FDA - Industry MDUFA III Reauthorization Meeting
November 3, 2011, 9:20 - 3:00 pm
FDA Switzer, Washington, DC
To discuss MDUFA III reauthorization.
|Malcolm Bertoni||Office of the Commissioner (OC)|
|Ashley Boam||Center for Devices and Radiological Health (CDRH)|
|Nathan Brown||Office of Chief Counsel (OCC)|
|Kate Cook||Center for Biologics Evaluation and Research (CBER)|
|David Miller||Office of Financial Management (OFM)|
|Don St. Pierre||CDRH|
|Ruth Watson||Office of Legislation (OL)|
|Susan Alpert||Medtronic (representing AdvaMed)|
|Hans Beinke||Siemens (representing MITA)|
|Faith Cristol||Quest Diagnostics (representing ACLA)|
|David Fisher||Medical Imaging Technology Alliance|
|John Ford||Abbott Laboratories (representing AdvaMed)|
|Donald Horton||Laboratory Corporation of America Holdings (representing ACLA)|
|Tamima Itani||Boston Scientific (representing MDMA)|
|Mark Leahey||Medical Device Manufacturers Association|
|Joseph Levitt||Hogan Lovells US LLP (representing AdvaMed)|
|David Mongillo||American Clinical Laboratories Association|
|Janet Trunzo||Advanced Medical Technology Association|
Meeting Start Time: 9:20 am
FDA feedback on the Commitment Letter
FDA indicated that they share Industry’s* support for the draft Commitment Letter which both parties have developed together. FDA provided some general comments on the outstanding issues in the Commitment Letter, and indicated that specific language in the form of a counter-proposal would be forthcoming.
In particular, FDA noted that it was developing some additional proposals relating to patient input on patient safety-related issues.
FDA also explained that MDUFMA I included goals for BLAs and BLA Efficacy Supplements; however, no such goals were included in MDUFA II. FDA has continued to meet the MDUFMA I goals for BLA and BLA Efficacy Supplements and suggested adding the goals back in MDUFA III. Industry signaled support for this proposal and said there had never been a conscious decision to omit these goals from MDUFA II.
Finally, FDA indicated concerns with the numerical goals associated with PMA and 510(k) average total time to decisions and with the progressive targets for FDA day goals in Industry’s counter-proposal. FDA noted challenges with meeting goals that reflect immediate improvements, given that additional resources would not be in place to support hiring and training prior to implementation of these goals and that program improvements would only just be initiated during the first year. FDA further stated that the specific quantitative goals must be considered in the context of resources.
FDA agreed to provide a counter-proposal which addresses all information outlined above. Both parties acknowledged that additional textual edits could be necessary prior to finalization of the Commitment Letter.
Justification of FDA Resource Estimates for MDUFA III
In the October 31, 2011 meeting, FDA presented their estimate of resources needed to support the October 26, 2011 proposal. At Industry’s request, FDA followed-up with additional information regarding this estimate, planned allocation of FTEs, and efficiency considerations.
Justification of FDA Resource Estimates for MDUFA III: Resource Estimation Approach
FDA’s approach for estimating resource needs involved the derivation of full-time equivalent (FTE) staffing levels from an evidence-based resource estimation model. The model was developed by Booz Allen Hamilton (BAH) for FDA as part of a broader study of medical product resource needs. The broader study was conducted in response to a series of GAO reports in 1989 and 2009, both of which indicated the Agency did not have an appropriate methodology to understand resources needed for all responsibilities. The goal was to develop a methodology for evidence-based estimation. Objectives were to: develop a framework of FDA’s medical product oversight responsibilities; calculate baseline and future workload associated with oversight responsibilities; develop models to enable calculation and sensitivity analysis of labor, materials, equipment, facilities, and contracted services; and, estimate baseline and future resources and assess functional and performance gaps.
The model utilizes five years of the following historical data: the number of submissions of each type received per year, the number of decisions made on each submission type per year, the number of decisions made within a target timeframe on each submission type per year, and time reporting measures of FTE effort on each submission type per year. Time reporting data included most time reporting codes related to the pre-market review process with the following codes spread proportionally across different submission categories: bioresearch monitoring, guidance development, standards assessment, training, outreach, and Center overhead. The target timeframes were MDUFA II goals where applicable and statutory timeframes for other submissions. From these data, the model calculates the accomplished rate (i.e., the percentage of decisions that were made within the appropriate timeframe) and the labor rate (the average number of FTEs required to complete one workload unit) for each submission type for each year. The model also projects future workload levels and labor rates for each submission type and calculates the FTEs needed to maintain performance as well as to meet a specified planned performance target. FDA shared the main underlying equations of the model with Industry but indicated they did not have clearance to share the model itself.
The model features were fine-tuned to focus on CDRH resource needs under MDUFA and results were cross-checked with alternate estimation methods. Additionally, FDA added 2010 data as model inputs when they became available. Model estimates were then entered into a financial model to obtain total cost estimates.
