Process Improvement for Interinstitutional Research Contracting

Abstract

Introduction

Sponsored research increasingly requires multiinstitutional collaboration. However, research contracting procedures have become more complicated and time consuming. The perinatal research units of two colocated healthcare systems sought to improve their research contracting processes.

Methods

The Lean Process, a management practice that iteratively involves team members in root cause analyses and process improvement, was applied to the research contracting process, initially using Process Mapping and then developing Problem Solving Reports.

Results

Root cause analyses revealed that the longest delays were the individual contract legal negotiations. In addition, the “business entity” was the research support personnel of both healthcare systems whose “customers” were investigators attempting to conduct interinstitutional research. Development of mutually acceptable research contract templates and language, chain of custody templates, and process development and refinement formats decreased the Notice of Grant Award to Purchase Order time from a mean of 103.5 days in the year prior to Lean Process implementation to 45.8 days in the year after implementation (p = 0.004).

Conclusions

The Lean Process can be applied to interinstitutional research contracting with significant improvement in contract implementation.

Introduction

Sponsored and/or translational research will increasingly require multiinstitutional collaboration and interinstitutional contracts will become an increasingly important prerequisite to such activities. Given institutional concerns about the fiscal solvency of research endeavors, safeguarding of intellectual property and personal health information, mitigating risk, improving compliance, adherence to sponsor requirements, and, maintaining academic visibility, the successful negotiation of interinstitutional research agreements that adequately address the concerns of all involved parties has become more complicated, frequently requiring several months or longer to complete.

Utah has one dominant urban region that provides an arguably unique opportunity to develop denominator‐based community research programs. The two dominant health care providers in Utah are Intermountain Healthcare (or Intermountain, a not for profit corporation and the state's largest healthcare provider) and the University of Utah Health Sciences Center (or UUHSC, the only academic health center). Both entities have significant research enterprises with substantial synergistic potential, in that each can bring an area of particular expertise (discovery research, community‐based health services research) to a collaboration. In fact, this prospect of enhanced research productivity was a major motivation for UUHSC and Intermountain to collaborate on the successful National Institutes of Health (NIH) Utah Clinical and Translational Science (CTSA) Award (named the Utah Center for Clinical and Translational Science, or CCTS).1 The CCTS has developed infrastructures for enhanced collaborative research that include a Master Collaboration Agreement, a Research Education and Training Core, and Clinical Research Core Facilities.1

The UUHSC Department of Obstetrics and Gynecology and the Intermountain Women and Newborns Clinical Program have long recognized these synergistic opportunities and have been jointly engaged in collaborative research projects, supported primarily through participation in NIH‐funded clinical trial networks, beginning well before the development of the current NIHCTSA format.2

However, multiple internal and external issues have evolved and gradually altered the contracting procedures at both institutions. These changes were usually not discussed or coordinated between the two institutions. As a result, the procedures required for interinstitutional funded research added multiple approvals and process requirements that have, on occasion, resulted in financial stresses and administrative misunderstandings for both entities.

In 2013, the UUHSC senior administration initiated an institution‐wide emphasis on Lean Process Improvement designed to engage the work force at all levels to improve efficiency and reduce waste throughout the institution. These efforts were discussed among the aforementioned research staff at UUHSC and Intermountain and their administrative supervisors. A collective decision was made to begin a Lean Process initiative to coordinate and streamline the contracting processes and to reduce the amount of time needed to finalize a contract by developing, mutually institutionalizing, and continuously improving upon, existing processes.

Methods

The Lean Process, which originally defined an incremental manufacturing improvement process, has historically been attributed to the post–World War II Japanese manufacturing sector and specifically to Toyota Motor Company.3, 4 It is a management practice directed at increasing quality and reducing costs through ongoing foci on (1) the creation of value (from the customer's perspective) and (2) reduction of waste and variation in all processes. It addresses management practices and thinking and both empowers and involves employees by providing opportunities to improve processes by utilizing established principles and rules.

