BUCKLEY, Circuit Judge:

Dr. Charles McCutchen filed a request under the Freedom of Information Act for a list of all investigations of scientific misconduct undertaken by the Department of Health and Human Services' Office of Scientific Integrity. The Department released this list, but in all cases where there had been no finding of wrongdoing, it deleted the names of both the investigated scientists ("respondents") and the individuals who had made the allegations of misconduct ("complainants"). After Dr. McCutchen sued to compel disclosure of these names, the district court ordered the Department to disclose those of the respondents but permitted it to withhold those of the complainants. The Department appeals and Dr. McCutchen cross-appeals from this order. Because we find that the privacy interests of both the respondents and the complainants outweigh the public interest in disclosure, we reverse the ruling requiring disclosure of the respondents' names and affirm the ruling approving the withholding of the complainants' names.

I. BACKGROUND

A. Freedom of Information Act

We recently observed that "[t]he purpose of the Freedom of Information Act, 5 U.S.C. § 552 (1988) ("FOIA"), is to facilitate public access to Government documents" and that "[FOIA] is meant to pierce the veil of administrative secrecy and to open agency action to the light of public scrutiny." Beck v. Dep't of Justice, 997 F.2d 1489, 1490 (D.C. Cir. 1993) (internal quotation marks and citations omitted). Despite its goal of broad disclosure, FOIA allows agencies to withhold documents that fit within one of nine exemptions. See 5 U.S.C. § 552(b). In the instant case, the Department of Health and Human Services ("HHS") invoked Exemptions 6, 7(C), and 7(D) to justify withholding the information requested by Dr. McCutchen. Exemption 6 protects "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy." Id. § 552(b)(6). Exemption 7(C) allows withholding of

records or information compiled for law enforcement purposes, but only to the extent that the production of such law enforcement records or information . . . could reasonably be expected to constitute an unwarranted invasion of personal privacy.

Id. § 552(b)(7)(C).

Comparing the language of these two provisions, the Supreme Court has noted:

[W]hereas Exemption 6 requires that the invasion of privacy be "clearly unwarranted," the adverb "clearly" is omitted from Exemption 7(C) . . . [and] whereas Exemption 6 refers to disclosures that "would constitute" an invasion of privacy, Exemption 7(C) encompasses any disclosure that "could reasonably be expected to constitute" such an invasion.

Department of Justice v. Reporters Comm. for Freedom of the Press, 489 U.S. 749, 756, (1989). "Thus, the standard for evaluating a threatened invasion of privacy interests . . . is somewhat broader" under Exemption 7(C) than under Exemption 6. Id.

Both Exemptions 6 and 7(C) call for a balancing of "the privacy interests that would be compromised by disclosure against the public interest in release of the requested information." Beck, 997 F.2d at 1491 (internal quotation marks and citations omitted). On the public interest side of the scale is "the citizens' right to be informed about what their government is up to." Id. (internal quotation marks and citations omitted).

Exemption 7(D), which HHS invoked only to justify withholding the complainants' names, shields "records or information compiled for law enforcement purposes, but only to the extent that the production of such law enforcement records or information . . . could reasonably be expected to disclose the identity of a confidential source. . . ." 5 U.S.C. § 552(b)(7)(D).

Under FOIA, the burden is on the agency to justify withholding requested information and the agency's refusal to disclose it is subject to de novo review by the district court. Id. § 552(a)(4)(B).

B. Office of Scientific Integrity

When this suit was filed in the district court, the Public Health Service Act ("PHSA"), 42 U.S.C. §§ 201 et seq. (1988), required HHS to demand that all applicants for research funding establish "an administrative process to review reports of scientific fraud" and "report to the Secretary [of HHS] any investigation of alleged scientific fraud which appears substantial." Id. § 289b(a). In addition, the Director of the National Institutes of Health ("NIH"), an HHS official, was required to "establish a process for the prompt and appropriate response to information . . . respecting scientific fraud in connection with projects" funded by the PHSA. Id. § 289b(b). In accordance with this mandate, HHS created the Office of Scientific Integrity ("OSI") to "oversee the implementation of all . . . policies and procedures related to scientific misconduct." 42 C.F.R. § 50.102. "Misconduct" includes "fabrication, falsification, plagiarism, or other practices that seriously deviate from those that are commonly accepted within the scientific community for proposing, conducting, or reporting research." Id.

After this suit was filed, Congress amended the PHSA, see National Institutes of Health Revitalization Act of 1993, Pub.L. No. 103-43 § 161 (1993); and HHS abolished OSI, transferring its functions to the Office of the Assistant Secretary for Health, see 57 Fed.Reg. 24,262 (1992). While still extant, however, OSI investigated research performed within NIH as well as research performed at outside institutions that receive PHSA funds, such as universities. Allegations of misconduct involving research by a recipient institution could be investigated by that institution or by OSI. OSI supervised all investigations conducted by outside institutions, however, and could perform its own inquiry if not satisfied. If OSI found misconduct, it could recommend sanctions, including disciplinary action against government researchers and termination of funding and debarment from future grants for outside researchers.

