right to apply

alternative procedures in some circumstances—particularly those “involving multiple

examinees.” Id. ¶ A.5.

B. Investigation into Exams Associated with Nepal

By early 2023, the Board had received several tips claiming that groups of examinees in

certain countries had distributed exam questions in advance of testing. See Ward Decl. ¶ 6. One

tip, for example, alleged that test-takers in India and Nepal were relying on banks containing live

USMLE questions to attain high scores. Id. Another stated that, in those two countries, test-

takers were “purchasing last six months question papers.” Id. Another, still, reported “concern”

that “USMLE graduates from Nepal have been scoring really high scores . . . [because] almost

all questions are out and [they get] repeated question[s] in their exam[s].” Id. ¶ 8. This tipster

also reported that there were “preparation libraries/reading rooms in Nepal where students gather

and go through these volumes of question[s].” Id.

4

The Board also discovered posts on social media and online chat rooms “suggesting that

groups of individuals in Nepal were collecting and sharing large amounts of secure exam

material in private groups.” Id. ¶ 7. Notably, it dispatched an individual to access an online

messaging group on an app called “Telegram,” where some participants claimed to have seen

questions on their exams that were identical to “Past Questions” (or “PQs”) shared within the

group. Id. ¶¶ 10–13. That undercover operative was required to furnish a USMLE testing permit

and documents showing some nexus to Nepal before being admitted to the group. Id. ¶ 10.

Access to the Telegram group confirmed the initial tips: Individuals were sharing

“recalls” from examinees of prior USMLE questions that were still in active use. Id. ¶ 12.

Multiple of the nearly 1,300 group members boasted that many of the questions they encountered

on their Step One exams were “PQs” shared within the group. Id. “I had my exam on 16th

October,” one poster wrote. “I had around 75% pqs even though most of our friends had around

90–95%.” Id. ¶ 13. “New question takes our time but PQs are our real SAVIOUR,” the poster

crowed. Id. Others echoed that sentiment. “I had my exam today,” another wrote. Id. “Most of

the questions were PQ or around PQ. I request all to go through 1000 pages (saviour). No

words can describe how grateful I am for this group. See you all on the other side. Complete set

loading soon . . . .” Id. At the same time, some group members expressed unease about the rate

at which test-takers were racing through their exams and answering questions that they had

already encountered. “No one is with that type of super power to finish that exam this much

early,” one member presciently observed after warning of the “consequences of talking about

pq.” Id. ¶ 14. “Really guys! [A]re you that much dumb or what?” Id. “[I] have been told that

people [taking] step 1 are finishing their exams 2–3 hours early and are coming out of the

5

prometric centre talking about pq,” another added. Id. “If you have friends or know the admin

of step 1 please forward a message requesting them to take their time during exams.” Id.

In response to this information and other anonymous tips, the USMLE program asked the

Board’s Psychometrics and Data Analysis (“PADA”) department to “analyze examinee

performance data for test centers in Jordan, Nepal, and Pakistan.” Opp’n, Ex. 2 (“Jurich Decl.”)

¶ 6. The results from Nepal were the most extreme. Id. ¶ 7. The country’s sole test center

produced the highest average scores in the world on the 2021 Step One and 2022 Step Two

exams. Id. Individuals who attended medical school there also received top marks globally on

all three steps in 2023—by a long shot. Jurich Decl., Attach. A. For Step One, for instance,

which is scored on a 300-point scale, the average score for a student who attended medical

school in Nepal was around 15 points higher than the average score of students from schools in

the next best performing country. Id. In other words, the home of Everest scaled equal heights

when it came to exam results. And its students moved at turbo speed. In 2022, examinees who

tested at the center in Nepal were among the fastest 5% for the Step One exam and 10% for Step

Two, out of all test-takers worldwide. Jurich Decl. ¶ 7. Further, the number of people taking the

Step One and Step Two exams at the Nepal test center had more than doubled in the span of

three years. Id.

Around April 2023, PADA ran an “agreement analysis” focused on all test centers in

Jordan, Nepal, Pakistan, and two in India to ascertain which examinees may have shared or

accessed secure exam content in advance of testing. Id. ¶ 8. That analysis was designed to

identify sets of exams that contained a “statistically improbable number of the same incorrect

response options to the same questions,” which could be indicative of prior, shared access to

those questions. Id. The results—which focused on Step One and Step Two exam data from

6

2021 and 2022—showed that “the vast majority of examinees with statistically significant

number of matching incorrect answers tested at the Nepal test center.” Id. ¶ 9. A re-run of that

analysis in July 2023, which “focused on examinees who tested at the Nepal test center and/or

were citizens of Nepal (based on information self-disclosed in their application to test),” yielded

similar results. Id. ¶¶ 11–12. As did an analysis of Step Three exams completed by individuals

who attended medical school in Nepal or self-reported as a citizen of Nepal. Id. ¶ 13. (Because

Step Three is offered only in the United States, there were no exams from the Nepali test center.)

PADA later developed criteria for identifying passing exam results of dubious validity,

including the results of incorrect-response agreement analysis, median response times, and

substantial differences in performance on different steps of the exam, among other factors. Id.

¶¶ 15, 17. Such criteria did not necessarily identify “irregular behavior” or intentional

“cheating,” but it provided a basis to question whether the test results genuinely reflected the

knowledge and skills of the examinees. Id. ¶ 16. After applying this analysis to test-score data

from the Nepal test center, self-reported citizens of Nepal, and graduates of Nepali medical

schools, NBME identified 832 examinees (or roughly forty percent of the sample group) from

2021 through 2023 with at least one exam result of questionable validity. Id. ¶¶ 18–19. As a

prophylactic measure—and in a departure from its usual procedures—the Board immediately

invalidated the suspect scores and notified the affected examinees. See Mechaber Decl. ¶¶ 19–

20.

C. Dr. Giri’s Case

Dr. Giri is a citizen and current resident of Nepal who graduated from the Kathmandu

University School of Medical Sciences in 2022. See Compl ¶ 27. In the hope of applying for

residency in the United States, she took the Step One exam in Kathmandu in February 2023, Step

7

Two in India three months later, and Step Three in Connecticut last September. Id. ¶ 28. She

has since received passing scores on all three and avers that she did not cheat. Id. ¶¶ 29–30.

