What do evidence-based secondary journals tell us about the publication of clinically important articles in primary healthcare journals?

1741-7015-2-33 1741-7015 Research article What do evidence-based secondary journals tell us about the publication of clinically important articles in primary healthcare journals? McKibbon Ann Kathleen mckib@mcmaster.ca Wilczynski L Nancy wilczyn@mcmaster.ca Haynes Brian Robert bhaynes@mcmaster.ca

Health Information Research Unit, Department of Clinical Epidemiology and Biostatistics, McMaster University Faculty of Health Sciences, Room 3H6 Health Sciences Center, 1200 Main Street West, Hamilton, Ontario, Canada. L8N 3Z5

Center for Biomedical Informatics, School of Medicine, 8084 Forbes Tower, 200 Lothrop Street, University of Pittsburgh, Pittsburgh, PA, USA. 15213-2582

Department of Medicine, McMaster University Faculty of Health Sciences, Health Sciences Center, 1200 Main Street West, Hamilton, Ontario, Canada. L8N 3Z5

BMC Medicine 1741-7015 2004 2 1 33 http://www.biomedcentral.com/1741-7015/2/33 15350200 10.1186/1741-7015-2-33 27 1 2004 06 9 2004 06 9 2004 2004 McKibbon et al; licensee BioMed Central Ltd. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Background

We conducted this analysis to determine i) which journals publish high-quality, clinically relevant studies in internal medicine, general/family practice, general practice nursing, and mental health; and ii) the proportion of clinically relevant articles in each journal.

Methods

We performed an analytic survey of a hand search of 170 general medicine, general healthcare, and specialty journals for 2000. Research staff assessed individual articles by using explicit criteria for scientific merit for healthcare application. Practitioners assessed the clinical importance of these articles. Outcome measures were the number of high-quality, clinically relevant studies published in the 170 journal titles and how many of these were published in each of four discipline-specific, secondary "evidence-based" journals (ACP Journal Club for internal medicine and its subspecialties; Evidence-Based Medicine for general/family practice; Evidence-Based Nursing for general practice nursing; and Evidence-Based Mental Health for all aspects of mental health). Original studies and review articles were classified for purpose: therapy and prevention, screening and diagnosis, prognosis, etiology and harm, economics and cost, clinical prediction guides, and qualitative studies.

Results

We evaluated 60,352 articles from 170 journal titles. The pass criteria of high-quality methods and clinically relevant material were met by 3059 original articles and 1073 review articles. For ACP Journal Club (internal medicine), four titles supplied 56.5% of the articles and 27 titles supplied the other 43.5%. For Evidence-Based Medicine (general/family practice), five titles supplied 50.7% of the articles and 40 titles supplied the remaining 49.3%. For Evidence-Based Nursing (general practice nursing), seven titles supplied 51.0% of the articles and 34 additional titles supplied 49.0%. For Evidence-Based Mental Health (mental health), nine titles supplied 53.2% of the articles and 34 additional titles supplied 46.8%. For the disciplines of internal medicine, general/family practice, and mental health (but not general practice nursing), the number of clinically important articles was correlated withScience Citation Index (SCI) Impact Factors.

Conclusions

Although many clinical journals publish high-quality, clinically relevant and important original studies and systematic reviews, the articles for each discipline studied were concentrated in a small subset of journals. This subset varied according to healthcare discipline; however, many of the important articles for all disciplines in this study were published in broad-based healthcare journals rather than subspecialty or discipline-specific journals.

Background

Evidence on the journal-reading habits of clinicians comes from three separate groups of publications. First, several surveys have been used to ascertain the reading habits of physicians. Fafard and Snell 1 assessed house staff who reported reading an average of 8.7 hours per week, with about half of their time spent reading for specific patient situations. Reading time for family practice residents was more than three hours per week 23 and ranged from 1–12 hours. Dermatology residents averaged 4.2 hours reading per week and read an average of seven journals, four of which were peer reviewed 4. Internists read an average of 4.4 hours per week 3, while surgeons reported an average reading time of 3.5 hours across 3–16 journals 5. This average of three-four hours of reading time per week is quite consistent across disciplines, level of education, time, and nationality.

