This is a time of dramatic changes in the economic, academic, and scientific context of clinical research within academic health centers (AHCs). The Balanced Budget Act of 1997 and the growth of managed care are two important economic influences being felt by both medical schools and AHCs; their combined impact is restricting available funds for teaching and research.1,2 Suggestions of decreasing research activities in highly managed care environments are being raised.3–6 while the American Medical Association and the Association of Academic Medical Colleges have described a “crisis” in the conduct of clinical research within AHCs over the past two decades.7–10 While this crisis undoubtedly has arisen for several reasons, one key factor in this complex equation might be a misalignment of the perceived value of and the incentives for the conduct of clinical research within AHCs. To evaluate this hypothesis, in 1999 we surveyed deans, academic administrators, department chairs, and faculty members across the membership of the University HealthSystem Consortium (UHC), a non-for-profit cooperative of 80 AHCs. The UHC represents a cross section of the nation's AHCs, with a membership of both eminent and emerging institutions from the private and public sectors; approximately two thirds of the 125 accredited allopathic medical schools in the United States are UHC members. In this article, we report the comments and opinions of prominent researchers from these institutions who responded to our survey.
We carried out the survey to (1) understand the perceived value of clinical research among AHC administrators and faculty; (2) identify incentives for faculty to become involved in clinical research at AHCs; and (3) gain insight into the motivations for and barriers to the conduct of clinical research at AHCs.
How Clinical Research Was Defined
For the purpose of this survey, clinical research was defined as a continuum of studies involving interactions with individual patients, diagnostic clinical materials or data, and/or populations of patients. Common to all of these studies is an interaction between an investigator and a patient/population. For the purposes of this survey, laboratory research using human material (e.g., cell culture research) was not considered clinical research.8 The value of and incentives for three distinct categories of investigation were independently assessed. (The definitions of those categories below are quoted from those used in the questionnaire.)
▪ Translational research—the bench-to-bedside transfer of knowledge, physiology, or pathophysiology, which usually focuses on a small number of patients. This subset would be referred to in the broad sense and may involve diagnostics, therapeutics, or devices.
▪ Clinical trials (investigator-driven and/or industrial in origin)—these employ larger numbers of patients, usually stress therapeutic, diagnostic, epidemiologic, or behavioral investigations, and may be single- or multi-center.
▪ Outcomes research and population-based research—this focuses on the largest populations and investigates the roles of therapeutic strategies, devices, and patient management practices in medical and/or economic outcomes.
Design and Data Analysis
We developed our questionnaire with constructive criticism from clinical research administrators and faculty at the University of Wisconsin—Madison and Loyola University—Chicago. In mid-May 1999, 22 copies of the finalized instrument were sent to specific kinds of faculty and administrators* at each of the 80 UHC institutions. Responses were no longer accepted after July 30, 1999. (Copies of the questionnaire are available from us.)
The data were compiled first for “all responders” and then for each category of responder, excluding the “other” category. Descriptive and statistical data analyses were conducted using standard software. An analysis of the number of responders reporting an “increase” over five years in the institutional investment in the three types of clinical research (broken down by the categories of responders) was conducted using the chi-square goodness-of-fit test, with significant observations confirmed by analysis of standardized residuals. Because the number of deans responding to this particular question was low, a second analysis was performed comparing the responses of the deans and hospital/health system administrators combined versus all other groups combined. Results were considered significant if p < .05.
WHAT WE LEARNED
A total of 358 returned questionnaires were reviewed and analyzed. There were 46 AHCs that were identified as the sources of the responders, a 57.5% institutional response. Responding institutions were represented by one to ten responses per institution (mean, 3.76; median, 3). However, 47% of the 358 responders did not identify their institutions as requested; hence the true response rate from institutions is probably considerably higher than 57.5% but cannot be calculated with certainty. The reason for this lack of identifiers is unclear, but several issues may have contributed. For example, some institutions did not distribute copies to their appropriate affiliates and/or institutional officials. Others lacked the appropriate offices or departments to comply. Some individuals in specific institutions had more than one role (e.g., as both the dean of research and the director of grants and contracts). In addition, some respondents wished to remain anonymous. As a result, both the number of institutions responding and the responses per institution are definitely underreported, and hence it is impossible to calculate a true response rate for these survey results. However, the institutional response rate, when corrected for anonymity, appears to be robust and reasonably representative of U.S. AHCs.
Table 1 shows the percentages of responders in the various academic positions. The total of the numbers of responders listed for the positions is more than the total number of responders because some responders indicated that they held more than one of the positions. Although the responders spanned all the categories requested, the majority of responders were associate professors (21%) or professors (57%) who had spent at least 21 years (52% of all responding faculty) at their institutions. Thus, those who responded represent the opinions of relatively senior faculty members deeply committed to their parent institutions and to the continued conduct of clinical research. In terms of frequency of response, those involved in clinical trials (39%) were followed by translational researchers (32%) and then outcomes investigators (26%). (It is interesting that the distribution of the responders' perceptions of their AHCs' main clinical research interests corresponded to the above distribution of responders' research involvements.)