On January 31, 2024, NBME notified her by email that all three of her Step scores were

“invalidated” based on “highly irregular patterns found in [her] exam data that are indicative of

prior and substantial unauthorized access to secure exam content.” Id. ¶ 32; Mot. Prelim. Inj.,

Ex. 1 (“Deb Decl.”), Attach. 1 (“USMLE Email”) at 2. More specifically, the Board found all

three exams displayed “[e]xtremely improbable answer similarity with other examinees testing

on the same form at similar times” and “[a]bnormal question response times.” USMLE Email at

2. Her Step One exam also displayed “[u]nusually high performance.” Id. The email indicated

that the likelihood of observing Dr. Giri’s data results “during normal testing conditions” was

roughly 1 in 100 million. See USMLE Email at 3.

The Board offered Dr. Giri three alternative forms of recourse. First, she could choose to

retake each step of the USMLE free of charge beginning in mid- to late-2024. See id. at 4.

Second, she could request reconsideration of its decision to invalidate her scores, which could

take ten weeks or longer to evaluate. Id. Third, she could take no action, in which case her

access to USMLE would be suspended for three years. Id. In any event, she was required to

report her choice within fifteen calendar days. Id. at 3.

That same day, the Board released a statement on the USMLE website explaining that it

had invalidated certain examinees’ test scores (Dr. Giri’s among them). It read, in part:

The USMLE program regularly monitors and analyzes examinees’

test performances for unusual score patterns or variations, and other

information that could raise questions about the validity of an

examinee’s results. As part of an ongoing investigation, the

USMLE program has identified a pattern of anomalous exam

performance associated with Nepal, which challenges the validity of

test results for a group of examinees. Highly irregular patterns can

be indicative of prior unauthorized access to secure exam content.

8

Examinees with results in question are being notified by the USMLE

Secretariat’s Office that their previous Step scores have been

invalidated and that they will be required to take a validation

exam(s). The USMLE program is working to notify examinees who

need to schedule validation exam(s) and to support score users and

other stakeholders impacted by the validation exam requirements.

Mechaber Decl., Attach. C. A week later, ECFMG sent a letter informing Dr. Giri that, without

valid USMLE scores, she no longer met the requirements for certification and was required to

destroy any previously issued ECFMG Certificate. See Deb Decl., Attach. 3 (“ECFMG Ltr.”) at

1.

Dissatisfied with these limited options, Dr. Giri filed a putative class action on February

12 on behalf of the 832 Nepal-associated examinees whose USMLE scores were invalidated.

See Compl. ¶¶ 57, 60. She alleges that NBME discriminated against her and other putative class

members on the basis of their Nepali national origin and ethnicity, in violation of Title VII and

42 U.S.C. § 1981. Id. ¶¶ 73–80. Alongside her Complaint, she filed the instant Motion for a

Preliminary Injunction ordering NBME to reinstate class members’ test scores in time for the

February 28 rank-order list deadline for residency matching. See Mot. Prelim. Inj. NBME

subsequently agreed to extend the deadline for putative class members to state which of the three

options they intend to pursue until the Court rules on the Motion. See Joint Stipulation ¶ 1; ECF

No. 10 (“Order Adopting Stipulation”).

The parties briefed the motion, as well as the accompanying class-certification motion, on

an expedited basis. The Court heard argument on February 21, after which it summarily denied

the motion from the bench. This opinion elaborates on the Court’s reasoning.

II. Legal Standards

“A preliminary injunction is an extraordinary remedy that should be granted only when

the party seeking the relief, by a clear showing, carries the burden of persuasion.” Cobell v.

9

Norton, 391 F.3d 251, 258 (D.C. Cir. 2004). To obtain a preliminary injunction, the moving

party must show: (1) that she is likely to succeed on the merits of her claim; (2) that she is likely

to suffer irreparable harm in the absence of preliminary relief; (3) that the balance of equities tips

in her favor; and (4) that a preliminary injunction is in the public interest. Winter v. Nat. Res.

Def. Council, Inc., 555 U.S. 7, 20 (2008).

Historically, these factors have “been evaluated on a ‘sliding scale.’” Davis v. Pension

Ben. Guar. Corp., 571 F.3d 1288, 1291 (D.C. Cir. 2009) (quoting Davenport v. Int’l Bhd. of

Teamsters, 166 F.3d 356, 361 (D.C. Cir. 1999)). In other words, if the movant makes an

“unusually strong showing on one of the factors, then it does not necessarily have to make as

strong a showing on another factor.” Id. at 1291–92. This Circuit has hinted, though not held,

that Winter—which overturned the Ninth Circuit’s “possibility of irreparable harm” standard—

establishes that “likelihood of irreparable harm” and “likelihood of success” are “independent,

free-standing requirement[s].” Sherley v. Sebelius, 644 F.3d 388, 392–93 (D.C. Cir. 2011)

(quoting Davis, 571 F.3d at 1296 (Kavanaugh, J., concurring)); see League of Women Voters v.

Newby, 838 F.3d 1, 7 (D.C. Cir. 2016) (declining to address whether the “sliding scale”

approach is valid after Winter). In any event, this Court need not resolve the viability of the

sliding-scale approach today, as it determines that “a preliminary injunction is not appropriate

even under the less demanding sliding-scale analysis.” Sherley, 644 F.3d at 393.

III. Analysis

Dr. Giri seeks an injunction ordering the Board to restore all class members’ USMLE

scores and maintain them as valid during the pendency of this litigation (and any further

investigation the Board may conduct). She wishes, additionally, to compel the Board to notify

all impacted parties—including affected test-takers, medical schools, and hospitals—that it has

10

done so and refrain from applying score-invalidation procedures to test-takers associated with

Nepal that differ from those generally applied to other test-takers. See Mot. Prelim. Inj. at 1; id.,

Ex. 3 (“Proposed Order”). The Court concludes that such extraordinary relief is not merited

under the Winter factors.