A second set of surveys and studies center on general information-seeking behaviors of clinicians. These studies show how journal reading fits in with the other types of information that clinicians use. Two systematic reviews have been done recently.

Researchers at the Australian National Institute of Clinical Studies 6 reviewed preferred information sources in many clinician groups, including physicians (primary care/general practice/family practice, hospitalists, rural physicians, diabetologists); nurses (hospital and occupational health nurses); physical therapists; dental hygienists; and policy-makers. They reviewed 34 studies and concluded that all groups used multiple information resources, with information needs answered most often by other people, followed by books and journals. Dawes and Sampson 7 evaluated 19 studies of physician information-seeking behavior. They placed books and journals in one category (print resources) and found this to be the most used information source, with colleagues being the second.

The third source of information on clinicians' use of information resources comes from marketing studies. The Association of Medical Publications 8 monitors physician use of printed journals and other information resources. Despite the rapid expansion of the Internet and all the information it contains, physicians continue to read and value journal articles, and their reliance on journals may be increasing. Data collected in 1983 and 1998 shows that physician reliance on journal literature as their main source of medical information increased from 61.8% to 76.3%, an absolute increase of 14.5% in 15 years.

The importance of reading journal articles for clinical care is evident. The increasing number of journals from which important and relevant articles are found, combined with the decreasing number of personal subscriptions 9, makes it more important than ever for physicians to choose carefully which journals to subscribe to and read. This decision should not be based on intuition alone, as shown in an important study by obstetricians and gynecologists. Weiner et al. 10 sought to determine which journals had published numerical data on the relation between oral contraceptive use and cancer–information they judged to be clinically important and readily available in their subspecialty journals. Assessing 3735 articles identified by MEDLINE searches, only 27 studies reported numerical data, of which 23 were published in mainstream general medical journals. Only four were published in obstetrics and gynecology journals.

Since the publication of the study by Weiner et al. 10, several groups have tried to determine targeted journal subsets that could provide the most important clinical information to physicians in different specialties. Birken and Parkin 11 assessed journals with pediatric content. Using data from pediatric-related systematic reviews in the Cochrane Database for Systematic Reviews for 1997, as well as policy statements from the American Academy of Pediatricians and the Canadian Paediatric Society, they determined that four general medical journals and three pediatric specialty journals provided access to most of the important advances: Archives of Diseases in Childhood, BMJ, JAMA, Journal of Pediatrics, Lancet, New England Journal of Medicine, and Pediatrics. Their results validate the findings of Weiner et al.–important studies in a discipline or specialty are often not published in specialty journals.

Gehanno and Thirion 12 used MEDLINE searches and the Science Citation Index (SCI) Impact Factors to identify journal subsets in occupational health. Eight journals provided coverage of 27% their discipline content; 38 journals increased this to 52%. Coverage needed to be expanded beyond their specialty journals for them to remain current in occupational health.

Lee et al. 13 sampled research articles from 30 randomly selected journals from a list of 107 general internal medicine journals defined by SCI. They found that journals with high citation rates, SCI Impact Factors, and circulation rates; low manuscript acceptance rates; and listing on the Brandon/Hill Library List 14 were predictive of higher article methodologic scores.

Ebell et al. 15 as well as our research group 16 present an alternative approach for clinicians to keep up to date with current literature. Both groups produce summaries of important advances in areas of clinical care so that individuals do not have to read primary journals and evaluate reports. Ebell et al. provided results of a hand search of 85 core journals of interest to family/general practice. Physicians read these journals for six months and identified articles that were considered to be POEMs (patient-oriented evidence that matters). A POEM addresses a clinical question encountered by a family physician at least once every two weeks, measures patient-oriented outcomes, and presents results that will likely affect practice. The report provides summaries of which journals publish important clinical advances for general/family practice.

