Keywords
ophthalmology match - medical education - holistic review - interview selection -
residency selection - preference signaling - program directors
Background
Each year, the number of residency applications submitted per student has increased
across all specialties including ophthalmology.[1] During the 2022 application cycle, successfully matched ophthalmology applicants
applied to an average of 81 programs—a 12.5% increase compared to ophthalmology applicants
during the 2017 application cycle.[2] On average, the number of applications submitted for successfully matched ophthalmology
applicants has increased by 1.5 each year over the last 6 years. In a 2021 survey
distributed to ophthalmology program directors (PDs) after the first application cycle
in which interviews were held virtually, 72% of participating ophthalmology PDs reported
an increased number of applications received by their respective programs compared
to previous years.[3] The rise in applications that programs received has limited the time available for
holistic review, and as a result, ophthalmology PDs rely on data including United
States Medical Licensing Examination (USMLE) scores, class rank, grades, and Alpha
Omega Alpha status.[4] While previously thought to be objective measures, these factors have been found
to be subject to bias and racial disparities, and the transition of USMLE step 1 scores
from numerical grades to pass/fail has made the review process more difficult.[5]
[6]
Starting with the 2021 to 2022 application cycle, many other medical specialties including
otolaryngology, dermatology, and urology have implemented preference signaling as
a possible solution to application overload.[7]
[8]
[9]
[10]
[11]
[12] Preference signaling is a standardized system in which residency applicants may
indicate to particular programs a sincere interest before application review and interview
selection.[9] Applicants are limited to the number of signals they are able to send. Signaling
aims to both mitigate the disparities inherent in resource inequities between applicants
while improving transparency by providing a formal, uniform system for expressing
interest. While preference signaling may come with benefits, it has only been recently
implemented in the residency application process.
The ideal number of signals per applicant remains unclear. Having too many tokens
would decrease their value, while limiting them would force applicants to choose between
their top programs. For the 2021 to 2022 application cycle, otolaryngology allowed
four signals, dermatology allowed three signals, and internal medicine, surgery, and
urology allowed five signals.[11]
[12]
[13] The number of programs and applicants varies between these specialties and ophthalmology,
so it is difficult to discern an ideal number of signals for ophthalmology applicants
based on these guidelines. This study aims to clarify the perception among ophthalmologists
involved in residency selection of integrating preference signaling into the Ophthalmology
Match.
Methods
An anonymous online 19-item questionnaire hosted by SurveyMonkey (Momentive Inc.,
Waterford, NY) was approved by the Association of University Professors of Ophthalmology
(AUPO) Data Resource Committee for distribution to 391 individuals from the AUPO Chair,
PDs, and Directors of Medical Student Education email listservs on August 1, 2022.
The email included a description of the study and a secure link to the SurveyMonkey
questionnaire. The study protocol was determined to be exempt from Institutional Review
Board (IRB) approval by the University of Maryland Baltimore IRB and University of
Miami IRB. The research performed adhered to the tenets of the Declaration of Helsinki.
Participants answered questions about demographic data, years of experience in residency
selection, and rated statements characterizing how ophthalmology faculty perceived
the use of preference signaling in the ophthalmology residency match ([Supplementary Material 1], available in the online version). Statements were rated based on a Likert scale
(strongly disagree, disagree, neither disagree or agree, agree, or strongly agree).
Responses to the questionnaire were collected over the course of 9 weeks between August
1 to October 1, 2022. Per AUPO Survey guidelines, no reminders for completing the
questionnaire were sent after the initial email. Statistical analysis was performed
using Microsoft Excel (Microsoft Corporation, Redmond, WA).
Results
Respondent Demographics
The questionnaire was sent to 391 individuals and a total of 96 (24.6%) ophthalmology
faculty responded. Participants were allowed to choose more than one option when answering
the question regarding positions held in the ophthalmology department. For example,
a participant may select “program director” and “professor” if they fall under both
the positions. Participants included PDs (n = 39, 40.6%), Chairs (n = 30, 31.3%), Directors of Medical Student Education (n = 26, 27%), Vice-Chairs (n = 12, 12.5%), and Associate/Assistant PDs (n = 9, 9.4%). Faculty rank included Professors (n = 15, 15.6%), Associate Professors (n = 13, 13.5%), and Assistant Professors (n = 12, 12.5%). Responses were distributed among programs from the North Central (n = 27, 28%), New England and Middle Atlantic (n = 24, 25%), South Atlantic (n = 19, 19.8%), South Central (n = 18, 18.8%), Pacific (n = 5, 5.2%), and Mountain (n = 3, 3%) regions. The median number of years for postresidency graduation was 15.5.
