| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Reports |
RR Richter, PT, PhD, is Associate Professor, Department of Physical Therapy, Doisy College of Health Sciences, Saint Louis University, 3437 Caroline St, St Louis, MO 63104-1111 (USA)
SL Schlomer, PT, DPT, is Staff Physical Therapist, BJC HealthCare, St Louis, Mo
MM Krieger, RN, MLIS, is Assistant Director for Information Services, Saint Louis University Medical Center Library, St Louis, Mo
WL Siler, PhD, is Associate Dean for Research Administration, Doisy College of Health Sciences, Saint Louis University
Address all correspondence to Dr Richter at: richterrr{at}slu.edu
Submitted September 8, 2006;
Accepted October 22, 2007
 | Abstract |
|---|
Subjects and Methods: A total of 194 programs listed on the Commission on Accreditation in Physical Therapy Education (CAPTE) Web site in the spring of 2004 were examined. The databases were searched for bibliographic citations of journal articles attributed to particular programs. The program characteristics of faculty size, offering of a research doctorate, and listing in the Carnegie Classification of Institutions of Higher Education (Carnegie Classification) were compared with the number of citations.
Results: A total of 169 programs had at least 1 attributed citation, 50.3% of the programs had fewer than 5 citations, and 3% had 44 or more citations. Rankings based on the number of citations changed when adjusted for faculty size. Of the 38 programs offering a research doctoral degree, 16 had 20 or more citations. Five programs with 44 or more citations were all categorized by the Carnegie Classification as doctoral intensive or extensive.
Discussion and Conclusion: A few programs had a large number of attributed bibliographic citations, but the majority of programs had limited publication productivity in the 5 years studied. These results may provide a baseline for studying the effectiveness of the relatively new CAPTE standards mandating scholarship by physical therapy faculty over time and the impact of the Doctor of Physical Therapy degree on research in physical therapy.
| Introduction |
|---|
 TopAbstract Introduction MethodResultsDiscussionConclusionReferences |
|---|
Beyond the expectations of academe, the continued maturation of a profession depends on members of that profession working to refine and expand the body of knowledge guiding practice in that profession.11–15 Moreover, the clinical world is placing greater emphasis on evidence-based practice. The evidence must exist before it can guide practice, making the publication of research findings essential in contemporary academic preparation and clinical practice.12–14,16–18
Despite scholarship being characterized as the most valuable tool available to any profession,17,19 schools of the health professions have historically emphasized teaching over scholarship20 and are generally characterized by limited scholarly productivity.20,21 Physical therapy is among the largest of the health care professions,22 but few studies have examined the scholarly productivity of faculty in programs for physical therapy education. Holcomb et al23 published findings regarding the scholarly productivity of physical therapy faculty 2 decades ago. In late 1986 and early 1987, Holcomb et al23 surveyed 130 physical therapy faculty members at 19 southeastern academic health science centers; 127 faculty members completed the survey. Respondents indicated the number of manuscripts published or accepted for publication in refereed journals (ie, research findings, descriptive reports of projects, testimonials, editorials, or reviews). Respondents had an overall publication rate of 0.94 article published for every 2 years of employment in higher education. Even when liberal definitions of publications were used, almost one third (30%) of the respondents could not claim authorship of a manuscript.
In the interim since the study of Holcomb et al,23 there have been substantive changes in physical therapy education. First, the professional (entry-level) degree moved from a bachelor's level to a master's level (MPT). Before the transition to the MPT degree was complete, a few programs began offering the Doctor of Physical Therapy (DPT) degree as a professional-level degree.24 In 2000, the House of Delegates of the American Physical Therapy Association (APTA) passed a vision statement, which stated in part that, "Physical therapy, by 2020, will be provided by physical therapists who are doctors of physical therapy ..." (RC 44–00).25 Since then, the number of programs offering a professional-level DPT degree has increased.24 As a growing number of programs made the transition from the MPT degree to the DPT degree, changing accreditation standards required that core full-time physical therapy faculty members demonstrate scholarly activity.26 Scholarship is now unavoidably an expectation in academic physical therapy programs. A critical question for the physical therapy profession is how these changes ultimately affect scholarly productivity.
