- Email: [email protected]
A Review of The Journal of Pediatrics: The First 75 Years
A Review of The Journal of Pediatrics : The First 75 Years Martha A. Hellems, MD, MS, Kelly K. Gurka, PhD, MPH, and Gregory F. Hayden, MD Objective To assess changes in the format and content of articles published in The Journal of Pediatrics to aid in understanding the current state of pediatric research and to anticipate what to expect in the coming years.
Study design We conducted a retrospective review of case reports, research articles, and review articles published in volume 150 of The Journal. We noted components of study authorship, content, and design and compared these data with those in earlier volumes. Results The numbers of authors and grants per article and the proportion of studies with international authorship have all increased. Print circulation has dropped, but The Journal is now available in electronic format. Research topics shifted, with fewer articles devoted to infectious disease and general pediatrics. An increase in articles about endocrinology reflects a surge of interest in pediatric obesity. Most articles contain original research, reporting analyses from observational studies with few clinical trials. Conclusions Recent trends suggest that readers of The Journal should expect more research articles with observational designs and increasing international authorship. The Journal’s electronic presence will likely increase and provide greater readership globally. (J Pediatr 2009;155:16-20). See editorials, p 3 and p 6
T
he Journal of Pediatrics celebrated its 75th anniversary in 2007 with the publication of volume 150. Since its inception, considerable changes have occurred in the field of pediatrics. Clinicians face a new set of challenges, including the care of extremely low birth weight infants, the recognition and treatment of an ever-expanding number of new diseases and syndromes, the judicious use of numerous new diagnostic tests, drugs, and vaccines, and compliance with new regulations and clinical practice guidelines. Just as the field of pediatrics has seen substantial changes, so have the fields of information technology, biostatistics, and publishing. Computers and the Internet have revolutionized how physicians access information, store and analyze data, and prepare manuscripts. Electronic search engines have changed the way literature is reviewed, and computing technology has allowed major advances in the application and development of statistical methods. Although print journals continue to exist, some now contain electronic pages not included in the print version, and purely electronic journals are becoming more prevalent. The Journal has established a prominent position in pediatrics by publishing important articles of interest to investigators and clinicians alike. Thus, The Journal can serve as 1 indicator of developments in the field of pediatrics during this era. We have characterized the format and content of articles appearing in The Journal at regular intervals in the past 75 years and have identified changes in both style and substance. These observations provide useful perspective so that we may understand how The Journal and the discipline have become what they are today and may anticipate what to expect in the coming years.
Methods We previously reviewed articles appearing in The Journal’s first 50 years.1 We used identical methods to extend this earlier analysis to the last 25 years. We reviewed all articles appearing in volume 150 of The Journal. We included review articles, case reports, and research articles in our analysis. Letters to the editor, editorial comments, clinical-pathological conferences, committee reports, and presidential addresses were excluded. We counted the number of articles and number of pages standardized to the print content of volume 1. We characterized the editorial board, print circulation, and subscription cost in both 2007 and 1932 (US $). We noted the number of authors, the geographic region of origin of the first author, and the presence of grant support and the number of grants acknowledged. In cases in which From the Departments of Pediatrics (M.H., G.H.) and authors included large study groups or research networks, we counted only inPublic Health Sciences (K.G.), University of Virginia, Charlottesville, VA dividual authors listed on the title page and not group members listed in apThe findings and conclusions in this paper are those of pendices. We noted the number of tables, figures, clinical illustrations, the authors, and do not necessarily represent the views of The Journal of Pediatrics, the Editors, or the Publisher, Elsevier, Inc. This is unfunded research with no potential conflict of interest.
AAP
American Academy of Pediatrics
0022-3476/$ - see front matter. Copyright Ó 2009 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.08.049
16
Vol. 155, No. 1 July 2009
Table I. General characteristics of The Journal of Pediatrics by volume number
Table II. Characteristics of authorship in The Journal of Pediatrics by volume number
Volume number and year of publication
Articles per volume (n) Standardized pages per volume (n) Format (per article)† Standardized pages (mean) Tables (mean) Figures (mean) Clinical illustrations (mean) References (mean) References (range) Editors, including assistant, associate, and consulting (n) Editorial board, including contributing editors (n) Approximate circulation Cost of annual subscription in US (nominal $) Cost of annual subscription in US (1932 $)
1
33
66
99
150
1932
1948
1965
1982
2007
67 569
93 603
111 920
225 1272
118 1012*
8.5
6.5
8.3
5.7
8.6
1.8 0.5 1.1
1.2 1.3 1.5
2.5 1.5 1.4
1.3 0.9 0.3
2.0 0.6 0.2
17 0–173 2
11 1–38 1
24 0–357 2
18 2–106 4
27 0–89 7
6
8
18
31
24z
NA $8.50
NA 11,600 31,000 5,575 $10.00 $15.00 $33.50 $189.00
$8.50
$5.68
$6.52
$4.76
Volume number and year of publication
Authorship Authors per article (mean) Articles with only one author (%) Articles with $3 authors (%) Grant funding Grants per article (mean) Articles with $1 grant (%) Region of first author (%) East South Midwest West International
1
33
66
99
150
1932
1948
1965
1982
2007
1.5
2.0
2.6
4.0
6.2
60
32
20
4
3
9
23
49
82
88
0.1
0.2
1.2
1.2
1.4
8
10
64
57
56
28 9 40 19 3
48 11 30 8 3
44 13 14 15 14
23 17 20 16 25
16 11 10 10 53
$12.49
NA, Data not available. *Excluding electronic-only pages. †Excludes electronic-only content. Electronic content included a mean 0.7 e-pages per article consisting of a mean of 1.1 tables, 0.4 figures, and 0.2 illustrations. zIncludes an International Advisory Panel.