A. Likelihood of Success

Dr. Giri brings discrimination claims under Title VII and 42 U.S.C. § 1981. The former

makes it unlawful for an employer to “discriminate against any individual with respect to his

compensation, terms, conditions, or privileges of employment, because of such individual’s race,

color, religion, sex, or national origin.” 42 U.S.C. § 2000e-2(a)(1). Section 1981 similarly

prohibits discrimination but affords its protections in the context of “making, performance,

modification, and termination of contracts, and the enjoyment of all benefits, privileges, terms,

and conditions of the contractual relationship.” 42 U.S.C. § 1981(b). Section 1981 also, as

relevant here, proscribes only discrimination on the basis of “race,” id. § 1981(a), which includes

discrimination on the basis of “ancestry or ethnic characteristics,” Saint Francis Coll. v. Al-

Khazraii, 481 U.S. 604, 613 (1987). The Court briefly notes two preliminary issues concerning

whether these statutes are an appropriate vehicle for Dr. Giri’s claims: extraterritoriality and

applicability to professional-licensing authorities. Then, assuming both statutes apply here for

present purposes, the Court explains why the Board likely did not violate either.

1. Extraterritoriality and Applicability

NBME first protests that the statutes Dr. Giri seeks relief under cannot be stretched so far

as to reach the conduct at issue here. To start, it submits that denying employment to a job

seeker who applies from abroad constitutes extraterritorial conduct to which the statutes do not

extend. Opp’n at 32–33 (citing Nakhid v. Am. Univ., No. 19-cv-3268, 2021 WL 4169355

11

(D.D.C. Sept. 14, 2021), aff’d on other grounds, 2022 WL 2678742 (D.C. Cir. July 12, 2022)).

The Board also reserves any argument it may have that Title VII cannot form the basis of Dr.

Giri’s claim because there is no employment relationship between the parties to speak of. Id. at

32 n.9. The Court need not say any more on these topics. Because another problem with Dr.

Giri’s case proves fatal to her claim at this stage, the Court will assume without deciding that the

statutes provide for liability in this context and march forward. See Morrison v. Nat’l Australia

Bank, Ltd., 561 U.S. 247, 254 (2010) (classifying extraterritorial application as “a merits

question” rather than one going to “a tribunal’s power to hear a case”) (cleaned up)).

2. Inference of Discrimination

The primary stumbling block to Dr. Giri’s claims—under both Title VII and § 1981—is

that the current record does not indicate that the Board’s actions were motivated by ethnicity or

national origin. Dr. Giri’s purported evidence of discriminatory treatment takes two forms. The

first is the group of people to which the Board’s actions applied. In NBME’s own words, it

invalidated the scores of “[e]xaminees with results in question” because of “a pattern of

anomalous exam performance associated with Nepal.” Compl. ¶ 41 (emphasis added). As the

Board has since spelled out, examinees “associated with Nepal” include those who self-identified

on exams as having Nepali citizenship. Opp’n at 13. As a result, Dr. Giri insists, NBME said

“the quiet part out loud” and acknowledged that its conduct was “facially discriminatory.” Mot.

Prelim. Inj. at 19, 21. Dr. Giri’s second source of evidence is the Board’s decision to focus on

cheating in Nepal over cheating that it suspects is ongoing in other countries, arguing that the

Board’s single-mindedness exposes its discriminatory purpose. Reply at 10. The Court will take

these pieces of evidence one at a time.

12

NBME’s statements do not amount to an admission of facial discrimination. To start, the

Board’s action applied, at least in theory, to test-takers who do not belong to the protected group

that was purportedly targeted. Recall that the Board conducted a statistical analysis of the

following examinees: (1) Step One and Two test-takers who sat for the exam in Nepal or self-

identified as a citizen of Nepal, and (2) Step Three test-takers who attended medical school in

Nepal or self-identified as a citizen of Nepal. Jurich Decl. ¶ 18. Conspicuously lacking from

these criteria is a limitation to those of Nepali ethnicity or national origin. Of course, Dr. Giri

alleges that the “overwhelming majority of these people—if not all of them—are Nepali by

citizenship, origin, ethnicity or all three.” Compl. ¶ 45. But even taking this unsupported

allegation as true, such overlap is not enough, on its own, to confirm that discriminatory animus

explains NBME’s decision. See Ky. Ret. Sys. v. EEOC, 554 U.S. 135, 142 (2008) (noting that

policies differentiating on the basis of pension status do not discriminate on the basis of age

absent additional “evidence of intent” even though age and pension status “typically go hand in

hand”). The overlap is especially insufficient to show that the Board took action on the basis of

ethnicity, as distinct from national origin, which Dr. Giri must do to make out her § 1981 race-

discrimination claim. See Ndondji v. InterPark Inc., 768 F. Supp. 2d 263, 273 (D.D.C. 2011).

Nor is the bare fact that citizenship was one of the inputs in defining the class of those

“associated with Nepal” enough to bring this case within the ambit of facial discrimination. Cf.

Ky. Ret. Sys., 554 U.S. at 143 (employer action discriminating on basis of pension status, where

advanced age was one of two ways to become pension eligible, did not “automatically

discriminate[] because of age”). NBME’s use of citizenship information, moreover, is readily

explained by the fact that an online group responsible for disseminating past test questions to

large numbers of people restricted its membership to test-takers who could show a “nexus to

13

Nepal,” which would include Nepali citizenship. Ward Decl. ¶ 10. The Board thus did not

target Nepalese people for heightened scrutiny. It analyzed the scores of the group of test-takers

who had physical and digital access to the locations in which “Past Questions” from prior exams

were being distributed.

What is more, the Board did not indiscriminately cancel the scores of every examinee

meeting its criteria. Only forty percent of test-takers associated with Nepal had scores stricken—

namely, the ones whose scores were individually flagged by NBME’s statistical analysis for

signs of invalidity based on metrics like answer-similarity agreement, performance differences

across different Step exams, and median answer response times. Jurich Decl. ¶¶ 17, 19. That an

allegedly discriminatory policy skips over many of its would-be targets can be good evidence

that discrimination is not, in fact, afoot. That is particularly so when the alleged discriminator

selects the subset of those affected according to a nondiscriminatory rationale. In short, because

the Board’s decision both swept in people who do not belong to the protected group and

excluded significant numbers of them in a way that bespeaks a nondiscriminatory motive, it

cannot be called “facially discriminatory,” at least on the current record.

