Regular Articles
Characteristics ofjournal Clubs in Psychiatric Training Joel Yager, M.D. Lawrence S. linn, Ph.D. Daniel K. Winstead, M.D. Barbara Leake, Ph.D. We surveyed the membership of the American Association of Directors of Psychiatric Residency Training (AADPRT) regarding the presence and characteristics of journal clubs in their general and child psychiatry residency training programs. Responses were obtained from 141 general residency programs and 76 child psychiatry programs. Eightysix percent (N=180) offered at least one required and/or voluntary journal club, and many offered multiple journal clubs. Higher effectiveness ratings were associated with journal clubs that had mandatory participation, met frequently, were held in a convenient location, reviewed articles on original research, emphasized and taught research methods, and had regular faculty participation. Higher attendance ratings were associated with daytime meetings, smaller residency programs, required clubs, clubs that met more often and for shorter lengths of time within the hospital, and clubs that show continuity in faculty participation. Descriptions of various formats and inducements reported may be useful to other programs wishing to establish or improve journal clubs.
J
ournal clubs are traditionally part of many clinical training programs in a variety of medical specialties (1~). Studies in internal medicine and family medicine have re-
Dr. Yager is Professor and Director of Residency Education at the Department of Psychiatry and Biobehavioral Sciences, Neuropsychiatric Institute, University of California at Los Angeles; and Associate Chief of Staff (Residency Education) at the West Los Angeles Veterans Administration Medical Center (Brentwood Division); Dr. Linn was Research Professor at the Department of Medicine and School of Public Health, University of California at Los Angeles (Dr. Linn died in July 1990); Dr. Winstead is Professor and Chair at the Department of Psychiatry and Behavioral Sciences, Tulane University; Dr. Leake is Senior Statistician, Department of Medicine and Neurology, UCLA. Address correspondence to Joel Yager, M.D., UCLA Neuropsychiatric Institute, 760 Westwood Plaza, Los Angeles, CA 90024. Copyright © 1991 Academic Psychiatry. ,
vealed that journal clubs are quite popular and that they serve a variety of functions. This format is also being increasingly employed in continuing medical education, with groups of practitioners coming together for ongoing study. For example, the Jour-
nal of the American Medical Association
(JAMA) has established programs whereby journal club attenders can obtain category I credit for their participation (5). With regard to psychiatric training, we were impressed by the wide variety of journal club types described to us over the years by colleagues in training programs around the country. We wondered how commonly journal clubs were used in psychiatric training programs, what their characteristics were, and what practices might exist to render some journal clubs particularly effective or ineffective. \ \ 1\ 1 \ II '
. " I \ II >I I{ , •
~ 1'1, I" (
""
METHODS
by training directors; the others were completed by other faculty (10%) or residents (11 %). University-affiliated programs provided 82% of the responses, whereas private hospital programs (4%), state hospital programs (5%), and other types of non-university-affiliated programs (9%) provided the rest. Available data do not permit us to estimate whether the proportion of university programs responding are representative of the general AADPRT membership or whether they constitute a sample biased toward university programs.