In this article we report on a survey of the contents of 170 core clinical journals for the publishing year 2000 to assess which journals publish the highest number of methodologically sound and clinically relevant studies in the disciplines of internal medicine, general/family practice, general practice nursing, and mental health. In the "Methods" section we describe our two-step article selection process for clinical importance and methodologic rigor, which is very similar to that used by Ebell et al. 15. The data we provide reflects the merit of individual journal titles from a clinical perspective; it may help clinicians to choose which journals to read, and health sciences libraries to include them in their collections.

Methods

The Health Information Research Unit of the Department of Clinical Epidemiology and Biostatistics at McMaster University, Ontario, Canada, publishes several secondary "evidence-based" journals, systematically selecting, summarizing and appraising articles in a broad range of primary clinical journals. In 2000 we prepared ACP Journal Club (ACP J Club) to support internal medicine , Evidence-Based Medicine (EBM) to support general/family practice, Evidence-Based Nursing (EBN) to support general care nursing, and Evidence-Based Mental Health (EBMH) to support mental health. To identify potential candidate articles for inclusion in these journals, six Masters-level trained staff read each article in the major general healthcare journals and those in the disciplines and subdisciplines related to the content of each abstract journal. The list of these journals (see Additional File 1) is comprised of titles suggested by librarians, clinicians, editors, and editorial staff; SCI Impact Factors; and systematic examination of the contents of each title for at least six months. More than 400 journal titles have been assessed since the abstract journals were started in 1991.

Additional File 1

Additional File 1 includes a list of the 170 journals read for 2000 along with the number of articles reviewed, the number and percentage that passed criteria, and the NNR (number of articles that are needed to be read to obtain one that is clinically relevant and has high-quality methods). The file name is "Publishing Important Articles Appendix.doc" and it is in Word 2000 format.

Click here for file

We consider the Cochrane Database of Systematic Reviews to be a separate journal that publishes systematic reviews of the literature on a quarterly basis. This is consistent with the U.S. National Library of Medicine's decision to index the Cochrane Database of Systematic Reviews as a separate journal. We evaluate only the new reviews and those that are substantially updated each quarter. We do not consider the rest of the database or protocols that describe reviews that are in progress or being planned.

Original and review articles are placed in one or more of seven categories of study type–therapy and prevention, screening and diagnosis, prognosis, etiology and harm, economics and cost, clinical prediction guides, and qualitative studies 16. All categories have a set of pass/fail rules for selection (see: http://www.acpjc.org/shared/purpose_and_procedure.htm), except for qualitative and cost studies. Basic inclusion criteria are that the articles i) are about the healthcare of humans; ii) have at least one clinically important outcome; and iii) use appropriate statistical analyses. As an example of category-specific criteria, an article on screening or diagnosis must meet these additional criteria:

• a spectrum of participants were included, some with the disease or condition of interest and some without

• objective diagnoses were made using the "gold" standard or current clinical standard for diagnosis of the disease or condition

• participants received both the new test and some form of the diagnostic standard

• the diagnostic standard was interpreted without knowledge of the test result and vice versa.

The basic inclusion criteria are based on study design and methodology principles for evidence-based healthcare. Their use identifies studies that have data related to patients or those at risk of disease, diseases and conditions, and real-life clinical settings. Therefore, a study or review article that meets the criteria can be considered to be appropriate for possible use in patient care decision-making. The article readers are trained and retested annually so that they can reliably apply these selection rules for inclusion in our evidence-based journals (kappa measuring chance-adjusted agreement > 80% for all categories) 16.

With a research grant from the U.S. National Library of Medicine we intensified our data collection from the reading process related to the evidence-based journals for the publishing year 2000. All articles in 170 journals were classified as to whether they were "of interest" to the healthcare of humans and, if so, whether they reported original data or were systematic review articles. These original studies and reviews were classified into all possible categories (where more than one category could apply, for example, a therapy article that included economic data), and were then given a pass or fail methodologic designation for each category.