Dr. Giri’s second contention—that NBME’s neglect of similar cheating outside of Nepal

belies its nondiscriminatory intent—falls flat as well. In fact, the Board’s entirely reasonable and

nondiscriminatory rationale for taking aim at examinees associated with Nepal counts as a point

in its favor. See Ky. Ret. Sys., 554 U.S. at 144 (noting “non-age-related rationale for the

disparity” as reason for rejecting charge of facial discrimination). Nothing in the present record

suggests that NBME went looking for a problem in Nepal out of ethnicity- or national-origin-

based suspicion; it followed the trail of evidence, including tips about organized cheating taking

place in medical schools and at a testing center located in Nepal, and on an online forum for

14

which a “nexus to Nepal” was a ticket to admission. Ward Decl. ¶¶ 7–10. In other words,

location was the crux of the Board’s investigation. Having received tips about several countries,

moreover, the Board investigated all of them. Jurich Decl. ¶¶ 6, 9. In narrowing and prioritizing

its efforts to test-takers associated with Nepal, NBME again relied on neutral inputs: Of the

places that had been the subject of tips, Nepal presented the clearest and most significant site of

potential organized cheating according to metrics such as the prevalence of statistically

anomalous answer similarity. Id. ¶¶ 6–13. NBME’s investigative process thus reflected a desire

to zero in on test-takers for their connection to a locus of cheating, not due to their ethnicity or

national origin.

Dr. Giri sees things differently. She contends that the Board has improperly singled

Nepalis out despite its awareness of a pervasive cheating problem worldwide and despite

confirmation that “similar conduct” occurred in India, Pakistan, and Jordan. Reply at 10–13. No

doubt, Dr. Giri is correct that the Board’s own behavior, such as setting up a tip line, signifies its

appreciation of cheating untethered to any particular place. Ward Decl. ¶ 4. But, according to

NBME’s declarants, that general knowledge is far afield from the specific tips and evidence of

organized, systematic, group-level cheating that prompted the Board to scrutinize test-takers

associated with Nepal. There is nothing unreasonable or untoward about choosing to expend

resources on a particular area of concern against the backdrop of more diffuse problems

elsewhere.

As for India, Pakistan, and Jordan, Dr. Giri is again right that the Board had reason to

investigate those locales in addition to Nepal. It did just that. But the investigation did not

reveal nearly as much cause for concern about cheating in those other countries as it did with

respect to Nepal. Jurich Decl. ¶ 3 (noting the “significantly larger magnitude” of statistical

15

abnormality in Nepal). Dr. Giri twists NBME’s words too far when she suggests that it

identified a “similar” problem in India, Pakistan, and Jordan. See Reply at 9–10. True, the

Board found a statistically significant level of suspicious answer similarity when comparing all

four countries to a baseline comparison group. Jurich Decl. ¶ 9. But it also discovered that “the

vast majority of examinees with a statistically significant number of matching incorrect answers”

within that combined group had “tested at the Nepal test center.” Id. That difference, along with

evidence that test scores in Nepal were dramatically higher than in any other country in the world

justifies a focus on examinees associated with Nepal over other possible sites of cheating. Id.

¶¶ 7, 9, 12. And although Dr. Giri would prefer the Board to act all at once with respect to any

suspected cheating it catches a whiff of, see Reply at 15–16, its decision to tackle the most

severe problem first holds up to scrutiny at this point.

Nor does the Board’s decision to implement a so-called “very punitive” appeal process

different from its usual approach indicate animus. Reply at 11. Indeed, NBME says, it has not

deviated from the norm because its policy always has stated that bespoke procedures may be

necessary for situations “involving multiple examinees.” USMLE Score-Validity Policy ¶ A.5;

cf. Walker v. Johnson, 798 F.3d 1085, 1092 (D.C. Cir. 2015) (listing “deviation[s] from

established procedures” as possible basis for inference of discrimination). This scenario,

implicating over 800 test-takers, is clearly such a situation calling for a procedure geared towards

efficiency. And there is no evidence in the record that the Board would not take similarly swift

and stringent action if confronted with evidence of cheating on a comparable scale elsewhere.

The Board’s Vice President represents, moreover, that to “the extent that issues have been or are

identified reflecting similar conduct by large numbers of examinees in other countries, [he]

expect[s] that those examinees will be subject to the same policies and procedures that have been

16

applied to the examinees at issue” here. Mechaber Decl. ¶ 23. Far from admitting that “similar

conduct” has in fact been identified elsewhere and simply ignored, as Plaintiff reads it, see Reply

at 10–11, this statement merely references the possibility of uncovering comparable conduct.

Nepal may have been the first. But until Dr. Giri can offer evidence of a similarly situated

example of mass localized cheating handled under the default procedure, that is not so

remarkable.

The aspect of the procedure to which Dr. Giri most strongly objects—the invalidation of

already-released scores pending further investigation—also appears sensible under the

circumstances. The Board reasonably sought to mitigate the risk involved in allowing these test-

takers to participate in the upcoming Match, which would, in its view, endanger patients by

allowing unqualified individuals to practice in residency programs while investigation proceeds.

Mechaber Decl. ¶ 22. That aim is as laudable as it is nondiscriminatory.

Dr. Giri does not attack that justification head-on, but several of her allegations indicate

her belief that invalidating first and asking questions later is not so wise for another reason: the

chances of canceling a perfectly innocent score. She uses her own circumstances as an example.

If given the opportunity to respond prior to the cancellation of her scores, Dr. Giri says, she

would have informed the Board that her high degree of answer similarity was explained by her

and her fellow Nepali test-takers’ exposure to the same (legitimate) preparation materials and

medical school curriculum. Compl. ¶ 49. In addition, she posits that her suspiciously quick

answer times were the product of either an innocent test-taking strategy of guessing on difficult

questions or the fact that she encountered technical difficulties during Step Three. Id. ¶¶ 50–51.