We surveyed the membership of the American Association of Directors of Psychiatric Residency Training (AADPRT), a national organization representing more than 90% of the training directors of general and child psychiatry residency programs. The survey asked questions regarding the presence and characteristics of journal clubs in the members' residency programs. Questionnaires were mailed to all members of the AADPRT, and after six weeks a second mailing was sent to all initial non-reMeasures of Success sponders. Responses were obtained from 141 general psychiatry residency programs (representing 67% of all 204 American gen- Dependent variables. Two measures were eral psychiatry residency programs listed in used to define successful journal clubs. The the most recent edition of the National Resi- first was based on respondents' ratings of dent Matching Plan book plus four Cana- effectiveness of their club using the followdian programs) and from 76 child and ing criteria: 1) helps residents keep up with adolescent psychiatry training programs the literature; 2) teaches residents research (representing 58% of American child psychi- methods and statistics; 3) trains residents to atry programs plus one Canadian program). make presentations; and 4) helps residents to Seven respondents answered for both the think critically about articles. Achievement general and child programs in a combined 'on each criterion was rated as Excellent-4, fashion. In the event that duplicate re- Good-3, Fair-2, or Poor-I. Responses to the sponses were received from a given pro- four items were summed, and a reliable gram, only one response per program was Likert scale was constructed (Cronbach's recorded-the response provided by the re- alpha reliability coefficient =0.73). The second measure of success was the spondent most closely involved with journal respondents' estimate of the percentage of clubs. In our introductory letter, we asked that their target audience that actually attended the questionnaires be completed by the their journal club. training director or by another person knowledgeable about the journal club at Independent variables. We investigated the retheir institution. In the event that a program lationship between the two measures of sucoffered more than one journal club, we asked cessful journal clubs and three categories of that the respondent complete the survey independent variables: structure, process, with respect to the best-attended journal and content. Structural variables included club. To supplement the formal question- whether attendance was required or volunnaire, we solicited additional narrative com- tary, time of day (A.M., noon, P.M., or evements and descriptions of unusual or ning), day of the week, location, type of particularly effective practices. Many of residency program (adult or child), size of these narrative comments are interwoven residency program (number of residents), with the quantitative results that we present frequency and length of journal club sesand discuss. Seventy-nine percent of the sions,and whether or notfood was routinely completed questionnaires were answered served.
Process variables included the type of faculty that attended at least 50% of the meetings (research, clinical, invited guests, department chair, training director, or research director), as well as the continuous presence or absence of a faculty member (the same person at most club meetings). We also identified who routinely organized, scheduled, selected, and presented articles (faculty, residents, or combination), who took primary responsibility for leading the discussion, and how enduring their present format had been. Respondents also characterized their residency program's expectations regarding resident participation in the club, as well as their use of authority figures and resident input to boost participation. Content variables included the type of articles predominantly selected (current literature, research methods, biological, psychological, psychotherapy), the focus of the articles (original research, reviews, essays), the number of articles discussed per session, and whether or not residents were provided with any formal instruction about how to read journal articles. The classification of variables into the categories of structure, process, and content was done to facilitate the presentation of findings. The assignment of certain variables to one category vs. another was quite arbitrary. The classification of variables into categories is therefore not meant to represent mutual exclusiveness nor were these variables thought to be all inclusive.
Data Analysis. Results of each of the two measures of journal club success were grouped into quartiles. All independent variables were also grouped into categories, and chi-square tests were run between dependent and independent variables. However, to keep presentation of findings manageable, bivariate associations presented in the ensuing tables divided the two dependent variables at the median. In the case of the effectiveness ratings, the data
represent the percentage of programs with higher effectiveness ratings (scale scores of 12 or more). For attendance, the data represent the percentage of programs with attendance estimates of greater than 75% of their target audience. A p value of 0.05 was used as the criterion for statistical significance.
RESULTS Basic Characteristics Half of the responding programs, including most of the child psychiatry programs, had 19 or fewer trainees, whereas 8% had 50 or more trainees. Fifty-nine percent of the programs characterized themselves as being strongly clinical in nature, 37% as having a balanced emphasis between clinical and research interests, and only 4% as being strongly and predominantly research oriented. Forty-seven percent of the programs had only required journal clubs, 27% only elective journal clubs, 12% both, and 14% offered no journal clubs. Many programs offered multiple journal clubs, most of which were based in specific services, such as the consultation-liaison or crisis-emergency service, or alternated with an ongoing seminar or case conference series on a regular basis. A few respondents considered that most of their didactic program was organized in a journal club-like manner, or should be, to reduce the "passivity" of training. The results reported below represent the 180 programs offering at least one journal club and focus on the best-attended journal club when more than one club existed. Table -1 lists the characteristics most frequently reported by respondents and the percentage of respondents reporting them. Structure, Scheduling, and Setting In 81 % of programs, the journal club was offered to all residents in the program together, in 8% to each cohort of residents
,
"
...""""'"
