2022-23378. Agency Information Collection Activities; Submission for Office of Management and Budget Review; Comment Request; Substances Generally Recognized as Safe: Best Practices for Convening a Generally Recognized as Safe Panel

AGENCY:

Food and Drug Administration, HHS.

ACTION:

Notice.

SUMMARY:

The Food and Drug Administration (FDA) is announcing that a proposed collection of information has been submitted to the Office of Management and Budget (OMB) for review and clearance under the Paperwork Reduction Act of 1995.

DATES:

Submit written comments (including recommendations) on the collection of information by November 28, 2022.

ADDRESSES:

To ensure that comments on the information collection are received, OMB recommends that written comments be submitted to https://www.reginfo.gov/​public/​do/​PRAMain. Find this particular information collection by selecting “Currently under Review—Open for Public Comments” or by using the search function. The title of this information collection is “Substances Generally Recognized as Safe: Best Practices for Convening a GRAS Panel.” Also include the FDA docket number found in brackets in the heading of this document.

FOR FURTHER INFORMATION CONTACT:

Rachel Showalter, Office of Operations, Food and Drug Administration, Three White Flint North, 10A-12M, 11601 Landsdown St., North Bethesda, MD 20852, 240-994-7399, [e-mail].

SUPPLEMENTARY INFORMATION:

In compliance with 44 U.S.C. 3507, FDA has submitted the following proposed collection of information to OMB for review and clearance.

I. Background

Best Practices for Convening a Generally Recognized as Safe Panel

OMB Control Number 0910-NEW

This information collection supports FDA's implementation of Agency guidance. In 2017, FDA developed and published for comment a draft guidance entitled “Best Practices for Convening a Generally Recognized as Safe Panel,” ( https://www.fda.gov/​media/​109006/​download) which, once finalized, would assist persons who choose to convene a panel of experts in support of a conclusion that the use of a substance in food is generally recognized as safe (GRAS).

The Federal Food, Drug, and Cosmetic Act (FD&C Act) requires that all food additives (as defined by section 201(s) (21 U.S.C. 321(s)) be approved by FDA for their intended use in food before they are marketed. Section 409 of the FD&C Act (21 U.S.C. 348) establishes a premarket approval requirement for “food additives.” Section 201(s) of the FD&C Act provides an exclusion to the definition of “food additive,” and thus from the premarket approval requirement, for uses of substances that are generally GRAS by qualified experts.

The GRAS provision of section 201(s) of the FD&C Act is implemented in 21 CFR part 170 for human food and 21 CFR part 570 for animal food. The regulations provide the criteria for when the use of a substance in food for humans or animals is GRAS. Part 170, subpart E and part 570, subpart E provide the procedure under which a person (also referred to as the “proponent” of a GRAS conclusion) may notify FDA about a conclusion that a substance is GRAS under the conditions of its intended use in human and/or animal food. This includes a standard format for the submission of a GRAS notice. The information submitted to us in a GRAS notice is necessary to allow us to administer efficiently the FD&C Act's various provisions that apply to the use of substances added to food; specifically, whether a substance is GRAS under the conditions of its intended use or whether it is a food additive subject to premarket review and approval by FDA. To support a GRAS conclusion, a proponent may convene a panel of qualified experts to provide evidence that generally available safety data and information about the intended use of the substance in food are generally accepted among experts, which is one of the criteria for eligibility for GRAS status (81 FR 54959 at 54975; August 17, 2016).

From 2008 to 2010, the Government Accountability Office (GAO) conducted a study related to ingredients used in human food on the basis of the GRAS provision in section 201(s) of the FD&C Act. In 2010, GAO issued a report (the GAO report ^[1] ) that included recommendations for FDA. Of relevance here, the GAO report recommended that FDA develop a strategy to minimize the potential for conflicts of interest among GRAS panel members, including issuing guidance for companies on conflict of interest, and we requested comment on issuing such a guidance in our reopening notice (75 FR 81536 at 81542; December 28, 2010). In the GRAS final rule, we stated our intent to issue such guidance (see Response 125, 81 FR 54959 at 55026). This guidance recommends an assessment for conflict of interest and the appearance of conflict of interest as part of the best practices for convening a GRAS panel and would address the final outstanding GAO recommendation for FDA from the 2010 report.

The guidance document recommends specific content elements pertaining to recordkeeping and third-party disclosure. The guidance explains a recordkeeping recommendation for proponents to develop a one-time, written GRAS panel policy record describing how it will convene a panel. The proponent creates the written policy to fit its needs. The guidance document discusses a third-party disclosure recommendation for prospective panel members to provide vetting information to proponents, to ascertain expertise and reduce risk of bias. The guidance document also explains a recordkeeping recommendation for proponents to document the application of the GRAS panel policy to each GRAS panel member as part of the vetting process. Respondents do not submit to FDA the recordkeeping or third-party disclosure information.

