Medicine is an ever-changing practice as a result of ongoing new scientific breakthroughs in both diagnosis and treatment options, requiring physicians to constantly pursue updates in published literature.1 Moreover, even while focusing on a certain subspecialty, clinicians still attempt to keep up with developments outside their chosen subspecialty. This is especially true in pediatric ophthalmology, because this is a broad area that has constant close interactions with many other subspecialties.
Peer-reviewed journals are seen as a major vehicle in the transmission of research findings to clinicians.2 Schein et al.3 found that traditional, peer-reviewed journals were considered by American surgeons to be the most important information source. However, although information has become readily available, the immense number of publications and journals presents a challenge for a practicing clinician. The category of ophthalmology in the Web of Science contains 59 journals. Some of these journals are aimed at subspecialists (eg, retina and cornea), whereas others are more general and designated for comprehensive ophthalmologists. The latter include Ophthalmology, the journal of the American Academy of Ophthalmology, which includes “clinical and basic science research and other relevant manuscripts that relate to the sense of sight,”4 and JAMA Ophthalmology, with a mission statement “to be the indispensable source of ophthalmic knowledge for the generalist, subspecialist, and trainee; to publish innovative, clinically relevant research for the vision scientist.”5 Thus, top-ranking journals dealing with general clinical ophthalmology serve as leading resources for medical education, publishing the latest and most important advances across the discipline.
In this study, we investigated publication trends of pediatric ophthalmology articles over the past two decades using the rate of pediatric-related articles in high-ranking general ophthalmology journals.
A list of all Q1 ophthalmology journals was obtained from the InCites Journal Citation Reports web site.6 The description of each journal was reviewed separately by two of the authors (ALeshno and DM-K) to categorize the journal as “specialized” if it focused on a specific subspecialty in ophthalmology or “general” if it claimed to encompass all fields of ophthalmology. Non-clinical or review journals were excluded. In case of disagreement between the first two reviewers, the journal category was determined by a third reviewer (AB). Among journals categorized as “general,” the five journals with the highest impact factor in 2017 were selected for analysis. Using a syntax search code on PubMed (Table A, available in the online version of this article), a list of all articles published in the selected journals between January 1, 1997, and December 31, 2016, was obtained. The list was cross-referenced with similar lists that were obtained using the same syntax, in addition to activating the age filter “child: birth-18 years” and/or article type filter “Clinical Trial” on PubMed, to classify publications as “pediatric” and/or “clinical trial,” respectively. The annual number of publications in each journal and overall was summarized and the percentage of pediatric publications was calculated from the total number of publications each year in every journal. The percentage of pediatric clinical trials was also determined from the total number of clinical trials published during the study period. The impact factor ratings for the selected journals between 1997 and 2016 were collected from the InCites Journal Citation Reports web site.
PubMed Syntax Search Code
Pearson correlations and multivariate linear regressions were used to detect associations between the number of publications and rates, as well as the journals' impact factor over time, and to exclude confounders.
Finally, because changes in pediatric publication rates in general ophthalmology journals might have occurred due to changes in publishing rates in pediatric and strabismus subspecialty journals, we reanalyzed the data after including publications from three main pediatric ophthalmology journals: Journal of the American Association for Pediatric Ophthalmology and Strabismus, Journal of Pediatric Ophthalmology & Strabismus, and Strabismus. The same age search filter was used to exclude adult-only strabismus articles. The rates and absolute number of annual pediatric publications were recalculated.
Ethical approval was not required for the study, and the research was performed in compliance with the tenets of the Declaration of Helsinki.
All analyses and calculations were preformed using SAS (SAS, Inc., Cary, NC) and MATLAB (MathWorks, Inc., Natick, MA) software. Graphs were generated with Excel (Microsoft Corporation, Redmond, WA) or MATLAB software. To estimate the relationships between the dependent variables as the rate of pediatric publications over the years, a linear mixed effects model was employed. In the mixed effects models, the ith journal at the jth measurement occasion:
where the fixed-effects contains the two first components of the equation and the random effects contains the last two components of the equation (for additional technical details, see Table A).7
Of 14 Q1 journals, the following five were categorized as “general” by both reviewers and were included in the analysis: Ophthalmology, JAMA Ophthalmology, American Journal of Ophthalmology, British Journal of Ophthalmology, and Acta Ophthalmologica. There was a complete consensus regarding journal selection. These journals had 37,181 publications between January 1, 1997, and December 31, 2016, of which 7,828 (21.1%) were classified as “pediatric” and 3,393 (9.1%) as “clinical trials.” The mean number of annual publications was 371.81 ± 103.48. Figure 1 presents the annual number of publications between 1997 and 2016 for each of the five “general” journals, as well as the total number of publications in all five journals.
