Solicitation by Spam: A Cross-Sectional Study of Predatory Publisher E-mails Received By Two Anesthesiologist Clinician-Scientists

Predatory journals (ie, entities that prioritize self-interest at the expense of scholarship) are believed to primarily acquire submissions via unsolicited e-mails sent to researchers in bulk.1 These solicitation e-mails place a substantial burden on researchers and clinicians,2,3 as an estimated US$1.1 billion is lost globally in wasted time every year.4 Moreover, predatory e-mails can deceive even senior researchers to submit their work to dubious entities, potentially harming their reputation and contributing to unethical dissemination of research involving people and animals.5,6 The content and burden of predatory e-mails has not been studied within the field of anesthesiology, although they have been characterized in other fields (eg, oncology, orthodontics, surgery).7–9 Given the thousands of articles that we have identified in predatory anesthesia journals,10 we believe that further characterization of predatory e-mails in the field of anesthesiology will help reduce submissions to these predatory journals. The objective of this study was therefore to characterize predatory e-mail solicitations received by 2 anesthesiologist clinician-scientists. METHODS The study protocol was published on the Open Science Framework (doi: 10.17605/OSF.IO/258WG). We collected all unsolicited e-mails from academic journals received by 2 anesthesiologist clinician-scientists at The Ottawa Hospital Research Institute (M.M.L. and D.I.M.) on their respective institutional accounts over a 28-day period (July 2021). M.M.L. and D.I.M. are both mid-career clinician-scientists with established publication records in anesthesia and biomedical journals, and have diverse research interests including translational research, publication science, perioperative care of older adults, and prehabilitation. E-mails were screened by a single reviewer (N.C.). We included all e-mails that solicited a journal submission, from both presumed predatory and non-predatory journals. We excluded all other e-mails. Data from the included e-mails were extracted by a single reviewer (N.C.) with a second reviewer (M.M.L.) auditing a random sample of 20%. Abstracted data included 3 domains: (1) e-mail characteristics, (2) solicitation characteristics, and (3) journal/publisher characteristics (see Supplemental Digital Content 1, Supplemental Table 1, https://links.lww.com/AACR/A565 for the full extraction sheet). Presumed predatory e-mails were defined as those from sources (ie, journals or publishers) not indexed in the Directory of Open Access Journals (DOAJ) or from sources not listed as a member of the Committee on Publication Ethics (COPE). Indexing in the DOAJ requires that a journal be peer-reviewed and open access, whereas membership in COPE requires the journal to demonstrate adherence to high standards in publication ethics. Following e-mail data extraction, the 2 anesthesiologist clinician-scientists (M.M.L. and D.I.M.) were each randomly assigned 30 e-mails addressed to them and asked to determine the relevance of the journal’s title to their respective research interests using a 5-point Likert scale. Counts and descriptive statistics were used to identify common characteristics of solicitations. Direct comparisons were made between presumed predatory and non-predatory e-mails using Firth’s logistic regression for 4 a priori selected objective e-mail characteristics: (1) incorrect naming of the recipient, (2) presence of the word “greetings,” (3) obvious grammar mistakes, and (4) requesting submission via e-mail. Likert scales were analyzed by median and interquartile range. RESULTS Five hundred forty-six unsolicited e-mails were received by the 2 anesthesiologist clinician-scientists over the 28-day period. These e-mails were sent from 84 unique publishers (Supplemental Digital Content 1, Supplemental Table 2, https://links.lww.com/AACR/A566). We analyzed 452 messages, including 430 presumed predatory e-mails (Supplemental Digital Content 1, Supplemental Figure 1, https://links.lww.com/AACR/A567). Presumed predatory e-mails had several common characteristics (Table). Of the 311 (72%) presumed predatory e-mails that named the recipient, 147 (47%) named the recipient incorrectly. The word “greetings” appeared in 167 (39%) of presumed predatory e-mails. The grammar of the predatory e-mails was generally poor as 405 (94%) contained at least one obvious grammatical error. The presumed predatory e-mails also frequently requested submission directly via e-mail (202, 47%). All 4 of these objective characteristics, when present, were significantly associated with the solicitation being from a presumed predatory journal (Supplemental Digital Content 1, Supplemental Figure 2, https://links.lww.com/AACR/A568). Another common characteristic was the use of explicit flattery, which was