Dr. Giri thus maintains that NBME’s score invalidations are hopelessly underinformed before an

examinee has the chance to fill in gaps.

17

The Board knocks down these explanations one by one, effectively demonstrating that a

pre-invalidation response from examinees is not the be-all, end-all guardrail Dr. Giri suggests.

Begin with her answer-similarity theory. NBME retorts that its analysis included examinees who

had attended the same medical school as part of the baseline control group, eliminating the

possibility that the observed answer similarity was attributable to studying closely together in a

small environment. Jurich Decl. ¶ 29. Giri’s explanations for her faster-than-normal response

times fare no better. If she had employed a strategy of guessing on hard questions to save time,

the Board explains, one would expect her performance on faster answers to fall in the 20–25%

correct range. Id. ¶ 31. Instead, she answered correctly on between 85% and 100% of such

questions across the three exams. Id. As for Dr. Giri’s technical problems during the computer-

based case simulation portion of Step Three, the Board explains that this segment was not

included within its review (which encompassed only single-best answer multiple-choice

questions) and therefore would not have impacted her score-validity results. Id. ¶ 33. Preferable

as it may be to consider an examinee’s explanations prior to invalidation, then, Dr. Giri has not

shown it was so crucial and that the Board’s prioritization of a timely and efficient resolution of

these scores was unreasonable or, more importantly, indicative of pretext for animosity towards

Nepalis.

Finally, contrary to Dr. Giri’s suggestion, NBME’s actions here bear little resemblance to

the racial profiling found unconstitutional in Floyd v. City of New York, 959 F. Supp. 2d 540

(S.D.N.Y. 2013). That case involved the singling out of members of particular racial groups for

deprivation of a constitutionally protected privacy right, through police “stop and frisks.” Id. at

558. The antidiscrimination statutes employed here, by contrast, protect against discriminatory

“adverse employment action[s].” Baloch v. Kempthorne, 550 F.3d 1191, 1196 (D.C. Cir. 2008).

18

As Dr. Giri does not, and could not, claim that the Board took an adverse action when deciding

to analyze its own exam data—the point at which it allegedly considered ethnicity and national

origin to narrow its field of vision—her theory of liability is quite distinct from that in Floyd.

In addition, Floyd’s own logic suggests that there is insufficient evidence of

discriminatory motive in this case. Floyd’s basic insight was that “it is impermissible to subject

all members of a racially defined group to heightened police enforcement because some

members of that group appear more frequently in criminal complaints.” 959 F. Supp. 2d at 603.

But as discussed previously, the Board’s decisions did not rely on a propensity inference based

on ethnicity or national origin. Rather, its decisions differentiated on the basis of test-takers’

connection to a geographic region where organized cheating occurred, an approach that Floyd

expressly sanctioned. Id. at 562–63 (recognizing the constitutionality of police “deploy[ing]

their limited resources to high crime areas” where “the need for policing is greatest,” so long as

race does not enter the picture when officers determine who to stop within those areas). And

although NBME’s criteria included examinees with Nepali citizenship, that decision was also

tied to a neutral consideration: whether the examinee could have proved the nexus with Nepal

necessary for gaining admission into the Telegram group where secured answers were being

distributed. Floyd was careful to leave room for such neutral decision making as well. See id. at

663–64 (distinguishing Brown v. City of Oneonta, 221 F.3d 329 (2d Cir. 2000), which held that

a police search focusing on Black suspects because the perpetrator was known to be Black was

not facially discriminatory).

In sum, the Court cannot say at this point that the record portends a likelihood of success

on Dr. Giri’s discrimination claims based on either Title VII or § 1981. Rather, it seems that

19

NBME relied on criteria that, while coinciding with national origin to a large extent, were used

to root out location-based cheating, a perfectly permissible and nondiscriminatory goal.

B. Irreparable Harm

Dr. Giri’s arguments regarding the likelihood of irreparable harm fare somewhat better.

Under this factor, the movant must show that “the injury complained of is of such imminence

that there is a ‘clear and present’ need for equitable relief to prevent irreparable harm.”

Chaplaincy of Full Gospel Churches v. England, 454 F.3d 290, 297 (D.C. Cir. 2006) (citation

and quotation marks omitted). That injury, put another way, must be “certain, great, actual,

imminent, and beyond remediation.” Save Jobs USA v. U.S. Dep’t of Homeland Sec., 105 F.

Supp. 3d 108, 112–13 (D.D.C. 2015). Here, Dr. Giri alleges that the proposed class “will suffer

inevitable harm to their employment opportunities, residency, and immigration status.” Mot.

Prelim. Inj. at 23. While she has offered no evidence that she or any other putative class member

has lost or would lose their visas because their Step exam scores were cancelled, the harm to

their ability to seek residency is obvious enough. A passing score on Steps One and Two of the

USMLE is required to apply for residency programs through Match. Compl. ¶¶ 11, 16, 42. The

final Match application deadline is February 28. Id. ¶ 18. If class members do not have valid

exam scores by then, they will be ineligible to participate in the Match program this year. Id.

¶ 54. While their scores are invalid, moreover, they will remain ineligible for a medical license

in many jurisdictions. See Johnson Decl. ¶ 12.

The loss of opportunity to pursue one’s chosen profession due to alleged discrimination is

widely recognized as constituting an irreparable injury. See, e.g., Ramsay v. Nat’l Bd. of Med.

Examiners, 968 F.3d 251, 254, 262–63 (3d Cir. 2020) (agreeing, partly for that reason, that the

Board’s refusal to provide extra time on an upcoming exam to accommodate plaintiff’s ADHD

20

and dyslexia was irreparable); Enyart v. Nat’l Conf. of Bar Examiners, Inc., 630 F.3d 1153,

1165–66 (9th Cir. 2011) (“[Plaintiff] demonstrated irreparable harm in the form of the loss of

opportunity to pursue her chosen profession. . . . If she fails the Bar Exam or scores too low on

the MPRE to qualify for admission, [she] cannot be licensed to practice law in California.”);

Maczaczyj v. State of New York, 956 F. Supp. 403, 408 (W.D.N.Y. 1997) (exclusion from

masters program “will most likely affect plaintiff’s ability to engage in the future employment of

his choice”); see also Tanner v. Fed. Bureau of Prisons, 433 F. Supp. 2d 117, 125 (D.D.C. 2006)

(citing Carson v. Am. Brands, Inc., 450 U.S. 79, 89 n.16 (1981), for the similar proposition that

“[t]he loss of specific job opportunities, training and competitive advantages can constitute

irreparable harm”).