An understanding of factors influencing scholarship would allow for the possible future development of a model of scholarly productivity in the physical therapy field similar to those created in the area of family medicine by Bland and colleagues27,28 and Brocato and Mavis.29 Those models incorporate factors that are related to the institution, program, and individual and that are known to influence scholarly productivity in academic family medicine departments.27,29 Specifically, research productivity among family medicine faculty members increases when the individual faculty members possess key characteristics, work within the context of an institutional environment conducive to research, and are affiliated with a department led by a research-oriented individual. However, few studies have examined scholarly productivity in the physical therapy field, and it was unknown how or whether these same factors would influence academic physical therapy program scholarly productivity. Through investigations of factors influencing scholarly productivity, prospective studies can be planned with the goal of providing for the development of empirically based strategies for increasing scholarship in the physical therapy field.
Given the need to assess the impact of the recent changes in physical therapy education on scholarly productivity and the desire to continue to increase scholarship in the physical therapy field, it was necessary to establish baseline measures. Because of the time lag between initiating a project and seeing the work published, a representative period of 5 years, 1998 to 2002, was studied. Because citations of peer-reviewed journal articles are publicly available in bibliographic databases, we chose to measure scholarly productivity by tallying article citations. Therefore, the purposes of this investigation were to describe the publication productivity of accredited academic physical therapy programs in the United States and Puerto Rico between 1998 and 2002 and to examine publication productivity in the context of selected program characteristics.
| Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
Measure of Program Productivity by Journal Publication
In previous reports on the scholarly productivity of faculty in allied health programs, investigator-developed surveys were used to collect data.17,23,32 The use of surveys to collect data relies on accurate recall by the respondents, which is difficult to verify. Furthermore, defining what constitutes a unit of publication may be difficult to convey within the survey directions. Therefore, for this study, we decided to search PubMed and the Cumulative Index to Nursing & Allied Health Literature (CINAHL) for bibliographic citations of journal articles attributed to accredited academic physical therapy programs. PubMed is a freely available database that contains more than 16 million bibliographic citations.33 CINAHL is a proprietary database that is available by subscription and contains more than 1.2 million34 records relevant to nursing and other health care professions. The use of these databases eliminated potential errors in recall.
The PubMed and CINAHL search strategies were designed to search the field of the bibliographic citation containing the institution and address of the first author, which is typically the only address listed. We used field tags to retrieve bibliographic citations with the terms "physical" and "therapy" in the affiliation field. We refined our search by using Boolean operators and limits. Because of differences in the scope and terminology of the databases, the search strategies used for PubMed and CINAHL varied slightly (Tab. 1).
|
After the searches were completed, the bibliographic citations were imported into EndNote (version 7.0),
a bibliographic management program. We used several methods to ensure that duplications were eliminated and that the final data set included only bibliographic citations of likely peer-reviewed journal articles from accredited programs.
First, we used EndNote and manual searching to identify bibliographic citations that were retrieved from both PubMed and CINAHL. These duplicate bibliographic citations were eliminated from the final data set. Second, manual searching was used to eliminate foreign addresses from the CINAHL results, because CINAHL does not have an option to limit searches to bibliographic citations from a particular country. Because we searched for the terms "physical" and "therapy" in the affiliation field, we knew that we would retrieve bibliographic citations of journal articles authored by people affiliated with physical therapy departments that were not academic units in accredited programs.
To determine whether a program was accredited, we used the CAPTE listing of professional physical therapy education programs. Accredited physical therapy education programs listed on the CAPTE Web site in the spring of 2004 were eligible for inclusion. At times, it was not clear whether the institution listed in the affiliation field was an accredited physical therapy education program or a hospital site at an accredited physical therapy education program. A decision rule and consensus were used when questions regarding the attribution of a bibliographic citation arose. The decision rule, simply stated, was that there had to be clear attribution to an accredited academic program. Records in which the attribution was not clear were deleted from the final data set. Attribution to a particular academic program was based on the institution listed in the affiliation field of the bibliographic citation. Although we limited the CINAHL search to retrieve bibliographic citations from peer-reviewed publications, we noted that bibliographic citations of abstracts for conference presentations were included in the retrieval results when these abstracts were published in peer-reviewed journals. Such bibliographic citations were eliminated from the final data set. In addition, we retrieved from both databases bibliographic citations that we believed were not likely peer reviewed. The decision to exclude such bibliographic citations was based on our interpretation of the title of the publication. For example, from CINAHL we retrieved bibliographic citations with the title "Preface," and from PubMed we retrieved bibliographic citations with the titles "Presidential Address" and "Mary McMillan Lecture." These types of bibliographic citations were excluded from the final data set. Finally, a small number of bibliographic citations retrieved from CINAHL were excluded from the final data set for miscellaneous reasons, such as the primary author being a visiting professor or the bibliographic citation being a reprint.