Industry asked if FDA balanced the additional effort up front for Pre-Submissions with efficiencies on the tail end of the process. FDA explained its analysis that Pre-Submissions do not actually move the review work forward significantly as they do not involve review of data. FDA acknowledged that the intention of the MDUFA III proposal is for Pre-Submissions and other enhancements to result in a reduction in the number of cycles; however, Pre-Submissions are not currently correlated to such a reduction and their future impact is difficult to predict. However, FDA did believe the Pre-Submission process as outlined in the Commitment Letter would provide value to industry. Industry responded that it believes Pre-Submissions would provide value to FDA as well as to industry, and that FDA’s model should account for anticipated agency resource savings. Industry asked if FTE estimates for current performance are based on how long it currently takes an FTE to complete a workload unit or an assessment of how long it should take an FTE to complete a workload unit. FDA replied that the estimate is based upon current performance. Moreover, the independent assessment should include an evaluation of the impact of Pre-Submissions.
Industry asked FDA if it utilized the data from the Booz Allen Hamilton report for the purposes of the PDUFA negotiations or for justifying its budget request to Congress. FDA indicated that it had not used the report for either purpose.
Justification of FDA Resource Estimates for MDUFA III: Allocation of FTEs
FDA provided Industry with a provisional allocation of the proposed 491 FTEs within the Agency and CDRH. ODE and OIVD would receive 295 and 120 reviewers, managers, medical officers, program operations staff, and project managers, respectively. An additional 25 FTEs would be assigned across the Center to address guidance development, finances, and training. CBER would receive 12 FTEs and Office of the Commissioner would receive 39 FTEs. FDA also mapped out how additional FTEs would be applied to the program improvements in the Commitment Letter at the beginning of MDUFA III (in FY 2013) and at the end of MDUFA III (in FY 2017). FDA anticipates most effort will go towards Pre-Submissions, Substantive Interactions, and management oversight. At Industry’s request, FDA clarified that management oversight includes both additional supervisors and project managers. Industry asked FDA for additional data that maps the projected program as a whole, including that currently covered by MDUFA, and that covered by increased fee revenues.
FDA provided additional details regarding how the additional resources would be targeted for reviewers and medical officers. ODE’s current program critical needs include: cardiovascular, anesthesia, orthopedics, neurology, dental, and plastics and reconstructive surgery. ODE plans to reorganize into seven review divisions, each with an average of 38 additional staff. Medical officers will be allocated to target areas where FDA is lacking or has minimal clinical support. Engineering, scientific, and statistical reviewer allocations will be based on normalized workload per reviewer ratios from FDA’s workload management model. OIVD’s current program critical needs include: radiology, companion diagnostics, diabetes care, cancer, biothreat/pandemic, 510(k) workload, and infectious diseases. OIVD will add 20-30 FTEs to each of four review divisions based on professional judgment of need. In terms of program operational support, FDA will target hiring of project managers to facilitate scheduling of meetings, track timeliness, etc., to help reviewers be more efficient in the review process; limited hiring of additional Program Operation Staff to provide program oversight and ensure policy adherence; and hiring of a small HR staff to recruit, hire, and support new staff.
Industry questioned why user fees should fund the reorganization in light of public statements by FDA that such reorganization would be accomplished in 2012. FDA stated that they do not have the resources to implement the reorganization and achieve the program performance Industry desires. FDA explained that they began plans for the reorganization in advance of MDUFA III so that it will be ready for implementation as resources become available. Industry also questioned the relationship between FDA’s resource allocations for Pre-Submissions and substantive interactions and the BAH model. FDA explained that Pre-Submissions were modeled as pre-IDEs with a 10% increase in labor rate to account for the meetings and timelines proposed. Substantive interactions were modeled in a way that recognizes that they are a component of the 510(k) and PMA workload categories. FDA estimated that substantive interaction goals would require an additional 10% effort for PMAs and 2.5% effort for 510(k)s.
FDA and Industry agreed that the Commitment Letter they developed together should improve the program in the long term. Each element of the proposal is intended to work together to create a high probability of overall success. For this reason, both parties would like to avoid having to subtract any elements to drive the resource needs down. FDA acknowledged that their good faith resource estimate includes professional judgment and conservative assumptions. FDA is willing to consider refining their assumptions, but doing so would necessitate risk mitigation measures so that FDA will not be left with insufficient resources to meet performance goals if assumptions about workload or resources prove to be too optimistic. Industry stated its belief that the assumptions used by FDA were unduly conservative, and that these assumptions were unnecessarily driving up FDA’s estimates of resource needs.
Justification of FDA Resource Estimates for MDUFA III: Efficiency Considerations
FDA described the manner in which the following potential efficiencies were considered in their derivation of the estimate of resource needs: submission acceptance criteria, e-copy, Pre-Submission, and Independent Assessment.
FDA and Industry agreed that submission acceptance criteria should lead to efficiencies in the long term as less reviewer time will be spent on incomplete applications. FDA explained that associated savings were incorporated into the model by adjustments to labor rates. For example, FDA estimated a reduced labor rate for 510(k)s by evaluating historical data on 510(k)s with single-cycle review and adding back a small amount of effort for screening. Industry questioned whether FDA had underestimated these efficiencies.