references, and pages of text to the nearest half-page. Electronic pages in volume 150 were counted separately. We noted the age of subjects (neonates alone, adolescents alone, or all other) and the primary topic addressed. Each article was classified as a research article, case report, or review article. We categorized the content of each research study as analytic or descriptive, disease- or non-disease-oriented, and people- or laboratory-centered. Study materials were classified as human or non-human. We determined whether statistical tests were used to assess for associations between variables. No attempt was made to quantify the significance of trends with statistical techniques. To test the reliability of the ratings assigned to articles in volume 150, we repeated the examination approximately 2 months after the initial review on a random sample of 10% of the articles.
Results General Trends In 1982, the number of articles per volume was the highest for the years studied, whereas the articles had the shortest mean length, and the numbers of tables, figures, and illustrations were low (Table I). By 2007, the number of articles per volume had declined, but the length of articles had increased. Each article in volume 150 included, on average, 0.7 pages of electronic-only content such as tables, figures, and appendices. Print circulation dropped substantially in 2007 compared
with 1982. The cost of a 1-year subscription has increased, but after adjusting for inflation,2 this increase is modest. Authorship and Format The number of authors per article has risen steadily in the last 75 years and increased >50% between volumes 99 and 150 (Table II). The proportion of articles in volume 150 with only 1 author remained low (3%), whereas most articles had $3 authors. Listed authors included large study groups or research networks in 10 articles in volume 150. The proportion of articles acknowledging at least 1 grant as a source of support continues to exceed one-half, but the region of origin of the first author has changed, with more than half the articles now originating from outside the United States. Topics of Articles The topics addressed by articles in The Journal have changed with time (Table III). For example, the proportion of articles dealing with infectious disease has steadily declined, whereas articles about metabolism and genetics have become more frequent. No articles in volume 150 were devoted to general pediatrics; every paper could be better categorized in another topic area. The proportion of articles addressing >1 topic continued to increase, from 24% in volume 99 to 52% in volume 150. Articles about nutrition accounted for approximately onetenth of articles in volume 1 and tended to concern malnutrition. In contrast, approximately the same proportion of articles in volume 150 concerned obesity (or over-nutrition). These articles were assigned a primary topic of endocrinology, because they focused on clinical and laboratory markers such as insulin resistance, lipid levels, and cardiovascular risk factors. 17
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Table III. Primary topic addressed in articles by volume number
Vol. 155, No. 1
Table IV. Characteristics of the research design by volume number
Percent of articles by volume number
Allergy/immunology Behavioral pediatrics Cardiology Endocrinology Gastroenterology General pediatrics Hematology/oncology Infectious disease Metabolism/genetics Neurology Nutrition Pharmacology Pulmonology Surgery Other
1
33
66
99
150
0 3 0 5 3 9 6 40 3 5 10 5 2 2 9
3 4 10 3 4 5 9 30 1 2 5 2 1 3 16
5 0 5 6 6 4 7 12 23 5 5 10 6 1 6
4 1 0.4 7 8 5 9 14 18 2 4 9 5 0.4 13
4 4 11 19* 4 0 5 5 20 3 0 0 6 0 20†
*Includes 9% of articles about obesity. †Includes 9% of articles about neonatology.
The proportions of papers devoted exclusively to the study of neonates remained stable since the earlier review at 27%. Approximately 9% of articles were focused on care of the sick or premature infant; because the earlier survey did not include ‘‘neonatology’’ as a topic, these articles are included in the ‘‘other’’ category in Table III. Research Design The number of case reports has decreased by approximately 50%, whereas the proportion of research articles has increased steadily (Table IV). The design of research studies has also changed, with a decrease in experimental designs and a corresponding increase in studies with an observational design. The proportion of articles classified as analytic more than doubled in the last 25 years, with a corresponding decrease in descriptive studies (Table IV). The proportion of articles that were disease-oriented and people-centered increased in the last 25 years. Articles involving exclusively human material have always comprised most articles since The Journal’s inception. The proportion of articles reporting statistical testing increased slightly from 76% in volume 99 to 86% in volume 150. Re-examination of a random sample of 10% of the reviewed articles showed the test-retest reliability to be generally excellent. Of the total 276 data points, 254 (92%) were concordant. Of the 22 disagreements, 9 (41%) represented errors in data collection or entry, and 13 (59%) represented discrepancies in interpretation. For example, a few articles included such an even mixture of clinical and laboratory data that it was unclear whether to code these articles as ‘‘people-centered’’ or ‘‘laboratory-centered.’’ Similarly, a few articles that addressed problems originating in the neonatal period included follow-up >30 days, so it was unclear whether to code the ages of the patients as ‘‘neonates only’’ or ‘‘all ages.’’ Taken in aggregate, however, the test-retest reliability was very good. 18
Percent of articles by volume number
Type of article Review Case reports Research Research study design Clinical trial Uncontrolled Controlled, randomized Controlled, non-randomized Observational cohort study Case-control study Cross-sectional study Research focus Analytic Descriptive Exposure allocation Experimental Observational Research orientation Disease Non-disease, state of wellness Focus of data collection People-centered Laboratory-centered Material Human Non-human
1
33
66
99
150
30 33 37
11 39 50
9 29 62
4 31 65
4 17 79
59 41 9 9 5 0 36
35 22 0 13 13 4 48
30 12 9 9 10 1 58
23 7 12 4 20 2 55
8 0 6 1 48 1 43
64 36
44 56
43 57
38 62
72 28
59 41
42 58
41 59
32 68
9 91
79 21
78 22
76 24
70 30
88 12
57 43
60 40
30 70
22 78
39 61
96 4
93 7
96 4
98 2
100 0
