Keywords
ophthalmology - pediatric ophthalmology and strabismus - fellowship - medical education
- online availability - coronavirus disease 2019 pandemic
There is a shortage of pediatric ophthalmologists in the United States.[1] Yet, approximately one-third of pediatric ophthalmology and strabismus (POAS) fellowship
positions go unfilled each year.[2] Fellowship program Web sites are the primary source of information used by applicants.[3]
[4] Difficulty obtaining accurate information to make informed decisions on where to
apply may increase the number of applications per candidate and deter well-suited
applicants.[5] Restrictions on in-person interactions due to the coronavirus disease 2019 pandemic
further highlight the importance of available fellowship information.[6]
A previous study of glaucoma fellowship programs found that program Web sites were
consistently missing information that applicants considered important.[7] However, data are limited regarding the availability and completeness of information
from POAS fellowship Web sites. This study examined the content on POAS fellowship
Web sites to understand what information is currently available online.
Methods
We analyzed the availability and content of all 45 POAS fellowship programs, which
were identified from the Association of University Professors of Ophthalmology (AUPO)
directory of programs in compliance (https://aupofcc.org/programs-in-compliance?institution=&state=All&sub_specialty%5B%5D=pediatric-ophthalmology) and the San Francisco Match (SFMatch) application listing. Program name, affiliated
institution, program size, and geographic location were abstracted from the AUPO and
SFMatch Web sites, respectively. We then assessed the availability and content of
each program's Web site.
We first designed a standardized search strategy to identify POAS program Web sites
prior to the start of data collection. Data were collected by two independent researchers,
with consensus meeting to resolve any varying assessments. To determine the availability
of Web sites, each POAS fellowship was searched on Google by: “program name” + “affiliated
institution” + “pediatric ophthalmology fellowship.” A POAS fellowship program was
considered available if the Web site could be identified from the Google search results.
If no Web site could be found by Google Internet search, “pediatric ophthalmology
fellowship” was queried using the search function of the affiliated parent institution's
Web site. If no Web site could be identified from either Google Internet search or
search of the affiliated institution's Web site, then that program was determined
to lack an available program Web site. Though we decided on this two-step search strategy
a priori, all Web sites could be identified by Google search, and the second step
of the search was not used.
The content of Web sites was evaluated for 10 recruitment and 10 program training
factors ([Table 1]). These criteria were selected based on AUPO certification criteria, similar studies,
and surveys of what content is important to applicants.[7]
[8] Recruitment factors were program description, list of teaching faculty, program
contact information, link to application, current fellows, prior fellows, alumni placement,
demographic characteristics of patient community, information about program's location,
and salary and benefits. Program training factors were overview of fellowship curriculum
and schedule, affiliated hospitals and clinics, surgical experience, research requirement
and opportunities, opportunities to present research at conferences and meetings,
resident and student teaching responsibilities, call schedule, grand rounds, journal
club, and fellow evaluation process. All Web site materials, including video content,
were reviewed to determine the presence or absence of each criterion.
Table 1
Proportion of fellowship program Web sites reporting each recruitment and training
criteria
|
Recruitment criteria
|
Training criteria
|
|
Criterion
|
Frequency (proportion)
|
Criterion
|
Frequency (proportion)
|
|
Program description
|
44 (97.8%)
|
Rotation overview
|
45 (100.0%)
|
|
Listing of teaching faculty
|
30 (66.7%)
|
Affiliated hospitals and sites
|
44 (97.8%)
|
|
Program contact name/email
|
28 (62.2%)
|
Surgical experience
|
41 (91.1%)
|
|
Link to application
|
25 (55.6%)
|
Research requirement and opportunities
|
39 (86.7%)
|
|
Salary and benefits
|
22 (48.9%)
|
Resident teaching role
|
32 (71.1%)
|
|
Current fellow names
|
13 (28.9%)
|
Journal club
|
29 (64.4%)
|
|
Information about city
|
12 (26.7%)
|
Academic conferences
|
26 (57.8%)
|
|
Prior fellow names
|
10 (22.2%)
|
Grand rounds
|
24 (53.3%)
|
|
Alumni placement
|
6 (13.3%)
|
Call schedule
|
19 (42.2%)
|
|
Information on patient demographic
|
2 (4.4%)
|
Fellow evaluation process
|
7 (15.6%)
|
Summation of the 20 criteria calculated a total score of each Web site's content (range:
0–20). Mean and standard deviation were estimated. We also examined differences in
fellowship Web site content, stratifying by geographic location (Northeast/Midwest/South/West
US, Canada) number of fellows (1/2+ fellows) and teaching faculty (≤8/≥9 faculty),
and affiliation with a top-ranked ophthalmology residency program. A top ophthalmology
residency program was defined as those ranked in the 2021 Top 12 Best Residency programs
by the Ophthalmology Times.[9] Kruskal–Wallis and Mann–Whitney U tests were used to assess the association with
Web site content.
All data were collected from program Web sites over a 2-day period in July 2022. Data
maintenance and analyses were conducted in SAS version 9.4 (SAS Institute, Cary, NC).
