ARCHIVED - NIH Regulatory Burden:
VIII. Analysis of Regulatory Issues
Factors Associated with Regulatory Burden
Several crosscutting factors associated with burden emerge in examining the five sets of regulations. In many instances these factors are interrelated and may be synergistic in a negative sense. These factors are important as they define the implementation approach for resolving the issues that have been identified. Those that emerged are as follows:
The specific instances of regulatory burden that have been identified in the five areas under examination are consistently associated with more than one of these factors. In some cases, all four factors may be at play. As NIH develops its strategy for relieving regulatory burden, it must take these factors into consideration.
- Regulations that dictate process often limit flexibility without enhancing results – Many of the regulatory requirements identified as burdensome dictate "how" the research institution is to proceed. Unfortunately, this "required way" was not always seen as the "best way" to achieve the program outcome. Examples are numerous in the areas that were studied. The research integrity regulations require that a case proceed through 3 defined stages, a requirement that institutions argue is inconsistent with its natural flow; requirements for when an IRB should perform a protocol review do not consider the lag that occurs in subject accruals nor do they allow accommodation for unforeseen scheduling difficulties; USDA's rigid caging requirements for group housing of animals is at odds with scientific judgment; and the EPA's rigid storage limits for hazardous wastes do not allow research institutions to accumulate sufficient quantities to make disposal cost-effective. A "one size fits all" requirement has been placed on institutions that is not optimal from their perspective in every case.
This type of regulatory burden is the inevitable consequence of regulatory systems that are designed to control outcome by defining the process that the research institution is to follow. While each of these restrictions undoubtedly has a well-intended purpose, the programs are too complex and the universe of research institutions is too varied to have rigid rules that limit adaptation to changing and variable conditions, and that stifle creativity. As Philip K. Howard argues in The Death of Common Sense it is the result that is important, not the process. Process should facilitate the program objective, not hinder it. President Clinton made the same point in his March 16, 1995 remarks at a reinventing government event focusing on regulatory reform: "…we've been far more obsessed … with process than results."
- Regulations administered by multiple agencies impose inconsistent requirements – In each of the five sets of regulations examined, at least two Federal agencies are involved. In each case, the different requirements imposed by the agencies were identified as burdensome. The differences range from technical annoyances to conflicting requirements having significant programmatic consequence.
The conflict of interest programs administered by NIH and NSF illustrate the point that it is difficult for two Federal agencies to develop completely congruent programs. The NIH regulations and the NSF policy on financial conflict of interest were developed in concert and published simultaneously in the Federal Register. Though most of the requirements are similar, differences still remain. The NIH, because of a provision in legislation, requires that an institution notify NIH when it has identified a conflict and is managing it; NSF, on the other hand requires no such notification if the conflict is being managed. In addition, the NIH regulations, because of provisions in the overall DHHS grants management policies, require that institutions also apply the conflict of interest provisions to subgrantees or subcontractors participating in the research; NSF has no such requirement in its overall grants management policies and therefore its financial conflict of interest provisions apply only to the grantee institution. Other differences also exist.
More burdensome are differences in the human subjects protections regulations promulgated by OPRR, located within NIH, and FDA. Both are agencies of DHHS and, although effort has been taken by the two agencies to reduce the conflicting provisions of the regulations and policies, differences exist and the programs are administered in distinctly different ways. OPRR requires an assurance and generally proceeds on the assumption that the institutions will operate in accordance with that assurance. FDA, on the other hand, assures compliance through inspections. Each approach is consistent with the culture of the Federal organization - OPRR's approach is consistent with NIH's generally collaborative relationship with research institutions and FDA's approach is consistent with the way it exercises regulatory oversight for the programs under its jurisdiction. The net result is two agencies within the same Department administering their human subjects programs consistent with their own culture, and thereby imposing two different processes on the research institutions.
These examples illustrate that even under the best of circumstances, two different agencies administering the same program have at least slightly different regulations and policies, resulting in regulatory burden. NIH and NSF, with a common science mission, and NIH and FDA, operating within the same Department, still impose at least some different requirements on research institutions.
- The regulation of science by non-science agencies often results in additional regulatory burden – An additional level of complexity and burden arises when science is regulated by a non-science agency, because their programs are poorly tailored to research. The five sets of regulations examined offer two striking examples - USDA, which jointly administers animal care and use regulations with OPRR, and EPA, which administers the hazardous waste regulations. Each is problematic for the research community. With regard to animal care and use protections, conflicting provisions exist such as different time requirements for performing continuing review of protocols and different yearly reporting requirements, a phenomenon similar to that noted above that different agencies regulating the same program will almost inevitably impose different requirements. In addition, and particularly troubling to the research community, are requirements that they believe to be inconsistent with scientific findings or inconsistent with how scientific research is conducted. USDA's caging requirements for paired animals are inconsistent with scientific conclusions regarding animal husbandry. The USDA requirements regarding written narratives for alternatives to painful procedures are seen as inconsistent with scientific methodology used in research. USDA inspectors were often cited as adding to regulatory burden because they are sometimes unfamiliar with scientific research and its methodology. Similarly, the hazardous waste disposal regulations administered by EPA were constructed to regulate the chemical industry and impose many requirements, previously enumerated, that are not well adapted to the scientific research environment and may not be the only way to assure the necessary environmental protections.
