| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
Rapid Responses to:
|
|
Rapid Responses published:
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif){kind=icon}
Comments on PT faculty publication productivity from a neighbor to the North
Misrepresentation of PT Program Research Productivity
Authors' Response
Re: Underestimation of publication productivity discredits the profession's contribution
Underestimation of publication productivity discredits the profession's contribution
|
|
|||
|
Susan R. Harris, Professor University of British Columbia
Send rapid response to journal:
susan.harris{at}ubc.ca Susan R. Harris
|
Having published an article on pediatric physical therapy publication trends in 1993,1 I was keenly interested in reading Richter and colleagues’ recent article on publication productivity in US academic physical therapy programs2 and the letters written in response to it.3-5 Kudos to Randy Richter and his colleagues for conducting this important study, despite some methodological shortcomings that were pointed out by the authors themselves2 and the respondents.3-5 I agree wholeheartedly with Christopher Maher that this study should not be ignored4! Having been an academic in physical therapy programs in two major research-intensive universities in the US from 1981 to 1989, I moved to Canada and the University of British Columbia’s (UBC) physical therapy program in 1990 and have remained there ever since. Not only are there far fewer physical therapist educational programs in Canada (n=13) than in the US, they are also much more standardized in their approach. All 13 are located in research-intensive universities and are part of major health sciences centers. Consequently, the pressure at most of these universities to conduct and publish research is very strong. For example, at UBC, our program is housed within the Faculty of Medicine, and we are expected to attain the same benchmarks as all other faculty within Medicine (ie, 2–3 peer-reviewed publications per year per faculty member). Because, as a Canadian physical therapy program, we were not included in the study by Richter et al, I decided to do a comparison of our own physical therapy faculty’s productivity during the same time period: 1998–2002. However, instead of doing a PubMed/CINAHL search, it was easier (and likely more accurate, as pointed out by Bennett and Ohtake5) to just ask my colleagues who were on faculty at that time to send me their CVs. From 1998 to 2002, we were a entry-level bachelor’s degree program. Although we also offered a research master’s degree (MSc), our PhD program did not begin until 2003. During 1998–2002, we had 8 full-time faculty members; however, 2 of the 8 were on teaching tracks versus academic tracks and, therefore, were not expected to produce research articles as part of their faculty mandate. During that 5-year period, we published 51 peer-reviewed articles, which would rank us as third (among the US programs) in Table 2 of the article by Richter et al.2 Dividing that number by 8 (and including the 2 teaching-track faculty members), our ratio of publications per faculty member was 6.4—which would place us second among US physical therapy programs (or perhaps first, based on Bennett and Ohtake’s very candid letter about the University at Buffalo program actually being at 4.3 articles per faculty member5). Although my US academic colleagues might accuse me of “tooting our own horn,” I am sharing this information with the hopes that it might be useful to CAPTE in their future accreditation efforts by realizing that a relatively small faculty (n=8) from an entry-level bachelor’s degree physical therapy program (at that time) can produce a very strong level of “scholarship,” even without a DPT program! And many other Canadian physical therapy programs likely meet or exceed our own research productivity. My sincere thanks to Richter and colleagues for providing this critical baseline on faculty productivity across physical therapy educational programs in the US. For the sake of our scholarly reputation as a profession, I hope that this article will provide a wake-up call to all of my academic colleagues “south of the border.” How can we possibly expect our physical therapist clinician colleagues to be evidence-based practitioners if we are not providing them with the evidence on which to base that practice? Susan R Harris SR Harris, PT, PhD, FAPTA, is Professor, Department of Physical Therapy, University of British Columbia, Vancouver, British Columbia, Canada References 1 Mulero Portela AL, Harris SR. Research in pediatric physical therapy: a comparison of publication trends from 1975 to 1988. Pediatr Phys Ther. 1993;5:183–187. 2 Richter RR, Schlomer SL, Krieger MM, Siler WL. Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002. Phys Ther. 2008;88:376–386. 3 Warden SJ. Letter to the editor. [Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002.] Phys Ther. 2008;88;538–539. 4 Maher CG. Letter to the editor. [Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002.] Phys Ther. 2008;88:539. 5 Bennett SE, Ohtake PJ. Letter to the editor. [Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002.] Phys Ther. 2008;88:539–540. |
|||
|
|
|||
|
Susan E Bennett, Interim Program Director Physical Therapy Program, University at Buffalo, Patricia J Ohtake
Send rapid response to journal:
sbennett{at}buffalo.edu Susan E Bennett, et al.