Articles passing methodologic criteria were assessed further for clinical interest by an editorial group of practicing clinicians for each abstract journal. These clinicians have expertise in methodology and specific areas of healthcare such as gastroenterology or neonatology nursing. At this point the clinician raters excluded all studies with preliminary results, interventions that were not readily available or proven useful, already known and applied findings, and topics addressing rare conditions or diseases. After review, often by a team of three-five clinicians (see http://hiru.mcmaster.ca/more/RatingFormSample.htm for a copy of the rating system that was used in paper format for this study) some articles were further processed. The editors chose articles to be abstracted that they considered to have the most important message for clinicians. The remaining pass articles were listed as "Other Articles Noted" if their content was of relevance to the disciplines covered by the abstract journals.

This dual selection process (methodologic rigor and clinical importance) provided insight into which journals yielded the highest numbers of pass articles. This had major implications for clinical practice at two levels of clinical relevance. The more stringent level includes articles that were summarized in each abstract journal. The second, less stringent level includes articles that are abstracted as well as those articles that are listed in the Other Articles Noted sections. Analysis was done by abstract journal title (ACP J Club, EBM, EBN, and EBMH) to ascertain which journal titles were most important to their target clinical audience (internal medicine, general/family practice, general practice nursing, and mental health, respectively). SCI Impact Factors were collected for each journal title for each discipline. If an SCI Impact Factor was not available we sought Social Science Index Impact Factors. These data were analyzed to determine if Impact Factors were related to yield of clinically important advances, as found by Lee et al. 13 and Gehanno and Thiron 12.

Results

For 2000, the 170 core journals we selected published 60,352 articles. The total number of pass articles was 3059 for original studies and 1073 for reviews. An article could be counted more than once if it passed for multiple categories. Six journals did not publish any pass articles. The complete list of journals and their yield appears in the Additional File 1.

The category breakdown of pass articles for original studies and review articles, respectively, was 1639 and 662 for therapy and prevention, 152 and 47 for screening and diagnosis, 195 and 22 for prognosis, 290 and 308 for etiology and harm, 35 and 10 for economics, 358 and 8 for qualitative studies, and 93 and 4 for clinical prediction guides.

The top 20 journals for yield of pass articles are included in Table 1. The titles varied considerably in both the total number and proportion of clinically relevant articles that they published. For example, 95.0% of the articles (all reviews) in the Cochrane Database of Systematic Reviews passed our criteria, while only 2.8% of the articles in the American Journal of Gastroenterology met standards for clinically applicable studies. (The American Journal of Gastroenterology is a specialty journal and a substantial proportion of its content is preclinical. These preclinical articles, by definition, did not meet the clinical criteria in this study.) Generally, a clinical reader would need to read in the range of 13–14 articles from these top 20 journals to obtain one that is directly clinically important in any healthcare area, although the range is substantial (1.1 to 36.9). We call this number the "number of articles needed to be read" or NNR.

Table 1

Number of high-quality, clinically relevant articles in the top 20 clinical journals for 2000

Journal title

Number of articles with abstracts (from MEDLINE)

Number of articles evaluated

Number of pass articles / number evaluated (% pass)

NNR* for number evaluated

SCI Impact Factor** for 2000

Cochrane Database of Systematic Reviews

1004

444

422 (95.0)

1.1

Not available

Lancet

669

3858

134 (3.5)

28.8

10.2

Journal of Clinical Oncology

445

650

100 (15.4)

6.6

8.8

BMJ

209

3428

93 (2.7)

36.9

5.3

Circulation

925

1351

92 (6.8)

14.7

10.9

Journal of Advanced Nursing

341

611

92 (15.1)

6.6

0.77

Obstetrics and Gynecology

389

478

88 (18.4)

5.4

2.0

JAMA

329

1930

87 (4.5)

22.2

16.4

New England Journal of Medicine

228

1530

83 (5.4)

18.4

29.5

Archives of Internal Medicine

340

620

81 (13.1)

7.7

6.1

Journal of the American College of Cardiology

514

707

76 (10.7)

