Our search yielded 5,954 hematology and oncology-related RCTs. The gender of the first or the senior author of 63 RCTs could not be confirmed. Therefore, the population sample for analysis comprised first and senior authors of 5,891 RCTs.
The majority of the first (85.0%, n = 5,009) and senior authors (85.7%, n = 5,049) in our study population were White; Asian authors comprised about 13% in both authorship positions (Table 1). Over two-thirds of the authors had an M.D. degree without a master's or Ph.D., and 22% had an M.D. with a Ph.D. in both authorship positions.
Table 1
Characteristics of the authors of hematology and oncology RCTs. All values are presented as n (%).
Variables | Overall (n = 5,891) | Female Authors |
First Author | Senior Author | First Author (n = 1,349) | Senior Author (n = 1,032) |
Race or ethnicity |
White | 5,009 (85.0) | 5,049 (85.7) | 1,153 (85.5) | 898 (87.0) |
Asian | 767 (13.0) | 722 (12.3) | 174 (12.9) | 114 (11.0) |
Hispanic | 86 (1.5) | 83 (1.4) | 14 (1.0) | 14 (1.4) |
Black | 18 (0.3) | 27 (0.5) | 6 (0.4) | 5 (0.5) |
Other | 11 (0.2) | 10 (0.2) | 2 (0.1) | 1 (0.1) |
Education |
Master | 12 (0.2) | 19 (0.3) | 7 (0.5) | 9 (0.9) |
Master, Ph.D., | 29 (0.5) | 27 (0.5) | 11 (0.8) | 10 (1.0) |
M.D. | 3,960 (67.2) | 3,975 (67.5) | 851 (63.1) | 673 (65.2) |
M.D., Master | 242 (4.1) | 188 (3.2) | 94 (7.0) | 39 (3.8) |
M.D., Master, Ph.D. | 20 (0.3) | 32 (0.5) | 7 (0.5) | 2 (0.2) |
M.D., Ph.D. | 1,299 (22.1) | 1,333 (22.6) | 240 (17.8) | 210 (20.3) |
Ph.D. | 242 (4.1) | 274 (4.7) | 102 (7.6) | 75 (7.3) |
Other | 31 (0.5) | 23 (0.4) | 15 (1.1) | 10 (1.0) |
N/A | 56 (1.0) | 20 (0.3) | 22 (1.6) | 4 (0.4) |
Country of Affiliated Institution (World Bank Income Classification)* |
Low income | 0 (0.0) | 1 (0.0) | 0 (0.0) | 0 (0.0) |
Lower middle income | 31 (0.5) | 27 (0.5) | 8 (0.6) | 4 (0.4) |
Upper middle income | 235 (4.0) | 235 (4.0) | 71 (5.3) | 44 (4.3) |
High income | 5625 (95.5) | 5628 (95.5) | 1270 (94.1) | 984 (95.3) |
Top 10 Journals based on Publication Count# |
Ann Oncol | 950 (16.1) | 950 (16.1) | 200 (14.8) | 161 (15.6) |
Haematologica | 134 (2.3) | 134 (2.3) | 41 (3.0) | 25 (2.4) |
J Clin Oncol | 2,057 (34.9) | 2,057 (34.9) | 491 (36.4) | 371 (35.9) |
J Natl Cancer Inst | 177 (3.0) | 177 (3.0) | 47 (3.5) | 34 (3.3) |
J Thorac Oncol | 219 (3.7) | 219 (3.7) | 47 (3.5) | 38 (3.7) |
JAMA Oncol | 126 (2.1) | 126 (2.1) | 30 (2.2) | 20 (1.9) |
Lancet | 215 (3.6) | 215 (3.6) | 38 (2.8) | 34 (3.3) |
Lancet Oncol | 610 (10.4) | 610 (10.4) | 133 (9.9) | 112 (10.9) |
Leukemia | 148 (2.5) | 148 (2.5) | 28 (2.1) | 23 (2.2) |
N Engl J Med | 413 (7.0) | 413 (7.0) | 75 (5.6) | 67 (6.5) |
*First authors represented 30 high-income, nine upper-middle-income, and seven lower-middle-income countries. Senior authors represented 32 high-income, 11 upper-middle-income, five lower-middle-income, and one low-income country. # RCTs meeting our inclusion criteria were published in 62 journals, 85.7% of which were published in 10 journals. Country and Journal-level datasets are presented in Supplementary Tables 1 and 2 |
A total of 49 countries were represented in our study sample. Authors in both authorship positions were predominantly (95.5%) affiliated with institutions in high-income countries (Table 1). At the country level, the majority of first and senior authors were affiliated with institutions in the U.S. (35.1%, n = 2069 and 36.3%, n = 2140), Germany (8.8%, n = 518 and 9.2%, n = 540), U.K. (8.7%, n = 512 and 8.4%, n = 494), and France (8.5%, n = 503 and 8.2%, n = 485). Complete data for authorship distribution at the country level is presented in Supplementary Table 1.
Although our overall sample included RCTs published in 62 journals, 85.7% of the RCTs were published in just ten journals: J Clin Oncol (34.9%, n = 2,057), Ann Oncol (16.1%, n = 950), Lancet Oncol (10.4%, n = 610), N Engl J Med (7%, n = 413), J Thorac Oncol (3.7%, n = 219), Lancet (3.6%, n = 215), J Natl Cancer Inst (3%, n = 177), Leukemia (2.5%, n = 148), Haematologica (2.3%, n = 134), and JAMA Oncol (2.1%, n = 126) (Table 1). Author distribution for all 62 journals is presented in Supplementary Table 2.