present in 213 (50%) of predatory e-mails (eg, descriptors such as “most esteemed”). Table. - Characteristics of Presumed Predatory and Non-Predatory E-mails. Predatory N (%) Non-Predatory N (%) E-mail Characteristics E-mail in English 429 (99.8) 22 (100.0) Who was the e-mail sent to? Dr Lalu 167 (38.8) 6 (27.3) Dr McIsaac 263 (61.2) 16 (72.7) E-mail flagged as potentially wanted/reputable 2 (0.5) 11 (50.0) Recipient named in e-mail 311 (72.3) 14 (63.6) Recipient named incorrectly 147 (47.3) 1 (7.1) Personalized subject line 74 (17.2) 2 (9.1) Unprofessional e-mail address 8 (1.9) 0 (0.0) Sent on behalf of a legitimate researcher 13 (3.0) 7 (31.8) “Greetings” used in the e-mail 167 (38.8) 0 (0.0) Content of the e-mail personalized 79 (18.4) 1 (4.5) Obvious flattery 213 (49.5) 3 (13.6) Grammatical mistakes 405 (94.2) 4 (18.2) Multiple fonts used in an unprofessional manner 136 (31.6) 1 (4.5) Country of origin listed 232 (54.0) 16 (72.7) United States 200 (86.2) 1 (6.3) Japan 18 (7.8) 0 (0.0) India 8 (3.4) 0 (0.0) Italy 3 (1.3) 0 (0.0) England 1 (0.4) 9 (56.3) Georgia 1 (0.4) 0 (0.0) Latvia 1 (0.4) 0 (0.0) China 0 (0.0) 1 (6.3) Germany 0 (0.0) 2 (12.5) Switzerland 0 (0.0) 3 (18.8) Street address listed 162 (37.7) 15 (68.2) E-mail includes option to unsubscribe 241 (56.0) 20 (90.1) E-mail uses graphics 12 (2.8) 18 (81.8) Solicitation Characteristics Nature of the E-mail Request for journal submission 430 (100.0) 22 (100.0) Advertising of journal 1 (0.2) 4 (18.2) Invitation to join editorial board 16 (3.7) 0 (0.0) Invitation to be guest editor 1 (0.2) 0 (0.0) Invitation to review 0 (0.0) 2 (9.1) Is there a submission deadline 223 (51.9) 12 (50.0) Median (range) days after e-mail receipt 15 (0–103) 165 (76–178) Solicitation advertises quick turn around 47 (10.9) 10 (45.5) Method of submission E-mail 128 (29.8) 0 (0.0) Online portal 60 (14.0) 13 (59.1) Either 74 (17.2) 0 (0.0) Not specified 168 (39.1) 6 (27.3) Not relevant 0 (0.0) 3 (13.6) Solicitation mentioned publication fee 55 (12.8) 11 (50.0) No fee 16 (29.9) 1 (9.1) Discounted fee 27 (49.1) 8 (72.7) Specific amount listed 10 (18.2) 2 (18.2) Median (range) [USD] 399 (149–608.78) 2385 (2332–2438) Fee is negotiable 2 (3.6) 0 (0.0) Journal and Publisher Characteristics Solicitation claims peer review 90 (20.9) 15 (68.2) Solicitation claims open access 84 (19.5) 13 (59.1) Solicitation claims indexing 60 (14.0) 9 (40.9) ISSN reported 214 (49.8) 2 (9.1) Website given 223 (51.9) 22 (100.0) Impact factor mentioned 137 (31.9) 12 (54.5) Journal/publisher in DOAJ 1 (0.2) 22 (100.0) Presumed predatory solicitations regularly used time limitations to pressure a quick response. These e-mails often gave a deadline to either submit an article, receive a discount, or reply to the e-mail (223, 52%). The median time between e-mail receipt and the deadline was 15 days (range 0–103). Similarly, 47 (11%) presumed predatory e-mails used language that advertise rapid turnaround of submissions. The sources of the presumed predatory e-mails usually lacked transparency, or made false/misleading claims, regarding their publishing practices. Ninety presumed predatory e-mails claimed peer review (21%), 84 (20%) claimed to be open access, and 60 (14%) claimed indexing. Some sources claimed indexing in “databases” that are not considered indexing services (eg, Crossref, International Committee of Medical Journal Editors). Of the 137 (32%) presumed predatory e-mails that presented an impact factor, only one presented a Clarivate-calculated impact factor. Across the 60 randomly selected e-mails assessed for relevance by the 2 anesthesiologist clinician-scientists, the median (IQR) Likert score was 2 (1–3), indicative of a journal title that is “irrelevant” to the respective researcher’s discipline. DISCUSSION We identified a substantial burden of predatory e-mails received by 2 anesthesiologist clinician-scientists. Our analysis provides a novel characterization of predatory e-mails in anesthesia and identifies 4 objective characteristics commonly found in these messages. Although these features have been noted in other fields (oncology,7 orthodontics,8 surgery9), our analysis confirms their significant association with presumed predatory journals. These objective characteristics can help researchers to effectively differentiate predatory e-mails from desired solicitations (Figure). Further, these characteristics can be used to inform the development of tools (eg, improved firewalls to block predatory e-mails) and educational resources to strengthen our collective response to predatory journals.11 To increase the generalizability of our findings, future studies should assess solicitation e-mails received by anesthesiologists, researchers, and trainees with differing backgrounds.Figure.: Is this solicitation from a potentially predatory source?This manuscript was handled by: Charles Emala, MS, MD.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.