Figure 1 illustrates the number of bibliographic citations included from both databases, the number of bibliographic citations excluded from the final data set, and the reason for the exclusion of a citation from the final data set.
|
Second, we determined whether a program offered a research doctoral degree in addition to a professional-level degree. We believed that the offering of a research doctoral degree could increase the likelihood of publication productivity because the presence of doctoral-degree students likely increases the amount of active research and subsequent publications. For the purposes of this study, we did not consider the DPT degree to be a research doctoral degree. We used 2 sources to determine whether a program offered a research doctoral degree: APTA's listing of postprofessional programs30 and individual program Web sites. Program Web sites were accessed during the spring of 2004. Programs were given credit for offering a research doctoral degree if such a degree was listed at either source.
Third, we noted the Carnegie Classification for each of the accredited programs.31 The Carnegie Classification provided a framework for classifying institutions of higher education within the United States. Because we wanted to examine the differences among degree levels offered by programs, we combined the master's I and II classifications into one composite category and the doctoral intensive and extensive classifications into another composite category.
Data Analysis
Counts and percentages were used to summarize the data. Because of nonnormality, nonparametric tests were used to examine differences in publication productivity. Specifically, a Mann-Whitney U test was used to examine differences in publication productivity between programs offering a research doctorate and those not offering a research doctorate. A Kruskal-Wallis test was used to examine differences in publication productivity across the collapsed Carnegie Classifications. Significant results were evaluated with Bonferroni-corrected pair-wise comparisons by use of Mann-Whitney U tests. Significance for the Mann-Whitney U and Kruskal-Wallis tests was set at P<.05.
| Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
|
|
|
| Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
While programs were being encouraged to move to the DPT degree, CAPTE standards requiring scholarly activity by core faculty were introduced. The purpose of the revised standards apparently was to provide a significant incentive for programs to increase attention and resources dedicated to scholarship. As mentioned earlier, Kaufman35 learned that nearly two thirds of academic physical therapy programs reviewed failed to meet CAPTE standards for scholarship during the same period (1998–2002) as that studied in the present investigation. Incorporating expectations for scholarship into accreditation standards is a substantial incentive for an increased focus on scholarship. Unfortunately, the success of that strategy will likely be mediated by a variety of internal and external environmental factors. In any event, the results of the present investigation effectively present a baseline measure of the publication productivity of academic physical therapy programs at the advent of transformative changes in the profession.
Although approximately half of the accredited academic physical therapy programs published one article per year or less, a select group of programs were far more productive: 11.3% of programs had 20 or more attributed citations. The work of Bland and colleagues27,28 suggested that scholarly activity by individual faculty members depends on the characteristics of those individuals, the environment provided by the institution, and the leadership of the department with which the individuals are affiliated. Our ability to identify which factors may set these highly productive departments apart from the majority is limited without a direct study of the departments. However, it may be possible to gain some insight into what sets highly productive departments apart by studying publicly available characteristics.
Accordingly, the secondary purpose of this investigation was to examine publication productivity in the context of selected publicly available program characteristics. Specifically, we examined the influence of faculty size, offering of a research doctorate, and Carnegie Classification of the institution on publication productivity. For example, expressing publication productivity as a ratio of the number of publications to the number of full-time faculty members may be an effective means of equating programs.37 When controlling for faculty size, we found that some programs improved in rank while others fell in rank. Wagner et al38 also found that some programs improved in publication productivity rank while others fell when adjusted for full-time faculty size. Several possibilities could account for these findings. First, one faculty member might account for the majority of publications. In a program with many faculty members, the effect of this one faculty member would be diluted when adjusting for number of full-time faculty members, accounting in part for a decrease in rank. Conversely, in a program with a small faculty, the effect of this one faculty member would be enhanced when adjusting for number of full-time faculty members, accounting in part for an increase in rank. For programs that are basically comparable in size, however, programs showing an increase in rank when faculty size is factored in are more efficient, for some reason, and those programs would be particularly interesting to study.