The guidance will assist respondents convening a GRAS panel. The information collection recommendations (establishment of a written GRAS panel policy, solicitation of information from prospective GRAS panel members about potential conflicts of interest and other sources of bias, and documentation of the application of the GRAS panel policy to each GRAS panel member) would help the respondent to identify GRAS panel members who have appropriate and balanced expertise and reduce the risk that bias (or the appearance of bias) will affect the credibility of the GRAS panel's output.

Description of Respondents: Respondents to this collection of information are persons (“proponents”) who are responsible for a conclusion that a substance may be used in food on the basis of the GRAS provision of the FD&C Act when such persons convene a GRAS panel to independently evaluate whether the available scientific data, information, and methods establish that the substance is safe under the conditions of its intended use in human food or animal food. Respondents would also include members and prospective members of GRAS panels. We estimate that there are 1,260 such respondents as discussed more fully below. The term “GRAS panel” is defined as a panel of individuals convened for the purpose of evaluating whether the available scientific data, information, and methods establish that a substance is safe under the conditions of its intended use in food.

In the Federal Register of November 16, 2017 (82 FR 53433), we published a draft guidance requesting public comment on the proposed collection of information. We received 13 comments with almost half being responsive to PRA topics. All comments were considered even if they were not fully captured by our paraphrasing in this document. Most comments communicated general support for the information collection. Comments articulated that the guidance would promote uniformity of practices for industry and mitigate potential conflicts and biases. One comment expressed that the recommendations for preparing a GRAS panel policy are not unusual or burdensome to proponents, although another comment believed that establishment and implementation of a written GRAS panel policy would be burdensome. However, the comment recognized that the proponent drafts the written policy, which allows the proponent to tailor the policy to itself and make the policy broad enough to cover the wide range of issues it may encounter. The comment further stated that many in industry already employ policies and procedures recommended in the guidance. Another comment believed that the time estimates to perform the information collections are reasonable. No other comments were received disputing the need for the information, the accuracy of our burden estimate, or ways to minimize burden. Although we are preparing to finalize the guidance document to clarify discussions around evaluating and managing conflicts of interest and appearance issues, to emphasize that a GRAS panel is not necessary, and providing additional background information regarding the value of a GRAS panel in providing evidence to support the “general acceptance” aspect of the criteria for eligibility for GRAS status through scientific procedures, none of the revisions pertain to the information collection recommendations discussed in our 60-day notice.

Since the publication of the 60-day notice, we have further considered the burden estimate and adjusted it based on updated information available to us from FDA's GRAS Notices Inventory and the Independent GRAS Conclusion Inventory Database (Refs. 1, 2, and 3).

The records recommended in the guidance to be maintained by a proponent include a one-time information collection burden pertaining to a written GRAS panel policy to govern the assembly and conduct of a GRAS panel. The records recommended in the guidance also include annual information collection burdens pertaining to documenting the application of the written GRAS panel policy to each member of a GRAS panel convened in a given year and collecting information from prospective members of the GRAS panel to conduct the vetting process as detailed in the written GRAS panel policy. Finally, the guidance recommends that a GRAS panel provides a written report of its findings; however, we consider a written GRAS panel report as customary business practice that is already being created by GRAS panels and, thus, we do not estimate an annual information collection burden for the creation of a GRAS panel report.

We estimate the burden of this collection of information as follows:

Burden Estimate for Written GRAS Panel Policy Recommendation

Table 1—Estimated One-Time Recordkeeping Burden ¹

Activity	Number of recordkeepers	Number of records per recordkeeper	Total annual records	Average burden per recordkeeping (in hours)	Total hours
Establishing written GRAS panel policy	696	1	696	40	27,840
1  There are no capital costs or operating and maintenance costs associated with this collection of information.

For the purpose of this analysis, we make the conservative assumption that all proponents who document a GRAS conclusion will create a written GRAS panel policy that would apply to GRAS panels convened in the first year that the final guidance would be in effect, as well as to GRAS panels convened in subsequent years. We also assume that these proponents will create a written GRAS panel policy regardless of whether they report the panel's documented GRAS conclusion to FDA in the form of a GRAS notice. Therefore, for the purpose of this analysis we: (1) calculated the number of proponents who have submitted at least one GRAS notice to FDA and (2) estimated the number of proponents who have documented at least one GRAS conclusion without reporting that documented GRAS conclusion to FDA in the form of a GRAS notice.