Two key tools from the original Lean Process lend themselves particularly well to the healthcare sector. Process Mapping is the procedure by which key stakeholders participate in developing a visual process map. This is a prerequisite for the subsequent identification of waste and delays, improvement in efficiency, and integration of new process steps, with resultant improved throughput.

The second key tool is the Problem Solving Report which provides a structured framework that identifies: (1) the current status of a process (or problem) as well as its root cause analysis, (2) the objective(s) that the process improvement hopes to achieve and the methods by which these objectives will be implemented, and, (3) follow‐up plans and results. This provides the basis for iterative process improvement, as previously unrecognized issues are commonly identified via these procedures. The Problem Solving Report Template used in this process is outlined in Figure 1.

Figure 1.

Representative Lean Process Problem Solving Report.

The primary outcome measure was the time, in days, from a Notice of Grant Award (NGA) by one institution to the issuance of a Contract Purchase Order (CPO). The means of these intervals were compared between the year before, and the year following, the initiation of the Lean Process Improvement. A two‐tailed Student's t‐test was used to compare mean differences, with a p value of 0.05 considered to be statistically significant.

Results

Team members

One of the two primary authors (MV) has served as a clinical researcher across both institutions over the previous two decades and was primarily responsible for identifying the issues to be evaluated by this process. The two primary authors (MV, JL), who were affiliated with the Program in Personalized Health Care (an administrative unit within the Utah CCTS), contacted the relevant research administration personnel at both institutions to discuss the feasibility of subjecting the interinstitutional research contracting process to a Lean Process evaluation. The list of relevant personnel included administration and contracting personnel at both the Department (UUHSC) and Clinical Program (Intermountain) level as well as institutional (UUHSC Office of Sponsored Projects and Intermountain Office of Research) levels, with the belief that all relevant stakeholders should have the opportunity to participate in the proposed process improvements.

Given that all parties expressed a willingness to participate, a series of ground rules for process development were proposed and subsequently accepted. Importantly, the involvement of trained process engineers was acknowledged and accepted by all participants. The following headings conform to those outlined on the Lean Process Problem Solving Report (Table 1).

Table 1.

Ground rules for process development

1. Transparency

2. Willingness to listen without interrupting

3. Willingness to consider all ideas (and there would be no “bad ideas”)

4. Adhere to Lean Process Development guidelines:

a. Focus on reducing waste and variation in all processes

b. Focus on continuous process improvement

5. Willingness and ability to share internal institutional procedures

Problem situation

At the initial group meeting the ground rules were reviewed and an overview of the Lean Principles was presented. This, and all subsequent, meetings were chaired by one of the primary authors (JL) who is not involved in perinatal research but who is familiar with clinical research and contracting procedures. Following iterative reviews of existing process steps and wait times, the group concluded that: (1) the research contacting process between the two institutions took more time than was necessary, (2) neither institutions’ representatives understood the processes in place at the other institution, and, (3) considerable reduction in the time required to transfer funding and to begin research projects should be achievable.

Current situation

Representatives of both institutions agreed to develop Process Maps of their current internal institutional research contract approval procedures. These are presented in Figure 2. Presentation of the two Process Maps to the team clearly demonstrated that both systems had developed complex processes in response to multiple internal and external issues. It was also apparent that neither institution had previously understood the issues facing their counterparts at the other institution. Finally, both institutions agreed that, after reviewing their own Process Maps, multiple opportunities existed for simplifying and streamlining their own procedures.

Figure 2.

Value stream mappings (VSMs) of internal institutional research contract approval procedures at project initiation.

Target/Goal

The group subsequently agreed that it should be possible to develop and mutually institutionalize processes that could reliably reduce the interval from the receipt of a NGA by one institution to the issuance of a CPO to fewer than 30 days.

Root cause analysis

By reviewing the baseline time intervals required for various contracting procedures, the group recognized that the largest single contributor to delay was the legal negotiations required for each individual contract (Figure 3 A). It was also noted that there was variable amounts of “dead time,” defined as intervals from the completion of one step to the active initiation of the next, between each step in the contracting process. Further analysis revealed that no standard contract templates existed between the two institutions and that each contract required de novo collection of background information, budget negotiation, and bilateral legal review.