A two-sided p-value less than 0.05 was considered statistically significant. The Northwestern University
Institutional Review Board determined this study was not human subjects research because
all data collected were publicly available online and the results are presented in
aggregate.
Results
Out of 45 identified POAS fellowships, all 45 programs had a Web site to provide fellowship
information identifiable from a Google Internet query. These programs included 22
POAS fellowship programs with one fellow (48.9%) and 23 programs with more than or
equal to two fellows (51.1%). The greatest number of programs were located in the
Midwestern United States (n = 19, 42.2%), followed by South (n = 10, 22.2%), West (n = 8, 17.8%), and Northeast regions (n = 5, 11.1%). We also identified and included three programs located in Canada.
The mean number of total criteria on POAS fellowship Web sites was 11.1 (standard
deviation 3.3, range: 3–18), which included a mean of 4.3 recruitment criteria (standard
deviation 1.9, range: 1–9) and 6.8 training criteria (standard deviation 2.1, range:
2–10). POAS fellowships reported fewer recruitment criteria than training criteria
(paired t-test, p < 0.0001).
The factors found most commonly on fellowship program Web sites were rotation overview
(100.0%), overall program description (97.8%), rotation overview (97.7%), and affiliated
hospitals and sites (97.8%), whereas patient demographic information (4.4%) and alumni
placement (13.3%) were reported least frequently ([Table 1]). Notably, 5 out of 10 training criteria were reported on more than 70% of POAS
Web sites, whereas only one recruitment criterion was reported on more than 70% of
Web sites.
There were no differences in number of overall criteria on Web sites by number of
fellows, number of faculty, geographic location, or affiliation with a top ophthalmology
residency (p ≥ 0.36) ([Table 2]). Similarly, we found no differences in the number of recruitment or training criteria
reported by the number of fellows, number of faculty, geographic location of the program,
or affiliation with a top residency program (p ≥ 0.09).
Table 2
Association of program size and geographic location with Web site reporting of criteria
|
Classification
|
Recruitment score
|
Training score
|
Total score
|
|
Mean (SD)
|
p-Value
|
Mean (SD)
|
p-Value
|
Mean (SD)
|
p-Value
|
|
Number of fellows
|
|
1 fellow
|
4.0 (1.8)
|
0.45
|
7.3 (2.3)
|
0.09
|
11.2 (3.1)
|
0.44
|
|
≥2 fellows
|
4.6 (2.1)
|
6.3 (1.9)
|
10.9 (3.5)
|
|
Number of faculty
|
|
1–8 faculty
|
4.5 (2.0)
|
0.16
|
7.2 (1.6)
|
0.69
|
11.7 (2.8)
|
0.36
|
|
≥9 faculty
|
5.6 (1.7)
|
7.2 (2.4)
|
12.8 (3.5)
|
|
Geographic region
|
|
Midwest
|
4.3 (1.9)
|
0.36
|
6.9 (2.1)
|
0.33
|
11.3 (3.1)
|
0.43
|
|
Northeast
|
5.0 (2.5)
|
7.4 (1.5)
|
12.4 (3.5)
|
|
South
|
4.0 (1.9)
|
5.6 (2.2)
|
9.6 (3.8)
|
|
West
|
4.8 (1.8)
|
7.3 (2.5)
|
12.0 (3.1)
|
|
Canada
|
2.3 (1.5)
|
7.7 (1.5)
|
10.0 (2.0)
|
|
Affiliated with top 12 residency program
|
|
No
|
4.3 (1.8)
|
0.72
|
6.8 (2.2)
|
0.81
|
11.1 (3.2)
|
0.70
|
|
Yes
|
4.3 (2.4)
|
6.7 (1.9)
|
11.0 (3.8)
|
Abbreviation: SD, standard deviation.
Discussion
This cross-sectional nationwide review of POAS fellowships found all 45 programs have
a Web site available by Google search. However, there was significant variation in
the content posted on these Web sites. The average Web site contained approximately
half the evaluated criteria, missing multiple pieces of information that applicants
considered important.[7]
[8] There were no differences in the Web site content scores by program size, affiliated
institution ranking, or geographic location. Taken together, these findings suggest
there are significant opportunities to improve the content of POAS program Web sites,
regardless of program size, affiliation, and geographic location. Medical educators
and administrators may consider working together to improve information availability
on program Web sites.
All POAS fellowships had an available Web site easily discovered by a Google search
of the Internet. The importance of having a functional Web site has been well-established
in literature for program recruitment. Orthopaedic surgery fellowship applicants used
the program Web site more often than any other online resource.[3] Of applicants to emergency medicine residency, 96% used the Internet to learn more
about programs, and 40 and 47% reported the program Web site was very important and
moderately important, respectively.[10] Similarly, 98% of applicants to the anesthesia residency at Stanford University
used the program Web site to learn about the program, and 56% used the Web site to
determine whether to apply to the program.[11] Previous studies found most emergency medicine residency applicants use the Web
site content to determine whether to apply to the program, and almost half did not
apply to a program because of the Web site.[12] However, these studies did not specifically survey prospective POAS fellows. Further
research is needed to validate the importance of program Web sites specifically among
POAS candidates. Nonetheless, POAS candidates likely benefit from available, informative
program Web sites when determining applications and rank list.
Fellowship program Web sites help applicants discern which programs to apply to and
create their rank list.[10] Inconsistent online information may contribute to candidates submitting applications
to more programs. A survey of surgery fellowship applicants found 62% would have removed
programs from their list if the program's Web site had additional information.[5] This results in higher applicant spending on application fees and greater time invested
by faculty to review applications and conduct interviews. Multiple solutions have
been proposed to address inefficiencies, including limiting the number of applications
and interviews and implementing tiered rank lists.[13] Moreover, without reliable information on Web sites, applicants may rely on anecdotal
sources, such as blogs and forums, to assess program fit.
Additionally, easily accessible information may increase a program's matching success.
Informative Web sites help candidates determine their suitability before submitting
an application, so programs are more likely to interview a cohort of applicants who
understand their fit with the program. Conversely, it is less likely a candidate will
realize post-interview that the program was a poor fit and rank the program poorly.[5] Studies are needed to better understand the association of fellowship Web sites
with program matching success and drive quality improvement efforts. Low match rates
are not limited to pediatric ophthalmology fellowships. Comparable fellowship match
rates were observed in all other pediatric surgical subspecialties, except for pediatric
general surgery.[14] Similar studies may be warranted to understand the online presence of fellowship
programs in other pediatric surgical subspecialties and optimize program match rates.
Notably, POAS fellowship Web sites listed fewer recruitment than training criteria,
highlighting that this could be an area for programs to target updating their Web
sites. Of the recruitment criteria examined in this study, less than 1 in 4 programs
provided information on current fellows, prior fellows, alumni placement, and information
about the city and patient demographic. Only six POAS fellowships reported alumni
placement, even though prospective applicants likely consider this important information.[15] Alumni placement offers insight into the program's track record of placement by
region and practice type (i.e., academic institution, private practice). This helps
applicants know whether the program's goals align with their desired career path.
A similar analysis of American Society of Ophthalmic Plastic and Reconstructive Surgery
(ASOPRS) fellowship Web sites found none of the 53 ASOPRS programs reported information
about alumni job placement.[16] Profiles of current and prior fellows can also help applicants better understand
the background of a fellow who will likely thrive in the program. Furthermore, information
about the city and patient demographic may allow prospective fellows to have a more
accurate impression of the surrounding community before applying or interviewing.
These are key areas that POAS fellowship Web sites can improve availability of information
prior to the interview.
We evaluated fellowship program Web sites using 10 recruitment and 10 training criteria
based on previous studies.[7]
[17] This allowed for comparison to analogous assessments of glaucoma and surgical retina
subspecialties. Content scores of POAS Web sites in this study were overall comparable
to those found in previous assessments of glaucoma and surgical retina fellowship
Web sites.[7]
[17] However, there are a few notable differences. Teaching faculty were listed on 66.7%
of POAS Web sites, compared with 87.7% of glaucoma and 88.2% of surgical retina Web
sites. Also, resident teaching responsibilities were reported on 71.1% of POAS Web
sites, compared with 82.5% of glaucoma and 76.5% of surgical retina program Web sites.
Lastly, program contact information could be identified on 62.2% of POAS Web sites,
which was similar to 68.4% of glaucoma Web sites, and lower compared with 94.1% of
surgical retina Web sites. POAS fellowships may consider including a list of teaching
faculty, resident teaching responsibilities, and program contact information to facilitate
information gathering for prospective POAS fellows.
Interestingly, there were no differences in fellowship Web site content by geographic
region, ranking of affiliated institution, or program size, consistent with a previous
study that examined program Web sites of glaucoma fellowships.[7] These findings suggest there are opportunities for improvement in Web site transparency
across programs, regardless of geography and size.
Strengths of this study include being a nationwide survey of all POAS fellowship programs
in the United States. However, there are also some limitations that merit mention.
First, there may be criteria that are important to applicants not included in our
evaluation. We selected our 20 study criteria based on previous studies, which allowed
for comparison and benchmarking with other subspecialties.[7]
[8] However, the selection of study criteria based on what is important to non-POAS
means there may be other factors that are uniquely important to POAS fellowships and
are not captured here. No previous studies have surveyed prospective POAS fellows
to identify the most important criteria for them in particular. Future studies are
needed to validate best-practice Web site criteria for POAS fellows. Second, program
Web sites are dynamic and may have changed since our evaluation. To reduce bias, all
programs were evaluated within a 2-day period in July 2022. Third, the cross-sectional
nature of the study design precluded any conclusions of trends over time.
Conclusions
In conclusion, all POAS fellowships have an available Web site, yet there are opportunities
for programs to improve the content on their Web sites. The mean POAS program Web
site contained approximately half of the evaluated content criteria, with lower recruitment
subscores than training subscores. Program size and location were not associated with
differences in Web site content score, suggesting that these opportunities exist across
programs. Informative program Web sites can attract qualified candidates well-suited
to that program, whereas Web sites lacking reliable information may deter qualified
candidates. As such, improvements have the potential to increase a program's matching
success within the existing application process.
