Blinding of Peer Review and the Impact on Geographic Diversity of Authors in the Medical Literature

Figures & data

Table 1 Characteristics of the Included Journals and Outcomes of Diversity Evaluation

Characteristic	Total (N=1054)	Single-Blinded (N=766)	Double-Blinded (N=288)	P value
Journal age (years)^a	28 [18–41]	28 [18–40]	28 [18–41]	0.969
Number of issues per year^b	6 [6–12]	8 [6–12]	6 [4–12]	< 0.0001
Journals publishing both clinical and basic sciences work	408 (38.7)	343 (44.8)	65 (22.6)	< 0.0001
OA status^c				< 0.0001
● Hybrid	745 (70.7)	523 (68.3)	222 (77.1)
● Gold OA	228 (21.6)	198 (25.8)	30 (10.4)
● OA without a fee	57 (5.4)	24 (3.1)	33 (11.5)
● Page charges	19 (1.8)	19 (2.5)	0 (0)
● Green OA	5 (0.5)	2 (0.3)	3 (1)
Publisher				< 0.0001
● Elsevier	168 (15.9)	115 (15)	53 (18.4)
● Springer	143 (13.6)	134 (17.5)	9 (3.1)
● Wiley	140 (13.3)	115 (15)	25 (8.7)
● Taylor & Francis	83 (7.9)	34 (4.4)	49 (17)
● Lippincott Williams & Wilkins	69 (6.5)	50 (6.5)	19 (6.6)
● Sage	51 (4.8)	28 (3.7)	23 (8)
● Oxford	30 (2.8)	21 (2.7)	9 (3.1)
● Karger	26 (2.5)	16 (2.1)	10 (3.5)
● BMJ	22 (2.1)	20 (2.6)	2 (0.7)
● Marry-Ann	12 (1.1)	11 (1.4)	1 (0.3)
● Cambridge	11 (1)	9 (1.2)	2 (0.7)
● Bentham	11 (1)	11 (1.4)	0 (0)
● Hindawi	8 (0.8)	8 (1)	0 (0)
● SciELO	8 (0.8)	3 (0.4)	5 (1.7)
● MedKnow	8 (0.8)	1 (0.1)	7 (2.4)
● Other	260 (24.7)	186 (24.3)	74 (25.7)
Specialty (top 15)				< 0.0001
● General medicine	134 (12.7)	85 (11.1)	49 (17)
● Cardiology	79 (7.5)	68 (8.9)	11 (3.8)
● Oncology	63 (6)	57 (7.4)	6 (2.1)
● Neurology/Neurosurgery	59 (5.6)	55 (7.2)	4 (1.4)
● Psychiatry	75 (7.1)	48 (6.3)	27 (9.4)
● Infectious diseases	47 (4.5)	37 (4.8)	10 (3.5)
● Dentistry	45 (4.3)	26 (3.4)	19 (6.6)
● Endocrinology	42 (4)	37 (4.8)	5 (1.7)
● Gastroenterology	38 (3.6)	30 (3.9)	8 (2.8)
● Nursing	38 (3.6)	6 (0.8)	32 (11.1)
● Pharmacy	32 (3)	26 (3.4)	6 (2.1)
● General surgery	30 (2.8)	22 (2.9)	8 (2.8)
● Orthopedics	29 (2.8)	17 (2.2)	12 (4.2)
● Radiology	24 (2.3)	13 (1.7)	11 (3.8)
● ENT	20 (1.9)	11 (1.4)	9 (3.1)
Journal’s country (top 10)				< 0.0001
● International	452 (42.9)	355 (46.3)	97 (33.7)
● United States	284 (26.9)	201 (26.2)	83 (28.9)
● United Kingdom	85 (8.1)	56 (7.3)	29 (10.1)
● Japan	30 (2.8)	28 (3.7)	2 (0.7)
● Canada	23 (2.2)	18 (2.3)	5 (1.7)
● Brazil	15 (1.4)	5 (0.7)	10 (3.5)
● Germany	13 (1.2)	12 (1.6)	1 (0.3)
● Italy	11 (1)	11 (1.4)	0 (0)
● China	11 (1)	10 (1.3)	1 (0.3)
● France	7 (0.7)	4 (0.5)	3 (1)
Journal’s country in the journal title	189 (17.9)	118 (15.4)	71 (24.7)	< 0.0001
Journal of a society	670 (63.6)	479 (62.5)	191 (66.3)	0.265
Indexed in SCIE^d	908 (86.1)	679 (88.6)	229 (79.5)	< 0.0001
Indexed in ESCI^e	993 (94.2)	732 (95.6)	261 (90.6)	0.002
Indexed in Scopus^f	1035 (98.2)	754 (98.4)	281 (97.6)	0.347
Indexed in Embase^g	961 (91.2)	722 (94.3)	239 (83)	< 0.0001
Impact factor^h	3.490 [2.414–5.411]	3.671 [2.595–2.952]	2.852 [2.032–4.192]	< 0.0001
CiteScore^i	4.5 [2.8–6.6]	4.8 [3.2–7.5]	2.5 [2.2–5.1]	< 0.0001
%diversity (%)	45 [30–60]	45 [35–60]	45 [30–55]	0.199
Simpson’s diversity index	0.84 [0.64–0.92]	0.84 [0.65–0.92]	0.82 [0.57–0.92]	0.128
Number of countries	9 [6–12]	9 [7–12]	9 [6–11]	0.199