Discussion The observed changes in the format and content of articles published in The Journal may reflect changes in pediatrics, in medical publishing, and in The Journal’s editors and editorial policies in these three-quarters of a century. Editorial influence may have been particularly important because The Journal receives many more acceptable articles than can be published and has had only 4 editorial regimes since its inception (Table V; available at www.jpeds.com). The length of published papers, for example, was no doubt reduced by the editors’ liberal use of the ‘‘red pencil’’ to abbreviate and otherwise modify accepted papers.3,4 Near the end of Garfunkel’s tenure, he studied an entire volume to compare the length of accepted articles as they were published with that when they were submitted; only 1 paper was longer.4 The pages saved by editorial suggestions and stipulations allowed The Journal to accept 7 additional articles of average length per month. The number of articles that could be published also depended on the number of journal pages allocated by the publisher, which in turn depended on financial considerations, such as advertising. During the Nelson/Garfunkel era, advertising revenues increased exponentially until the recession of the 1980s, then plummeted (Joseph Garfunkel, personal communication 2008). An important aspect of advertising in The Journal not addressed in this review relates to the content of advertisements and their interspersion throughout the issue. Hellems, Gurka, and Hayden
July 2009 The Journal is now published in both print and electronic formats; the most recent volume reviewed includes content only available electronically. Most journals are now available electronically, and many have used the electronic format as a means of expanding the number of articles published in each issue.5 Many libraries no longer maintain print collections of recent volumes of many journals; readers must instead access the electronic version. Print circulation of The Journal in 2007 declined to 18% of what it was in 1982, but because many readers now access The Journal electronically, it is challenging to compare current readership with that of the past. Electronic publishing provides efficient access to international readers and might explain in part the noted increase in international authorship and the addition of an International Advisory Panel to the Editorial Board. The average number of authors per article and the proportion of manuscripts with multiple authors have continued to increase, a trend noted in many other journals.6 Published articles with multiple authors are cited more frequently, a marker which has been correlated with high research quality, implying that this trend represents an increase in productive collaboration rather than a manifestation of academic gamesmanship. The number of authors can also be influenced by editorial policy. As the number of authors grew, Garfunkel began to request justification when the number exceeded 5. With a challenge, it was common to have 1 to 3 names removed from the revised paper (Joseph Garfunkel, personal communication, 2008). The increase in the percentage of first authors from outside the United States noted in volume 150 has also been reported in other scientific fields.7 This trend may be caused by changes in research funding in the United States and abroad, improvement in the quality of research in other nations, changes in the perceived desirability of publishing research in the United States and abroad, and ease of electronic communication. The increase in listed grants may be caused in part by the requirement by many journals that all sources of support be identified so that conflicts of interest will be more obvious. Of the research articles in volume 150, only 8% were clinical trials, whereas most reported the results of observational cohort or cross-sectional studies. This trend may reflect the increasing accessibility of electronic databases and the computing and statistical power to analyze large amounts of data. Indeed, the complexity of the research design and data analysis in studies published in The Journal has increased since 1982, with fewer reviews, case reports, and descriptive studies. Most articles in volume 150 were analytic and used a varied array of complex statistical techniques, in agreement with parallel trends seen in other medical journals.8,9 Although the increased complexity of study design and analysis suggests increased scientific rigor, the articles may be less easily understood by clinician readers with little research experience or statistical training. One notable trend in The Journal has been the decrease in articles devoted to infectious disease and general pediatrics. Fewer articles about infectious disease may reflect the better A Review of The Journal of Pediatrics: The First 75 Years
ORIGINAL ARTICLES prevention and treatment of many infectious diseases through immunization and antimicrobial treatment. Trends in the research topics presented in The Journal may also reflect competition from other publications. Journals devoted to pediatric infectious disease, ambulatory pediatrics, and virtually every other pediatric subspecialty have been founded since the inception of The Journal. Do the changes noted in this survey of The Journal better serve its readership? According to a recent study,10 a typical American Academy of Pediatrics (AAP) member reads approximately 150 articles per year, spends only approximately 20 minutes per article, reads primarily for current clinical information, and usually reads print journals at home from personal subscriptions. Younger pediatricians and pediatricians doing research are more likely to use electronic journals. The low proportion of review articles, the absence of general pediatric articles, the complexity of some statistical analyses, and the shift to electronic publishing may make The Journal less attractive to practicing general pediatricians. Nelson previously has described The Journal’s creation in 1932 as the official organ of the AAP and the schism between the publisher and the AAP that led to the establishment of Pediatrics in 1948.11 In 2001, The Journal became affiliated with the Association of Medical School Pediatric Department Chairs and has published a series of related reports, policy discussions, and commentary in a special section of The Journal. This collaboration builds on The Journal’s longstanding leadership role in pediatric training and workforce issues. It is challenging to predict what the observed trends in The Journal in the last 75 years portend for the future. If recent trends are any indication, however, readers can expect more research articles with correspondingly fewer reviews and case reports, more authors per article and an increase in the number of international authors, and fewer experimental designs, including clinical trials, with correspondingly more observational designs, especially observational cohorts. The print circulation may decline, but The Journal’s electronic presence will likely increase and provide greater readership. The Journal’s impact factor is very high (4.017 in 2007), suggesting great importance to the field of pediatrics. If The Journal’s editors continue to monitor and respond to the needs and desires of its readers, there is every reason to expect that The Journal will continue to be an important source of information for pediatricians in the years ahead. n Submitted for publication April 18, 2008; last revision received July 31, 2008; accepted Aug 28, 2008. Reprint requests: Martha A. Hellems, MD, Department of Pediatrics, University of Virginia School of Medicine, PO Box 800386, Charlottesville, VA 22908-0386. E-mail: [email protected].