The median number of years involved in both the interview selection process and ranking
process was 10. Of the participants, 61.5% were male (n = 59), 36.5% were female (n = 35), and 2.1% preferred not to specify (n = 2; [Table 1]).
Table 1
Demographic data of survey participants from ophthalmology residency programs
|
Characteristics
|
Number of responses (%)
|
|
Position
|
|
|
Program Director
|
39 (40.6)
|
|
Associate/Assistant Program Director
|
9 (9.4)
|
|
Chair
|
30 (31.2)
|
|
Vice-Chair
|
12 (12.5)
|
|
Director of Medical Student Education
|
26 (27.1)
|
|
Professor
|
15 (15.6)
|
|
Associate Professor
|
13 (13.5)
|
|
Assistant Professor
|
12 (12.5)
|
|
Gender
|
|
|
Male
|
59 (61.5)
|
|
Female
|
35 (36.5)
|
|
Prefer not to specify
|
2 (2.1)
|
|
Years postresidency graduation, median (IQR)
|
15.5 (9.8–26.3)
|
|
Years involved in interview selection, median (IQR)
|
10 (5–15)
|
|
Years involved in ranking process, median (IQR)
|
10 (4–15)
|
|
Region
|
|
|
New England and Middle Atlantic
|
24 (25)
|
|
North Central
|
27 (28.1)
|
|
South Atlantic
|
19 (19.8)
|
|
South Central
|
18 (18.8)
|
|
Mountain
|
3 (3.1)
|
|
Pacific
|
5 (5.2)
|
Abbreviation: IQR, interquartile range.
Perceptions of Preference Signaling
A majority of participants agreed or strongly agreed that preference signaling should
be implemented in the ophthalmology application system (n = 76, 79.2%; [Table 2]). They also believed that implementing preference signaling will allow for more
holistic reviews of applications (n = 55, 57.3%), benefit applicants who do not have connections to home programs or
faculty that can reach out to desired programs (n = 81, 84.4%), improve the distribution of interviews to applicants (n = 76, 79.2%), decrease geographical bias when offering interviews (n = 62, 64.6%), and will be a better alternative to program-specific essays for the
initial application review phase (n = 59, 61.5%). Most participants agreed or strongly agreed that applicants who have
signaled interest in their program will receive preference when offering interviews
(n = 59, 61.5%) and those signals will be used as a tiebreaker for similar applications
(n = 75, 78.1%). However, only 39.6% (n = 38) of participants felt that applicants who have signaled interest in their program
will receive preference when determining a rank list and only 16.7% (n = 16) viewed absence of a signal as a sign of disinterest. Additionally, most participants
agreed or strongly agreed that implementing preference signaling would de-emphasize
numerical cutoffs in the application review process (n = 49, 51.0%), but a majority did not agree or strongly agree that signals would be
effective alternatives to away rotations for applicants to demonstrate interest in
a program (n = 43, 44.8%). In terms of number of preferences signals that applicants should receive,
the majority of respondents answered either three to four signals (n = 35, 36.0%) or five to six signals (n = 29, 30.2%; [Fig. 1]).