,_
.....,,,_..
'. ... ( ,! 1\"
....
54
Daytime meetings at the institution were most common, and most were scheduled during the lunch period, from 11:301:30 (41 %), with the rest divided roughly evenly among morning (24%), afternoon (16%), and evening (19%) hours. Lunchtime meetings often included meals being brought in, sometimes provided by pharmaceutical representatives. Evening meetings were usually held at a faculty member's or resident's home and were always accompanied by refreshments, if not a meal; in a few programs, evening journal clubs were held at a private dining room in a restaurant or faculty club. Evening journal clubs were roughly divided among those held at a single faculty member's home (9%), rotating among various faculty's homes (5%), alternating faculty and residents' homes (3%), and other sites (2%). Ordinarily the longer meetings took place during the evening, and these journal clubs were more often elective than required. Journal clubs were scheduled every day of the week, most commonly on Wednesday (29%) and least commonly on Mondays (12%), with roughly equal distributions on Tuesdays (19%), Thursdays (20%), and Fridays (19%).
29
Nature of the Readings
83
Current literature was emphasized by 83% of the journal clubs, whereas 17% focused either on classic literature or alternated between current and classic literature. Twenty-four percent of respondents indicated that research methods, experimental design, or statistics were the main emphases of their journal clubs. Thirty-eight percent of journal clubs were portrayed as broad-based with regard to content, whereas 27% were described as biologically focused, 22% as psychosocially or psychologically focused, and 13% as focused in a specific specialty direction. One respondent noted their tradition of whenever possible selecting one biological, one psychological, and one social
separately, in 4% to specialists with some particular interest, and in 7% to some other mix of participants. Journal clubs met monthly (or less frequently) in 53% of programs, every other week/twice per month in 18%, once per week or more often in 28%, and at some other frequency in 2% of programs. One program adhered to a strict 50 meetings per year schedule, rescheduling those that were cancelled. The meetings were most frequently scheduled for one hour (54%), but meetings of one and one-half hours (29%), two hours (11 %), and longer than two hours (6%) were not uncommon. TABLE L Most frequently reported journal club characteristics and percentage of programs reporting them Characteristic
Percent of Respondents
Attendance is required, not elective "Very strong" expectations that all residents will attend Meetings held at the institution (not other locations) All residents in program attend the same journal club
59
monthly for one hour in midday across the noon hour most often on Wednesdays Feature current articles mixture of research and review articles Broad-based focus (Le. biopsychosocial) Two articles assigned Articles photocopied and distributed in advance All attendees expected to have read articles in advance Organized by various faculty other than training director Articles selected by presenter A single resident presents the articles Discussions led by a single faculty member Two or three faculty members usually present At least one faculty member is a continuous participant Specific instructions on critical reading not rovided
53
Scheduled
:
46
67 81
41
60 38
37 90
86
35 42
51 47 37 76 62
emphasis paper on each given topic. Highly specific formats included journal clubs dedicated to computers in psychiatry, philosophy, neuropsychology and neuropsychiatry, and personality assessment, to cite but a few of those mentioned. Some servicebased journal clubs selected articles to correspond with the types of patients being seen on the service at any given time, e.g., painrelated articles in the context of a consultation-liaison service. One weekly journal club systematically devoted one meeting each month to a non-psychiatric topic, e.g., neurology, anthropology, sociology, etc., and one to a "classic" article or book review. One program favored articles written by its own faculty. Another emphasized the importance of selecting