This is true even where, as here, the barrier to the plaintiff’s pursuit is only temporary. In

Bonnette v. District of Columbia Court of Appeals, for example, a legally blind aspiring lawyer

was denied certain assistive technology she requested to take the July 2011 Multistate Bar Exam

(“MBE”). 796 F. Supp. 2d 164, 167 (D.D.C. 2011). She moved for a preliminary injunction to

avoid having to either take the exam “under discriminatory conditions” or “wait until at least the

February 2012 administration while her claim is litigated.” Id. at 186. The Court found that she

faced irreparable harm because “any delay in taking the MBE deprives her of time to practice her

chosen profession,” and the “substantial time and effort” she expended in preparing for the July

2011 exam would have been “effectively wasted if she must wait to take the test at a later date.”

Id. at 186–87. Similarly, in Doe v. Pennsylvania State University, a student in a seven-year pre-

med program was temporarily barred from participating after another student accused him of

sexual assault, and he moved to preliminarily enjoin the sanction. 276 F. Supp. 3d 300, 302 &

n.1, 306 (M.D. Pa. 2017). The court concluded that, even if his suspension “were limited to two

21

years,” as the University had argued, “this gap would constitute irreparable harm as he would

forever be forced to explain his lengthy tenure within this program and, ultimately, his delayed

entry in the professional workforce.” Id. at 315. So too here. That Dr. Giri may pursue the

Board’s appeal procedures or retake the exam at a later date does not eliminate the harm of being

unable to participate in this year’s Match or pursue the practice of medicine in the interim.

NBME’s contrary arguments do not persuade the Court that it should break ranks from

this steady line of precedent. First, while it is true that purely economic harms that may later be

compensated monetarily are not irreparable, see Opp’n at 36, “[t]he lost opportunity to engage in

one’s preferred occupation goes beyond monetary deprivation,” Bonnette, 796 F. Supp. 2d 186.

Second, the Board marshals several reasons why it is somewhat speculative whether Dr. Giri or

any class member would ultimately match with a residency program—including that no evidence

suggests they received residency interviews or that any program would rank them. See Opp’n at

36–37. But Dr. Giri attests that she “was preparing to enter the Match [in January 2024] and

hoping to enter residency in the summer of 2024,” Compl. ¶ 31, and her lack of USMLE scores

guarantees that she will be ineligible for this year’s Match. That lost eligibility and opportunity

to compete is a cognizable and irreparable harm in itself. See Tanner, 433 F. Supp. 2d at 125

(finding loss of eligibility to participate in vocational training program was irreparable, though

the benefits of participation itself and any job opportunities to be derived therefrom were too

speculative). Third, the Board suggests that Dr. Giri may apply for U.S. residency programs

directly, outside of the Match process, or seek licensure in other countries. See Opp’n at 37. But

the former path still requires valid USMLE scores, see Feddock Decl. ¶ 27; Compl. ¶ 16, and is

available to a “relatively small number of medical school graduates” in any event, Feddock Decl.

¶ 21. And the latter option is no substitute for practicing medicine in the United States, which

22

Dr. Giri clearly wishes to do—that, after all, is why she took the United States Medical

Licensing Exam.

NBME’s cases concerning professional harm do not warrant a different conclusion. Doe

v. Ohio State University, for example, is easily distinguished. No. 15-cv-2830, 2016 WL 692547

(S.D. Ohio Feb. 22, 2016), adopted and aff’d, 2016 WL 1578750 (S.D. Ohio Apr. 20, 2016).

There, a student was dismissed from the Ohio State University College of Medicine following

accusations of sexual assault, and he subsequently enrolled in a foreign medical school. Id. at

*5. Hoping to apply for residency in his final year at that new institution, and fearful that his

prior dismissal would undermine his candidacy, the plaintiff moved for a preliminary injunction

ordering Ohio State to reinstate him as a student in good standing. Id. at *1, 5. In finding no

likelihood of irreparable harm, the court noted that plaintiff “may well obtain residency” despite

his record, and that it was “far from clear that . . . he will be unable to complete his education or

to pursue his chosen profession.” Id. at *11. Here, by contrast, the cancellation of class

members’ scores goes well beyond diminishing their competitiveness. It makes them ineligible

for the Match and medical licensure this go around.

Mahmood v. National Board of Medical Examiners is slightly more in tension with Dr.

Giri’s irreparable-harm arguments, but it too is distinguishable, and its reasoning is less

persuasive in any event. No. 12-cv-1544, 2012 WL 2368462 (E.D. Pa. June 21, 2012). There, a

test-taker was suspended from the USMLE for three years after starting a fire in the restroom

during a test administration, id. at *1, and she later sought to enjoin that punishment on the

ground that it “preclude[d] her ability to graduate from medical school within seven years,” id. at

*5. The court did not consider the consequences for her ability to apply for residency or obtain

medical licenses—the harms at issue here. Rather, it merely noted the absence of evidence that

23

she could not transfer schools or obtain a graduation-deadline extension. Id. It further reasoned

categorically that “delays in testing or education services do not constitute irreparable harm.” Id.

However, that notion conflicts with subsequent Third Circuit case law. See Ramsey, 968 F.3d at

262 (favorably citing a district court case for the principle that a “gap in medical school

education and likelihood that the student could not gain acceptance to another school constituted

irreparable harm”). Consequently, Mahmood’s persuasiveness within that jurisdiction, let alone

outside of it, is in doubt.

In short, the likely harm that Dr. Giri and the putative class will face absent an injunction

is sufficient to satisfy the irreparable-injury requirement. This factor alone, however, does not

carry the day.

C. Balance of the Equities

Even if Dr. Giri has met her burden of demonstrating irreparable injury, that harm must

be weighed against NBME’s competing interests. The balance of equities does not clearly weigh

in favor of granting the preliminary injunction here.