References 1. Hayden GF, Saulsbury FT. A review of The Journal of Pediatrics: the first 50 years. J Pediatr 1982;101:5-11. 19
THE JOURNAL OF PEDIATRICS
www.jpeds.com
2. Calculated from the U.S Department of Labor Bureau of Labor Statistics Consumer Price Index (updated 1/16/2008), Available at: http://www. bls.gov/cpi/#data. Accessed Feb 8, 2008. 3. Balistreri WF. Building on a tradition. J Pediatr 1996;128:97-8. 4. Garfunkel JM. Reflections of an editor on his retirement. J Pediatr 1995; 127:947. 5. Anderson K, Lucey JF. Pediatrics electronic pages: looking back and looking ahead. Pediatrics 1998;102:124-8. 6. Wuchty S, Jones BF, Uzzi B. The increasing dominance of teams in production of knowledge. Science 2007;316:1036-9.
Vol. 155, No. 1 7. Rahman M, Fukui T. A decline in the U.S. share of research articles. N Engl J Med 2002;347:1211-2. 8. Horton NJ, Switzer SS. Statistical methods in the journal. N Engl J Med 2005;353:1977-9. 9. Hellems MA, Gurka MJ, Hayden GF. Statistical literacy for readers of Pediatrics: a moving target. Pediatrics 2007;119:1083-8. 10. Tenopir C, King DW, Clarke MT, Na K, Zhou X. Journal reading patterns and preferences of pediatricians. J Med Libr Assoc 2007;95:56-63. 11. Nelson WE. The first 50 years of The Journal of Pediatrics. J Pediatr 1982; 101:1-4.
50 Years Ago in THE JOURNAL OF PEDIATRICS Hyperbilirubinemia in Premature Infants: A Follow-up Study Koch CA, Jones DV, Dine MS, Wagner EA. J Pediatr 1959;55:23-9
I
n this interesting article, the authors describe the outcomes of preterm infants with varying degrees of hyperbilirubinemia. Serial serum bilirubin concentrations were measured in 100 consecutive preterm infants. Each infant’s condition at birth was rated clinically, and the postnatal course was documented. The infants were invited back for clinical assessment at age 2 to 3 years. Of the 100 infants studied, 8 died before discharge and 42 returned for assessment. Seven infants were assessed after discharge at clinics where charts were accessible and were also included in the study. Of the 8 infants who died, 2 had kernicterus diagnosed at autopsy. Of the 49 infants assessed after discharge, 5 had clinical features suggestive of kernicterus. All of the infants with kernicterus had bilirubin levels above 20 mg/dL. Three of the 4 infants with a bilirubin concentration of 30 to 40 mg/dL exhibited evidence of kernicterus. Importantly, none of those infants subsequently diagnosed with kernicterus exhibited clinically specific features before discharge. The authors concluded that exchange transfusion is indicated in all infants who have or are expected to develop a serum bilirubin concentration > 20 mg/dL. This article provides a glimpse into just how different a preterm infant ’s journey was 50 years ago. There are many important differences between then and now. Instead of using an ad hoc clinical score, neonatologists have now universally adopted the Apgar score. The Bloxsom airlock, considered an innovation in neonatal management 50 years ago, is now only found in museums. Back then, conjugated and unconjugated bilirubin could not be easily differentiated, whereas now they can be routinely measured in the modern chemical pathology laboratory as part of the clinical service. With the advent of phototherapy in the 1960s, the use of exchange transfusions to treat hyperbilirubinemia has decreased dramatically. Therefore, the conclusions of Koch et al are no longer applicable in the modern era. But despite these dramatic advances, many questions remain; for example, factors predisposing the human infant brain to the development of kernicterus are still poorly understood. Given the relatively safe option of phototherapy to effectively treat hyperbilirubinemia and the potentially disastrous consequences of kernicterus, ethical constraints make it unlikely that these questions will ever be answered completely. Hugh S. Lam, MRCPCH Department of Pediatrics The Chinese University of Hong Kong New Territories, Hong Kong 10.1016/j.jpeds.2009.01.053
20
Hellems, Gurka, and Hayden
ORIGINAL ARTICLES
July 2009
Table V. Editors of The Journal of Pediatrics Years
Editor(s)
1932-1948 1948-1959 1959-1977 1977-1995 1996-present
Borden S. Veeder and Hugh McCulloch Borden S. Veeder Waldo E. Nelson Joseph M. Garfunkel William F. Balistreri
A Review of The Journal of Pediatrics: The First 75 Years
20.e1
Study design We conducted a retrospective review of case reports, research articles, and review articles published in volume 150 of The Journal. We noted components of study authorship, content, and design and compared these data with those in earlier volumes. Results The numbers of authors and grants per article and the proportion of studies with international authorship have all increased. Print circulation has dropped, but The Journal is now available in electronic format. Research topics shifted, with fewer articles devoted to infectious disease and general pediatrics. An increase in articles about endocrinology reflects a surge of interest in pediatric obesity. Most articles contain original research, reporting analyses from observational studies with few clinical trials. Conclusions Recent trends suggest that readers of The Journal should expect more research articles with observational designs and increasing international authorship. The Journal’s electronic presence will likely increase and provide greater readership globally. (J Pediatr 2009;155:16-20). See editorials, p 3 and p 6