articles relevant at the PGY2 level of residency. Several programs in which faculty were journal editors or editorial board members used the journal club to ask the residents to analyze papers submitted for review, putting participants in the role of critical editorial reviewer. Most journal clubs reviewed a mixture of research and review articles (60%), while 24% examined research articles and 16% examined reviews and essays exclusively. Thirty percent of journal clubs assigned only one article per meeting, 37% assigned two, 21 % three, and 12% four or more. In 86% of the journal clubs, all attenders were expected to have read the articles in advance, whereas in 12% only the presenters were expected to read the articles (2% listed another type of expectation regarding reading). To assist the participants, articles were photocopied and distributed in advance by 90% of the journal clubs. Only one journal club that expected all the participants to read the articles in advance did not photocopy and distribute copies in advance. With respect to how many of those attending in fact actually read the articles prior to the meetings, the respondents' impressions were quite varied, with individual estimates ranging from 10% to 100%; 48% estimated that up to about 50% of attenders actually read the 2~
articles, and an additional 45% estimated that between 51 % and 90% actually read the articles prior to the meetings. One respondent noted that residents in his program complained that the faculty did not read the articles in advance of the meetings, thereby diminishing the value of the club. Journals Reviewed We asked the respondents to list up to six journals whose articles were most frequently reviewed in the journal clubs. Sixtyfour separate journals were listed. Usable responses to this question were provided by 151 respondents. Among the general psychiatry programs (N =100) and the child psychiatry programs (N=51), journals cited by 10% or more are noted in Table 2. Organization The journal clubs were organized by training directors (30%), other faculty members (35%), chief residents (35%), other residents (16%), or other persons, usually a journal club committee (5%). Selection of the articles to be discussed was carried out by a variety of individuals, often a mixture within a given journal club. These included the person presenting (42%), but also the faculty leader (35%), chief resident (10%), other permanent coordinator (7%), rotating residents (30%), or other persons (8%). Often those presenting scheduled their presentation dates, topics, and discussants at the beginning of the academic year. Some respondents commented that in their programs giving residents control of the journal club improved participation and attendance. Instruction in Reading Journal Articles We asked whether residents received any instruction in how to read and evaluate journal articles. Sixty-two percent provided \ (11 L \11 I:;. '-.l.
\II)II~
I • "1'111\ :.
!'I\I'\.\
,I'
Characteristics ofjournal Clubs in Psychiatric Training Joel Yager, M.D. Lawrence S. linn, Ph.D. Daniel K. Winstead, M.D. Barbara Leake, Ph.D. We surveyed the membership of the American Association of Directors of Psychiatric Residency Training (AADPRT) regarding the presence and characteristics of journal clubs in their general and child psychiatry residency training programs. Responses were obtained from 141 general residency programs and 76 child psychiatry programs. Eightysix percent (N=180) offered at least one required and/or voluntary journal club, and many offered multiple journal clubs. Higher effectiveness ratings were associated with journal clubs that had mandatory participation, met frequently, were held in a convenient location, reviewed articles on original research, emphasized and taught research methods, and had regular faculty participation. Higher attendance ratings were associated with daytime meetings, smaller residency programs, required clubs, clubs that met more often and for shorter lengths of time within the hospital, and clubs that show continuity in faculty participation. Descriptions of various formats and inducements reported may be useful to other programs wishing to establish or improve journal clubs.