These examples illustrate the difficulties that arise for research institutions that are regulated by agencies such as USDA and EPA. These agencies do not have a predominant scientific research mission and employ a different mix of expertise than do NIH, NSF, or FDA. As such, it is possible that the impact of their actions on scientific research may not be as well recognized. More critical, however, is that these agencies are responding to a different set of constituencies and a different set of priorities, so that the impact of their regulations on scientific research may be only a secondary consideration. They are regulating conditions on a much broader scale such that research institutions comprise a very small portion of the regulated community. Furthermore, in many cases, the requirements imposed by these agencies that are considered burdensome by the research community are legislatively mandated so that flexibility of the agency to design regulations that are accommodating to research institutions, if they are so disposed, is restricted.
- Better communication between Federal agencies and research institutions can be expected to reduce regulatory burden – Communication, for the purpose of this discussion, encompasses the ability of members of the research community to participate in the formulation of policies that affect them, to have easy access to Federal officials, and to receive clear and consistent interpretations of policy. Members of the research community consistently cited the need for more communication, the need to establish formal mechanisms for communication, e.g., advisory groups, forums, etc., and the need for training of Federal staff to assure consistent interpretation. In fairness, it would appear that in many cases efforts are being undertaken and the problem is, at least partially, the difficulty of communicating across a broad system of research institutions, rather than a lack of effort. Regardless, better communication will result in policy that fully considers its impact on the research institution. It can also reduce the uncertainty and mistrust that was sometimes reported by the research community. It was reported that this uncertainty and mistrust leads some institutions to unnecessarily administer its programs slightly more conservatively than the regulations require in order to avoid a negative review by the Federal agencies.
Next steps - Implementation Principles
A reading of this document confirms that the current system of regulation for each of the five areas is in need of change, and in some cases, dramatic change. At the same time, NIH faces a difficult challenge as it undertakes this initiative. The factors that cause this burden are multiple and systemic. Addressing them will require concerted action and sustained attention if this initiative is to be successful.
The solutions offered by the research community require changes to legislation, regulation, policy, and practice. NIH's strategy for implementation should be guided by at least three considerations - the merit of the proposal, the difficulty of implementing change, and the degree to which NIH, or its parent organization DHHS, has the authority to effect this change. Though perspectives may vary on the relative merit of each proposal presented in this paper, each represents the considered judgment of thoughtful members of the research community and deserves similar serious consideration by the Federal agencies and, where appropriate, the Congress.
Of those judged to be meritorious, some can be implemented more quickly than others. Changes to NIH or DHHS policies and practices, while requiring sustained attention on the part of agency leadership for successful implementation, would seemingly offer the best possibility for immediate success. Legislative and regulatory changes upon which there is consensus on the need for change will take longer, but the fact that it will be a lengthy process is no reason for delay; in fact, it may argue for immediate action given the urgency for change.
Special consideration should be given to legislative, regulatory, or policy changes recommended for programs administered by other Federal departments, i.e., animal care and use and hazardous waste disposal programs. As noted above, these are the most intractable issues. When burdensome requirements such as these are imposed in response to constituency pressures or institutional priorities other than the scientific mission, NIH or DHHS is at a substantial disadvantage since it has little leverage to effect resolution. In addition, legislative and/or regulatory changes need to be carefully considered if the current legislation or regulation is essentially a political compromise between competing interests and constituencies. In such cases, the argument that seeking to amend legislation or regulations opens the possibility of even further restrictions is not trivial. These issues should be carefully weighed and political judgments should be made by those knowledgeable of potential outcomes to assure the optimal result for research. On issues such as these, NIH, as the advocate for the research mission, should enlist the support of key Executive and Legislative Branch allies to provide the necessary leverage and political perspective.
With this in mind, the following principles are offered as NIH begins its program of regulatory reform.
- Take both immediate and long-term action to focus regulatory oversight on results, not process.
- In the short-term, take action to relieve burden resulting from process restrictions that do not facilitate program results; in doing so, act quickly to create momentum for change. Candidates for immediate change are those under the direct control of the NIH or DHHS. A number of candidates exist in the human subjects protections program – reengineer the assurance process; define parameters and then allow IRBs flexibility as to when they conduct protocol reviews; and eliminate restrictions on the use of audio teleconferencing. Policy changes recently published in the Federal Register significantly expand the ability of IRBs to use expedited review procedures for continuing review; an evaluation of the impact of these changes would provide an informed basis for determining whether further action is required.