Using the data in our inventories of GRAS notices submitted for substances intended for use in human food (Ref. 1) and animal food (Ref. 2) during the time period of April 17, 1997, through October 2, 2020, we calculate that 466 proponents submitted at least one GRAS notice for a substance intended for use in human food, and 20 proponents submitted at least one GRAS notice for a substance intended for use in animal food. For the purpose of this analysis, we make the conservative assumption that there will be no overlap between proponents who submit GRAS notices for substances intended for use in human food and proponents who submit GRAS notices for substances intended for use in animal food. Therefore, the total number of proponents who have submitted at least one GRAS notice to FDA is 486 (466 human food proponents + 20 animal food proponents).

We have very little information about the number of proponents who have documented a GRAS conclusion without reporting that GRAS conclusion to FDA in the form of a GRAS notice. To estimate the number of such proponents, we used a publicly available database entitled “Independent GRAS (Generally Recognized As Safe) Conclusion Inventory Database” (Ref. 3), which is a compilation of the results of a consulting company's search of publicly available information in industry trade journals about documented GRAS conclusions for substances intended for use in human food. The oldest entry is for the year 1995. We received the first GRAS notice for substances intended for use in human food in 1998 and, thus, the database covers the entire timeframe during which FDA has been receiving GRAS notices for substances intended for use in human food. As of October 2, 2020, that database recorded that there had been a total of 213 documented GRAS conclusions, with 41 of those documented GRAS conclusions reported to FDA as a GRAS notice and 172 of those documented GRAS conclusions not reported to FDA as a GRAS notice. In contrast, as of October 2, 2020, FDA's inventory of GRAS notices shows that the number of GRAS conclusions reported to FDA during this timeframe was 937, not 41 (Ref. 1). We assume that the reduced number of documented GRAS conclusions that the database recorded as being reported to FDA is due to the mechanism by which the database searches for documented GRAS conclusions ( i.e., publications in industry trade journals). For example, there could be less incentive for a business that reports its documented GRAS conclusion to FDA to publicize that GRAS conclusion through industry trade journals, because the business can publicize FDA's response to the GRAS notice in other ways.

The database attributes the 172 documented GRAS conclusions not reported to FDA to 146 different proponents. However, 62 of these proponents have also submitted a GRAS notice to FDA and, thus, we calculate that the database attributes documented GRAS conclusions to 84 proponents who have not submitted a GRAS notice to FDA (146 proponents listed in the database—62 proponents whom we already counted because they submitted a GRAS notice to FDA). We also make the conservative assumption that the number of proponents who have documented GRAS conclusions without reporting them to FDA since FDA began receiving GRAS notices is twice as high as recorded in the database— i.e., 168 proponents (84 proponents listed in the database × 2).

The publicly available database does not record documented GRAS conclusions for substances intended for use in animal food. In the burden estimate for the approved information collection “Substances Generally Recognized as Safe: Notification Procedure” (OMB control number 0910-0342), we estimated that 100 GRAS notices would be submitted to FDA for substances intended for use in human food and that 25 GRAS notices will be submitted to FDA for substances intended for use in animal food (86 FR 64943; November 19, 2021). For the purpose of our current analysis, we use that 25 percent ratio to estimate that the number of proponents who have documented GRAS conclusions for substances intended for use in animal food without reporting those GRAS conclusions to FDA is 25 percent of the number of proponents who documented GRAS conclusions for substances intended for use in human food without reporting those GRAS conclusions to FDA— i.e., 42 proponents (168 estimated proponents who have documented GRAS conclusions without reporting those GRAS conclusions to FDA × 0.25). We estimate that the total number of proponents who documented GRAS conclusions without reporting those GRAS conclusions to FDA is 210 proponents (168 estimated proponents who have documented GRAS conclusions for substances intended for use in human food + 42 estimated proponents who have documented GRAS conclusions for substances intended for use in animal food).

To estimate the total number of proponents, we are adding 210 estimated proponents who have not reported their documented GRAS conclusions to FDA to the 486 proponents who have already submitted at least one GRAS notice to FDA for a total of 696 proponents who will document a GRAS conclusion (210 non-reporting proponents + 486 reporting proponents). As already stated, for the purpose of this analysis we make the conservative assumption that all of these proponents who document GRAS conclusions ( i.e., 696 proponents) will create a written GRAS panel policy. We estimate that it will take 40 hours to create a written GRAS panel policy, including 8 hours to review relevant, publicly available policies that address conflict of interest and 32 hours to tailor a GRAS panel policy specific to the proponent, using relevant information from such existing policies as appropriate to the needs of the proponent. As shown in table 1, the total one-time burden to create a written GRAS panel policy is 40 hours per proponent × 696 proponents = 27,840 hours.

