NOT PRECEDENTIAL

UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ___________

No. 23-1601 __________

JIN-PYONG PETER YIM, Appellant

v.

NATIONAL INSTITUTES OF HEALTH

____________________________________

On Appeal from the United States District Court for the District of New Jersey (D.C. Civil Action No. 3:21-cv-07031) District Judge: Honorable Zahid N. Quraishi ____________________________________

Submitted Pursuant to Third Circuit LAR 34.1(a) September 26, 2023 Before: HARDIMAN, PORTER, and FREEMAN, Circuit Judges

(Opinion filed: October 13, 2023) ___________

OPINION* ___________

* This disposition is not an opinion of the full Court and pursuant to I.O.P. 5.7 does not constitute binding precedent. PER CURIAM

Jin-Pyong Peter Yim, an independent journalist, filed a Freedom of Information

Act lawsuit against the National Institutes of Health. NIH issues non-binding guidelines

for treating patients with COVID-19, with the approval of a panel of experts. Yim

believes that NIH issued a guideline on the use of ivermectin, an antiparasitic drug,

without an approving vote by the panel. His FOIA suit is aimed at getting NIH to admit

that it did so. The District Court held that NIH’s response to Yim’s FOIA request was

satisfactory and granted summary judgment to NIH. We agree and will affirm.

I.

In response to the COVID-19 pandemic, the National Institutes of Health created

the COVID-19 Treatment Guidelines, a set of non-binding recommendations for

clinicians treating patients with COVID-19. NIH develops the guidelines through its

COVID-19 Treatment Guidelines Panel, a panel of experts appointed by the agency. “To

be included in the Guidelines, a recommendation statement must be endorsed by a

majority of [the Panel’s] voting members; this applies to recommendations for and

against treatments and cases where there is insufficient evidence to recommend either for

or against treatments. Updates to existing sections that do not affect the rated

recommendations are approved by Panel co-chairs without a Panel vote.”1

1 COVID 19 Treatment Guidelines: Guidelines Development, National Institutes of Health (last updated December 1, 2022), https://www.covid19treatmentguidelines.nih.gov/about-the-guidelines/guidelines- development/ 2 On January 14, 2021, NIH released a recommendation statement about the

antiparasitic drug ivermectin.2 The statement explained the reasoning behind the

recommendation, briefly summarized the then-existing clinical data, and concluded:

“[t]he COVID-19 Treatment Guidelines Panel (the Panel) has determined that currently

there are insufficient data to recommend either for or against the use of ivermectin for the

treatment of COVID-19.”3, 4

Independent journalist Jin-Pyong Peter Yim is convinced that NIH added this

2021 ivermectin recommendation to the Guidelines without the required vote of the

Panel. To test his theory, he filed a request for records under the Freedom of Information

Act. His request sought “All updates to the Coronavirus Disease 2019 (COVID-19)

Treatment Guidelines that were endorsed by a vote of the Panel. (Date Range for Record

Search: From 01/01/2012 to 01/28/2021)”. The only update to NIH’s guidelines during

that time was the recommendation on ivermectin. So the ivermectin recommendation

would be responsive only if it had received a Panel vote.

2 The COVID-19 Treatment Guidelines Panel’s Statement on the Use of Ivermectin for the Treatment of COVID-19, National Institutes of Health (last updated January 14, 2021), https://files.covid19treatmentguidelines.nih.gov/guidelines/archive/statement-on- ivermectin-01-14-2021.pdf 3 Id. 4 NIH’s current ivermectin Guideline states that “The Panel recommends against the use of ivermectin for the treatment of COVID-19”. COVID 19 Treatment Guidelines: Ivermectin, National Institutes of Health (last updated March 6, 2023), https://files.covid19treatmentguidelines.nih.gov/guidelines/section/section_94.pdf 3 NIH missed its statutory deadline to answer Yim’s request, and Yim filed a private

FOIA lawsuit on March 31, 2021 to compel the agency to respond. Yim apparently

expected NIH to inform him that it had no responsive records, thereby tacitly admitting

that its ivermectin recommendation was made without a Panel vote. But to Yim’s

consternation, when NIH responded to the request by email on April 23, it stated that the

information he had requested not only existed but was publicly available on the NIH

website. The email included a link to the agency’s 2021 ivermectin recommendation.

NIH also offered to print out and send Yim the recommendation if he was unable to

access it online.

Yim protested, responding by email that “NIH must confirm that the record I

requested does not exist.” The Assistant U.S. Attorney representing NIH then emailed

Yim: she reiterated that the information Yim sought in his FOIA request did in fact exist

and was publicly available, explained again where to find it, and offered to help him

access it if needed. Undeterred, Yim replied that “NIH’s response is not acceptable. I

insist that NIH confirm that the record that I requested does not exist.” The AUSA again

assured him that it did and offered to help him access it. But Yim stated, “I remain

convinced that my case against NIH is strong.”

Yim and the AUSA continued to communicate but were unable to resolve the

matter to Yim’s satisfaction. Finally, NIH moved for summary judgment. The District

Court found that NIH had conducted a reasonable and good-faith search for records

4 responsive to Yim’s FOIA request and that it had produced all non-exempt records it

found. So the District Court granted summary judgment to NIH. Yim appeals.5

II.