T
he Journal of Pediatrics celebrated its 75th anniversary in 2007 with the publication of volume 150. Since its inception, considerable changes have occurred in the field of pediatrics. Clinicians face a new set of challenges, including the care of extremely low birth weight infants, the recognition and treatment of an ever-expanding number of new diseases and syndromes, the judicious use of numerous new diagnostic tests, drugs, and vaccines, and compliance with new regulations and clinical practice guidelines. Just as the field of pediatrics has seen substantial changes, so have the fields of information technology, biostatistics, and publishing. Computers and the Internet have revolutionized how physicians access information, store and analyze data, and prepare manuscripts. Electronic search engines have changed the way literature is reviewed, and computing technology has allowed major advances in the application and development of statistical methods. Although print journals continue to exist, some now contain electronic pages not included in the print version, and purely electronic journals are becoming more prevalent. The Journal has established a prominent position in pediatrics by publishing important articles of interest to investigators and clinicians alike. Thus, The Journal can serve as 1 indicator of developments in the field of pediatrics during this era. We have characterized the format and content of articles appearing in The Journal at regular intervals in the past 75 years and have identified changes in both style and substance. These observations provide useful perspective so that we may understand how The Journal and the discipline have become what they are today and may anticipate what to expect in the coming years.
Methods We previously reviewed articles appearing in The Journal’s first 50 years.1 We used identical methods to extend this earlier analysis to the last 25 years. We reviewed all articles appearing in volume 150 of The Journal. We included review articles, case reports, and research articles in our analysis. Letters to the editor, editorial comments, clinical-pathological conferences, committee reports, and presidential addresses were excluded. We counted the number of articles and number of pages standardized to the print content of volume 1. We characterized the editorial board, print circulation, and subscription cost in both 2007 and 1932 (US $). We noted the number of authors, the geographic region of origin of the first author, and the presence of grant support and the number of grants acknowledged. In cases in which From the Departments of Pediatrics (M.H., G.H.) and authors included large study groups or research networks, we counted only inPublic Health Sciences (K.G.), University of Virginia, Charlottesville, VA dividual authors listed on the title page and not group members listed in apThe findings and conclusions in this paper are those of pendices. We noted the number of tables, figures, clinical illustrations, the authors, and do not necessarily represent the views of The Journal of Pediatrics, the Editors, or the Publisher, Elsevier, Inc. This is unfunded research with no potential conflict of interest.
AAP
American Academy of Pediatrics
0022-3476/$ - see front matter. Copyright Ó 2009 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2008.08.049
16
Vol. 155, No. 1 July 2009
Table I. General characteristics of The Journal of Pediatrics by volume number
Table II. Characteristics of authorship in The Journal of Pediatrics by volume number
Volume number and year of publication
Articles per volume (n) Standardized pages per volume (n) Format (per article)† Standardized pages (mean) Tables (mean) Figures (mean) Clinical illustrations (mean) References (mean) References (range) Editors, including assistant, associate, and consulting (n) Editorial board, including contributing editors (n) Approximate circulation Cost of annual subscription in US (nominal $) Cost of annual subscription in US (1932 $)
1
33
66
99
150
1932
1948
1965
1982
2007
67 569
93 603
111 920
225 1272
118 1012*
8.5
6.5
8.3
5.7
8.6
1.8 0.5 1.1
1.2 1.3 1.5
2.5 1.5 1.4
1.3 0.9 0.3
2.0 0.6 0.2
17 0–173 2
11 1–38 1
24 0–357 2
18 2–106 4
27 0–89 7
6
8
18
31
24z
NA $8.50
NA 11,600 31,000 5,575 $10.00 $15.00 $33.50 $189.00
$8.50
$5.68
$6.52
$4.76
Volume number and year of publication
Authorship Authors per article (mean) Articles with only one author (%) Articles with $3 authors (%) Grant funding Grants per article (mean) Articles with $1 grant (%) Region of first author (%) East South Midwest West International
1
33
66
99
150
1932
1948
1965
1982
2007
1.5
2.0
2.6
4.0
6.2
60
32
20
4
3
9
23
49
82
88
0.1
0.2
1.2
1.2
1.4
8
10
64
57
56
28 9 40 19 3
48 11 30 8 3
44 13 14 15 14
23 17 20 16 25
16 11 10 10 53
$12.49
NA, Data not available. *Excluding electronic-only pages. †Excludes electronic-only content. Electronic content included a mean 0.7 e-pages per article consisting of a mean of 1.1 tables, 0.4 figures, and 0.2 illustrations. zIncludes an International Advisory Panel.