Burden Estimate for Records Documenting the Application of the GRAS Panel Policy to GRAS Panel Members

Table 2—Estimated Annual Recordkeeping Burden ¹

Activity	Number of recordkeepers	Number of records per recordkeeper	Total annual records	Average burden per recordkeeping (in hours)	Total hours
Application of written GRAS panel policy to GRAS panel members	94	6	564	16	9,024
1 There are no capital costs or operating and maintenance costs associated with this collection of information.

Based on the number of annual GRAS notices submitted to FDA in recent years, we previously estimated that 100 GRAS notices will be submitted to FDA for substances intended for use in human food and that 25 GRAS notices will be submitted to FDA for substances intended for use in animal food (OMB control number 0910-0342; 86 FR 64943), for a total number of 125 GRAS notices submitted to FDA each year. We count each GRAS notice as a single GRAS conclusion, and, for the purpose of our analysis, we assume that a different proponent submits each of these GRAS notices. Therefore, we estimate that the total number of documented GRAS conclusions submitted to FDA on an annual basis is 125 GRAS conclusions and that these GRAS conclusions are submitted by 125 proponents.

We have not previously estimated the annual number of documented GRAS conclusions that are not reported to FDA as a GRAS notice. To estimate such GRAS conclusions, we used the same database (Ref. 3) that we used to estimate the total number of proponents who document GRAS conclusions without reporting the GRAS conclusions to FDA in the form of a GRAS notice. As already stated, the oldest recorded entry in the database is for the year 1995. However, with the exception of that single entry for 1995, the remaining entries are for the years 2001 and beyond. Therefore, we use 20 years ( i.e., from 2001 through 2020) as the number of years covering those documented GRAS conclusions that are not reported to FDA. For the purpose of calculating the annual number of documented GRAS conclusions that are for substances intended for use in human food and are not reported to FDA, we estimate that there are 171 such GRAS conclusions (172 documented, unreported GRAS conclusions for substances intended for use in human food minus 1 GRAS conclusion reported before 2001). We calculate that, on average, the annual number of documented, unreported GRAS conclusions for substances intended for use in human food and recorded in the database is 9 (171 documented, unreported GRAS conclusions/20 years = 8.55 documented, unreported GRAS conclusions per year recorded in the database, rounded up to 9). As with our analysis of the total number of proponents, we conservatively assume that the annual number of documented, unreported GRAS conclusions for substances intended for use in human food could be twice as high as the annual number of documented, unreported GRAS conclusions recorded in the database— i.e., 18 documented, unreported GRAS conclusions for substances intended for use in human food each year (9 documented, unreported GRAS conclusions recorded in the database on an annual basis × 2). As with documented GRAS conclusions that are reported to FDA, we assume that a different proponent is responsible for each documented GRAS conclusion not reported to FDA and, thus, on an annual basis there are 18 proponents who do not report their documented GRAS conclusions for substances intended for use in human food to FDA. We previously estimated that 100 GRAS notices will be submitted to FDA for substances intended for use in human food and that 25 GRAS notices will be submitted to FDA for substances intended for use in animal food (OMB control number 0910-0342; 84 FR 29216). Using that ratio, we conservatively assume that the annual number of documented, unreported GRAS conclusions for substances intended for use in animal food is 25 percent of the annual number of documented, unreported GRAS conclusions for substances intended for use in human food— i.e., 5 documented, unreported GRAS conclusions for substances intended for use in animal food on an annual basis (18 documented, unreported GRAS conclusions for substances intended for use in human food × 0.25 = 4.5, rounded up to 5). We also calculate that there are a total of 23 documented, unreported GRAS conclusions each year (18 documented, unreported GRAS conclusions for substances intended for use in human food + 5 documented, unreported GRAS conclusions for substances intended for use in animal food). We therefore calculate that there are 148 proponents who document a GRAS conclusion on an annual basis (125 proponents who submit their documented GRAS conclusions to FDA in a GRAS notice + 23 proponents who do not submit their documented GRAS conclusions to FDA in a GRAS notice).