Figure 3.

Time lines for Research Contracting Consensus, with diameters corresponding to average time required. (A) Pre‐LEAN Project, with subcontract legal negotiations representing the longest average delay. (B) Revised research contract process incorporating standard template language.

A second analysis focused on the vendor–customer business model for which the Lean Process was originally developed. It became apparent that “The Business” in the current analysis was comprised of research administrators from both institutions and that “The Customers” were the investigators attempting to conduct clinical research across both institutions. This important recognition, more than any other, accelerated subsequent process improvements.

Countermeasures

Following identification of these issues, the group sought to develop processes by which they could be improved or resolved. The group recognized that optimizing production efficiency would require a mutually acceptable baseline template for “standard work.” Over several meetings the group developed a revised and simplified research contract process that was acceptable at both departmental and institutional levels (Figure 3 B). The new standardized contract templates and language enabled the group to map out a shared and transparent method by which standardized interinstitutional research contract templates and language would be accepted by both institutions with the intent of decreasing, to the extent possible, the time required for budget negotiations and legal reviews (Figure 4).

Figure 4.

Interinstitutional Decision Tree for contract template development and/or implementation.

A chain of custody template was developed within both business infrastructures in the form of a spreadsheet that has a single editor (JW) and all appropriate research administration personnel from both entities are regularly updated. The ability to collectively track time intervals for each step was considered vital for continuing process improvement activities.

Finally, the recognition that “The Business” actually represented the combined institutional research infrastructures provided substantial incentive to develop and refine processes across the institutions. It was agreed that a collective focus on facilitating The Customers’ ability to initiate, conduct, and conclude clinical research across both institutions was in the best interest of not only the institutions and the researchers but of additional customers, including: (1) funding agencies, (2) those interested in licensing and commercializing new technologies, and, perhaps most importantly, (3) healthcare consumers. Research administrative personnel from both institutions now conduct joint monthly meetings to identify and promptly resolve collaborative research issues.

Implementation

As noted in Figure 1, the Implementation phase of the Lean Process requires decisions about who, what, where, and when are necessary to ensure the implementation and sustainability of the developed processes. In this case study, the mutually acceptable baseline template for standard work, as well as the standardized interinstitutional research contract templates and language, have been developed collectively by the UUHSC Office of Sponsored Projects and the Intermountain Office of Research in conjunction with their respective legal counsels and input from department‐level research personnel.

A chain of custody template, as well as the development and refinement of processes to facilitate research implementation, have been developed by the collective department research personnel (“The Business”) and put to use. In each of these areas, time lines and deadlines have been instituted to ensure that progress continues.

Follow‐up

Sustainability planning is critical for process improvement. It was apparent early in our process that the level of transparent communication between the two research infrastructures had been insufficient for a considerable length of time. The ability to track document progress via regular activity reports not only enhances communication but also provides important data by which to track future progress and look for areas amenable to further improvement.

Regularly scheduled meetings of department‐level researchers across the institutions are continuing. In addition, less frequent, but still vitally important, meetings of the larger group (department‐ and institutional‐level administrators from both institutions) are being scheduled and outcomes are being tracked.

A comparison of intervals (in days) from NGAs to CPOs between calendar 2013 (the year prior to process implementation) and calendar 2014 reveals significant improvement, with mean intervals decreased by over 57% (2013 mean ± SD = 103.5 ± 9.2 (N = 2); 2014 mean ± SD = 45.8 ± 19.8 (N = 8); two‐tailed p = 0.004). However, the mean NGA to CPO interval is still almost 46 days (with an almost 20‐day standard deviation) and bespeaks the opportunity for continued improvement.

These process improvements are now available to other collaborative disciplines across the Utah CCTS. The ability to increase collaboration between our CCTS partners represents an important objective of our CTSA's efforts to enhance research collaboration between our partners and between CTSAs.