Annual number of publications in five top ranking general journals.
As demonstrated in the figure, there was a slight increase in the annual number of publications over time, with the exception of two periods (between the years 2000–2002 and 2007–2008) in which the annual number of publications declined. These fluctuations resulted mainly from changes in the number of publications in two journals (Ophthalmology and JAMA Ophthalmology). However, in the linear mixed effects model, the change in the number of publications over time was not statistically significant (P = .18). In contrast, the total number of pediatric publications in all five “general” journals remained relatively constant throughout the study period, exhibiting a negative yet non-significant slope (P = .66) in the linear mixed effects model.
The total annual rates of pediatric publications in all five journals together are depicted in Figure 2. It is evident that, despite some fluctuations, the rates continually decreased over time: from 24.3% in 1996 to 21.7% in 2006 and further decreased to 16.3% in 2016. An exceptional unique increased rate of 23.5% in pediatric publications occurred between 2009 and 2010, mainly due to a high number of pediatric-related publications in JAMA Ophthalmology, after which the rate decreased steadily. The linear mixed model revealed a statistically significant negative trend in pediatric publication rates with an overall annual decrease of 0.23% (P = .01), yielding a reduction of the annual publication rate by nearly 18% over the two decades.
Combined annual rates of pediatric publications in the five top-ranking general journals.
Figure 3 depicts the rate of pediatric publications between 1997 and 2016 in each journal individually. As shown in Figure 3A, fluctuations in pediatric publication rates occurred in all journals over time. Notably, the British Journal of Ophthalmology had the greatest median pediatric publication rate among the journals and Acta Ophthalmologica had the widest range (Figure 3B).
(A) Annual rates of pediatric publications in each of the five top-ranking general journals and (B) boxplot summary.
There was a non-significant increase in the number of clinical trials publications over time (P = .61) (Figure 4). The rate of pediatric clinical trials, calculated from the total number of clinical trials published, did not change significantly during the study period (P = .30).
Annual number of clinical trial publications in the five top-ranking general journals.
Although the absolute number of pediatric publications and the total absolute number of publications in general ophthalmology journals did not change significantly over the 20-year period, there was a significant annual increase of 3.22 publications every year (P = .033) in the number of publications in pediatric ophthalmology and strabismus journals. Combining the data from top-ranking general ophthalmology journals and pediatric ophthalmology journals resulted in an overall marginal, non-significant change in the rate of pediatric publications over the past two decades (P = .80). The annual pediatric publication rate of the combined data is presented in Figure 5. As shown, the rates of pediatric publications in the combined model fluctuate without a clear trend, yet a gradual decrease in the past 5 years of the study period is apparent.
Combined annual rates of pediatric publications in the five top-ranking general journals and three main pediatric ophthalmology journals.
The current study analyzed the change in the rate of publication of pediatric-related articles in leading general ophthalmology journals over the past two decades. We found that although there was an increase in the number of overall publications, the percentage of pediatric articles significantly diminished.
In 2015, Mimouni at al.8 compared the absolute number of publications between pediatric ophthalmology and adult ophthalmology over a 15-year period, including all types of publications and all journals cited by PubMed. Not surprisingly, they showed that although the absolute number of pediatric-related publications increased, the rate of increase was slower compared to publications on adult patients. Based on their data, we calculated that the annual pediatric publication rate gradually decreased from 30.4% in 1999 to 26% in 2012 (mean 0.33% annual decrease).