Table 2
Survey participants' perspectives of preference signaling on the ophthalmology match
|
Statement
|
Strongly disagree
|
Disagree
|
Neither agree/disagree
|
Agree
|
Strongly agree
|
|
Preference signaling should be implemented in the ophthalmology application system
|
6 (6.3%)
|
4 (4.2%)
|
10 (10.4%)
|
25 (26.0%)
|
51 (53.1%)
|
|
Implementing preference signaling will allow for more holistic reviews of applications
|
6 (6.3%)
|
10 (10.4%)
|
25 (26.0%)
|
25 (26.0%)
|
30 (31.3%)
|
|
Implementing preference signaling will de-emphasize numerical cutoffs in the application
review process
|
7 (7.3%)
|
20 (20.8%)
|
20 (20.8%)
|
21 (21.9%)
|
28 (29.2%)
|
|
Implementing preference signaling will benefit applicants who do not have connections
to home programs or faculty that can reach out to desired programs
|
2 (2.1%)
|
4 (4.2%)
|
9 (9.4%)
|
39 (40.6%)
|
42 (43.8%)
|
|
Implementing preference signaling will improve the distribution of interviews to applicants
(avoiding a subset of applicants receiving a larger number of interviews per person)
|
5 (5.2%)
|
1 (1.0%)
|
14 (14.6%)
|
40 (41.7%)
|
36 (37.5%)
|
|
Implementing preference signaling will decrease geographical bias when offering interviews
|
5 (5.2%)
|
13 (13.5%)
|
16 (16.7%)
|
33 (34.4%)
|
29 (30.2%)
|
|
Preference signaling will be an effective alternative to away rotations for applicants
to demonstrate interest in a program
|
8 (8.3%)
|
20 (20.8%)
|
25 (26.0%)
|
30 (31.3%)
|
13 (13.5%)
|
|
Preference signaling will be a better alternative to program specific essays for the
initial application review phase
|
1 (1.0%)
|
13 (13.5%)
|
23 (24.0%)
|
33 (34.4%)
|
26 (27.1%)
|
|
An applicant who has signaled interest in my program will receive preference when
offering interviews
|
2 (2.1%)
|
3 (3.1%)
|
22 (22.9%)
|
54 (56.3%)
|
15 (15.6%)
|
|
An applicant who has signaled interest in my program will receive preference when
determining a rank list
|
5 (5.2%)
|
10 (10.4%)
|
43 (44.8%)
|
34 (35.4%)
|
4 (4.2%)
|
|
Signals will be used as a tiebreaker for similar applications
|
6 (6.3%)
|
1 (1.0%)
|
14 (14.6%)
|
54 (56.3%)
|
21 (21.9%)
|
|
Absence of a signal will be viewed as a sign of disinterest
|
18 (18.8%)
|
31 (32.3%)
|
31 (32.3%)
|
13 (13.5%)
|
3 (3.1%)
|
Fig. 1 Percentage distribution of participant responses to the question, “How many preference
signals should applicants be given?”.
Discussion
Overall, preference signaling is viewed favorably among ophthalmology faculty members
that responded to this questionnaire. The results of this study align with existing
literature suggesting that signaling is a way to provide applicants a standardized
system for expressing interest to desired programs.[7]
[8]
[11]
[13]
[14] Survey studies in dermatology, otolaryngology, and urology have found considerable
advantages to using signaling such as improving the distribution of interview offers
among applicants and allowing seriously interested applicants to standout to desired
programs.[7]
[15]
[16] PDs in otolaryngology reported that signals were used as a tiebreaker for similar
applications and as part of the initial application review algorithm.[7] The rate at which applicants received interview offers was significantly higher
for programs they signaled compared to nonsignaled programs in the otolaryngology,
dermatology, and urology residency match.[7]
[15]
[16] These specialties reported that signals were most commonly used as part of the initial
application review to extend interviews and as a tiebreaker for similar applications,
but not for determining a rank list, which is consistent with our findings.[7] Additionally, orthopaedic PDs surveyed ranked signaling only below away rotation
performance and personal knowledge of the applicant in terms of importance for residency
selection.[17] Given the promising results in other specialties' matches, the American Orthopaedic
Association's Council of Orthopaedic Residency Directors announced in March 2022 its
endorsement for a preference signaling program for the 2022 to 2023 application cycle.
Traditionally, applicants would use opportunities such as away rotations, advocacy
from faculty mentors, and preinterview communication (emails, phone calls, etc.) to
demonstrate interest in desired programs.[7] While these methods can be effective, they may exacerbate inequities in the application
process for those who do not have home ophthalmology programs or students with limited
financial means who cannot afford the cost of an away rotation. A survey distributed
to medical students who participated in away rotations reported that the average cost
of a single rotation was $958.[18] Previous studies have shown that ophthalmology applicants without home residency
programs are at a considerable disadvantage. The presence of a home ophthalmology
residency program was associated with a 1.4-fold increase in likelihood of matching.