We have information about the percentage of proponents who convene a GRAS panel for a documented GRAS conclusion and also submit a GRAS notice to FDA. During the time period April 17, 1997, through October 2, 2020, on average, 57 percent of proponents who submitted a GRAS notice for a substance intended for use in human food, and 55 percent of proponents who submitted a GRAS notice for a substance intended for use in animal food, convened a GRAS panel. We therefore estimate that, on an annual basis, 57 proponents will convene a GRAS panel and submit a GRAS notice to FDA for substances intended for use in human food (57 percent × 100 proponents), and 14 proponents will convene a GRAS panel and submit a GRAS notice to FDA for substances intended for use in animal food (55 percent × 25 proponents). We calculate that the total number of proponents who will convene a GRAS panel and submit a GRAS notice to FDA is 71 proponents (57 human food proponents + 14 animal food proponents). We also assume that all proponents will document the application of a written GRAS panel policy to each member of the GRAS panel.

We have very little information about the percentage of proponents who convene a GRAS panel for a documented GRAS conclusion but do not report their documented GRAS conclusions to FDA in a GRAS notice. For the purpose of this analysis, we make the conservative assumption that all 23 proponents who annually document GRAS conclusions without reporting them to FDA will convene a GRAS panel. Taking into account the estimated number of proponents who convene a GRAS panel and submit a GRAS notice to FDA, and the estimated number of proponents who convene a GRAS panel but do not submit a GRAS notice to FDA, we calculate that the total number of proponents who will convene a GRAS panel and document the application of the written GRAS panel policy to each member of a GRAS panel on an annual basis is 94 proponents (71 proponents who submit GRAS notices to FDA + 23 proponents who do not submit GRAS notices).

Based on the recommendations in the guidance, we assume that all GRAS panels will include at least 3 panel members and that some GRAS panels will include as many as 6 panel members. We assume that a GRAS panel will include 5 panel members on average. We also assume that the proponent will reject at least one individual with applicable expertise due to a conflict of interest and, thus, that 94 proponents will document the application of the written GRAS panel policy to 6 individual GRAS panel members, for a total of 564 documentations of the application of the written GRAS panel policy (94 proponents × 6 panel members). As shown in table 2, we estimate that it will take the proponent 16 hours to document the application of the written GRAS policy to each panel member, for a total of 9,024 hours (564 documentations × 16 hours).

Burden Estimate for Disclosures by GRAS Panel Members to Proponents of GRAS Conclusions

Table 3—Estimated Annual Third-Party Disclosure Burden ¹

Activity	Number of respondents	Number of disclosures per respondent	Total annual disclosures	Average burden per disclosure (in hours)	Total hours
GRAS panel members provide information to the proponents of GRAS conclusions	564	1	564	4	2,256
1 There are no capital costs or operating and maintenance costs associated with this collection of information.

As shown in table 3, we assume that all 564 individuals who are being considered as members of a GRAS panel will each need 4 hours to provide information related to the panel selection and vetting process to the proponent, as detailed in the written GRAS panel policy, for a total of 2,256 hours (564 individuals × 4 hours).

FDA plans to consolidate this collection with OMB control number 0910-0342, “Substances Generally Recognized as Safe: Notification Procedure” which contains the regulatory procedures under which a person may notify FDA about a conclusion that a substance is GRAS under the conditions of its intended use in human and/or animal food and includes a standard format for the submission of a GRAS notice. The revision will add 39,120 burden hours and 1,260 respondents.

This guidance also refers to previously approved FDA collections of information. The collections of information in 21 CFR parts 170 and 570 have been approved under OMB control number 0910-0342.

II. References

The following references are on display with the Dockets Management Staff (see ADDRESSES ) and are available for viewing by interested persons between 9 a.m. and 4 p.m., Monday through Friday; they are also available electronically at https://www.regulations.gov. FDA has verified the website addresses, as of the date this document publishes in the Federal Register , but websites are subject to change over time.

1. FDA (2020). GRAS Notices. Available at https://www.accessdata.fda.gov/​scripts/​fdcc/​?set=​GRASNotices.

2. FDA (2020). Current Animal Food GRAS Notices Inventory. Available at https://www.fda.gov/​animal-veterinary/​generally-recognized-safe-gras-notification-program/​current-animal-food-gras-notices-inventory.

3. AIBMR Life Sciences, Inc. (2020). Independent GRAS (Generally Recognized As Safe) Conclusion Inventory Database. Available at http://aibmr.com/​natural-products-industry-compliance-consultation/​gras-generally-recognized-as-safe-safety-studies/​.

Dated: October 20, 2022.

Lauren K. Roth,

Associate Commissioner for Policy.

Footnotes

1.  United States Government Accountability Office (2010). “Report to Congressional Requestors on Food Safety: FDA Should Strengthen Its Oversight of Food Ingredients Determined to Be Generally Recognized as Safe (GRAS),” Report No. GAO-10-246. Available at https://www.gao.gov/​assets/​gao-10-246.pdf.