On Dr. Giri’s side of the ledger, the equities of the putative class members vary. For Dr.

Giri herself, while her injury may be “irreparable” for the reasons stated above, it is nonetheless

speculative. Many applicants who participate in the Match are not placed with a program. Last

year’s match rate was 59.4% for non-U.S. citizen graduates of international medical schools.

Feddock Decl. ¶ 38. But that number is not evenly distributed. The application portal opens in

September, and programs review applications and interview candidates from then until the end of

January. Id. ¶ 36. “If a residency program applicant is not invited to interview with a residency

program, it is a virtual certainty that the applicant will not be listed on the program’s Rank Order

List for the Main Residency Match.” Id. ¶ 26. At no point has Dr. Giri claimed that she

24

interviewed with any program—even though NBME expressly raised this point in its opposition.

See Opp’n at 25, 27. The only fair inference, then, is that Dr. Giri has not been invited for an

interview. Her purported injury from being excluded from the Match this go around is therefore

conjectural.

Other putative class members appear to be differently situated. Surely some have

received interviews, and, of that segment, it is fair to assume a fraction would have matched if

their scores had not been invalidated given that over half of foreign applicants are paired with a

program. At the same time, though, it is questionable whether those figures would hold true for

individuals who already have had their scores scratched. The residency programs are almost

certainly aware that these applicants’ exams have been flagged as suspect, and no decision from

this Court would change that fact or prevent the Board from invalidating these scores again at a

later date pursuant to proper procedures. Accepting these applicants would therefore be a risky

proposition—one that a great many programs may not be willing to wager.

The segment with the most at stake, by far, are any putative class members who already

matched in a prior year and are at risk of being removed from their programs. Beyond disrupting

their education, if removed from their programs, these individuals are at risk of losing their J1 or

HB1 visas, both of which require the person maintain participation in an educational program.

See Compl. ¶ 20. For now, though, no named plaintiff falls into this category. And the Court

lacks any visibility into just how many people might fit this mold.

Turning to the Board, Dr. Giri gives short shrift to its interest in this action when

asserting that the emergency relief would not harm the group’s pocketbook or impose real

administrative burdens. See Mot. Prelim. Inj. at 24–26. The Board has something much more

valuable at stake, however: ensuring the integrity of the results that it reports to residency

25

programs and licensing authorities. These entities rely on the Board to provide accurate

assessments of applicants’ knowledge and abilities. To perform this “valuable service,” the

Board must be able to “assure itself of the validity of students’ scores through internal review

procedures.” Murray v. Educ. Testing Serv., 170 F.3d 514, 517 (5th Cir. 1999); see also San

Mateo Union High Sch. Dist. v. Educ. Testing Servs., No. C 13-3660 SBA, 2013 WL 4711611,

at *15 (N.D. Cal. Aug. 30, 2013) (“Forcing Defendants to validate AP exam scores resulting

from improperly administered tests would place them in the untenable position of having to act

contrary to their obligations in the AP Bulletin, and also would result in colleges and universities

being less likely to rely on the integrity of such scores.”). A judicial order interfering with the

Board’s review procedures and forcing it to validate tainted scores that may not reflect the test-

takers’ actual medical acumen would risk severe damage to the organization’s brand—a harm

that could not be remedied after the fact. It is often said that reputation arrives on foot but leaves

on horseback. Once the test results exit the stable and are relied upon by residency programs for

matching, any subsequent effort to correct the records may be too little, too late for NBME’s

reputation. The damage will have been done. Cf. Rothberg v. L. Sch. Adm. Council, 102 Fed.

App’x 122, 125 (10th Cir. 2004) (noting “the LSAC cannot hope for relief once the injunction

issues”).

To be sure, the strength of the Board’s equities oscillates depending on the category of

putative class member. Its interest reaches its highest peak when it comes to applicants who are

likely to match with resident programs this cycle because swift action on the front end saves the

Board the embarrassment and reputational hit of having to invalidate exam results after a match

has been made and a residency program has tendered an irretrievable investment. By contrast,

the Board’s stake in the matter is at its lowest ebb when it comes to individuals who will not

26

match regardless, either because they never applied or were never interviewed, because no

program is likely to rely on the exam scores in the interim. Putative class members who have

matched in past cycles and are in residency fall somewhere in the middle. But even then, the

Board has a considerable interest in correcting the record as soon as possible while programs

wait in limbo pending a final resolution of this matter and implicated residents continue treating

patients.

Weighing these varied interests, the Court cannot conclude that the balance categorically

and unequivocally falls in favor of Dr. Giri. She certainly has not carried her burden when it

comes to her personally, and a more fine-grained analysis is required for other putative class

members. This lack of uniformity only underscores the troubles with provisionally certifying a

class for emergency relief where the putative class is splintered into different categories and the

equities for each group diverge.

D. Public Interest

Beyond the interests of the two parties in this litigation, the public interest plainly weighs

against granting the request for emergency relief. Indeed, this factor alone justifies denying the

preliminary injunction because the Court must give due weight to the vital interests of patients

who might receive health care from unqualified residents; other residency applicants who may be

rejected in favor of persons whose scores will later be invalidated again; residency programs that

will devote resources to the aspiring doctors they supervise; and hospitals that will be left in a

lurch if scores are cancelled once residents already have started on the job. See Winter, 555 U.S.

at 376–77 (“In exercising their sound discretion, courts of equity should pay particular regard for

the public consequences in employing the extraordinary remedy of injunction.” (quoting

Weinberger v. Romero-Barcelo, 456 U.S. 305, 312 (1982)).

27

First and foremost is the overriding interest in public safety. This is a case about the

credentials of doctors applying to medical residency programs. Resident physicians provide

medical care and prescribe drugs. See Feddock Decl. ¶¶ 12–13; Johnson Decl. ¶ 15. In doing so,

they take patients’ lives into their hands. Granting the preliminary injunction would create an

unacceptable risk that individuals who lack the requisite knowledge and skills they purport to

possess because they achieved their exam scores fraudulently will be administering medical care

to unsuspecting patients across the nation. See Feddock Decl. ¶ 35.a. That is a danger the public

should not be forced to shoulder. And, in addition to the immediate peril to patients’ health, the

perception alone that the licensing process is not functioning properly could erode confidence in

the medical profession, further jeopardizing public safety.