Representation of females in first and senior authorship positions
The proportion of females in our study population was 22.9% (n = 1,349) in the first authorship and 17.5% (n = 1,032) in the senior authorship positions (Table 1). Like the overall study population, females in first (85.5%, n = 1,153) and senior (87.0%, n = 898) authorship positions were predominantly White, held an M.D. degree without master's or Ph.D. (63.1%, n = 851 and 65.2%, n = 673), and were affiliated with institutions in high-income countries (94.1%, n = 1,270 and 95.3%, n = 984) (Table 1). Most female first authors in our study population were affiliated with institutions in the U.S. (24%, n = 496), followed by the Netherlands (40.3%, n = 108), France (19.9%, n = 100), Italy (23.4%, n = 98), and U.K. (18.8%, n = 96) (Supplementary Table 1). Female senior authors were most commonly affiliated with institutions in the U.S. (19.7%, n = 422), U.K. (17.2%, n = 85), France (15.3%, n = 74), Germany (11.5%, n = 62), and Italy (15%, n = 60) (Supplementary Table 1). The top 10 journals accounted for 83.4% of RCTs with females in the first authorship position and 85.7% with females in the senior authorship position (Table 1).
We also noted a significant upward trend in the percentage of female authors between 2000 and 2022, but only in the first authorship position (16.3% in 2000 to 32.4% in 2022, p = 0.001; Fig. 1A).
Trends based on the gender of the first and senior author
Nearly two-thirds (64.9%, n = 3,825) of the RCTs in our study sample had males in both first and senior authorship positions, 17.5% (n = 1,034) had a female first author with a male senior author, 12.1% (n = 717) had a male first author with a female senior author, and 5.3% (n = 315) had females in both first and senior authorship positions.
However, there was a significant decline in the percentage of RCTs with males in both first and senior authorship positions, while the percentage of RCTs with a female in the first authorship position and a male senior author increased significantly between 2000 and 2022 (p = 0.0025 and 0.0009; Fig. 1B). There was a small but significant increase in the percentage of RCTs with females in both first and senior authorship positions (p = 0.0014). In contrast, the percentage of RCTs with a male in the first authorship position and a female in the senior position remained unchanged (Fig. 1B).
Trends based on the gender of the authors and the income classification of the country of their affiliated institution
In 97.8% of the included RCTs, the first and senior authors were affiliated with institutions in countries with matching World Bank income classification. However, most author dyads were from high-income countries (94.4%), followed by upper-middle (3.1%) and lower-middle (0.3%) income countries. There were no author dyads from low-income countries.
Two-thirds of the RCTs with males in first and senior authorship positions were from lower-middle and high-income countries. In contrast, only half of the RCTs from upper-middle-income countries had males in first and senior authorship positions (Fig. 2A). Although the percentage of RCTs with females in both first and senior authorship positions was highest in high-income countries (5.4%; n = 302), the percentage of RCTs with First Female-Senior Male (29.5%, n = 54) and First Male-Senior Female (15.8%, n = 29) dyads were highest in upper-middle-income countries (Fig. 2A). As a result, the odds of the females in the first authorship position with a female senior author were significantly higher when both authors came from high-income countries [OR = 1.702 (95%CI 1.461–1.984; p < 0.0001]. Although the odds of females in the first authorship position with a male senior author when both authors came from upper-middle-income countries was numerically higher, it did not reach statistical significance [OR = 2.406 (95%CI 1.004–6.278; p = 0.0512].
In 68 RCTs (1.2% of the included RCTs), the first author was affiliated with institutions in countries with a lower income classification than the senior author; in all instances, the senior author was from a high-income country. The percentage of the RCTs with the First Female-Senior Male dyad was higher when the first author was affiliated with an institution in a lower-middle income country (26.7%, n = 4) compared to when affiliated to an institution in an upper-middle income country (15.4%, n = 8). At the same time, the First Male-Senior female dyad was more common when the first author was affiliated with an institution in an upper-middle income country (6.7%, n = 1 vs. 15.8%, n = 29) (Fig. 2B). However, the odds of females in the first authorship position with a female senior author when the first author was affiliated to an institution lower income country than the senior author was insignificant [OR = 1.333 (95%CI 0.2478–6.617; p = 0.66].
In the remaining 62 RCTs (1.1% of the included RCTs), the first author was affiliated with institutions in countries with a higher income classification (in all instances, a high-income country) than the senior author. In these RCTs, the dyad of male first and male senior authors was most common (69.4%, n = 43), followed by dyads of First Female-Senior Male (12.9%, n = 8), First Male-Senior Female (11.3%, n = 7), and both females (6.5%, n = 4) authors (Fig. 2B). However, the odds of females in the first authorship position with a female senior author when the first author was affiliated to an institution higher income country than the senior author was insignificant [OR = 3.071 (95%CI 0.8325–11.29; p = 0.19].
Trends based on the gender of the authors in the top ten journals
While the dyad of male first and male senior authors was predominant in all ten journals, J Natl Cancer Inst (61%, n = 108) and Haematologica (61.2%, n = 82) had the lower percentage of males in both authorship positions followed by J Clin Oncol (63.7%, n = 1311), JAMA Oncol (65.1%, n = 82), J Thorac Oncol (65.3%, n = 143), Ann Oncol (65.6%, n = 623), Lancet Oncol (65.9%, n = 402), Leukemia (68.9%, n = 102), N Engl J Med (69.5%, n = 287), and Lancet (70.7%, n = 152) (Fig. 3).
Interestingly, even though the top 10 journals also accounted for the majority of females in the first and senior authorship positions, the bottom 52 journals had even greater odds of a female in the first authorship position with a female senior author [OR = 2.084 (95%CI 1.427–3.041; p = 0.0002] than the top ten journals [OR = 1.553 (95%CI 1.320–1.828; p < 0.0001].