We review a district court’s summary judgment order in a FOIA case using a two-

tiered process. Abdelfattah v. U.S. Dep’t of Homeland Sec.,

, 182 (3d Cir.

2007). First, we decide “whether the district court had an adequate factual basis for its

determination.” Id., quoting McDonnell v. United States,

(3d Cir. 1993)

(citations omitted). If it did, we “must then decide whether that determination was clearly

erroneous.” Abdelfattah, 488 F.3d at 182 (citations omitted). Under this standard, we will

reverse only “if the findings are unsupported by substantial evidence, lack adequate

evidentiary support in the record, are against the clear weight of the evidence or where

the district court has misapprehended the weight of the evidence.” Id. (quoting Lame v.

U.S. Dep’t of Justice,

(3d Cir. 1985)).

III.

To meet its disclosure obligations under the FOIA, an agency must (1) conduct a

reasonable search for responsive records and (2) produce the non-exempt records that it

finds. See Abdelfattah,

, 182–86. Yim does not contend that NIH is

withholding responsive records; he disputes whether the records he received are in fact

responsive to his FOIA request. So our first task is to decide whether the District Court

5 We have jurisdiction under

. 5 had an adequate factual basis for its determination that NIH’s search for responsive

records was reasonable.

For an agency’s search to be reasonable, the agency must show that it made a good

faith effort to conduct a search for the requested records, using methods which can be

reasonably expected to produce the information requested. Oglesby v. U.S. Dep’t of

Army,

(D.C. Cir. 1990).6 Here, NIH has searched for and produced the

only record that could possibly be responsive to Yim’s request: the 2021 recommendation

statement about ivermectin. If the statement was endorsed by a vote of the Guidelines

Panel, it is responsive to Yim’s request.

We conclude that the NIH FOIA Officer had sufficient reason to believe that the

ivermectin statement was responsive to Yim’s request. On its face, the statement appears

responsive because it purports to be a recommendation of the COVID-19 Treatment

Guidelines Panel. See note 2, above. And because it was a new recommendation finding

insufficient evidence for or against a particular treatment, NIH procedures would have

required that the ivermectin recommendation be subject to a majority vote of the panel.

See note 1, above. Under these circumstances—and without any evidence provided by

6 Agencies typically meet their burden by submitting affidavits or declarations describing how the search was conducted—such as the locations that were searched and the search terms that were used—and affirming that all files likely to contain responsive materials were searched. Abdelfattah, 488 F.3d at 182. NIH’s declaration does not contain such details. See NIH Supp. App’x, ECF. No. 20 at SA 038–042. But the parties agree that the only record that might even possibly be responsive to Yim’s request is the one that NIH searched for and produced: the 2021 recommendation statement about ivermectin. So the lack of these details in NIH’s declaration does not bear on the reasonableness of its search. 6 Yim to suggest otherwise—the NIH FOIA Officer was reasonable in believing that the

record was what it purported to be, and in assuming that NIH had followed its own

procedures in creating it. So the District Court had an adequate factual basis for its

determination that NIH’s search for records was reasonable.

Because the District Court had an adequate factual basis for its determination, we

will disturb that determination only if Yim has shown that it was clearly erroneous.

Abdelfattah, 488 F.3d at 182. He has not.

First, Yim argues that the District Court erred “because there is uncertainty as to

the existence of the Requested Record.” ECF No. 10 at 12. As evidence of this, he states

that (1) at least one email from NIH contained an incorrect hyperlink to the 2021

ivermectin recommendation; (2) a link to a website is not an acceptable form of

producing a record because a website’s contents can change; and (3) there are slight

differences in wording between the cover letters and emails that NIH sent to him—such

as a cover letter that notes the date range of his FOIA request and an email that does not.

See id. at 12–15. These arguments are unavailing. Yim admits that on September 1, 2021,

NIH sent him the record he requested, in the form that he had requested it. Joint App’x,

ECF No. 8 at JA 54 (“NIH provided the record in the format requested by Yim”). Under

the FOIA, that is all that the agency need do. Second, Yim argues that the trivial

differences between NIH’s cover letters and emails constitute a genuine issue of material

fact. ECF No. 10 at 15–16. As we have just explained, they do not. Finally, Yim argues

that the District Court erred when it ruled that its authority under the FOIA was limited to

7 requiring an agency to produce records. ECF No. 10 at 16–17. Yim appears to argue that

a court can also compel an agency to confirm to a FOIA requester that it has no records

responsive to a FOIA request. We need not decide that issue because NIH has produced

records in response to Yim’s request.

Ultimately, Yim has produced no evidence suggesting that the agency’s search for

records was unreasonable, or that the records it produced are unresponsive to his request.

His underlying belief—that NIH is deceiving the public about whether its 2021

ivermectin recommendation was approved by its panel of experts—is based on

speculation, not evidence. Under the Freedom of Information Act, more than speculation

is required.

IV.

For these reasons, we will affirm the judgment of the District Court.

8