J
ournal clubs are traditionally part of many clinical training programs in a variety of medical specialties (1~). Studies in internal medicine and family medicine have re-
Dr. Yager is Professor and Director of Residency Education at the Department of Psychiatry and Biobehavioral Sciences, Neuropsychiatric Institute, University of California at Los Angeles; and Associate Chief of Staff (Residency Education) at the West Los Angeles Veterans Administration Medical Center (Brentwood Division); Dr. Linn was Research Professor at the Department of Medicine and School of Public Health, University of California at Los Angeles (Dr. Linn died in July 1990); Dr. Winstead is Professor and Chair at the Department of Psychiatry and Behavioral Sciences, Tulane University; Dr. Leake is Senior Statistician, Department of Medicine and Neurology, UCLA. Address correspondence to Joel Yager, M.D., UCLA Neuropsychiatric Institute, 760 Westwood Plaza, Los Angeles, CA 90024. Copyright © 1991 Academic Psychiatry. ,
vealed that journal clubs are quite popular and that they serve a variety of functions. This format is also being increasingly employed in continuing medical education, with groups of practitioners coming together for ongoing study. For example, the Jour-
nal of the American Medical Association
(JAMA) has established programs whereby journal club attenders can obtain category I credit for their participation (5). With regard to psychiatric training, we were impressed by the wide variety of journal club types described to us over the years by colleagues in training programs around the country. We wondered how commonly journal clubs were used in psychiatric training programs, what their characteristics were, and what practices might exist to render some journal clubs particularly effective or ineffective. \ \ 1\ 1 \ II '
. " I \ II >I I{ , •
~ 1'1, I" (
""
METHODS
by training directors; the others were completed by other faculty (10%) or residents (11 %). University-affiliated programs provided 82% of the responses, whereas private hospital programs (4%), state hospital programs (5%), and other types of non-university-affiliated programs (9%) provided the rest. Available data do not permit us to estimate whether the proportion of university programs responding are representative of the general AADPRT membership or whether they constitute a sample biased toward university programs.
We surveyed the membership of the American Association of Directors of Psychiatric Residency Training (AADPRT), a national organization representing more than 90% of the training directors of general and child psychiatry residency programs. The survey asked questions regarding the presence and characteristics of journal clubs in the members' residency programs. Questionnaires were mailed to all members of the AADPRT, and after six weeks a second mailing was sent to all initial non-reMeasures of Success sponders. Responses were obtained from 141 general psychiatry residency programs (representing 67% of all 204 American gen- Dependent variables. Two measures were eral psychiatry residency programs listed in used to define successful journal clubs. The the most recent edition of the National Resi- first was based on respondents' ratings of dent Matching Plan book plus four Cana- effectiveness of their club using the followdian programs) and from 76 child and ing criteria: 1) helps residents keep up with adolescent psychiatry training programs the literature; 2) teaches residents research (representing 58% of American child psychi- methods and statistics; 3) trains residents to atry programs plus one Canadian program). make presentations; and 4) helps residents to Seven respondents answered for both the think critically about articles. Achievement general and child programs in a combined 'on each criterion was rated as Excellent-4, fashion. In the event that duplicate re- Good-3, Fair-2, or Poor-I. Responses to the sponses were received from a given pro- four items were summed, and a reliable gram, only one response per program was Likert scale was constructed (Cronbach's recorded-the response provided by the re- alpha reliability coefficient =0.73). The second measure of success was the spondent most closely involved with journal respondents' estimate of the percentage of clubs. In our introductory letter, we asked that their target audience that actually attended the questionnaires be completed by the their journal club. training director or by another person knowledgeable about the journal club at Independent variables. We investigated the retheir institution. In the event that a program lationship between the two measures of sucoffered more than one journal club, we asked cessful journal clubs and three categories of that the respondent complete the survey independent variables: structure, process, with respect to the best-attended journal and content. Structural variables included club. To supplement the formal question- whether attendance was required or volunnaire, we solicited additional narrative com- tary, time of day (A.M., noon, P.M., or evements and descriptions of unusual or ning), day of the week, location, type of particularly effective practices. Many of residency program (adult or child), size of these narrative comments are interwoven residency program (number of residents), with the quantitative results that we present frequency and length of journal club sesand discuss. Seventy-nine percent of the sions,and whether or notfood was routinely completed questionnaires were answered served.
Process variables included the type of faculty that attended at least 50% of the meetings (research, clinical, invited guests, department chair, training director, or research director), as well as the continuous presence or absence of a faculty member (the same person at most club meetings). We also identified who routinely organized, scheduled, selected, and presented articles (faculty, residents, or combination), who took primary responsibility for leading the discussion, and how enduring their present format had been. Respondents also characterized their residency program's expectations regarding resident participation in the club, as well as their use of authority figures and resident input to boost participation. Content variables included the type of articles predominantly selected (current literature, research methods, biological, psychological, psychotherapy), the focus of the articles (original research, reviews, essays), the number of articles discussed per session, and whether or not residents were provided with any formal instruction about how to read journal articles. The classification of variables into the categories of structure, process, and content was done to facilitate the presentation of findings. The assignment of certain variables to one category vs. another was quite arbitrary. The classification of variables into categories is therefore not meant to represent mutual exclusiveness nor were these variables thought to be all inclusive.