9.3

7.1

Pediatrics

548

811

76 (9.4)

10.7

4.8

American Journal of Cardiology

631

850

72 (8.5)

11.6

2.7

American Journal of Obstetrics and Gynecology

539

704

72 (10.2)

9.8

2.5

Critical Care Medicine

340

977

70 (7.2)

14.0

3.8

Chest

589

882

66 (7.5)

13.4

2.5

Stroke

402

609

59 (9.7)

10.3

6.0

Neurology

814

1334

58 (4.3)

23.0

4.8

American Journal of Gastroenterology

474

923

56 (2.8)

16.6

Diabetes Care

263

529

55 (10.4)

9.6

5.0

Average

7.3%

13.8

*The NNR is the number of articles that would have to be read in each journal to identify one with high quality methods that is clinically relevant; **The SCI Impact Factor is the Science Citation Index Impact Factor (rating of how important each journal is in relation to citations). Data are for 2000. Articles have not been screened for direct clinical relevance beyond basic criteria of having at least one clinically important outcome.

The number of pass articles did not correlate with SCI Impact Factors for the top 20 journals (correlation coefficient 0.29, P = 0.24). Analysis of the top 50 titles showed a weak correlation (correlation coefficient 0.41, P = 0.004) for the same analysis.

The breakdown by discipline was done using the total number of articles that were selected for inclusion in each of the four abstract journals–internal medicine, general/family practice, general care nursing, and mental health (Tables 2,3,4,5). Both the total number of articles abstracted and the total number of articles in each journal (abstracted and "Other Articles Noted") are included in Tables 2, 4, and 5 giving a two-level assessment of "clinical worth". EBM does not publish an "Other Articles Noted" Section.

Table 2

ACP J Club (internal medicine) journal-specific content of high-quality, clinically relevant articles

Journal title

Number articles reviewed in 2000

Number abstracted (% included in ACP J Club)

NNR for abstracted article

Number abstracted or listed (% included in ACP J Club)*

NNR for abstracted or listed**

New England Journal of Medicine

1530

25 (16.9)

61.2

67 (6.7)

22.8

JAMA

1930

25 (16.9)

77.2

53 (5.4)

36.4

Lancet

3858

22 (14.9)

175.4

62 (6.6)

62.2

Cochrane Database Systematic Reviews***

444

11 (7.4)

40.4

130 (13.1)

3.4

Annals of Internal Medicine

602

8 (5.4)

75.3

33 (3.3)

18.2

Archives of Internal Medicine

620

6 (4.1)

103.3

57 (5.8)

24.8

BMJ

3428

5 (3.4)

685.6

50 (5.1)

68.6

Circulation

1351

5 (3.4)

270.2

33 (3.3)

40.9

AHRQ/AHCPR Reports***

N.A.

4 (2.7)

N.A.

9 (1.0)

N.A.

American Journal of Gastroenterology

923

4 (2.7)

225.6

21 (2.1)

44.0

American Journal of Medicine

435

3 (2.0)

130.5

21 (2.1)

20.7

CMAJ (formerly Canadian Medical Association Journal)

1007

3 (2.0)

335.7

12 (1.2)

83.9

Diabetic Medicine

188

3 (2.0)

62.6

13 (1.3)

14.5

Thorax

336

3 (2.0)

112.0

8 (0.8)

42.0

Annals of Emergency Medicine

294

2 (1.4)

147.0

11 (1.1)

26.7

Journal of the American Geriatrics Society

384

2 (1.4)

192.0

18 (1.8)

21.3

Journal of Vascular Surgery

544

2 (1.4)

272.0

6 (0.6)

9.0

American Journal of Cardiology

850

1 (0.7)

850.0

14 (1.4)

60.7

Archives of Neurology

313

1 (0.7)

313

2 (0.2)

156.5

British Journal of Surgery

402

1 (0.7)

402

6 (0.6)

67.0

CCOHTA Reports****

N.A.

1 (0.7)

N.A.