However, only 53% of U.S. medical schools are associated with an ophthalmology residency
program.[19]
[20] Medical students who have an ophthalmology department may have advantages in terms
of access to resources and mentors who can advocate on their behalf during the residency
application process. For these reasons, it is notable that the statement “implementing
preference signaling will benefit applicants who do not have connections to home programs
or faculty that can reach out to desired programs” had the highest percentage of people
agree or strongly agree (84.4%) in this questionnaire. This strongly supports that
signaling may make the residency application process more equitable for these students.
Previous studies also reported that approximately 60% of applicants matched to the
same geographic region as their medical school.[20] A majority of participants in our study believed that signaling may decrease geographical
bias when offering interviews, which means that it has the potential to bring more
visibility to applicants looking to move to certain regions that may have been previously
inaccessible to them. For example, a medical student may have chosen to go to an in-state
school for the lower cost of tuition rather than a school in a desirable location
that was out-of-state.
Our analysis also shows that signaling may streamline the application review and interview
distribution process of the ophthalmology residency match. A majority of ophthalmology
faculty believe that signaling will allow for more holistic review of applications
and improve the distribution of interviews to applicants. A common problem outlined
across different competitive surgical subspecialties was that a disproportionate number
of interviews were being given to a subset of “high-tier” applicants.[9]
[21] This may be in part due to the increasing number of applications submitted per student.
To continue completing holistic review of applicants in a reasonable amount of time,
residency programs may rely more on filters like USMLE scores and class rank which
may exclude a cohort of applicants. Signaling may improve the efficiency of the interview
invitation process by allowing programs to assign interviews to truly interested applicants
and potentially decrease the number of interview cancellations, especially with the
interview cap of 15 for the 2023 ophthalmology application cycle. Allowing a formal
system for applicants to signal preferences may also decrease the amount of time spent
by applicants sending and program leadership responding to informal communications
like letters of interest.
It is important to note that the majority of participants reported that an absence
of a signal would not be viewed as a sign of disinterest in their program. This is
a valuable perspective to keep in mind when determining the number of signals that
should be available for ophthalmology applicants because having too many signals may
lead to programs viewing the lack of a signal as an indication of disinterest. From
our survey, the ideal number of signals is between three and six, which is comparable
to other competitive specialties.[11]
[12]
[13] During the 2021 to 2022 application cycle, there were 643 otolaryngology, 1,019
dermatology, 604 urology, and 748 ophthalmology applicants. The average applications
per program were 427 for otolaryngology, 566 for dermatology, and 350 for urology.[22] Although there are no data for the average applications per programs for ophthalmology,
the number of applicants is closest to otolaryngology which allowed four signals.
Given this comparison and the results of our survey, four signals may be the ideal
number for the ophthalmology residency match. The perception of signaling as an effective
alternative to away rotations remains contentious among our respondents. This may
be because away rotations require a much larger time and financial commitment. However,
as board examinations move towards pass–fail scoring, ophthalmology PDs reported that
they will place increased importance on applicants attending a rotation in their department
and believed applicants will attend more away rotations.[23] As mentioned before, medical students with financial barriers may already be geographically
limited due to attending an in-state versus an out-of-state school. These same applicants
may also be unable to afford multiple away rotations which puts them at a significant
disadvantage, so preference signaling may be an equitable alternative.
Strengths of this study include a relatively large, diverse sample size with participants
across different positions, levels of experience, and geographical regions. Limitations
of this study include potential responder bias as ophthalmology faculty with stronger
opinions about the topic may be more likely to respond. Another limitation is that
this survey only explored the potential benefits of preference signaling. Future studies
will need to explore potential drawbacks and strategies applicants should consider
when using the signaling system, such as using signals on less competitive programs
to secure a “safety” interview or rank spot.[10]
Conclusion
In summary, a majority of ophthalmology faculty surveyed agreed that preference signaling
should be implemented in the ophthalmology residency match. If implemented successfully,
preference signaling has the potential to make it easier for programs to sort through
applications while simultaneously increasing equity within the process for applicants.
Based off of other specialties and the results from this survey, we recommend giving
four signals to each applicant.