Similarly, opportunities exist to simplify and improve the investigations of scientific misconduct. Allowing institutions to have, without ambiguity, the primary responsibility and the flexibility to administer research integrity programs would eliminate the duplication of effort and the possibility for institutional conflict that occurs with the current system. At the same time, institutions should be allowed to structure their investigations to facilitate a successful outcome, without being constrained by a requirement that they proceed strictly through three separately defined phases. Initiating action to eliminate the legal disincentives for those individuals or institutions pursuing allegations of scientific misconduct is also essential.
Finally, a systematic examination of the effectiveness and efficiency of the current thresholds for financial disclosure in identifying potential conflicts of interest could set the groundwork for thoughtful improvements to that program. Other changes might also be appropriate.
These actions should be taken within an overall objective of "do no harm" to the achievement of the protections that are intended. However, unless a requirement is directly related to, or facilitates, a program outcome, it should be eliminated. In doing so, institutions are being given authority commensurate with their responsibility to achieve the intended protections.
- Over the long term, foster paradigm changes to implement results-oriented oversight. The longer-term corollary to the actions discussed above, is to foster the development of alternatives that replace process-oriented systems of Federal regulations. The goal of such an effort should be to reduce regulatory burden while ensuring that systems of protections are being achieved. A variety of possibilities have been discussed. The hazardous waste disposal demonstration project that is being proposed by several northeastern universities would test the replacement of selected process-oriented regulatory provisions with a system of performance standards. This could serve as a model for a new regulatory paradigm for hazardous wastes. Members of the research community argue for performance standards in lieu of engineering standards for animal caging. Third-party accreditation systems, preferably results-oriented, might be explored in other areas. Regardless of the model, the goal should be to replace the current process-oriented regulations that have been examined. Unless the philosophy of regulation is changed to become results-oriented, the inflexible process-based restrictions noted above, if eliminated, will only be replaced by others in a relatively short period of time.
While the establishment of such systems should be encouraged, the difficulty of developing them should not be underestimated. In addition, it is counterproductive if new systems do not replace the aspects of the current system that are unnecessarily burdensome. The ultimate outcome must be a reduction in regulatory burden. It is important, however, when making this "burden reduction calculation" to distinguish between the unnecessary burden imposed by inefficient or ineffective regulation and the necessary burden that comes with responsibility. Institutions should be given authority commensurate with their responsibilities. With this responsibility, institutions must also have systems in place that demonstrate that they are successfully providing the intended protections.
- Take concerted action to eliminate conflicting requirements. While holding research institutions responsible for the programs of protection that they administer, all elements of the Federal government have a commensurate responsibility to assure that laws, regulations, policies, and practices place consistent requirements on the institutions.
- NIH/DHHS should take action to eliminate conflicting requirements for programs under its control. High priority should be given to resolving operational, regulatory, and policy differences that exist between OPRR and FDA. Active intervention by agency leadership will be required to further minimize or eliminate regulatory differences that exist. In addition, NIH should closely monitor adherence to its policy directive to assure that DSMBs are being established to monitor data of multi-site clinical trials where patients are at risk, and that aggregated and interpreted data are being provided to IRBs. FDA should also take similar action. Finally, NIH should accord high priority to eliminating the different requirements that are imposed on IRBs by the NIH-sponsored cooperative groups. FDA should also take similar action with regard to the CROs.
- OSTP and the Congress must take action where conflicting requirements cannot be resolved by NIH/DHHS. Particularly problematic are requirements imposed by other Federal departments. This includes USDA's animal care and use requirements that conflict with OPRR's and the restrictions on the transportation of biological materials imposed by the Department of Interior's Fish and Wildlife Service. Similarly, NIH has little independent leverage with EPA and its administration of the hazardous waste disposal program. These are non-scientific agencies that are regulating science. Assistance from within the Executive Office might be required from OSTP to facilitate, or broker, resolution. Particularly encouraging is OSTP's recent decision to become actively engaged in facilitating a dialogue between universities, Federal and industrial research laboratories, Federal and State regulators, and Federal science agencies on the issues surrounding the disposal of hazardous wastes. The recommendations contained in section VII provide a substantial agenda for these discussions.