Our findings show that it is possible for programs with a small number of full-time faculty members to have a relatively large number of publications. That is, assembling a large faculty is not sufficient for increasing scholarly productivity. Similarly, the finding that programs with a relatively small faculty may be very productive in the realm of scholarship reinforces a perception that scholarly productivity is a complex issue incorporating factors beyond the number of faculty members available to share the work of the program. This finding may be reassuring as concerns about the availability of faculty members increase because it implies that the nature of the institution and the faculty may be more important than the number of faculty members with regard to attempts to increase the scholarly productivity of a program. Further investigation is needed to determine how programs with a small faculty achieve high levels of publication productivity. The model of Bland and colleagues27,28 suggests possibilities for future investigations regarding interactions among individual faculty members, the institution, and the leadership of the program.
In keeping with the model of Bland and colleagues,27,28 we found that the nature of the institution influences publication productivity. Programs housed in research-intensive and -extensive institutions produced more publications than programs housed in any other setting, with the exception of specialized medical schools and medical centers. This greater productivity could be associated with the possibility that institutions in those settings are characterized by cultures in which scholarship is an expectation. Those are also the categories of institution most likely and most able to provide the resources necessary to encourage and facilitate scholarship. Given that institutional characteristics appear to be critical in publication productivity, the ability of programs housed in institutions without the resources or culture of research institutions or academic health centers to satisfy scholarly expectations reflected in the CAPTE standards may raise questions about their future.
Although the programs housed in research institutions produced more publications than any other category, with the exception of specialized medical schools and medical centers, there was marked variation among the programs in this category. Because the Carnegie Classification is given to each institution as a whole, it is not surprising that there was variation in the number of attributed bibliographic citations among doctoral intensive or extensive programs. At least 3 factors might account for this variation. First, the general priorities of a large institution, including scholarship, might not be consistent with those of an individual program.39 Second, resources for scholarship from an institution (eg, seed funds, release time, and sabbaticals) might not be equally distributed among the individual academic units. Third, a program might not be held accountable to the institutional priority for scholarship. The interaction of these factors might lead to differences in publication productivity among physical therapy programs located in institutions with the same Carnegie Classifications.
Increased publication productivity also appeared to be related to the offering of a research doctorate in a program. However, some programs offering a research doctorate had a very low level of publication productivity. Hence, the offering of a research doctorate may be necessary but not sufficient for a high level of publication productivity. We do not know whether a high level of productivity was a prerequisite to offering a research doctorate program or whether the addition of a research doctorate program was an attempt to enhance publication productivity. In either case, the failure of those programs to produce substantive numbers of publications again suggests that scholarly productivity is a complex issue reaching beyond the existence of an approved research degree.
Models of scholarly productivity examining relevant general and specific factors could be used to identify strategies to enhance productivity. However, there are no tested models for physical therapy that describe factors affecting scholarly productivity. Two models27,29,40 have been used to attempt to examine publication productivity in academic family medicine.
As mentioned previously, Bland and colleagues developed40 and tested27 a theoretical model of research productivity in academic family medicine that included 3 components: individual characteristics, institutional characteristics, and leadership characteristics of the departmental head. Each of these 3 components had characteristics that predicted individual and departmental research productivity. With regard to individual characteristics, motivation was a significant predictor.27 Resources, rewards, sufficient work time, clear coordinating goals, mentoring, culture, communication, research emphasis, communication with professional networks, and assertive-participative governance were significant institutional characteristics predicting productivity.27 The size, experience, and expertise of an institution, involving input from members with various degree levels, approaches to problems, and discipline backgrounds, also proved to be significant institutional factors predicting productivity.27 Finally, the use of an assertive-participative leadership style, in which the leader held regular meetings with clearly stated objectives, made high-quality information readily available, expected active participation of members, and showed that ideas from members were valued, was a significant predictor of productivity.27
Brocato and Mavis29 also developed and tested a theoretical model of research productivity of faculty in academic family medicine departments. Time spent on research, as well as psychological and cognitive factors such as improved research training, presence of a research agenda, motivation to do research, concurrent research projects, in-depth knowledge of research specialty area, presence of professional networks, and clear expectations of promotion and tenure in relationship to productivity, were significant predictors of research productivity.
