Trends in Academic Productivity Among Radiologists During the COVID-19 Pandemic

Abstract

Purpose

There is a scarcity of literature examining changes in radiologist research productivity during the COVID-19 pandemic. The current study aimed to investigate changes in academic productivity as measured by publication volume before and during the COVID-19 pandemic.

Methods

This single-center, retrospective cohort study included the publication data of 216 researchers consisting of associate professors, assistant professors, and professors of radiology. Wilcoxon’s signed-rank test was used to identify changes in publication volume between the 1-year-long defined prepandemic period (publications between May 1, 2019, and April 30, 2020) and COVID-19 pandemic period (May 1, 2020, to April 30, 2021).

Results

There was a significantly increased mean annual volume of publications in the pandemic period (5.98, SD = 7.28) compared with the prepandemic period (4.98, SD = 5.53) (z = −2.819, P = .005). Subset analysis demonstrated a similar (17.4%) increase in publication volume for male researchers when comparing the mean annual prepandemic publications (5.10, SD = 5.79) compared with the pandemic period (5.99, SD = 7.60) (z = −2.369, P = .018). No statistically significant changes were found in similar analyses with the female subset.

Discussion

Significant increases in radiologist publication volume were found during the COVID-19 pandemic compared with the year before. Changes may reflect an overall increase in academic productivity in response to clinical and imaging volume ramp down.

Visual Abstract

Introduction

The COVID-19 pandemic has had significant impacts on health care practices worldwide. Initial periods of the pandemic demonstrated drastic decreases in hospital volume and admissions, massive delays to elective surgical procedures, and, within the radiological sphere, significant drops in imaging volumes [1, 2, 3, 4, 5]. Amid these disruptive changes, however, there has been a lack of literature characterizing whether there is an associated change in academic productivity among physician researchers during the pandemic—specifically, assessing changes in radiologist participation in academia as measured by research publications before and during the pandemic.

In the current scarce literature, there are variable data reporting both increased, decreased, and maintained academic productivity from pre- to postpandemic periods across all medical specialties worldwide [6]. As expected, different changes in productivity seem to be specific to types of research, with research areas such as biochemistry and cellular biology being disproportionately negative affected by shutdowns owing to their dependence of in-person labor compared with other disciplines such as computer sciences and statistics [7]. For radiology specifically, there is evidence via survey data of decreased research activities during the COVID-19 shutdown, possibly owing to the rapid ramp down of elective imaging studies by which many research studies depend on for data collection, although objective data in this space are notably limited [8]. Interestingly, there has also been prior literature to suggest that there may additionally be gender disparities in research during the pandemic, with female researchers experiencing disproportionate declines in research productivity compared with male researchers [9]. The present study assessed changes in academic productivity during the COVID-19 pandemic among radiology faculty at a major academic radiology department. It was hypothesized that, with the disruptions in clinical activity and imaging caused by the pandemic, there would be an observed increase in academic productivity as measured by publication volume during the pandemic.

Methods

Design

The study use of aggregate publishing data was compliant with HIPAA and was approved with exemption by our institutional review board.

This single-center, retrospective cohort study included the publishing data of 216 radiology faculty at a large academic medical center in the Northeast United States. The study included all assistant professors, associate professors, and professors of radiology at the academic center. Recent faculty members who were residents or fellows transitioning to aforementioned job positions during the periods of interest were excluded. Systematic searches for the entire study period were conducted in PubMed and Medline and cross-referenced with an institutional database to compile publishing data of included researchers from May 1, 2019, to April 30, 2021. The prepandemic period of publication was determined to be the 1-year span from May 1, 2019, to April 30, 2020, and the pandemic period was determined to be May 1, 2020, to April 30, 2021. Time dates were set based on prior literature reporting median time from submission to review and acceptance to be roughly 2 to 3 months in radiological journals [10]. Periods were additionally defined over an entire year to allow for the same month of the prepandemic and pandemic period to be compared, as well as account for any potential seasonal factors as a cause of publication discrepancies. Authorship inclusion into any publication in a peer-reviewed journal was used to determine publishing volume, and there were no exclusion criteria according to the type of study or publication.

Statistical Analysis

Month-to-month publishing data were gathered over the 2-year study period. Wilcoxon’s signed-rank test was used to compare changes in publishing data between the same months of the pre- and postpandemic periods. Tests were additionally performed to compare total academic productivity in publishing volume between these two periods. Statistical analyses were repeated for female and male researchers separately to determine if underlying gender differences in productivity changes were observed. All statistical analyses were performed in SPSS 26 [11]. Additional subgroup analysis was additionally done for the assistant and associate professors individually to determine if seniority would have an impact on findings. Independent samples t test was used to compare mean change of publication volume between male and female researchers to elicit if one gender had a greater observed change in publication volume compared with the other.