Aside from the grave risks to the public writ large, granting the requested relief would

saddle third parties with unwarranted burdens. Other residency candidates would be the first

collateral damage. The Match is a competitive, zero-sum process in which there are many more

qualified candidates than open slots. Anyone who matches based on fraudulent exam results is

necessarily taking a spot from a student who honestly earned his or her marks. See id. ¶¶ 29–33,

35.c. That is the furthest thing from fair or equitable. And the damage does not stop there.

Medical residency programs invest considerable resources recruiting and training residents. See

id. ¶¶ 34, 35.d. If a resident’s scores are later invalidated due to cheating, all of that time,

money, and effort will have been wasted. Hospitals and health care facilities, meanwhile, would

have to pick up the slack for staff members whose scores have been invalidated while reviewing

all past patient encounters handled by the unqualified resident. See id. ¶¶ 35.b, 35.d. As any

doctor could tell you, though, there is hardly any slack in the first place. The natural result

28

would be staffing shortfalls, longer shifts, and morale problems in the facilities where the

discredited residents once worked. See id. ¶ 35.e.

Dr. Giri “acknowledge[s that] there is a public interest in ensuring doctors treating

patients are well-trained, and that residency programs and resident applicants are not

disadvantaged by cheaters.” Reply at 22. However, she contends that it is antithetical to the

public interest to address these issues by targeting a protected class of test-takers given the

paramount importance of vanquishing invidious discrimination. See id. at 22; Mot. Prelim. Inj.

at 26. But there is scant evidence here that NBME discriminated against anyone. As explained

above, the record instead indicates that NBME took appropriate action when responding to well-

documented allegations of cheating centered in one country and invalidated scores only after

conducting individualized statistical analyses. The public interest is not well served by requiring

NBME to ignore this evidence of a concentrated problem, and the creed “like cases should be

treated alike” has limited purchase where the record shows that there was nowhere like Nepal

when it came to concerns about cheating. Nor is the common welfare furthered by reinstituting

suspect scores and allowing unqualified candidates to match with programs and begin practicing

medicine. Any vision of antidiscrimination law that suggests otherwise has taken a wrong turn

somewhere.

Finally, despite her prior acknowledgment, Dr. Giri also attempts to discount the

exorbitant public costs detailed above. In her telling, any concerns that ill-equipped doctors who

cheated on their exams may match with residency programs is muted by the fact that “residents

are likely supervised in practice and there will be many opportunities to prevent incompetent

practitioners from harming patients.” Reply at 23. She takes solace in the fact that NBME long

has suspected some test-takers may be cheating but, despite never taking such stringent action

29

before, cannot identify “a single . . . harm.” Id. The Court finds little comfort in this contention,

and it doubts many patients would either. Medical licensing exams are not pointless acts of

paper pushing and scantron penciling: Almost every jurisdiction in the United States relies on

them to ensure that prospective doctors have the requisite competencies to administer safe and

effective health care before practicing medicine. See Johnson Decl. ¶ 12. To permit those who

have fraudulently achieved their marks to provide direct care to patients poses a safety risk that is

as obvious as it is intolerable.

* * *

In sum, Dr. Giri has not shown that she is likely to succeed on the merits of her claims,

and the balance of equities and public interest both weigh against granting this extraordinary

relief. The Court will accordingly deny Dr. Giri’s request for a preliminary injunction. Because

the motion for provisional class certification was trained at the request for emergency relief,

which the Court has denied, the Court will also deny that motion, without prejudice to renewal at

an appropriate time should the case proceed in this Court.

Before wrapping up, there are two additional items that the Court must address. At the

close of the motion hearing on this matter, Dr. Giri requested the Court grant an injunction

pending appeal pursuant to Federal Rule of Appellate Procedure 8(a). In determining whether to

grant such an injunction, courts consider four factors: (1) likelihood of success on the merits of

the appeal; (2) irreparable injury; (3) substantial harm to other parties; and (4) the public interest.

See Hilton v. Braunskill, 481 U.S. 770, 776 (1987). If those four factors ring a bell, that’s

because the “standards for evaluating a motion for injunction pending appeal are substantially

the same as those for issuing a preliminary injunction.” Republican Nat’l Comm. v. Pelosi, No.

22-cv-659 (TJK), 2022 WL 1604670, at *2 (D.D.C. May 20, 2022) (quotation marks omitted).

30

Unsurprisingly, then, the Court will deny the motion for an injunction pending appeal for the

reasons above.

Dr. Giri also asked the Court to extend the deadline for putative class members to

respond to the Board’s email by selecting one of the three available options: request

reconsideration, retake the exams free-of-charge, or stand pat. At a minimum, Dr. Giri requests

that the Court enter an order waiving the requirement that applicants forfeit their right to sue if

they select the first or second options. Yet she cites no authority empowering the Court to enter

such relief, and the Court is aware of no such authorization. The parties previously stipulated

that the Board would push the deadlines for responding until after the Court’s decision on the

preliminary-injunction motion. See Joint Stipulation at 1. The Board, of course, did not have to

agree to that deal. And now that the Court has entered its decision and the stipulation has

terminated, the Court lacks power to extend its terms.

IV. Conclusion

For these reasons, it is hereby

ORDERED that [Dkt. No. 3] Plaintiff’s Motion for Preliminary Injunction is DENIED;

it is further

ORDERED that [Dkt. No. 11] Plaintiff’s Motion to Certify Class is DENIED without

prejudice to renewal; it is further

ORDERED that Plaintiff’s motion for an injunction pending appeal is DENIED; it is

further

ORDERED that Plaintiff’s motion to continue the deadlines for responding to the

Board’s score-invalidation notice or, alternatively, waive the requirement that individuals forfeit

31

their right to sue if they choose to appeal their invalidation or retake the exams for free is hereby

DENIED.

SO ORDERED.

CHRISTOPHER R. COOPER

United States District Judge

Date: February 23, 2024

32