references, and pages of text to the nearest half-page. Electronic pages in volume 150 were counted separately. We noted the age of subjects (neonates alone, adolescents alone, or all other) and the primary topic addressed. Each article was classified as a research article, case report, or review article. We categorized the content of each research study as analytic or descriptive, disease- or non-disease-oriented, and people- or laboratory-centered. Study materials were classified as human or non-human. We determined whether statistical tests were used to assess for associations between variables. No attempt was made to quantify the significance of trends with statistical techniques. To test the reliability of the ratings assigned to articles in volume 150, we repeated the examination approximately 2 months after the initial review on a random sample of 10% of the articles.
Results General Trends In 1982, the number of articles per volume was the highest for the years studied, whereas the articles had the shortest mean length, and the numbers of tables, figures, and illustrations were low (Table I). By 2007, the number of articles per volume had declined, but the length of articles had increased. Each article in volume 150 included, on average, 0.7 pages of electronic-only content such as tables, figures, and appendices. Print circulation dropped substantially in 2007 compared
with 1982. The cost of a 1-year subscription has increased, but after adjusting for inflation,2 this increase is modest. Authorship and Format The number of authors per article has risen steadily in the last 75 years and increased >50% between volumes 99 and 150 (Table II). The proportion of articles in volume 150 with only 1 author remained low (3%), whereas most articles had $3 authors. Listed authors included large study groups or research networks in 10 articles in volume 150. The proportion of articles acknowledging at least 1 grant as a source of support continues to exceed one-half, but the region of origin of the first author has changed, with more than half the articles now originating from outside the United States. Topics of Articles The topics addressed by articles in The Journal have changed with time (Table III). For example, the proportion of articles dealing with infectious disease has steadily declined, whereas articles about metabolism and genetics have become more frequent. No articles in volume 150 were devoted to general pediatrics; every paper could be better categorized in another topic area. The proportion of articles addressing >1 topic continued to increase, from 24% in volume 99 to 52% in volume 150. Articles about nutrition accounted for approximately onetenth of articles in volume 1 and tended to concern malnutrition. In contrast, approximately the same proportion of articles in volume 150 concerned obesity (or over-nutrition). These articles were assigned a primary topic of endocrinology, because they focused on clinical and laboratory markers such as insulin resistance, lipid levels, and cardiovascular risk factors. 17
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Table III. Primary topic addressed in articles by volume number
Vol. 155, No. 1
Table IV. Characteristics of the research design by volume number
Percent of articles by volume number
Allergy/immunology Behavioral pediatrics Cardiology Endocrinology Gastroenterology General pediatrics Hematology/oncology Infectious disease Metabolism/genetics Neurology Nutrition Pharmacology Pulmonology Surgery Other
1
33
66
99
150
0 3 0 5 3 9 6 40 3 5 10 5 2 2 9
3 4 10 3 4 5 9 30 1 2 5 2 1 3 16
5 0 5 6 6 4 7 12 23 5 5 10 6 1 6
4 1 0.4 7 8 5 9 14 18 2 4 9 5 0.4 13
4 4 11 19* 4 0 5 5 20 3 0 0 6 0 20†
*Includes 9% of articles about obesity. †Includes 9% of articles about neonatology.
The proportions of papers devoted exclusively to the study of neonates remained stable since the earlier review at 27%. Approximately 9% of articles were focused on care of the sick or premature infant; because the earlier survey did not include ‘‘neonatology’’ as a topic, these articles are included in the ‘‘other’’ category in Table III. Research Design The number of case reports has decreased by approximately 50%, whereas the proportion of research articles has increased steadily (Table IV). The design of research studies has also changed, with a decrease in experimental designs and a corresponding increase in studies with an observational design. The proportion of articles classified as analytic more than doubled in the last 25 years, with a corresponding decrease in descriptive studies (Table IV). The proportion of articles that were disease-oriented and people-centered increased in the last 25 years. Articles involving exclusively human material have always comprised most articles since The Journal’s inception. The proportion of articles reporting statistical testing increased slightly from 76% in volume 99 to 86% in volume 150. Re-examination of a random sample of 10% of the reviewed articles showed the test-retest reliability to be generally excellent. Of the total 276 data points, 254 (92%) were concordant. Of the 22 disagreements, 9 (41%) represented errors in data collection or entry, and 13 (59%) represented discrepancies in interpretation. For example, a few articles included such an even mixture of clinical and laboratory data that it was unclear whether to code these articles as ‘‘people-centered’’ or ‘‘laboratory-centered.’’ Similarly, a few articles that addressed problems originating in the neonatal period included follow-up >30 days, so it was unclear whether to code the ages of the patients as ‘‘neonates only’’ or ‘‘all ages.’’ Taken in aggregate, however, the test-retest reliability was very good. 18
Percent of articles by volume number
Type of article Review Case reports Research Research study design Clinical trial Uncontrolled Controlled, randomized Controlled, non-randomized Observational cohort study Case-control study Cross-sectional study Research focus Analytic Descriptive Exposure allocation Experimental Observational Research orientation Disease Non-disease, state of wellness Focus of data collection People-centered Laboratory-centered Material Human Non-human
1
33
66
99
150
30 33 37
11 39 50
9 29 62
4 31 65
4 17 79
59 41 9 9 5 0 36
35 22 0 13 13 4 48
30 12 9 9 10 1 58
23 7 12 4 20 2 55
8 0 6 1 48 1 43
64 36
44 56
43 57
38 62
72 28
59 41
42 58
41 59
32 68
9 91
79 21
78 22
76 24
70 30
88 12
57 43
60 40
30 70
22 78
39 61
96 4
93 7
96 4
98 2
100 0