FDA asserted that mandatory use of e-copy should lead to modest time savings at the time the review is initiated and substantial cost savings from contractor services to scan paper documents. FDA noted this will require some initial investment in IT to create a system for housing e-copies, maintaining embedded links and bookmarks, and facilitating efficient reviewer access.
FDA suggested that a more structured Pre-Submission program should provide significant efficiencies and predictability for applicants, but not necessarily for the Agency. Specifically, significant effort is required for FDA to provide increased transparency and consistency under the proposed Pre-Submission process. Industry asked if FDA believes this will improve the quality of submissions. FDA explained that when the program is utilized and FDA’s advice is followed, it will improve submission quality; however, the program is voluntary. Industry asserted that Pre-Submissions should help FDA significantly by reducing the number of cycles needed per review. FDA agreed that some efficiencies would be gained and stated that modeling for 510(k)s omitted all submissions with more than three cycles as a surrogate for eliminating one cycle on 510(k)s with Pre-Submissions. Industry expressed concern that FDA was underestimating the efficiencies and overestimating added costs. Industry expressed concern with the costs FDA associates with Pre-Submissions relative to the efficiencies they expect to gain. FDA explained that the cost is driven by assumptions regarding Pre-Submission quantities; alternative means for addressing workload unknowns could drive the cost down.
FDA stated that the planned independent assessment is an important part of the program; however, FDA cannot predict precisely where, when, and to what extent efficiencies will be realized.
When all of the above efficiencies are considered together in the model, their net impact was a savings of 38 FTEs while meeting planned performance goals. Industry requested itemization of costs, savings, and net benefit for each efficiency. FDA explained that the modeling did not consider the marginal cost or savings of each individual potential efficiency, but instead considered all potential efficiencies collectively. FDA urged caution in looking at the marginal improvement of particular commitments, as the program is a comprehensive set of interlocking initiatives. FDA also reiterated their willingness to reconsider the conservative assumptions in the model, provided associated risks are mitigated. Industry again expressed concern that FDA’s overly conservative assumptions were unnecessarily driving up FDA’s estimated costs, and that FDA’s additional costs would likely be much lower.
FDA provided a high-level comparison of the April 13 th and October 24 th proposals to explain why the latter proposal package requires more resources. While the April 13 th proposal included substantive interactions and modest performance improvements to FDA day goals, it did not include a structured Pre-Submission program, improvements to FDA day goals, “no submission left behind” treatment of submissions that miss goals, or a commitment to shared outcome goals on total time to decision.
FDA presented 510(k) and PMA performance data since 2001 to illustrate their perspective that the program has been operating under stress for a long time. Data show that when FDA focussed on improving 510(k) performance in 2004 - 2006, PMA performance declined. When FDA focussed on improving PMA performance in recent years, 510(k) performance declined. FDA believes that this behavior reflects a system that is under-resourced and that capacity must increase to improve consistency, predictability, and transparency. Industry asserted that the recent decline in 510(k) performance is due in large part to FDA’s expending resources to conduct an evaluation of the program. FDA replied that the 510(k) program analysis was conducted in response to concerns. Industry also asked if FDA reflected their assessment that the program is under-funded in their request for appropriations, noting their dissatisfaction with being asked to bear the burden of expected budget authority shortfalls. FDA responded that user fees should be additive and result in performance that has a payoff for Industry. FDA believes that the package on the table will have a benefit for industry.
FDA closed with their overall objectives for MDUFA III. FDA believes that the program is not currently on a sound financial footing despite efforts in MDUFA II. FDA and Industry have taken a creative and collaborative approach to try to identify and address the root causes that lead to increases in total time to decision. FDA expressed willingness to discuss assumptions and work with Industry to reach agreement on the right resource level and structure for the program; however, reverting to MDUFA II will not achieve a program that satisfies the Agency or the industry. Rather, FDA would like to make sure MDUFA III represents a good return on investment.
Discussion and Next Steps
Industry indicated their need for additional, detailed information regarding the BAH model. FDA offered the following overview: FDA used a model that was independently developed as a result of a GAO audit. FDA tailored the model for this purpose by focusing on the aspects relevant to medical device user fees. FDA looked at the accomplished rate and labor rate using actual data from the last five years and made some adjustments based on professional judgment regarding how efficiencies may impact the labor rate. FDA assumed low growth in workload. FDA then used the model to estimate resource needs for the performance dictated by the FDA day goals in the MDUFA III proposal.
Industry indicated they still need additional details regarding the model, modifications made to it with associated justifications, and assumptions. FDA therefore suggested that the most expeditious path forward would be for Industry to outline their specific questions in writing to which FDA could respond.
Meeting End Time: 3:00 pm
* For purposes of these minutes only, the term industry means AdvaMed, MITA, and MDMA and does not include ACLA unless specifically noted.