|
It is with interest that we read the article in the March issue of PTJ in which Richter and colleagues1 examined the publication productivity of accredited academic physical therapy programs in the United States and Puerto Rico between 1998 and 2002. The authors examined publication productivity in the context of selected program characteristics, including the size of the faculty. Although the intent of this investigation is to be applauded, we do not agree that this article “effectively present[s] a baseline measure of the publication productivity of academic physical therapy programs.”1(p381) The finding that there is a wide variation in the level of faculty productivity is important; however, the quantification of the productivity is problematic, especially when using these data to establish baseline performance numbers. Furthermore, if specific physical therapy programs are going to be identified, it is imperative that the data be an accurate representation of publication productivity. Although the authors acknowledge that there were methodological limitations, we feel that these flaws compromised the accuracy of the data collected and thus the productivity of the physical therapy programs identified is probably inaccurate. The specific methodological problems relate to capturing citations based on the requirement to have the terms “physical” and “therapy” in the institution field. This approach has 3 important limitations: (1) it includes citations of authors who are not also included as faculty members in the CAPTE database; (2) it did not identify programs that do not use the terms “physical” and “therapy” in their identification; and (3) it did not identify physical therapist authors whose affiliation is not indicated in the article. As members of the Physical Therapy Program at the University at Buffalo (UB) (identified as the State University of New York at Buffalo in the manuscript), we would like to use our own situation as an example of the flaws in the data collection of this study. Our program was identified as being #2 in the total number of publications and #1 in the number of publications per faculty member. This is an overestimation of our productivity directly related to the methodological issues raised above. With respect to the first search strategy limitation we identified, our physical therapy program was a program in the Department of Physical Therapy, Exercise and Nutrition Science during the period of the search. We performed the searches of PubMed and CINAHL as published and identified 60 citations. Of these, 12 were from physical therapy faculty, whereas 48 were authored by our colleagues in the programs of Exercise Science and Nutrition. Because our colleagues in these other programs would not have been identified as faculty according to CAPTE, the report that we averaged 8 publications per faculty member is an overestimation. Concerning the second limitation we raised about the search strategy, publications by UB physical therapy faculty would not have been identified at all if our current affiliation was used at that time (we are now a program in the Department of Rehabilitation Science). Finally, the issue of not identifying articles by physical therapy faculty members if their affiliation was not in the article is important. The actual productivity for faculty in UB Physical Therapy Program during the period studied was 30 articles. This number of articles yields 4.3 articles per faculty member, 50% lower than reported. Richter and colleagues’ conclusion that, between 1998 and 2002, only 11% of physical therapy programs published more than 20 articles may be true; however, we suspect that the productivity of many physical therapy programs is misrepresented as was ours. Based on the erroneous reporting of UB’s productivity, we suggest that the publication productivity of the programs identified in this article, as well as the other numbers derived for performance measures, be viewed with caution as they may actually be higher or lower than reported. References 1 Richter RR, Schlomer SL, Krieger MM, Siler WL. Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002. Phys Ther. 2008;88:376-386. |
|||
|
|
|||
|
Randy R. Richter, Associate Professor Saint Louis University Program in Physical Therapy, Sarah L. Schlomer, Mary M. Krieger, William Siler
Send rapid response to journal:
richterrr{at}slu.edu Randy R. Richter, et al.