NIH should continue to work at the staff level with USDA (and AAALAC and FWS, as appropriate) to resolve animal care and use issues, particularly those that require only changes in policy. Candidates are policy changes reducing the number of redundant program and facility reviews and further streamlining of the various annual reports. At the same time, efforts to have OSTP broker discussions with USDA and FWS may offer a more productive outcome and should be pursued as necessary, particularly where regulatory or legislative requirements are in conflict between the agencies, or where requirements reflect perspectives or priorities other than research. The ideal outcome for the research community would be to gain agreement that one organization – OPRR(NIH) - provide oversight over research institutions where the two jurisdictions intersect. Finally, it should be emphasized that for both sets of requirements, many of the most burdensome aspects are legislatively mandated in USDA's Animal Welfare Act and EPA's Resource Conservation and Recovery Act. In these cases, Congressional action will ultimately be required if regulatory relief that will enhance the research mission is to be provided.
- Refocus the role of Federal agencies to give higher priority to partnership functions. Organizations charged with overseeing systems of regulation must continue to maintain appropriate levels of oversight. At the same time, an increasing focus on activities not always seen as traditional elements of an oversight function, will ultimately yield enhanced programmatic outcome. While many of the organizations providing oversight over the regulations that were examined are carrying out some of these partnership functions, the resources that are devoted to them should be proportionally increased.
- Foster model programs of education and training. The need for a broad spectrum of education and training activities consistently emerged in discussions with the research community. While the education and training program should be tailored to the individual program's needs, a specific and comprehensive focus on education and training, including Federal staff training, should be an integral part of most. To be successful, however, the education and training initiatives should be the result of a partnership effort between the Federal government and the research community. As a partner, the Federal government can serve a valuable role as a stimulator and facilitator of this effort. It can also serve as a central resource to develop, or facilitate the development of, common materials identified by the research community as necessary for a successful program.
In developing these programs, it is obviously important for the subject matter to be comprehensive and to reach the widest possible audience. A point emerging consistently from various discussions is the need to reach all parties who administer, or are affected by, the various programs of protections.
The education programs should also be multi-modal and include as many learning strategies as are appropriate to the topic. This could include group learning situations such as workshops, case-based seminars, and formal instruction at the research institution. The use of technology such as videotapes, CD-ROMs, computer-based learning programs, including access via the World Wide Web, or satellite technology might be appropriate to best utilize the time of busy professionals and reach the widest possible audience. Leadership by, and collaboration with, professional organizations is particularly valuable if these efforts are to be successful.
Finally, the program of training and education should be regularly and formally evaluated to assess its effectiveness. This evaluation should be a broad-based assessment of those participating to assure that the intended objectives are being met.
- Encourage, identify, and disseminate best practices. The diversity of the research community makes it an ideal laboratory for developing "best practices." An active component of the Federal role in administering systems of regulation should be the encouragement of innovative models, and the identification and dissemination of "best practices." A reallocation of resources so as to expand the role of the Federal agencies to facilitate this type of effort should be a goal.
- Facilitate paradigm shifts. Federal agencies should facilitate, not unilaterally develop, proposed paradigm shifts. It is essential that these be collaboratively developed with the research community. Professional organizations might be particularly helpful in this regard. Models developed might be tested through the Federal Demonstration Partnership or through other pilot testing. A specific allocation of resources should be identified by Federal agencies for the specific purpose of facilitating these changes.
- Expand existing channels of communication between the research community and the organizations providing oversight. Establishing more effective forums for communication is an essential element of partnership but is separately identified to emphasize both its importance and the priority accorded to it by the research community. It is recommended that committees or advisory bodies, whose members are drawn from the research community, be formed and act in an advisory capacity to the organizations administering each of the sets of regulations that have been identified. These could be either formal or informal depending on the circumstances. Where multiple agencies are involved, they should be advisory to the multiple agencies, if feasible. The purpose of these committees or advisory bodies would be to act as a resource as the agency formulates policy and to serve as a policy advisor as it administers its oversight function. Such committees or advisors might be charged with assisting in the streamlining efforts being undertaken in each of the areas.
Other venues should also be explored to increase interaction between Federal agency staff and the research community such as additional conferences, workshops, and short-term personnel exchanges (IPAs). These additional communication links are an essential element of a trust relationship that must exist if the Federal agencies and the research community are to collaborate successfully to reduce regulatory burden.
- Formally recognize institutions that have established superior programs of protection. Current regulatory systems have a tendency to formally impose the same requirements on all research institutions, regardless of their performance, and to set requirements based on the lowest performer. Instances of formal recognition exist, e.g., when a research institution that is not accredited by AAALAC files an assurance with OPRR, it must submit the latest semi-annual report of its IACUC evaluation; AAALAC accredited institutions do not have to provide this material. However, such instances of formal recognition are the exception rather than the rule. In streamlining the existing systems, priority should be placed on techniques that formally reward those institutions that are excelling. For example, the frequency of inspections by USDA animal care inspectors or EPA inspectors should be decreased for those institutions that have demonstrated superior performance. This not only provides an incentive for superior performance but also allows the Federal agency to allocate its resources more effectively.