Notes: Data are presented as median [IQR] or n (%). ^aAge was calculated by subtracting the year in which the journal was founded from year 2022. ^bThe number of issues published annually. 89 journals publish articles continuously without assigning them to volumes and/or issues; hence, were not included in this comparison. ^cGold OA indicates that journals require a certain fee for publishing of an article after acceptance. Green OA indicates the availability of the full text of an article after a certain embargo period. Page charges are typically less expensive than gold OA as the charges are based on the number of published pages. ^dn=906; 679 single-blinded and 227 double-blinded. ^eThis includes journals indexed in both ESCI and SCIE as of June 2022; hence, appear on the Web of Science website (n=993; 732 single-blinded and 261 double-blinded). ^fn=1035; 754 single-blinded and 281 double-blinded. ^gn=961; 722 single-blinded and 239 double-blinded. ^hBased on Journal Citation Reports 2021 data that were released in June 2022. ⁱBased on Scopus CiteScore data that were released in July 2022.

Abbreviations: BMJ, British Medical Journal; ENT, ear, nose, and throat; ESCI, Emerging Sources Citation Index; OA, open-access; SCIE, Science Citation Index Expanded.

Figure 1 Initial evaluation of MEDLINE-indexed journals for inclusion in the study.

Figure 2 Count of journal original countries (n=603; 452 of the 1054 journals are international journals; hence, were not included in this chart).

Table 2 Effect of Different Factors on %diversity Using Multivariate Linear Regression

Factor	β Coefficient ± SE	P value
Blinding
● Single blinding	Ref
● Double blinding	1.01 ± 0.01	0.363
Journal age	0.99 ± 0.0002	0.020
Number of issues per year^a	1.03 ± 1.25	0.832
Publishing both clinical and basic sciences work	1.03 ± 0.01	0.004
OA status
● Hybrid	Ref
● Gold OA	0.95 ± 0.01	< 0.0001
● OA without a fee	0.88 ± 0.02	< 0.0001
● Page charges	0.93 ± 0.04	0.031
● Green OA^b	0.51 ± 0.1	< 0.0001
Journal’s country in the journal title	0.78 ± 0.01	< 0.0001
Journal of a society	0.98 ± 0.01	< 0.0001
Indexing in SCIE	1.05 ± 0.02	0.004
Indexing in ESCI	1.06 ± 0.03	0.039
Indexing in Scopus	1.28 ± 0.04	< 0.0001
Indexing in Embase	1.04 ± 0.02	0.032
Impact factor	1.09 ± 0.12	0.456
CiteScore	1.33 ± 0.12	0.015

Notes: ^aThis included 965 journals with regular number of volumes and issues. The remaining 89 journals publish articles continuously without assigning them to volumes and/or issues. ^bCould not be properly calculated due to very small sample size (n=5). Publishers, specialties, and countries could not be included in the model due to very small sample size in each category, which resulted in losing the goodness-of-fit of the model.