1 (0.1)

N.A.

Critical Care Medicine

977

1 (0.7)

977

24 (2.4)

40.7

Diabetes Care

529

1 (0.7)

529

14 (1.4)

37.8

Gastroenterology

543

1 (0.7)

543

5 (0.5)

108.5

Gut

446

1 (0.7)

446

6 (0.6)

74.3

Health Psychology

1 (0.7)

3 (0.3)

26.6

Journal of Clinical Psychopharmacology

162

1 (0.7)

162

1 (0.1)

81.0

Journal of Family Practice

263

1 (0.7)

263

10 (1.0)

26.3

Journal of Infectious Diseases

760

1 (0.7)

760

6 (0.6)

126.7

Neurology

1334

1 (0.7)

1334

12 (1.2)

111.2

Spine

604

1 (0.7)

604

14 (1.4)

43.1

Stroke

609

1 (0.7)

609

26 (2.6)

101.5

Acta Obstetrica et Gynecologica Scandinavica

242

0 (0.0)

1 (0.1)

242.0

Addiction

295

0 (0.0)

Infinity

4 (0.4)

63.8

Age and Ageing

767

0 (0.0)

Infinity

6 (0.6)

127.8

Alimentary Pharmacology and Therapeutics

N.A.

0 (0.0)

N.A.

4 (0.4)

N.A.

American Journal of Epidemiology

362

0 (0.0)

Infinity

18 (1.8)

20.1

American Journal of Obstetrics and Gynecology

704

0 (0.0)

Infinity

2 (0.2)

352.20

American Journal of Psychiatry

508

0 (0.0)

Infinity

1 (0.1)

508

American Journal of Public Health

363

0 (0.0)

Infinity

5 (0.5)

72.6

American Journal of Respiratory and Critical Care Medicine

783

0 (0.0)

Infinity

9 (0.9)

87.0

Annals of Rheumatic Diseases

266

0 (0.0)

Infinity

1 (0.1)

266.0

Annals of Surgery

301

0 (0.0)

Infinity

3 (0.3)

100.3

Archives of Family Medicine (no longer published)

230

0 (0.0)

Infinity

6 (0.6)

38.3

Archives of General Psychiatry

161

0 (0.0)

Infinity

2 (0.2)

80.5

Archives of Physical Medicine and Rehabilitation

337

0 (0.0)

Infinity

4 (0.4)

84.3

Archives of Surgery

330

0 (0.0)

Infinity

3 (0.3)

103.3

Australian and New Zealand Journal of Psychiatry

214

0 (0.0)

Infinity

1 (0.1)

214.0

British Journal of General Practice

453

0 (0.0)

Infinity

3 (0.3)

151

British Journal of Psychiatry

402

0 (0.0)

Infinity

2 (0.2)

201

Canadian Journal of Gastroenterology

145

0 (0.0)

Infinity

3 (0.3)

48.3

Canadian Journal of Infection Control

0 (0.0)

Infinity

1 (0.1)

31.0

Canadian Respiratory Journal

0 (0.0)

Infinity

2 (0.2)

34.0

Cancer

786

0 (0.0)

Infinity

3 (0.3)

262.0

Chest

882

0 (0.0)

Infinity

21 (2.1)

42.0

Heart

450

0 (0.0)

Infinity

6 (0.6)

75.0

Heart and Lung

0 (0.0)

Infinity

1 (0.1)

59.0

Hypertension

419

0 (0.0)

Infinity

6 (0.6)

69.8

International Journal of Geriatric Psychiatry

169

0 (0.0)

Infinity

6 (0.6)

28.2

Journal of Affective Disorders

154

0 (0.0)

Infinity

3 (0.3)

51.3

Journal of the American Board of Family Practice

121

0 (0.0)

Infinity

1 (0.1)

121.0

Journal of the American College of Cardiology

707

0 (0.0)

Infinity

29 (2.9)

24.4

Journal of Bone and Joint Surgery (US)

360

0 (0.0)