|
We appreciate the interest in our work and the insights and commentary offered by Warden, Maher, and Bennett and Ohtake. We also appreciate this opportunity to respond. As Warden and Maher report, we acknowledged our limitations in the manuscript. We believe the heart of the issue, is the conflict implicit in the responses by Warden and Maher, respectively. That conflict seems to hinge around whether this work fell short because it failed to sufficiently acknowledge research productivity within physical therapy or succeeded in demonstrating that programs vary dramatically in research productivity. Our purpose in this study was to establish a baseline for research productivity for the profession at a critical juncture in its history. Bennett and Ohtake do not believe we have met our purpose and point out that our method misattributed citations to their program. In the manuscript we acknowledge that attribution of a citation was at times difficult. Had we examined our findings by author and cross-referenced the results with CAPTE listings we may have been able to more finely differentiate our results. However, this approach also has limits. For example identical surnames or inconsistent use of initials may make it difficult to identify a particular author. This approach may also eliminate citations when the first author is a student at an academic program and result in an under-representation of that program’s publication productivity. Ultimately, our desire is to use models such as those identified in our paper to inform strategies for growing research across programs. We see no evidence suggesting that our methodology was selectively biased against programs with little to no publication history in the time period studied. We can only reaffirm the message reiterated by Maher: publication productivity varies widely amongst academic programs and many programs are apparently struggling to contribute to the body of knowledge of the profession. Absolutely the profession should be proud of its investment in research. We submit, however, that the profession benefits more from a focus on how we can continue to grow rather than on why we should be proud. Although we recognize the limitations, we believe science is built on replication and refinement, we hope that our work at the very least provides a foundation for replication and refinement. |
|||
|
|
|||
|
Chris G Maher, Professor University of Sydney, Australia
Send rapid response to journal:
C.Maher{at}usyd.edu.au Chris G Maher
|
Stuart Warden is correct in pointing out that Richter and colleagues’ search strategy means they have likely underestimated the research productivity of physical therapists in their article.1 I did a quick check on Web of Science, which allows you to search the affiliation field of all authors, and found about 20% more articles. This is pretty close to the error of 30% suggested by Warden based on a hand search of PTJ. I also repeated my search on Web of Science, but this time for the period 2002-2007, and found twice as many papers. Given these errors we can just ignore Richter and colleagues’ paper; or can we? I don’t think we should ignore the article. The finding that does not seem to change materially with the different search strategies is that the publication productivity of the physical therapy institutions seems to vary enormously. Some institutions publish a lot, some very little, and some not at all; a result that suggests that there are quite different cultures of scholarship within the various institutions. With the APTA Vision 2020 statement emphasising a culture of research and scholarship in the physical therapist of 2020, Richter and colleagues’ article1 flags a potential problem for the profession. It would seem that not all institutions are equally equipped to foster this culture in their students. Chris Maher, PT, PhD Reference 1 Richter RR, Schlomer SL, Krieger MM, Siler WL. Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002. Phys Ther. DOI:10.2522/ptj.20060266. |
|||
|
|
|||
|
Stuart J. Warden, Assistant Professor and Director of Research Department of Physical Therapy, Indiana University
Send rapid response to journal:
stwarden{at}iupui.edu Stuart J. Warden
|
I read with interest the report in Physical Therapy by Richter et al1 regarding publication productivity in academic physical therapy programs in the United States and Puerto Rico. Evidence-based practice is a current ideal in health care and involves the integration of individual clinical expertise with the best available scientific evidence. In order for physical therapy to meet this ideal there is a need not only to perform quality research, but also to make it available to the widest possible audience. The dissemination of research via peer review and subsequent publication forms the cornerstone of this process. Physical Therapy does a superb job of disseminating information regarding the profession to a wide and broad audience. It does this by using not only traditional means (ie, printed journal issues), but also more recent methods of distribution (ie, the Internet), which permit greater globalization and rapid circulation. Richter and colleagues'1 report of publication productivity in academic physical therapy programs is timely for a number of reasons. First, the previous review of scholarly productivity in physical therapy was performed 2 decades ago,2 prior to the introduction of “Vision 2020” by the American Physical Therapy Association and the transition of the majority of academic programs within the United States to the Doctor of Physical Therapy degree. Second, there is increasing expectation for those within the academic realm to demonstrate productivity in order to achieve promotion and tenure and meet standards set by the Commission on Accreditation in Physical Therapy Education (CAPTE). Third, there are increasing threats to the profession from complementary and alternative health care professions, all of whom have also recognized the importance of publication productivity.3-6 Based on these facts, determining the overall publication productivity