Discussion The observed changes in the format and content of articles published in The Journal may reflect changes in pediatrics, in medical publishing, and in The Journal’s editors and editorial policies in these three-quarters of a century. Editorial influence may have been particularly important because The Journal receives many more acceptable articles than can be published and has had only 4 editorial regimes since its inception (Table V; available at www.jpeds.com). The length of published papers, for example, was no doubt reduced by the editors’ liberal use of the ‘‘red pencil’’ to abbreviate and otherwise modify accepted papers.3,4 Near the end of Garfunkel’s tenure, he studied an entire volume to compare the length of accepted articles as they were published with that when they were submitted; only 1 paper was longer.4 The pages saved by editorial suggestions and stipulations allowed The Journal to accept 7 additional articles of average length per month. The number of articles that could be published also depended on the number of journal pages allocated by the publisher, which in turn depended on financial considerations, such as advertising. During the Nelson/Garfunkel era, advertising revenues increased exponentially until the recession of the 1980s, then plummeted (Joseph Garfunkel, personal communication 2008). An important aspect of advertising in The Journal not addressed in this review relates to the content of advertisements and their interspersion throughout the issue. Hellems, Gurka, and Hayden
July 2009 The Journal is now published in both print and electronic formats; the most recent volume reviewed includes content only available electronically. Most journals are now available electronically, and many have used the electronic format as a means of expanding the number of articles published in each issue.5 Many libraries no longer maintain print collections of recent volumes of many journals; readers must instead access the electronic version. Print circulation of The Journal in 2007 declined to 18% of what it was in 1982, but because many readers now access The Journal electronically, it is challenging to compare current readership with that of the past. Electronic publishing provides efficient access to international readers and might explain in part the noted increase in international authorship and the addition of an International Advisory Panel to the Editorial Board. The average number of authors per article and the proportion of manuscripts with multiple authors have continued to increase, a trend noted in many other journals.6 Published articles with multiple authors are cited more frequently, a marker which has been correlated with high research quality, implying that this trend represents an increase in productive collaboration rather than a manifestation of academic gamesmanship. The number of authors can also be influenced by editorial policy. As the number of authors grew, Garfunkel began to request justification when the number exceeded 5. With a challenge, it was common to have 1 to 3 names removed from the revised paper (Joseph Garfunkel, personal communication, 2008). The increase in the percentage of first authors from outside the United States noted in volume 150 has also been reported in other scientific fields.7 This trend may be caused by changes in research funding in the United States and abroad, improvement in the quality of research in other nations, changes in the perceived desirability of publishing research in the United States and abroad, and ease of electronic communication. The increase in listed grants may be caused in part by the requirement by many journals that all sources of support be identified so that conflicts of interest will be more obvious. Of the research articles in volume 150, only 8% were clinical trials, whereas most reported the results of observational cohort or cross-sectional studies. This trend may reflect the increasing accessibility of electronic databases and the computing and statistical power to analyze large amounts of data. Indeed, the complexity of the research design and data analysis in studies published in The Journal has increased since 1982, with fewer reviews, case reports, and descriptive studies. Most articles in volume 150 were analytic and used a varied array of complex statistical techniques, in agreement with parallel trends seen in other medical journals.8,9 Although the increased complexity of study design and analysis suggests increased scientific rigor, the articles may be less easily understood by clinician readers with little research experience or statistical training. One notable trend in The Journal has been the decrease in articles devoted to infectious disease and general pediatrics. Fewer articles about infectious disease may reflect the better A Review of The Journal of Pediatrics: The First 75 Years
ORIGINAL ARTICLES prevention and treatment of many infectious diseases through immunization and antimicrobial treatment. Trends in the research topics presented in The Journal may also reflect competition from other publications. Journals devoted to pediatric infectious disease, ambulatory pediatrics, and virtually every other pediatric subspecialty have been founded since the inception of The Journal. Do the changes noted in this survey of The Journal better serve its readership? According to a recent study,10 a typical American Academy of Pediatrics (AAP) member reads approximately 150 articles per year, spends only approximately 20 minutes per article, reads primarily for current clinical information, and usually reads print journals at home from personal subscriptions. Younger pediatricians and pediatricians doing research are more likely to use electronic journals. The low proportion of review articles, the absence of general pediatric articles, the complexity of some statistical analyses, and the shift to electronic publishing may make The Journal less attractive to practicing general pediatricians. Nelson previously has described The Journal’s creation in 1932 as the official organ of the AAP and the schism between the publisher and the AAP that led to the establishment of Pediatrics in 1948.11 In 2001, The Journal became affiliated with the Association of Medical School Pediatric Department Chairs and has published a series of related reports, policy discussions, and commentary in a special section of The Journal. This collaboration builds on The Journal’s longstanding leadership role in pediatric training and workforce issues. It is challenging to predict what the observed trends in The Journal in the last 75 years portend for the future. If recent trends are any indication, however, readers can expect more research articles with correspondingly fewer reviews and case reports, more authors per article and an increase in the number of international authors, and fewer experimental designs, including clinical trials, with correspondingly more observational designs, especially observational cohorts. The print circulation may decline, but The Journal’s electronic presence will likely increase and provide greater readership. The Journal’s impact factor is very high (4.017 in 2007), suggesting great importance to the field of pediatrics. If The Journal’s editors continue to monitor and respond to the needs and desires of its readers, there is every reason to expect that The Journal will continue to be an important source of information for pediatricians in the years ahead. n Submitted for publication April 18, 2008; last revision received July 31, 2008; accepted Aug 28, 2008. Reprint requests: Martha A. Hellems, MD, Department of Pediatrics, University of Virginia School of Medicine, PO Box 800386, Charlottesville, VA 22908-0386. E-mail: [email protected].