Although there is no faculty scholarship model for physical therapy, Kaufman recently described the nature of the academic physical therapy discipline "as a soft, applied field with the complex array of integrated knowledge, connections, and agendas of a human services discipline."35(p6) Kaufman35 proposed that the diversity of physical therapy is reflected in the subsequent diversity of advanced degrees and clinical specializations held by physical therapy faculty. Although this diversity may be advantageous for physical therapy instruction, Kaufman35 believes that it may hinder the development of networks supporting scholarship. Kaufman's premise35 was supported by Bland et al,27 who reported that networking was a significant predictor of research productivity. The work of Kaufman,35 Bland et al,27 and Brocato and Mavis29 could be used to develop a model of physical therapy research productivity that accounts for the internal and external characteristics that must be considered in relationship to scholarship in this diverse field. There is little doubt that success in scholarly productivity is based on a complex interaction between individual faculty members and the departmental culture in the context of the global institution.
As noted previously, CAPTE recently mandated that all core faculty members in academic physical therapy programs have a scholarship agenda with documented accomplishments.4 It is unclear whether a mandate for scholarship will change the scholarly productivity of faculty members in academic physical therapy programs. Although a mandate may lead to changes in departmental aspects, the institution seems beyond the influence of CAPTE, and it is unclear how the mandate may lead to changes in individual characteristics related to scholarship. Because the inherent characteristics of an individual are difficult to alter, strategic hiring appears to be critical for the future growth of scholarship. Strategic hiring also may offer the advantage of the slow evolution of a departmental culture that is more conducive to scholarship.
The conclusions that we can draw from this research may be limited by a variety of factors. For example, our method of attributing bibliographic citations of programs is a potential limitation of our study. Because we determined program affiliation on the basis of the address field of the citation and because the address is typically listed for the first author only, we expected to miss some citations when faculty members at accredited programs were second authors publishing with first authors who were not affiliated with accredited physical therapy programs (eg, the first author was affiliated with a nursing or public policy program or a clinical site).
Similarly, we used publicly available sources instead of directly contacting individual physical therapy programs to determine individual program characteristics. By using this approach, we were unable to obtain some data regarding individual program characteristics. For example, we were unable to obtain the number of full-time faculty members at 2 institutions with large numbers of attributed bibliographic citations. In addition, we did not ascertain the accuracy of the data from these public sources. We believe that our approach is justified because consumers, such as prospective students, may use these sites when researching academic physical therapy programs. Finally, the program characteristics gathered at the inception of the study may not reflect the characteristics during the publication years that we examined.
Finally, publication productivity as measured by number of bibliographic citations of journal articles is only one possible measure of publication productivity. We selected this method because of the availability of databases that index bibliographic citations. Although we recognize that there are other forms of publication productivity, such as books and book chapters, we are unaware of a systematic and comprehensive database that indexes these types of publications in a manner that would allow the retrieval of book chapter citations for clear attribution to an individual academic physical therapy program.
| Conclusion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
| Footnotes |
|---|
The study protocol was granted exempt status by the Institutional Review Board of Saint Louis University.
Portions of this article were presented at the Combined Sections Meeting of the American Physical Therapy Association; February 1–5, 2006; San Diego, Calif.
* SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606. 
Thomson ResearchSoft, 3501 Market St, Philadelphia, PA 19104.
| References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionConclusionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  L. G Portney Invited Commentary Physical Therapy, March 1, 2009; 89(3): 216 - 219. [Full Text] [PDF]
|
|
|
|
|
|
R. R Kaufman Career Factors Help Predict Productivity in Scholarship Among Faculty Members in Physical Therapist Education Programs Physical Therapy, March 1, 2009; 89(3): 204 - 216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R Harris On "Journal publication productivity..." Richter et al. Phys Ther. 2008;88:376-386. Physical Therapy, June 1, 2008; 88(6): 791 - 791. [Full Text] [PDF]
|
|
|
|
|
|
S. J Warden On "Journal publication productivity..." Richter et al. Phys Ther. 2008;88:376-386. Physical Therapy, April 1, 2008; 88(4): 538 - 539. [Full Text] [PDF]
|
|
|
|
|
|
C. Maher On "Journal publication productivity..." Richter et al. Phys Ther. 2008;88:376-386. Physical Therapy, April 1, 2008; 88(4): 539 - 539. [Full Text] [PDF]
|
|
|
|
|
|
S. E Bennett and P. J Ohtake On "Journal publication productivity..." Richter et al. Phys Ther. 2008;88:376-386. Physical Therapy, April 1, 2008; 88(4): 539 - 540. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
