Comparative Analysis of Pediatric Mailing Lists on the Internet Angel A. Herna´ndez-Borges, MD*; Luis G. Pareras, MD‡; and Alejandro Jime´nez, PhD§ ABSTRACT. Objective. To analyze quantitative as- he diffusion of information is crucial to the pects and the relative quality of various pediatric discus- progress of medicine. The great profusion of sion groups on the Internet and to contrast them with qual- conferences, courses, and specialized journals itative aspects of a selected number of pediatric journals. T allows a reasonably fast dissemination of knowl- Material and Methods. An extensive number of mail- edge. Nevertheless, electronic mail offers a low-cost ing lists on the Internet of interest to pediatricians was compiled. Twelve of them concerned with pediatric spe- and faster system of information exchange that has cialties were selected. Six representative journals of pe- been used extensively in the academic environment. diatric specialties were also analyzed and compared with The growth of the Internet during the last few years the corresponding mailing lists. From the list of subscrib- has enhanced this way of sharing information, be- ers we studied the potential quality of each mailing list. coming a popular way of communication among The postings sent by each member to the on-line discus- nonacademic professionals. sions were also analyzed. As an estimate of the standing Mailing lists are discussion groups usually focused as author of each list member as well as of each first on a subject, whose members or subscribers inter- author of the selected journals, we calculated several change messages by electronic mail. Some mailing indexes of quality using the 1995 Medline database and the impact factors of the biomedical journals reported by lists are manually maintained. Others are maintained the 1994 Science Citation Index. by a number of commercial and public domain soft- Results. The most popular lists were NICU-NET ware packages including Listserv, Majordomo, List- and PICU, both having more than 1100 subscribers. proc, and others. Postings sent to the group are re- PEDPATH and PEDIHEART had the highest percentage ceived by computers which run these softwares or of subscribers who were published authors, and their list servers, and automatically send them to all the papers also yielded the highest impact factors. The most members of the mailing list. Thus, the comments active lists were NICU-NET and PICU. The most partic- may be read and discussed by the other subscribers, ipative ones were CHILD-NEURO and PED-LUNG. and an electronic dialogue of comment, reply, and CHILD-NEURO had the highest percentage of authors further replies is established as an international con- among the participants. PEDPATH and CHILD-NEURO 1–9 had the authors with the highest impact factors among ference on a topic held daily. The subscription to a the people who participated in the discussions. These mailing list is free. All that is required is a standard latter two lists also showed the highest impact factor per electronic-mail account and a simple piece of com- posting. Those which had the highest yield (highest ac- munication software. tivity with highest quality per posting) were CHILD- The degree of control during postings to a mailing NEURO and PEDIHEART. The average impact factor per list server varies widely. The lowest level of control first author of the analyzed journals was always higher is restricting postings to members of the list. This than the average impact factor per participant of the lists. prevents nonmembers, particularly advertisers, from Conclusions. The electronic-mail discussion groups on posting to the mailing list. The next level is to subject the Internet are new nonacademic forums in which knowl- edge and experience in pediatrics can be shared. They can- each posting to a process in which only messages not replace but they complement other more academic that the manager of the mailing list approves are sources such as medical journals. Pediatrics 1997;100(2). available by the members of the mailing list. The URL: http://www.pediatrics.org/cgi/content/full/100/2/e8; highest level of control includes both the posting Internet, mailing lists, pediatric information sources. approval and its edition as needed. The more control a moderator exercises on a list the more work is ABBREVIATIONS. IPQ, Index of Potential Quality; IFp, impact involved. Although the quality of a moderated list is factor per participant; IFe, impact factor per posting; IFa, average generally better it is at the expense of some sponta- impact factor. neity. The arrival of the Internet and its electronic forums of discussion in the field of biomedicine are generat- From the *Departments of Pediatrics and ‡Neurosurgery, and the §Research ing concern as to whether the opinions expressed in Unit, University Hospital of Canary Islands, Tenerife, Canary Islands, this media are sufficiently reliable.2 Consequently it Spain. Received for publication Aug 12, 1996; accepted Mar 3, 1997. has been suggested that certain measures of control Reprint requests to (A.A.H.