Results

The sample consisted of 216 radiologists, including 158 (73%) male and 58 (27%) female radiologists, 138 (63%) assistant professors of radiology, 73 (34%) associate professors of radiology, and 6 (3%) professors or distinguished professors of radiology, as summarized in Table 1 . A total of 1,076 research articles were published in the prepandemic period, and 1,291 research articles were published during the pandemic period. There was a significantly increased volume of publications in the pandemic period (median = 4, mean = 5.98, SD = 7.28, 95% confidence interval [CI] [5.01-6.95]) compared with the prepandemic period (median = 3, 4.98, SD = 5.53, 95% CI [4.24-5.72]) (z =−2.819, P = .005), as outlined in Table 2 . This equated to a 20.08% (95% CI [7.15%-32.45%]) increase from the prepandemic period to the pandemic period. In comparing individual months from the prepandemic period with the pandemic period, there were additionally significantly increased academic productivity in the pandemic period in June, October, and February compared with the corresponding month in the prepandemic period.

Table 1.

Demographic characteristics of the study population according to designation and gender

Designation	Male, n	Female, n	Total, n
Assistant professor of radiology	93	45	138
Associate professor of radiology	59	13	73
Professor of radiology or distinguished professor of radiology	6	0	6
Total	158	58	216

Table 2.

Results from Wilcoxon’s signed-rank test comparing publications month by month in each designated period

Month	Prepandemic Period, Mean (SD)	Pandemic Period, Mean (SD)	% Change	P Value
May	0.58 (0.93)	0.53 (1.01)	−8.62	.589
June	0.23 (0.56)	0.38 (0.69)	65.22	.027
July	0.44 (0.83)	0.49 (0.81)	11.36	.426
August	0.41 (0.86)	0.45 (0.98)	9.76	.630
September	0.37 (0.73)	0.45 (1.02)	21.62	.436
October	0.36 (0.75)	0.60 (1.05)	66.67	.003∗
November	0.53 (1.01)	0.46 (0.84)	−13.21	.236
December	0.33 (0.69)	0.44 (0.97)	33.33	.123
January	0.44 (0.83)	0.65 (1.19)	47.73	.029†
February	0.29 (0.62)	0.47 (0.81)	62.07	.004∗
March	0.48 (0.98)	0.56 (1.03)	16.67	.379
April	0.44 (0.9)	0.47 (0.81)	6.82	.473
Entirety	4.98 (5.53)	5.98 (7.28)	20.08	.005∗

^∗P < .01.

^†P < .05.

Further subgroup analyses by gender and designation were performed. A similar observation was made when comparing research production for male researchers. Specifically, there was an increased number of mean yearly publications in the pandemic period (median = 4, mean = 5.99, SD = 7.60, 95% CI [4.80-7.18]) compared with the prepandemic period (median = 3, mean = 5.10, SD = 5.79, 95% CI [4.20-6.00]) (z = −2.369, P = .018), equating to a 17.45% (95% CI [3.55%, 31.23%)] increase in mean annual publications between the two periods. Similar statistically significant increases in monthly productivity were seen for the months of September (59.4%, P < .05), October (51.3%, P < .05), and February (58.6%, P < .05). A 28.07% (95% CI [−1.94% to 57.67]) increase in publication volume between the mean prepandemic (median = 3, mean = 4.63, SD = 4.76, 95% CI [3.40-5.86]) to the pandemic (median = 4, mean = 5.93, SD = 6.38, 95% CI [4.29-7.57]) period for female researchers was observed, although this was not statistically significant. This is demonstrated in Table 3 . Independent samples t test to compare between group differences in change over the two periods did not demonstrate a statistically greater mean increase among female researchers (mean = 1.29, SD = 5.36) compared with male researchers (mean = 0.89, SD = 4.55) (P = .580) (mean difference = 0.41; 95% CI [−1.04 to 1.85]).

Table 3.