Discussion

We used the Lean Principles model to identify and implement improvements in the interinstitutional research contracting process between two member institutions in a CTSA partnership. The Process Mapping exercise was particularly instructive in this case, as it provided substantial insights into the infrastructure functions of the partner institutions. This exercise identified several obvious inefficiencies; addressing these has led to the development of mutually acceptable baseline templates for standard work (contract amendments, purchase orders, etc.) as well as standardized interinstitutional research contract templates and language. The latter procedures were developed in conjunction with both institutions legal counsels.

Although this process was initiated by the two primary authors in response to an improvement opportunity that was identified between the two CCTS participants, the engagement of research administration personnel from both institutions was necessary and critical to the project's success. It has been noted that process inefficiencies sometimes result from well‐intended high‐level decisions that were made without thorough understanding of their actual impact on work flow.5 This bespeaks the importance of the support and acknowledgement of the research team's administrative superiors who allow them to participate in these exercises.

The development of a chain of custody template was facilitated by the collective realization that the actual business entity is the combination of research administrators from both institutions and that the customers are the clinical investigators who wish to conduct research across the CCTS member institutions. The implementation of a template that can be viewed and updated by research administrators from both institutions is important both for tracking current timelines and deadlines and for identifying future opportunities for continued process improvement.

Another benefit of these collaborative exercises is the opportunity to address real and perceived imbalances in the control of power and resources between academic and community partners. These collaborations have the additional benefit of facilitating the dissemination of new knowledge and thus its translation into practice and policies. Although academic–community research partnerships can also be based on consultant and mentoring models,6 we believe that a collaborative approach is the most likely to be sustained long term and to produce the desired synergistic outcomes. Involvement of trained process engineers was critical to our project's success and should be included in other projects whenever possible.

The development and maintenance of an enhanced “customer service” focus will be critical for facilitating subsequent interinstitutional research contracting processes. A readily accessible and up‐to‐date “one stop source” for research information, along with coordinated educational and advocacy efforts among participating institutions will help jump‐start new projects by minimizing start‐up inertia caused by complex processes and facilitating engagement of more Customers, particularly new researchers.

Consistent with the Lean Process paradigm, we must continue our pursuit of additional opportunities for process improvement. Although progress has been made toward the initial goal of reducing the NGA to CPO interval to <30 days, that objective has, at least as of the time of submission of this manuscript, not yet been achieved. We are developing process flow diagrams and “smart” tip sheets and will be expanding the use of agreed‐upon templates to other interinstitutional research collaboration, with a particular emphasis on the multiple “internally funded” (alternatively described as “unfunded”) projects that currently lie outside the scope of both institution's research administrations. In addition, continuation of the process will integrate into both institutions an acknowledgement of the requirements of time, communication, and commitment to ensure that small issues are recognized and addressed early and do not escalate into larger, more time consuming obstacles.

We also recognize that research ideas that lead to ad hoc collaborations frequently develop with short timelines. Groups of institutions that comprise research networks (including CTSA institutions) or that have frequent collaborations across multiple disciplines should consider developing standardized templates for interinstitutional contracting to facilitate such collaborations.  

Likewise, research networks or multidisciplinary research institutions that have any reasonable likelihood of industry collaboration could consider developing either federated verbiage (consensus contractual boilerplate language to be used by all network members) or a centralized contacting entity that could negotiate on behalf of all network members. This latter centralized contacting entity could be either a network member—most frequently a data coordinating center—or a separate administrative entity.

Because there are no “national standard” interinstitutional research contract templates, research institutions should consider cataloging research sponsor preferences, including contract templates, major historic negotiation issues (and their resolutions), F & A rate certifications, unallowable expenses, startup cost requirements, progress and financial progress report requirements, invoicing requirements, and payment histories. Collective access to these, and doubtless additional, resources can be useful for facilitating interinstitutional research contracting progress.

In conclusion, we believe that this case study confirms the applicability of the Lean Process to the interinstitutional research contracting process. Particularly important in any scenario is the willingness of upper‐level management to support both these processes and the products that they generate. While the procedures described in this report worked well for our two institutions, we recognize that different approaches would likely be required for other locations and situations, including academic–industry collaborations.

Conflicts of Interest

The authors report no conflicts of interest.

Funding

NIH/NCATS 1UL 1TR000105.