References 1. Hayden GF, Saulsbury FT. A review of The Journal of Pediatrics: the first 50 years. J Pediatr 1982;101:5-11. 19
THE JOURNAL OF PEDIATRICS
www.jpeds.com
2. Calculated from the U.S Department of Labor Bureau of Labor Statistics Consumer Price Index (updated 1/16/2008), Available at: http://www. bls.gov/cpi/#data. Accessed Feb 8, 2008. 3. Balistreri WF. Building on a tradition. J Pediatr 1996;128:97-8. 4. Garfunkel JM. Reflections of an editor on his retirement. J Pediatr 1995; 127:947. 5. Anderson K, Lucey JF. Pediatrics electronic pages: looking back and looking ahead. Pediatrics 1998;102:124-8. 6. Wuchty S, Jones BF, Uzzi B. The increasing dominance of teams in production of knowledge. Science 2007;316:1036-9.
Vol. 155, No. 1 7. Rahman M, Fukui T. A decline in the U.S. share of research articles. N Engl J Med 2002;347:1211-2. 8. Horton NJ, Switzer SS. Statistical methods in the journal. N Engl J Med 2005;353:1977-9. 9. Hellems MA, Gurka MJ, Hayden GF. Statistical literacy for readers of Pediatrics: a moving target. Pediatrics 2007;119:1083-8. 10. Tenopir C, King DW, Clarke MT, Na K, Zhou X. Journal reading patterns and preferences of pediatricians. J Med Libr Assoc 2007;95:56-63. 11. Nelson WE. The first 50 years of The Journal of Pediatrics. J Pediatr 1982; 101:1-4.
50 Years Ago in THE JOURNAL OF PEDIATRICS Hyperbilirubinemia in Premature Infants: A Follow-up Study Koch CA, Jones DV, Dine MS, Wagner EA. J Pediatr 1959;55:23-9
I
n this interesting article, the authors describe the outcomes of preterm infants with varying degrees of hyperbilirubinemia. Serial serum bilirubin concentrations were measured in 100 consecutive preterm infants. Each infant’s condition at birth was rated clinically, and the postnatal course was documented. The infants were invited back for clinical assessment at age 2 to 3 years. Of the 100 infants studied, 8 died before discharge and 42 returned for assessment. Seven infants were assessed after discharge at clinics where charts were accessible and were also included in the study. Of the 8 infants who died, 2 had kernicterus diagnosed at autopsy. Of the 49 infants assessed after discharge, 5 had clinical features suggestive of kernicterus. All of the infants with kernicterus had bilirubin levels above 20 mg/dL. Three of the 4 infants with a bilirubin concentration of 30 to 40 mg/dL exhibited evidence of kernicterus. Importantly, none of those infants subsequently diagnosed with kernicterus exhibited clinically specific features before discharge. The authors concluded that exchange transfusion is indicated in all infants who have or are expected to develop a serum bilirubin concentration > 20 mg/dL. This article provides a glimpse into just how different a preterm infant ’s journey was 50 years ago. There are many important differences between then and now. Instead of using an ad hoc clinical score, neonatologists have now universally adopted the Apgar score. The Bloxsom airlock, considered an innovation in neonatal management 50 years ago, is now only found in museums. Back then, conjugated and unconjugated bilirubin could not be easily differentiated, whereas now they can be routinely measured in the modern chemical pathology laboratory as part of the clinical service. With the advent of phototherapy in the 1960s, the use of exchange transfusions to treat hyperbilirubinemia has decreased dramatically. Therefore, the conclusions of Koch et al are no longer applicable in the modern era. But despite these dramatic advances, many questions remain; for example, factors predisposing the human infant brain to the development of kernicterus are still poorly understood. Given the relatively safe option of phototherapy to effectively treat hyperbilirubinemia and the potentially disastrous consequences of kernicterus, ethical constraints make it unlikely that these questions will ever be answered completely. Hugh S. Lam, MRCPCH Department of Pediatrics The Chinese University of Hong Kong New Territories, Hong Kong 10.1016/j.jpeds.2009.01.053
20
Hellems, Gurka, and Hayden
ORIGINAL ARTICLES
July 2009
Table V. Editors of The Journal of Pediatrics Years
Editor(s)
1932-1948 1948-1959 1959-1977 1977-1995 1996-present
Borden S. Veeder and Hugh McCulloch Borden S. Veeder Waldo E. Nelson Joseph M. Garfunkel William F. Balistreri
A Review of The Journal of Pediatrics: The First 75 Years
20.e1