Results from Wilcoxon’s signed-rank test comparing publications month by month in each designated period for female and male researchers

Month	Female Researchers (n = 58)				Male Researchers (n = 158)
	Prepandemic Period, Mean (SD)	Pandemic Period, Mean (SD)	% Change	P Value	Prepandemic Period, Mean (SD)	Pandemic Period, Mean (SD)	% Change	P Value
May	0.69 (1.08)	0.66 (0.98)	−4.34	.839	0.54 (0.87)	0.48 (1.03)	11.11	.597
June	0.21 (0.52)	0.40 (0.72)	90.45	.058	0.30 (0.57)	0.38 (0.67)	26.67	.163
July	0.40 (0.82)	0.50 (0.78)	25.00	.384	0.46 (0.83)	0.49 (0.83)	6.52	.678
August	0.31 (0.68)	0.47 (0.79)	51.61	.200	0.46 (0.92)	0.45 (1.04)	−2.17	.896
September	0.50 (0.76)	0.29 (0.70)	−42.00	.067	0.32 (0.72)	0.51 (1.11)∗	−59.38	.470
October	0.28 (0.59)	0.66 (1.13)∗	135.71	.015	0.39 (0.81)	0.59 (1.03)∗	51.28	.043
November	0.45 (0.75)	0.34 (0.65)	−24.44	.309	0.56 (1.09)	0.50 (0.90)	10.72	.451
December	0.29 (0.56)	0.40 (0.77)	37.93	.353	0.35 (0.74)	0.46 (1.04)	31.43	.212
January	0.26 (0.64)	0.50 (1.08)	92.31	.155	0.51 (0.88)	0.70 (1.24)	37.25	.083
February	0.31 (0.71)	0.52 (0.86)	67.74	.059	0.29 (0.57)	0.46 (0.79)∗	58.62	.022
March	0.40 (0.97)	0.64 (1.19)	60.00	.187	0.52 (0.98)	0.54 (0.97)	3.85	.960
April	0.55 (1.01)	0.57 (0.90)	3.63	.767	0.41 (0.86)	0.43 (0.78)	4.87	.497
Entirety	4.63 (4.76)	5.93 (6.38)	28.08	.119	5.10 (5.79)	5.99 (7.61)∗	17.45	.018

^∗Significant change between the two periods, P < .05.

In subgroup analysis according to seniority, assistant professors demonstrated a similar trend to large sample analysis, demonstrating a statistically significant (P = 0.029) 20.40% (95% CI [2.93%-36.97%]) increase in publication volume from the prepandemic (median = 3, mean = 4.30, SD = 4.26, 95% CI [3.59-5.01]) to the pandemic period (median = 3, mean = 5.18, SD = 5.45, 95% CI [4.27-6.09]), as demonstrated in Table 4 . Despite exhibiting a similar trend, the increase in publication volume for associate professors from the prepandemic (median = 4, mean = 6.01, SD = 6.93, 95% CI [4.42-7.60]) to the pandemic period (median = 4, mean = 7.09, SD = 9.23, 95% CI [4.97-9.21]), the 17.97% (95% CI [−2.82% to 38.76%] increase was not statistically significant (P = .13).

Table 4.

Results from Wilcoxon’s signed-rank test comparing publications month by month in each designated period for assistant and associate professors

Month	Assistant Professors (n = 138)				Associate Professors (n = 73)
	Prepandemic Period, Mean (SD)	Pandemic Period, Mean (SD)	% Change	P Value	Prepandemic Period, Mean (SD)	Pandemic Period, Mean (SD)	% Change	P Value
May	0.59 (0.94)	0.42 (0.75)	−0.29	.117	0.58 (0.94)	0.70 (1.28)	0.21	.375
June	0.17 (0.45)	0.3 (0.53)	0.76∗	.029	0.45 (0.66)	0.52 (0.86)	0.16	.458
July	0.35 (0.66)	0.46 (0.75)	0.31	.155	0.58 (1.05)	0.58 (0.92)	0.00	.933
August	0.38 (0.83)	0.37 (0.78)	−0.03	.869	0.49 (0.91)	0.55 (1.17)	0.12	.941
September	0.3 (0.59)	0.35 (0.69)	0.17	.549	0.52 (0.93)	0.59 (1.35)	0.13	.911
October	0.35 (0.77)	0.57 (0.98)	0.63∗	.015	0.36 (0.67)	0.67 (1.20)	0.86	.084
November	0.41 (0.71)	0.36 (0.65)	−0.12	.526	0.70 (1.30)	0.59 (1.03)	−0.16	.414
December	0.22 (0.48)	0.36 (0.73)	0.64	.048	0.51 (0.93)	0.45 (0.81)	−0.12	.653
January	0.41 (0.69)	0.56 (0.94)	0.37	.102	0.48 (1.01)	0.81 (1.58)	0.69	.11
February	0.32 (0.61)	0.41 (0.69)	0.28	.181	0.27 (0.62)	0.62 (1.00)	1.30∗	.006
March	0.38 (0.89)	0.6 (1.06)	0.58∗	.031	0.63 (1.03)	0.51 (0.98)	−0.19	.31
April	0.43 (0.77)	0.43 (0.74)	0.00	.798	0.45 (1.09)	0.52 (0.88)	0.16	.44
Entirety	4.30 (4.26)	5.18 (5.46)	20.40∗	.029	6.01 (6.94)	7.09 (9.23)	17.97	.125

^∗Significant change between the two periods, P < .05.

Discussion

In this study, retrospective publication data from academic radiologists were used to determine whether there was a significant, observable change in academic productivity during the COVID-19 pandemic. Specifically, a statistically significant 20.08% increase in academic productivity as measured via publications was observed for all researchers and a 17.45% increase in productivity in male researchers. To our awareness, this is the first study to directly compare changes in radiology research productivity using publishing data from the pandemic and prepandemic period and also one of the few studies to report an increased in research productivity during the COVID-19 pandemic as measured by mean publications per faculty member per year.

Several factors may influence the observed increase in publications. For one, a notable decline in imaging volume across multiple modalities was appreciated during multiple periods of the pandemic, secondary to elective procedure ramp downs and overall decreased elective hospital presentations [3,4,12, 13, 14, 15, 16]. This is believed to play a role in increasing the remaining time expenditure of radiologists to dedicate to other activities, such as academic involvement via research. This is in concordance with prior literature that has typically found an inverse relationship between clinical and academic productivity or workload [17,18]. Interestingly however, this trend does not seem to be present across all research disciplines. It is speculated that the clinical ramp down as a result had allowed radiology researchers greater time to both finish pre-existing projects already in process and start new research projects to answer the myriad of new academic questions posed by the COVID-19 pandemic.

In past literature, there has been a demonstration that female researchers seemed to be disproportionately affected by the pandemic, with a decline in academic productivity by female physicians and scientists of different fields [9,19,20]. Although findings were not statistically significant, there was no redemonstration of this trend in the present study, because female researchers were observed to have a greater relative change in productivity than male researchers. The t test additionally did not reveal any significant between group differences in publication volume changes between the two genders during across the two periods. Further investigation with a larger sample size is encouraged for this in the future, to further clarify any potential gender-related differences in academic productivity changes for academic radiologists.

Finally, subgroup analysis examining any association between seniority and productivity changes demonstrated a statistically significant increase in publication volume for the assistant professor group but not the associate professors, despite having an overall higher publishing volume in both the prepandemic and pandemic period. Similar to gender subgroup analyses, this seemed to be attributed to a relatively small sample size for the associate professors rather than lack of an increase in publishing volume. A greater SD in publishing volume was observed in the associate professor group compared with assistant professors. In prior literature, it has been documented that seniority in medicine often has a positive relationship with research performance, likely owing to factors such as the accumulative advantage effect [21]. Interestingly however, variation in research productivity additionally increases with seniority as well, theorized to be attributed to greater variability of individual motivation with age due to changes in personal lives, lack of further promotion opportunities, and other factors. It is possible that these factors may explain some of the observed trends in the present study, although further investigation into this is warranted.

Limitations

Several limitations are identified for the present study. Firstly, because submission dates for articles are not ubiquitously available across all journal platforms, only publication dates could be tracked for the study sample included. Using publication dates as a means of academic productivity will unfortunately introduce heterogeneity in other factors such as review and processing times after submissions. As a result, there is limited interpretation that can be made when addressing differences in month-to-month productivity. Future studies are encouraged to determine whether there is a clear relationship between month-to-month imaging volume and corresponding academic productivity. Small sample sizes for subgroup analyses by gender and seniority also impacted the power of the present study.

In conclusion, researchers of radiology were observed to demonstrate a significantly increased research output during the pandemic period compared with before the pandemic. Study findings provide unique insight into trends related to academic productivity during the COVID-19 pandemic and may be used to inform how research production changes in times of drastic health care disruption during pandemics.

Take-Home Points

• ▪Among radiology researchers, there seemed to have been a significant increase in publishing volume during the COVID-19 pandemic.
• ▪The increase of academic productivity observed may be reflective of the clinical and imaging volume ramp down experienced by radiologists during the COVID-19 pandemic.
• ▪Both male and female academic radiologists were observed to have similar increases in academic productivity during the pandemic.