Institutional Investments in Clinical Research
Responders were asked to report whether their institutions' investments in clinical research had increased, decreased, or remained the same over the past five years (by research type). Regarding “overall investment,” the majority of responders (50% to 82% across the various academic and administrative titles) noted an increase for each type of research (Table 2). However, “the same” was the largest response for almost all the investment areas (space, protected time, capital expenditures, and staff recruitment) across all three categories of clinical research, with the noteworthy exception of the fact that “increased” was the largest response for clinical-trial administrative support. This finding may possibly have been because the “overall investment” question was asked first, and after further reflection on the other investment types, the responders reported a lesser likelihood of a documentable “increase.” Of interest, protected faculty time was noted as “decreased” across the three types of research by 41 to 49% of the responders.
There were also important differences in the responses of the various subgroups of responders (Table 2). Deans and hospital/health system administrators tended to report greater percentages of “increased” investment than did their corresponding academic department chairs and faculty. This striking difference held across all three types of clinical research. After combining the “increased” responses of the deans and hospital/health system administrators versus all other groups combined, notable differences were observed regarding the following investment types for clinical trials; space (56% versus 34%, respectively), administrative support (81% versus 52%, respectively), and patient recruitment (61% versus 40%, respectively), all p < .05. No statistically significant difference in response rates, with combined groups or not, was seen with respect to the translational or outcomes research types. Protected faculty time was the investment type rated lowest by all respondents for each type of clinical research. Across all three types of clinical research, decreases in protected faculty time were reported 24-46% of the time by the deans and hospital/health system administrators, compared with 36-58% of the time by the chairs and academic faculty.
Institutional Benefits of Clinical Research
Responders were asked to rate selected institutional benefits from the conduct of the three types of clinical research (Table 3). Ratings of 1 to 2 were interpreted to represent “less value,” with scores of 3 to 5 representing “more value.”
In general, the responders felt that all three types of clinical research benefited their institutions in almost all categories evaluated, but to varying degrees. A notable difference in values was observed for the category “reduction of pharmacy costs in [an] increasing managed care environment,” in that outcomes research and clinical trials scored higher (59% and 53% “more value,” respectively, after combining responders) than did translational research (38%). Clinical trials were noted to provide greater value than either translational or outcomes research in terms of patient recruitment (90%, 74%, 74%, respectively), retention (88%, 74%, 77%, respectively), and diversity (80%, 70%, 70%, respectively). Of interest, in the categories “improved local and national prestige,” “public and community relations,” “improved national institutional ranking,” “recruitment and retention of clinical and research faculty,” no discernible difference in values was observed for the three types of clinical research. (These categories are not shown in Table 3). The majority of respondents (more than 60% in all cases) also reported greater value with all three types of clinical research in terms of
▪ technology transfer as key institutional mission,
▪ financial support of research infrastructure,
▪ quicker introduction of new therapies into patient care,
▪ focus on outcomes research in operational improvement,
▪ provision of seed money for future research,
▪ improved disease management and resource utilization algorithms, and
▪ improved network referral patterns.
There was no notable difference in the perceived values among the various subgroups of responders, within and across categories of institutional benefits. Thus, there was general agreement that clinical research activities were important assets for AHCs, and most responders made subtle but meaningful choices (as indicated in Table 3 in the variety of percentages for each type of faculty across benefits), which showed that they had given weight to these choices.
Priorities and Incentives for Clinical Research
Overall, the majority of responders rated priority and incentives as high (3 to 5 on the rating scale) for industry-funded, government-funded, investigator-initiated/industry-funded, and investigator-initiated/government-funded types of research contracts. Industry-funded subtypes were noted to have somewhat lower priority and incentives compared with their government-funded counterparts. Investigator-initiated/government-funded studies had the highest percentage of high (3 to 5) ratings for both priority (87% average) and incentives (73% average) in each type of research. There was no notable difference in the priority and incentives ratings among the various subgroups of responders, within and across all subtypes of clinical research. Of note, the percentages of high ratings for incentives (66% average) were lower than the corresponding priority ratings (81% average) for each subtype of clinical research.
Incentives for Academic Faculty
Responders were asked to indicate which incentives in a list of incentives their institutions offered academic faulty to conduct industry- and government-funded types of clinical research. All the listed incentives received some “true” answers, but the percentages of various types of faculty indicating “true” varied (see Table 4). For industry-funded clinical research, four of the seven listed incentives received “true” ratings from more than 50% of the responders as being offered in their institutions across the three types of clinical research. These incentives were consideration in criteria for promotion, consideration in criteria for tenure, support of a portion of base salary, and grant excess (all or %) revenues versus expenses to support research of the investigator. However, only 38% of the responders said that for industry-funded research, their institutions provided salary support above base. The responders affirmed that their institutions offered academic promotion and tenure to conduct industry-funded translational research (73-81% indicated “true”) somewhat more frequently than they did for conducting either outcomes (65-77%) or clinical trial research (56-69%). There was no notable difference in the choices among the various subgroups of responders within and across all subtypes of clinical research.