Although the reduction in pediatric publication rates depicted in our results is similar to that described in Mimouni et al.'s report, it is impossible to directly compare between the two studies due to significant methodology differences. First, their analysis was based on broad and heterogeneous data because they included all publication types and all journals cited in PubMed regardless of journal quality or rank, whereas our study focused only on high-ranking general journals, considered to be opinion leaders. Because Jones et al.2 demonstrated that most physicians are exposed to only a few key journals, we believe that our analysis reflects the changes in exposure of ophthalmologists across subspecialties to pediatric ophthalmology content more accurately. Second, due to their study design, Mimouni et al.'s results might have been affected by additional factors, such as the increase in number of open access journals and the increase in publications on specific topics, as occurred with the introduction of antivascular endothelial growth factor treatment that led to a significantly higher proportion of publications regarding the retina. Third, the mixed model analysis employed herein diminished the effect of any single journal and episodic fluctuations, giving a less biased and more reliable result.
Kumar et al.9 reviewed pediatric ophthalmology publication trends in seven leading general ophthalmic journals over a 5-year period (2005–2009). Although they found a relatively low annual rate of pediatric publications (6.6% to 9.6%), they failed to detect a significant trend during this period.
There are several possible explanations for these observations regarding the reduced rates of pediatric publications. First, it is in fact complicated to perform and conduct clinical trials with minors. In such trials, it is harder to obtain institutional review board approval and informed consent; moreover, clinical data collection is more challenging for both patients and clinicians and researchers. Nonetheless, the publication rate of pediatric clinical trials remained constant during the study period.
An alternative explanation for the decrease in annual rates of pediatric publications could be the process of subspecialization, which has led to “coalescence” of types of publications in specific journals. To examine this hypothesis, we reanalyzed the data after including publications from three main pediatric ophthalmology journals: Journal of the American Association for Pediatric Ophthalmology and Strabismus, Journal of Pediatric Ophthalmology & Strabismus, and Strabismus. We found that although pediatric ophthalmology journals had a significant annual increase in the number of publications, the combined model resulted in a marginal non-significant change in the pediatric publication rate over 20 years. These findings suggest a shift in publication of pediatric articles from the general top-ranking journals to pediatric subspecialty journals.
This shift could also be due to the preference of pediatric specialists to submit and have their articles published in pediatric ophthalmology journals rather than in general journals. As such, they are also more likely to be read by the target audience (pediatric specialists). In addition, there are better chances for acceptance for publication in these journals due to relatively less potential competition. As in most subspecialties, pediatric ophthalmologists will seek updates mainly in pediatric ophthalmology and strabismus journals. Other explanations are that general ophthalmology journals (and their editors) are less interested in articles dealing with pediatric ophthalmology or that the advances reported were not substantial enough to be given priority. However, to verify such an assumption would require access to data that we do not possess, such as acceptance rates and editors' considerations before acceptance for publication.
A decrease in publication of pediatric articles in general ophthalmology journals has several important implications for the field of pediatric ophthalmology. Medical publications are the most important tools for staying up to date in both general medicine and a specialty. Because there is an exponential increase in knowledge and data, many physicians use the journal's ranking as a guideline for their reading strategies. Because nearly all practicing ophthalmologists are required to treat patients with pediatric problems (either as children or growing adults), less exposure to newly developed treatments and diagnoses might result in reduced quality of care and a greater burden on pediatric ophthalmology specialists.
Lower exposure to pediatric ophthalmology in journals with a high impact factor might also lead to a vicious cycle of less exposure to the pediatric field, and therefore less interest in it. One possible consequence of less interest in pediatric ophthalmology was reported by Dotan et al.,10 who found that the supply of fellowship positions in pediatric ophthalmology and strabismus in the United States remained consistently greater between the years 2000 and 2015 than demand, with the highest percentage of unfilled positions among ophthalmology subspecialties.
In addition, the relative decrease in pediatric publications has significant implications for all researchers investigating ophthalmic conditions in children. Academic advancement often depends on publication of articles in journals with high ranks and high impact factors. Pediatric ophthalmology journals are not cited often because pediatric ophthalmology is a relatively small niche, so these researchers have less of a chance or might wait longer to achieve academic promotion. This might also deter ophthalmology residents from applying for pediatric ophthalmology fellowships. Furthermore, the risk of excluding studies by pediatric surgeons from exposure in more general journals may hinder potential beneficial collaborations with surgeons in other subspecialties.
It should be noted that our observations might be partially affected by dependence on private industry funding. Several studies have shown an association between industry funding and academic bibliometrics.11,12 It is possible that pediatric ophthalmology is a less attractive field for private companies, due to its relatively small market value and the difficulty of performing studies. Unfortunately, evaluation of this assumption is beyond the scope of this study.