-B.) Valencia, 21. 38005 Santa Cruz de Tenerife, should be introduced into these ever-changing, Tenerife (Canary Islands, Spain). somewhat chaotic groups that have emerged as a The authors’ e-mail addresses are as follows: Dr Herna´ndez-Borges: result of computer networks. One possible way to [email protected]; Dr Pareras: [email protected]; Dr Jime´nez: [email protected] evaluate opinions that are expressed on-line is to PEDIATRICS (ISSN 0031 4005). Copyright © 1997 by the American Acad- check up on the credentials as authors of the partic- emy of Pediatrics. ipants by looking at their inclusion in bibliographical http://www.pediatrics.org/cgi/content/full/100/2/Downloaded from www.aappublications.org/newse8 by guestPEDIATRICS on September Vol. 29, 2021 100 No. 2 August 1997 1of7 databases. This is the approach we took to analyze 5. The IFa per participant (IFp), that is the sum of the individual both the potential and real quality of a subset of impact factor of the published authors posting messages di- vided by the total number of people posting messages whether pediatric Internet discussion groups. At the same published authors or not. time we have also compared these groups with the 6. The IFa per posting (IFe), which was calculated by the sum of more traditional forum of medical journals in which the individual impact factors of participants in the discussions the latest medical findings are presented. multiplied by the number of postings of each individual. This sum was then divided by the total number of postings during MATERIALS AND METHODS the period of observation. 7. Finally, the Yield Index of the group which is calculated by From March 1 to June 30, 1996 we analyzed a subset of unmod- multiplying the IFe by the number of postings generated each erated and professional discussion groups (mailing lists) special- day during the period of observation. izing in pediatric topics. We reviewed the lists of subscribers, some general statistics supplied by automated mailing list servers We also analyzed the number of postings each group member and the messages sent by members on-line. sent. This allowed us to assess which participants carried the weight of the discussions in each list and what their reputation Calculation of IFa Among Subscribers on Each Mailing was in terms of published articles. List Finally, a selected subset of journals specializing in pediatrics was contrasted with the mailing lists on the Internet. The journals After randomizing the whole list of subscribers of each group selected were published in the same months as those in which we (N) we selected a representative sample from each one as previ- monitored the mailing lists. These journals were: 10 ously described. Exploratory random samples of nx size from each list of subscribers were obtained to calculate the IFa and the 1. Journal of Pediatric Surgery, April 1996. quasivariance for each group (Sx). Under these conditions, the IFa 2. Journal of Pediatrics, January, February, March 1996 (the section of each exploratory sample follows a t-Student distribution (with on Neonatal and Fetal Medicine). nx-1 degrees of freedom) instead of a normal distribution. The 3. Pediatric Nephrology, April 1996. representative sample size (n) is derived from the quasivariance 4. Journal of the American College of Cardiology, February, April 1996 (Sx) and the estimation error (e): (the section on pediatric cardiology). 5. Journal of Child Neurology, May 1996. S e 5 Ks2 5 K x 6. Pediatric Pulmonology, December 1995. x n1/2 The original articles and short communications from each jour- 2 in which sx is the population variance, which is estimated by nal were analyzed. The first authors were searched in Medline and 1/2 Sx/n . their publications in 1995 were evaluated in terms of their impact Leaving n from this equation: factors in the Science Citation Index as mentioned above. The individual impact factor for each first author was calculated and K2S2 e2 5 x, their sum divided by the number of articles analyzed gave a n coefficient for each publication. This coefficient was contrasted 2 2 with the IPQ of the lists, and also with the IFa per participant (IFp) K Sx n` 5 2 on the lists on the Internet. e Statistical analysis was performed with the Statistica software (Statsoft Inc, Tulsa, OK, 1994). in which n` is assumed for infinite populations. K was obtained from the t-Student table with nx-1 degrees of freedom for a specific probability (in our case a probability level of RESULTS 95%, with type I error equaling .05). Then, an iterative process of We reviewed an extensive list of electronic-mail increasing the degrees of freedom was performed until two equal discussion groups that centered their discussions on consecutive n` values were found.