Similarly, for government-funded clinical research, four of the five listed incentives received “true” ratings from more than 50% of the responders as being offered in their institutions, across the three types of clinical research, the exception to this generalization being a percentage of indirect cost recovery returned to the investigator/laboratory/department for professional development. There was no notable difference in the choices among the various subgroups of responders, within and across all three subtypes of clinical research.
The incentives “criteria for promotion” and “criteria for tenure” received strikingly higher percentages of “true” responses for government-funded clinical research (88-97%) than for industry-funded clinical research (56-81%). In addition, the responders indicated “true” to the assertions that “a percentage of indirect cost recovery to the investigator/laboratory/department to support their research” and “support of base salary” occurred more frequently for government-funded (56-76%) than for industry-funded (46-63%) clinical research. Authorship of papers regarding all three types of clinical research was considered for promotion by 94% or more of the responders.
Clinical and/or Volunteer Faculty
Data for volunteer clinical faculty are not shown, as their responses were generally consistent with those outlined above. However, two aspects of their responses are noteworthy. First, only approximately 25% of the responders indicated that compensation for patient recruitment was available to non-investigators as an incentive for conducting all types of clinical research in their institutions. Second, the volunteer clinical faculty were reported to more frequently serve as sub-investigators (71% “true”) than as principal investigators (39%).
Subjective Comments from Responders
Responders were invited to add additional comments at the end of the questionnaire. Representative comments echoed by several included “Managed care has focused all attention on clinical activity, and there is no time for academic research,” and “Research is essentially dead in our department.” This responder further cited other problems, including “patient lockout” from clinical research by managed care, increasing hiring of “clinician educators” who are not research-oriented, horrendous paperwork and oversight of trials, and difficulty in successfully competing for clinical research grants. While these issues were not addressed in the survey, they were perceived by several of the responders to be important barriers to the conduct of clinical research.
WHAT DO THESE FINDINGS MEAN?
In general, the responses to our questionnaire are considerable good news for the clinical researcher. There was general agreement among most respondents that their institutions benefited in many ways from clinical research and that the benefits are well understood. These include prestige and publication relations for their centers, enhanced patient referrals to their hospitals and health care delivery systems, facilitated faculty recruitment and subsequent retention, improved intellectual quality of life, and increased research revenues. Given these perceived benefits, the highest priorities were placed on government-funded and investigator-initiated projects, these types of research were associated with greater incentives. Moreover, performance of clinical research weighed considerably in promotion and tenure decisions, establishment of base salaries, and, in several institutions, a sharing of the financial benefits of these efforts with the investigators and their laboratories. In addition, the large number of responding associate and full professors with more than 20 years in their institutions strongly suggests that successful clinical investigators are stable, well-funded, and loyal institutional faculty members and full participants in the academic enterprise. Finally, several institutions had made significant investments in their infrastructure to support clinical investigators even at this most difficult and precarious financial juncture in their histories. Thus, these features are ones that indicate the potential for a bright future for clinical research in AHCs. We can report that our own institutions have made significant investments in the area of clinical research, demonstrating that such opportunities and commitments are possible where the priorities are appropriate.
Nonetheless, most of the responders also noted a disparity between the extent of the alleged enthusiasm for clinical research among their administrations and the availability of more tangible incentives to promote its conduct. Although the overall investment in clinical research within these AHCs over the last five years was perceived by their leadership to be increased, upon specific queries for more objective assessments of the specifics of these investments, discrepancies became apparent. These differences between the alleged increased emphasis on clinical research and the realities of actual investment were most notable (and achieved statistical significance) in the responses of deans and AHC leaders compared with those of department heads and academic faculty. A unique and somewhat concerning subset of responses that bolstered this impression of a decreased investment occurred regarding protected faculty time. These decreases could be the result of managed care and other contemporary changes in an AHC that, taken together, have reduced their flexibility in all aspects of teaching and research. These data are compatible with recent articles documenting decreased research productivity in areas of highest managed care competition.3,4 The negative implications of this trend, if continued, for the future conduct of clinical research is obvious.
Additional suggestions from the responders for promoting clinical research included limiting access to experimental therapies (i.e., therapies still under investigation) for their patients as an incentive to participate in clinical trials; creating a pool of funds from indirect recoveries to support new research projects; decreasing the variations in research incentives across departments; and uniformly supporting all types of clinical trials (oncology trials being favored).
This survey also identified significant opportunities for AHCs to provide a wider range of incentives for clinical research to their faculties. Expanding the incentives for clinical investigation in an AHC is clearly possible based on the documented availability and use of these various incentives at other AHCs. It is also clear that industry-funded clinical research, although noted to benefit an institution and its patients, will require that AHCs speed up the usual processes for implementing such research, since it is often so time-sensitive.
Despite the encouraging aspects of our findings, those findings also make clear that much needs to be done to strengthen the place of clinical research in AHCs. We hope this report will stimulate further research on the topic and alert institutional leaders to the kinds of initiatives they should promote to foster all forms of clinical research.