We reviewed the publication trends of pediatric articles in general ophthalmology journals with high ranks and high impact factors over 20 years and found a significant decrease in pediatric publication rates. Because many factors are involved in the journals' selection and approval process toward publication, it is impossible to determine a single explanation for this observation. Regardless of the causes, although there is a growing shortage of pediatric ophthalmology specialists,10 this phenomenon may reduce the exposure of residents to the field and the incentive to choose such fellowships. We have thus highlighted the need for integration of the pediatric field into the mainstream journals because it would be beneficial to both pediatric ophthalmology and general ophthalmology.
- Steinbrook R. Searching for the right search—reaching the medical literature. N Engl J Med. 2006;354(1):4–7. doi:10.1056/NEJMp058128 [CrossRef]
- Jones TH, Hanney S, Buxton MJ. The journals of importance to UK clinicians: a questionnaire survey of surgeons. BMC Med Inform Decis Mak. 2006;6(1):24. doi:10.1186/1472-6947-6-24 [CrossRef]
- Schein M, Paladugu R, Sutija VG, Wise L. What American surgeons read: a survey of a thousand Fellows of the American College of Surgeons. Curr Surg. 2000;57(3):252–258. doi:10.1016/S0149-7944(00)00177-X [CrossRef]
- Ophthalmology. Ophthalmology aims. https://www.aaojournal.org/content/aims. Accessed October 31, 2018.
- JAMA Ophthalmology. https://jamanetwork.com/journals/jamaophthalmology/pages/for-authors. Accessed October 2, 2018.
- Thomson Reuters. InCites Journal Citation Reports. Web of Science. https://jcr-clarivate-com.rproxy.tau.ac.il/JCRHomePageAction.action?. Accessed June 23, 2018.
- Fitzmaurice GM, Laired N, Ware J. Applied longitudinal analysis. Journal of the American Statistical Association. 2011;100:709–710. doi:10.1002/9781119513469 [CrossRef]
- Mimouni M, Segal O, Mimouni FB, Nemet AY. Trends in pediatric versus adult ophthalmology publications over 15 years. J Pediatr Ophthalmol Strabismus. 2015;52(4):239–244. doi:10.3928/01913913-20150520-02 [CrossRef]
- Kumar A, Cheeseman R, Durnian JM. Subspecialization of the ophthalmic literature: a review of the publishing trends of the top general, clinical ophthalmic journals. Ophthalmology. 2011;118(6):1211–1214. doi:10.1016/j.ophtha.2010.10.023 [CrossRef]
- Dotan G, Karr DJ, Levin AV. Pediatric ophthalmology and strabismus fellowship Match outcomes, 2000–2015. J AAPOS. 2017;21(3):181.e1–181.e8. doi:10.1016/j.jaapos.2017.01.004 [CrossRef]
- Ruan QZ, Cohen JB, Baek Y, et al. Does industry funding mean more publications for subspecialty academic plastic surgeons?J Surg Res. 2018;224:185–192. doi:10.1016/j.jss.2017.12.025 [CrossRef]
- Svider PF, Bobian M, Lin HS, et al. Are industry financial ties associated with greater scholarly impact among academic otolaryngologists?Laryngoscope. 2017;127(1):87–94. doi:10.1002/lary.26027 [CrossRef]
PubMed Syntax Search Code
|The following syntax was used for obtaining the list of all publications between January 1, 1997, and December 31, 2016, from the five selected journals (Ophthalmology, JAMA Ophthalmology, American Journal of Ophthalmology, British Journal of Ophthalmology, and Acta Ophthalmologica):
The following age filter was added in order to capture only pediatric publications:
“Ophthalmology”[Journal] OR “JAMA Ophthalmol”[Journal] OR “Am J Ophthalmol”[Journal] OR
“Br J Ophthalmol”[Journal] OR “acta ophthalmol”[Journal] OR “acta ophthalmol scand”[Journal]
OR “Arch Ophthalmol”[Journal]) AND (“1997/01/01”[PDAT] : “2016/12/31”[PDAT])”
The article types filter was added in order to capture only clinical trial studies:
“AND (“infant”[MeSH Terms] OR “child”[MeSH Terms] OR “adolescent”[MeSH Terms])
“AND (Clinical Trial[ptyp])”