Abbreviations: ESCI, Emerging Sources Citation Index; OA, open-access; SCIE, Science Citation Index Expanded; SE, standard error.

Table 3 Effect of Different Factors on Simpson’s Diversity Index Using Multivariate Linear Regression

Factor	β Coefficient ± SE	P value
Blinding
● Single blinding	Ref
● Double blinding	1 ± 0.02	0.905
Journal age	1 ± 0.0003	0.375
Number of issues per year^a	1 ± 0.002	0.377
Publishing both clinical and basic sciences work	1.01 ± 0.15	0.363
OA status
● Hybrid	Ref
● Gold OA	0.96 ± 0.02	0.040
● OA without a fee	0.94 ± 0.32	0.064
● Page charges	0.98 ± 0.52	0.691
● Green OA	0.72 ± 0.1	0.001
Journal’s country in the journal title	0.87 ± 0.02	< 0.0001
Journal of a society	0.93 ± 0.01	< 0.0001
Indexing in SCIE	1.09 ± 0.03	0.002
Indexing in ESCI	0.98 ± 0.04	0.577
Indexing in Scopus	1.11 ± 0.05	0.039
Indexing in Embase	1.04 ± 0.03	0.086
Impact factor	1 ± 0.002	0.134
CiteScore	1 ± 0.002	< 0.0001

Notes: ^aThis included 965 journals with regular number of volumes and issues. The remaining 89 journals publish articles continuously without assigning them to volumes and/or issues. Publishers, specialties, and countries could not be included in the model due to very small sample size in each category, which resulted in losing the goodness-of-fit of the model.

Abbreviations: ESCI, Emerging Sources Citation Index; OA, open-access; SCIE, Science Citation Index Expanded; SE, standard error.

Table 4 Effect of Different Factors on the Number of Countries Using Multivariate Linear Regression

Factor	β Coefficient ± SE	P value
Blinding
● Single blinding	Ref
● Double blinding	1.01 ± 0.02	0.684
Journal age	0.99 ± 0.0005	0.299
Number of issues per year^a	1 ± 0.002	0.798
Publishing both clinical and basic sciences work	1.03 ± 0.02	0.200
OA status
● Hybrid	Ref
● Gold OA	0.95 ± 0.03	0.044
● OA without a fee	0.88 ± 0.05	0.016
● Page charges	0.93 ± 0.08	0.335
● Green OA^b	–	–
Journal’s country in the journal title	0.78 ± 0.03	< 0.0001
Journal of a society	0.98 ± 0.02	< 0.0001
Indexing in SCIE	1.05 ± 0.04	0.197
Indexing in ESCI	1.06 ± 0.06	0.357
Indexing in Scopus	1.28 ± 0.09	0.010
Indexing in Embase	1.04 ± 0.04	0.337
Impact factor	1 ± 0.002	0.399
CiteScore	1 ± 0.002	0.004

Notes: ^aThis included 965 journals with regular number of volumes and issues. The remaining 89 journals publish articles continuously without assigning them to volumes and/or issues. ^bCould not be properly calculated due to very small sample size (n=5). Publishers, specialties, and countries could not be included in the model due to very small sample size in each category, which resulted in losing the goodness-of-fit of the model.

Abbreviations: ESCI, Emerging Sources Citation Index; OA, open-access; SCIE, Science Citation Index Expanded; SE, standard error.