Data Analysis. Results of each of the two measures of journal club success were grouped into quartiles. All independent variables were also grouped into categories, and chi-square tests were run between dependent and independent variables. However, to keep presentation of findings manageable, bivariate associations presented in the ensuing tables divided the two dependent variables at the median. In the case of the effectiveness ratings, the data
represent the percentage of programs with higher effectiveness ratings (scale scores of 12 or more). For attendance, the data represent the percentage of programs with attendance estimates of greater than 75% of their target audience. A p value of 0.05 was used as the criterion for statistical significance.
RESULTS Basic Characteristics Half of the responding programs, including most of the child psychiatry programs, had 19 or fewer trainees, whereas 8% had 50 or more trainees. Fifty-nine percent of the programs characterized themselves as being strongly clinical in nature, 37% as having a balanced emphasis between clinical and research interests, and only 4% as being strongly and predominantly research oriented. Forty-seven percent of the programs had only required journal clubs, 27% only elective journal clubs, 12% both, and 14% offered no journal clubs. Many programs offered multiple journal clubs, most of which were based in specific services, such as the consultation-liaison or crisis-emergency service, or alternated with an ongoing seminar or case conference series on a regular basis. A few respondents considered that most of their didactic program was organized in a journal club-like manner, or should be, to reduce the "passivity" of training. The results reported below represent the 180 programs offering at least one journal club and focus on the best-attended journal club when more than one club existed. Table -1 lists the characteristics most frequently reported by respondents and the percentage of respondents reporting them. Structure, Scheduling, and Setting In 81 % of programs, the journal club was offered to all residents in the program together, in 8% to each cohort of residents
,
"
...""""'"
,_
.....,,,_..
'. ... ( ,! 1\"
....
54
Daytime meetings at the institution were most common, and most were scheduled during the lunch period, from 11:301:30 (41 %), with the rest divided roughly evenly among morning (24%), afternoon (16%), and evening (19%) hours. Lunchtime meetings often included meals being brought in, sometimes provided by pharmaceutical representatives. Evening meetings were usually held at a faculty member's or resident's home and were always accompanied by refreshments, if not a meal; in a few programs, evening journal clubs were held at a private dining room in a restaurant or faculty club. Evening journal clubs were roughly divided among those held at a single faculty member's home (9%), rotating among various faculty's homes (5%), alternating faculty and residents' homes (3%), and other sites (2%). Ordinarily the longer meetings took place during the evening, and these journal clubs were more often elective than required. Journal clubs were scheduled every day of the week, most commonly on Wednesday (29%) and least commonly on Mondays (12%), with roughly equal distributions on Tuesdays (19%), Thursdays (20%), and Fridays (19%).