within physical therapy is a necessity, and Richter and colleagues should be applauded for their work. Unfortunately, there are numerous serious caveats to the report by Richter et al. These ultimately result in the data providing an inaccurate representation of the profession’s true publication productivity. The investigators rightly searched for peer-reviewed journal articles within publicly available bibliographic databases. However, the search execution by using field tags with the terms “physical” and “therapy” in the affiliation field is overly simplistic. As both PubMed and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) report only the first affiliation of the first author for each paper, Richter et al’s search strategy retrieved articles only where the first author listed their first affiliation as being “physical” and “therapy.” It did not retrieve publications in which a physical therapist: (1) was working as part of a multidisciplinary research team, but was not the first author; (2) was senior or last author and the first author was not affiliated with an academic physical therapy program; (3) teaches within an academic physical therapy program which does not contain the terms “physical” and “therapy” in its department’s name (for instance, when physical therapy is a division within a broader department); or (4) had an external affiliation that was listed within a publication prior to their academic physical therapy program affiliation. Richter et al1 acknowledge the possible limitation of their search strategy; however, a full appreciation of how the strategy grossly underestimates the actual productivity within academic physical therapy programs was not provided for readers. To illustrate the limitations in Richter and colleagues' search strategy, consider the following examples. I performed a search of PubMed using my own name and the terms “physical” and “therapy” as affiliation field tags. Using Richter and colleagues' search strategy, I am accredited with 12 publications during my time affiliated with an academic physical therapy program within the United States (August 2004-January 2008). However, when the affiliation field tags is removed, I am accredited with the more accurate number of 24 publications during this time period in which my academic physical therapy program affiliation was clearly listed in the publication. To demonstrate that this is not a finding related to an individual, I looked in Physical Therapy for the 2007 calendar year. Physical Therapy published 37 original research reports during this time that were co-authored by at least one investigator affiliated with an academic physical therapy program in either the United States or Puerto Rico. Using the search strategy usedby Richter et al, 30% (11 out of 37) of these publications were not located using either PubMed or CINAHL. In addition to grossly underestimating the contribution of the profession to health care in terms of total number of publications, the search strategy implemented by Richter et al also discredited important research contributions by the profession. As an example, recent research from the Extremity Constraint-Induced Therapy Evaluation (EXCITE) trial showed that constraining the nonparetic upper limb following stroke improves functional outcome of the paretic upper limb and reduces self-perceived hand function difficulty for up to 12 months.7 Physical therapists were the principal investigators in this trial and lead authors on the resulting publications; however, this influential body of work would not have been credited to the profession using the search strategy of Richter et al, because the first author’s affiliation did not include the terms “physical” and “therapy.” A further limitation of the data contained within Richter and colleagues' report is the fact that the search period was for publications between 1998 and 2002. The rationale for performing a search with a concluding time point of 2002 (5 years ago) was that there is “a time lag between initiating a project and seeing the work published.” While this is true, it does not explain why more recent publications were not included. A body of work that has been published has already experienced the time lag between project initiation and publication. Thus, it is not clear why Richter and colleagues did not conclude their search with a date closer to the initial submission of their manuscript (September 2006). This would have provided more up-to-date, rather than historical, information. In summary, I applaud Richter et al for their foresight in recognizing the need to establish journal publication productivity in academic physical therapy programs. While their intent was genuine, their methods were not, and I am concerned that the data obtained does not provide an accurate representation of the actual contribution of the profession. References 1 Richter RR, Schlomer SL, Krieger MM, Siler WL. Journal publication productivity in academic physical therapy programs in the United States and Puerto Rico from 1998 to 2002. Phys Ther. [published online December 20, 2007 as doi:ptj.20060266]. 2 Holcomb JD, Selker LG, Roush RE. Scholarly productivity: a regional study of physical therapy faculty in schools of allied health. Phys Ther. 1990;70:118-124. 3 Hoskins W, Pollard H, Reggars J, et al. Journal publications by Australian chiropractic academics: are they enough? Chiropr Osteopat. 2006;14:13. 4 Paul S, Liu Y, Ottenbacher KJ. Research productivity among occupational therapy faculty members in the United States. Am J Occup Ther. 2002;56:331-334. 5 Rose RC, Prozialeck WC. Productivity outcomes for recent grants and fellowships awarded by the American Osteopathic Association Bureau of Research. J Am Osteopath Assoc. 2003;103:435-440. 6 Starkey C, Ingersoll CD. Scholarly productivity of athletic training faculty members. J Athl Train. 2001;36:156-159. 7 Wolf SL, Winstein CJ, Miller JP, et al. Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial. JAMA. 2006;296:2095-2104. |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
Rapid Responses: