Keywords
H-index - student author - mentor - ophthalmology
When students seek out opportunities for research with faculty members, they are often
interested in learning about a specific field, gaining a potential mentor/advisor,
and establishing a network that may help them in future endeavors, in addition to
gaining experience and skills in scientific inquiry. As competition for residency
positions continues to rise, scholarly work plays an even more integral role in the
medical student's residency application. This is especially the case in competitive
surgical specialties such as ophthalmology. Data from the San Francisco Match Program,
the United States-based residency match program, shows that the number of applicants
exceeded the number of available positions by 141 to 163% in any given match cycle
from 2013 to 2022.[1] In the same time period, the average Step 1 score increased from 239 to 247, highlighting
the increasing competition to successfully match into an advanced position.[1] With the United States Medical Licensing Examination Step 1 transition to pass/fail
scoring in January 2022, program directors of various surgical subspecialties reported
an increased importance of published research on residency applications.[2]
[3]
[4]
[5]
The H-index (Hi) is a cumulative measure of the scholarly impact and productivity of an author and
was developed by physicist Jorge E. Hirsch in 2005.[6]
[7] The Hi is influenced by both quantity of publications and frequency of citations, and it
is an author-level metric that has been validated as predictive of future scientific
achievement.[7] Because Hi measures author productivity only, it is not influenced by journal impact factor.
In this study, we use Hi to evaluate the effect of student authorship on the scholarly impact of corresponding
authors (CAs) in ophthalmology. We hypothesize that publishing with student authors
(SAs) increases the scholarly impact of CAs, compared to publishing without SA.
Materials and Methods
We identified all authors from articles published in Ophthalmology during the years of 2008, 2012, and 2016. Articles from these three particular years
were sampled because they are relatively recent to this present study, yet the time
elapsed from the publication year is long enough for Hi to fluctuate. As the Hi is directly related to time since publication and depends on an author's citations
in other works, a more recently publishing author will have a lower H index. Given
this relationship, we withheld data collection after 2016 as CA's H-indices may not
be well established for statistical analysis between our two groups. We chose Ophthalmology as our source to identify the pool of CAs due to the journal's high impact. Each
author's degree was noted and ascribed in a database, with designations assigned as
either CA, SA, or other author. SA was defined as an author with a nondoctoral, (e.g.,
nurse practitioner, registered nurse) degree; authors with bachelor's degrees, master's
degrees, and PhD candidates were considered SAs.
Data abstraction was completed twice, in October 2018 and March 2021. For each of
the sampled years, Hi values were identified for the CA at the time of publication of the article, as determined
by Scopus.[8] For CAs who had publications in multiple years, the Hi was recorded once and the number of SAs was tabulated over time. There were no repeated
observations per CA. SAs were divided into groups based upon whether they were first
authors, second authors, or other authors. The CAs were designated as having student
co-authors if at least one of their publications included a SA. The Hi for each CA was recorded in October 2018 and March 2021.
We compared CA groups, inclusion of SAs (StA) or no SAs (nStA) over time, based on
the mean Hi of the CA using a linear regression analysis. We wanted to determine if the Hi of CAs who worked with SAs changed by a higher degree from 2018 to 2021 than those
without SAs, so we set the change in Hi from October 2018 to March 2021 as the outcome variable, with the CA groups as the
predictor variable, and adjusted for the covariates of baseline Hi, number of research items published in October 2018, the academic appointment of
the CAs, and the year the CA published his or her article. A secondary linear regression
analysis was completed with the total number of SAs per CA as the predictor variable
and the change in Hi from October 2018 to March 2021 as the outcome variable, adjusted
for the covariates of baseline Hi, number of research items published in October 2018,
and the academic appointment of the CAs. The number of research items was determined
from author profiles on ResearchGate, and includes published articles, chapters, conference
papers, data, preprints, and full-texts. Statistical significance was determined by
a p-value <0.05. Descriptive statistics were computed to describe the study cohort prior
to excluding repeated CAs. Descriptive results were expressed as mean (standard deviation)
and median for continuous measures, and counts with relevant percentages for categorical
variables.
Results
When comparing the mean Hi from data recorded in October 2018 and March 2021, the Hi for each CA in both StA and nStA demonstrate a general increase over time ([Table 1]). After excluding repeats, there were a total of 683 unique CAs, and 269 (39.4%)
of these CAs published with SAs. This observation confirms the direct relationship
between the time since publication and Hi. The number of SAs increased from 168 in 2008 to 192 in 2016, while the number of
students as first or second authors increased from 41 in 2008 to 69 in 2016 ([Table 1]). The number of publications including at least one SA also increased in this time,
from 105 student-authored publications in 2008 to 119 student-authored publications
in 2016 ([Table 1]).
Table 1
Descriptive analysis of articles from Ophthalmology 2008–2016
|
2008
|
2012
|
2016
|
Total
|
|
Total number of articles
|
291
|
318
|
293
|
902
|
|
Number of student authors
|
168
|
150
|
192
|
510
|
|
Number of all authors
|
1,653
|
1,848
|
2,163
|
5,664
|
|
Number of students listed as first or second authors
|
41
|
42
|
69
|
152
|
|
Number of articles with student authors
|
|
None
|
186
|
226
|
174
|
586
|
|
≥1 student author
|
105
|
92
|
119
|
316
|
|
1
|
66
|
59
|
72
|
197
|
|
2
|
25
|
20
|
32
|
77
|
|
3 or more
|
14
|
13
|
15
|
42
|
|
Number of authors per paper
|
|
1–5 authors
|
151
|
110
|
102
|
363
|
|
6–10 authors
|
124
|
142
|
110
|
376
|
|
11–15 authors
|
13
|
18
|
40
|
71
|
|
16 or more
|
3
|
48
|
3
|
54
|
When data abstraction was completed in October 2018, the mean and median Hi were higher for CAs who published with students versus those who published without
students for all years combined, as shown in [Table 2]. A similar trend was seen when data abstraction was again completed in March 2021,
when the average change in Hi in the StA group was significantly higher than the average change in Hi in the nStA group. Analyses at both the October 2018 and March 2021 timepoints were
adjusted for the number of published research items, baseline Hi, the academic appointment of CAs, and the year the CA published his or her article.
Furthermore, in the secondary analysis, it was found that as the number of SAs increased,
the change in Hi increased linearly for all years combined (regression coefficient = 1.70, p-value <0.0001).
Table 2
Median Hi, mean Hi, SD, and change in Hi by corresponding author groups
|
Articles with student authors,
median Hi/mean Hi (SD)
|
Articles without student authors,
median Hi/mean Hi (SD)
|
|
|
October 2018
|
March 2021
|
Change
|
October 2018
|
March 2021
|
Change
|
p-Value
|
|
28/32.6 (23.3)
|
40/43.6 (26.6)
|
7/11.0 (14.7)
|
22/25.0 (18.6)
|
27/31.3 (20.9)
|
4/6.2 (6.2)
|
<0.0001
|
Abbreviations: Hi, H-index; SD, standard deviation.
Discussion
Existing literature uses the Hi to analyze the effects of student authorship in other fields of medicine, namely
radiation oncology, general surgery, and otolaryngology.[6]
[7]
[9]
[10]
[11]
[12]
[13] Our study aimed to analyze student contribution to research within the field of
ophthalmology. We found that the scholarly impact, as measured by Hi, was higher for CAs who published with students compared to those who did not, achieving
significance when combining data on CAs who published in 2008, 2012, and 2016. We
also found that as the number of SAs increases, the change in Hi also increases linearly for all years combined. Our data indicate that the student–faculty
partnership resulting in publications increases the Hi, and thus the scholarly impact, of CAs. Additionally, the impact of student authorship
on the Hi of CAs may be stronger in ophthalmology compared to other fields of medicine previously
studied.[9]
[10]
[11]
[12] In contrast to our findings, the change in mean Hi was not significantly different between nStA and StA in the fields of general surgery,
radiation oncology, and internal medicine.[9]
[10]
[11]
[12] The difference in results may be due to differences in statistical analysis in our
work compared to the studies in other fields, rather than intrinsic differences between
medical fields. Unlike previously published works, our analysis was adjusted for potential
confounders including the number of published research items, baseline Hi, the academic appointment of CAs, and the year the CA published their article.
Indicators for success in a surgical residency, such as teamwork, collaboration, effective
communication, work ethic, and initiative, are promoted through research.[13]
[14]
[15] Importantly, research projects open opportunities for strong mentoring relationships,
which play a major role in attracting students to surgical specialties.[15]
[16]
[17] In addition to these benefits, students have cited that their motivations to pursue
research are in part influenced by the desire to build a stronger curriculum vitae
to appear more competitive for residency applications.[18] While the motivation to pursue research projects may vary, authorship has historically
been challenging to obtain for students.[19] Despite their interest, students often contend with barriers, such as lack of dedicated
and funded time to pursue research projects and difficulty in finding mentors willing
to guide students through a project, and these barriers can stymie student productivity.[20]
[21]
[22]
Moreover, medical school curricula seldom emphasize ophthalmology topics or offer
dedicated ophthalmology exposure through clerkships.[23]
[24]
[25] Given this, student participation in research can also serve as a means to gain
exposure to the field, increase their visibility, and develop mentoring relationships
early. All these factors will be helpful as students apply to residency programs,
particularly as competition for residency positions rises.
Alternately, for faculty, as academic teaching institutions place significant emphasis
on publications as a metric for faculty promotion and a barometer for national and
international reputation,[26] these findings lend credence to the body of literature suggesting that the student–faculty
partnership can be a synergistic and positive relationship for both.[27] In fact, surgical faculty in ophthalmology may have the most to gain in terms of
research productivity by collaborating with SAs. It was shown that nonsurgical ophthalmology
subspecialties, including uveitis, neuro-ophthalmology, and medical retina, have a
higher Hi and number of published studies compared to surgical ophthalmology subspecialties
such as cataract and refractive surgery.[28] Overall, CAs who collaborate with students are intrinsically motivated to mentor
students and further contribute to the field. Given these motivations, they would
be more likely to make choices, such as co-authoring with students, that increase
their research productivity and Hi, compared to CAs who are less concerned with student mentorship.
There are some limitations to this study, some of which are inherent to the Hi calculation. The Hi is a composite value derived from all of a given author's publications in any journal,
over time, and is not an individual calculation for each publication. Therefore, a
“control group” of multiple Hi for the same author is not possible. However, comparing Hi of authors who did not work with students to Hi of authors who did work with students at the time of publication controls for time.
Additionally, several factors may confound our reported number of SAs. Considering
all nondoctorate authors as SAs may have overestimated the number of SAs. On the other
hand, the time between research and publication may have led to graduate degrees for
those who performed their research as students, which may have underestimated the
number of StA. While most SAs were from the United States, other countries may have
unique and variable barriers to publication for students, such as a high cost of publication
or lack of funding, or students may be classified differently, impacting this variable
in our analysis. Furthermore, it may be possible that students are seeking research
mentors with higher Hi or ongoing research which may have overestimated Hi in the StA group; however, there is no literature to support this as a confounding
variable and our analysis was adjusted for baseline Hi and academic appointment. Qualitative information to gauge students' choices in research
mentors may provide greater insight into the possibility that students may seek out
more prolific researchers and could augment the correlation between high Hi and student authorship. Furthermore, the data were collected from one journal in
the field, which does not capture the full extent to which students participate in
ophthalmology research. However, with a 5-year impact factor of 12.08, and a broad
range of disciplines within ophthalmology represented, we felt Ophthalmology was an appropriate journal to evaluate student authorship in this surgical field.[29]
In summary, our study shows the number of SAs publishing in Ophthalmology is increasing while benefitting the scholarly advancement of CAs. We hope this study
encourages students to involve themselves in research and encourages faculty to seek
students for projects as a means to inspire, encourage, and maintain interest in ophthalmology
while furthering their own careers and scientific discovery. Development of programs
to improve faculty mentorship of student research as well as to better integrate students
into ophthalmology research early in training may further encourage students' pursuit
of a career in ophthalmology.