Infinity

1 (0.1)

360.0

Journal of Clinical Epidemiology

173

0 (0.0)

Infinity

8 (0.8)

21.6

Journal of Epidemiology and Community Health

205

0 (0.0)

Infinity

6 (0.6)

34.2

Journal of General Internal Medicine

155

0 (0.0)

Infinity

9 (0.9)

17.2

Journal of Internal Medicine

177

0 (0.0)

Infinity

4 (0.4)

44.3

Journal of Neurology and Neurosurgery and Psychiatry

478

0 (0.0)

Infinity

7 (0.7)

68.3

Journal of Psychosomatic Research

118

0 (0.0)

Infinity

6 (0.6)

19.7

Journal of Rheumatology

657

0 (0.0)

Infinity

4 (0.4)

164.3

Journal of Trauma Injury Infection and Critical Care

562

0 (0.0)

Infinity

2 (0.2)

281.0

Medical Care

162

0 (0.0)

Infinity

7 (0.7)

32.1

Medical Journal of Australia

598

0 (0.0)

Infinity

1 (0.1)

598.0

Pain

269

0 (0.0)

Infinity

7 (0.7)

38.4

Patient Education and Counseling

0 (0.0)

Infinity

1 (0.1)

13.3

Pediatrics

811

0 (0.0)

Infinity

2 (0.2)

405.5

Psychology and Aging

0 (0.0)

Infinity

1 (0.1)

55.0

Psychological Medicine

142

0 (0.0)

Infinity

4 (0.4)

35.3

Psychosomatic Medicine

106

0 (0.0)

Infinity

1 (0.1)

106.0

Qualitative Health Research

0 (0.0)

Infinity

1 (0.1)

60.0

Radiology

654

0 (0.0)

Infinity

1 (0.1)

654.0

Rheumatology

339

0 (0.0)

Infinity

9 (0.9)

37.7

Social Science and Medicine

302

0 (0.0)

Infinity

2 (0.2)

151.0

Western Journal of Nursing Research

0 (0.0)

Infinity

1 (0.1)

99.0

TOTAL

148 (100)

990 (100)

Correlation with SCI Impact Factors-correlation coefficient (P-value)

N.A.

0.788 (< 0. 001)

N.A.

0.688 (< 0.001)

N.A.

*Articles abstracted are those that pass methodological criteria and are deemed to be the most important by practicing internists. The other articles listed are articles with the same high-quality methods but are considered to be slightly less important clinically by practicing clinicians; **The number of articles needed to read (NNR) is a measure of the ratio of number of relevant articles (abstracted or combined abstracted or listed) divided into the total number of articles for each journal title; ***SCI Impact Factors not available for analysis; ****Canada Coordinating Office for Health Technology Assessment Reports; N.A. Not applicable. Data are for 2000. Note that the AHRQ/AHCPR and CCOHTA reports were not considered journal titles and read as such for this report.

Table 3

EBM (general/family practice) journal-specific content of high-quality, clinically relevant articles^†

Journal title

Number articles in 2000 in journal

Number abstracted (% included in EBM)

NNR for abstracted article

Number abstracted or listed (% included in EBM)*

NNR for abstracted or listed**

JAMA

1930

18 (12.5)

107.2

N.A.

BMJ

3428

17 (11.8)

201.6

N.A.

Lancet

3858

17 (11.8)

226.9

N.A.

New England Journal of Medicine

1530

13 (9.0)

117.6

N.A.

Cochrane Database of Systematic Reviews***

444

8 (5.6)

55.0

N.A.

Annals of Internal Medicine

602

7 (4.9)

86.1

N.A.

AHRQ/AHCPR Reports***

6 (4.2)

N.A.

Archives of Disease in Childhood

392

4 (2.8)

98.0

N.A.

American Journal of Medicine

435

4 (2.8)

108.8

N.A.

Archives of Family Medicine (no longer published)

230

3 (2.1)

76.7

N.A.

Journal of Family Practice

263

3 (2.1)

87.7

N.A.