29
Nature of the Readings
83
Current literature was emphasized by 83% of the journal clubs, whereas 17% focused either on classic literature or alternated between current and classic literature. Twenty-four percent of respondents indicated that research methods, experimental design, or statistics were the main emphases of their journal clubs. Thirty-eight percent of journal clubs were portrayed as broad-based with regard to content, whereas 27% were described as biologically focused, 22% as psychosocially or psychologically focused, and 13% as focused in a specific specialty direction. One respondent noted their tradition of whenever possible selecting one biological, one psychological, and one social
separately, in 4% to specialists with some particular interest, and in 7% to some other mix of participants. Journal clubs met monthly (or less frequently) in 53% of programs, every other week/twice per month in 18%, once per week or more often in 28%, and at some other frequency in 2% of programs. One program adhered to a strict 50 meetings per year schedule, rescheduling those that were cancelled. The meetings were most frequently scheduled for one hour (54%), but meetings of one and one-half hours (29%), two hours (11 %), and longer than two hours (6%) were not uncommon. TABLE L Most frequently reported journal club characteristics and percentage of programs reporting them Characteristic
Percent of Respondents
Attendance is required, not elective "Very strong" expectations that all residents will attend Meetings held at the institution (not other locations) All residents in program attend the same journal club
59
monthly for one hour in midday across the noon hour most often on Wednesdays Feature current articles mixture of research and review articles Broad-based focus (Le. biopsychosocial) Two articles assigned Articles photocopied and distributed in advance All attendees expected to have read articles in advance Organized by various faculty other than training director Articles selected by presenter A single resident presents the articles Discussions led by a single faculty member Two or three faculty members usually present At least one faculty member is a continuous participant Specific instructions on critical reading not rovided
53
Scheduled
:
46
67 81
41
60 38
37 90
86
35 42
51 47 37 76 62
emphasis paper on each given topic. Highly specific formats included journal clubs dedicated to computers in psychiatry, philosophy, neuropsychology and neuropsychiatry, and personality assessment, to cite but a few of those mentioned. Some servicebased journal clubs selected articles to correspond with the types of patients being seen on the service at any given time, e.g., painrelated articles in the context of a consultation-liaison service. One weekly journal club systematically devoted one meeting each month to a non-psychiatric topic, e.g., neurology, anthropology, sociology, etc., and one to a "classic" article or book review. One program favored articles written by its own faculty. Another emphasized the importance of selecting articles relevant at the PGY2 level of residency. Several programs in which faculty were journal editors or editorial board members used the journal club to ask the residents to analyze papers submitted for review, putting participants in the role of critical editorial reviewer. Most journal clubs reviewed a mixture of research and review articles (60%), while 24% examined research articles and 16% examined reviews and essays exclusively. Thirty percent of journal clubs assigned only one article per meeting, 37% assigned two, 21 % three, and 12% four or more. In 86% of the journal clubs, all attenders were expected to have read the articles in advance, whereas in 12% only the presenters were expected to read the articles (2% listed another type of expectation regarding reading). To assist the participants, articles were photocopied and distributed in advance by 90% of the journal clubs. Only one journal club that expected all the participants to read the articles in advance did not photocopy and distribute copies in advance. With respect to how many of those attending in fact actually read the articles prior to the meetings, the respondents' impressions were quite varied, with individual estimates ranging from 10% to 100%; 48% estimated that up to about 50% of attenders actually read the 2~
articles, and an additional 45% estimated that between 51 % and 90% actually read the articles prior to the meetings. One respondent noted that residents in his program complained that the faculty did not read the articles in advance of the meetings, thereby diminishing the value of the club. Journals Reviewed We asked the respondents to list up to six journals whose articles were most frequently reviewed in the journal clubs. Sixtyfour separate journals were listed. Usable responses to this question were provided by 151 respondents. Among the general psychiatry programs (N =100) and the child psychiatry programs (N=51), journals cited by 10% or more are noted in Table 2. Organization The journal clubs were organized by training directors (30%), other faculty members (35%), chief residents (35%), other residents (16%), or other persons, usually a journal club committee (5%). Selection of the articles to be discussed was carried out by a variety of individuals, often a mixture within a given journal club. These included the person presenting (42%), but also the faculty leader (35%), chief resident (10%), other permanent coordinator (7%), rotating residents (30%), or other persons (8%). Often those presenting scheduled their presentation dates, topics, and discussants at the beginning of the academic year. Some respondents commented that in their programs giving residents control of the journal club improved participation and attendance. Instruction in Reading Journal Articles We asked whether residents received any instruction in how to read and evaluate journal articles. Sixty-two percent provided \ (11 L \11 I:;. '-.l.
\II)II~
I • "1'111\ :.
!'I\I'\.\
,I'