C. Procedural History

On February 7, 1990, Dr. McCutchen, a scientist, submitted a FOIA request to HHS for "a list of all cases closed by the Office of Scientific Integrity." On March 6, 1990, HHS turned over final reports for the four investigations in which OSI had found misconduct. By letter dated July 5, 1990, HHS notified Dr. McCutchen that it was withholding information on investigations in which OSI had made no finding of misconduct because disclosure of such information would constitute "a clearly unwarranted invasion of the subject individual or individuals [sic] personal privacy" under FOIA Exemption 6.

After exhausting his administrative remedies, Dr. McCutchen filed suit in the district court on January 23, 1991. According to his complaint, he sought to compel disclosure of "information listing cases closed by OSI where there was no finding of scientific misconduct." On March 29, 1991, HHS released to Dr. McCutchen a list of OSI's closed cases consisting of six columns: (1) the case number, (2) the institution involved, (3) the complainant, (4) the respondent, (5) the OSI staff member assigned to the case, and (6) the date the case was closed. Deleted from this list, however, were the names of all noninstitutional complainants, the names of all respondents other than those who had died or been found guilty of wrongdoing, and the name of any institution whose disclosure might have enabled Dr. McCutchen to discover the identity of a respondent. In an accompanying letter, HHS asserted that the deletions were based on FOIA Exemptions 6, 7(C), and 7(D).

Both Dr. McCutchen and HHS moved for summary judgment, and the district court granted each motion in part and denied each in part, McCutchen v. HHS, C.A. No. 91-142 (D.D.C. filed Aug. 24, 1992) ("Mem. op.") at 14-15, holding that Exemption 7(C) permitted HHS to withhold the names of the complainants but not those of the respondents. The court determined that the requested information had been "compiled for law enforcement purposes" and thus satisfied the threshold requirement of Exemption 7. Id. at 7. Turning to whether Exemption 7(C) protected the names of the respondents, the district court decided that the public interest in disclosure outweighed the privacy interests. The court found that the public interest was "significant" because "[t]here has been growing concern in recent years over scientific misconduct, and allegations . . . that OSI's handling of these cases has been inadequate," id. at 10, and because obtaining the respondents' names would enable Dr. McCutchen to assess the thoroughness of OSI's investigations. Id. at 11.

The court recognized that the respondents had an interest in remaining anonymous but ruled that "the privacy interest in one's identity is not nearly so strong when one's professional activities, rather than matters concerning personal conduct, are at issue." Id. at 8. The court also found that Exemption 6 did not justify withholding the respondents' names because it was narrower than Exemption 7(C). Id. at 12 n. 6.

The court held, however, that the complainants' names were shielded by Exemption 7(C). It found a significant privacy interest because "[a complainant] could become the subject of harassment and retaliation if his identity were revealed," id. at 13, and found no public interest in disclosure. Id. at 13-14. The court declined to address whether Exemption 7(D) supplied additional support for non-disclosure. Id. at 12-13.

II. DISCUSSION

A. Scope of FOIA Request

We must first determine the scope of the FOIA request. Dr. McCutchen argues on appeal that he requested not just a list of closed cases but OSI's final reports for those cases. He concedes that the request itself mentioned only a "list," but he insists that the Department understood his request to encompass OSI's final reports because it released such reports in the four cases in which OSI had found misconduct.

Dr. McCutchen did not raise this argument in the district court. Neither his complaint nor the three briefs he filed there suggested that he sought anything but a list of closed cases. Indeed, in one of his briefs, Dr. McCutchen stressed that he wanted, above all, the names of the respondents and argued that OSI's final reports would be worthless without them:

Without the names of individuals investigated but found not to have engaged in misconduct, other information tells the public nothing about whether whitewashing is occurring in closed cases. . . . In fact, even if the Government were to release the case studies of investigations in which scientific misconduct was not found with individuals' names deleted, the public would still be unable to determine whether a whitewash or inadequate investigation had occurred. . . . Thus . . . the names of the individuals investigated are the most critical pieces of information to assist the public in determining whether whitewashing has occurred and to fulfill the purpose of the FOIA.

Plaintiff Dr. Charles W. McCutchen's Memorandum of Points and Authorities in Opposition to Defendant's Cross-Motion for Summary Judgment at 14 (emphasis in original). Not surprisingly, then, the district court thought Dr. McCutchen was "seek[ing] access to a list of all cases closed" by OSI. Mem. op. at 1 (emphasis added).

We generally refuse to consider issues not raised in the district court. District of Columbia v. Air Florida, 750 F.2d 1077, 1084 (D.C. Cir. 1984) ("It is well settled that issues and legal theories not asserted at the District Court level ordinarily will not be heard on appeal.") (footnote omitted); see also American Fed'n of Gov't Employees, Local 2782 v. Dep't of Commerce, 907 F.2d 203, 208 (D.C. Cir. 1990) ("It is our office . . . to review the decision of the district court in the case that the appellants . . . there presented for decision"). Although we will on occasion depart from this rule when "the nature of the issues and the posture of the case [so] justify," Walker v. Jones, 733 F.2d 923, 928 n. 7 (D.C. Cir. 1984), we see no reason for doing so here. Accordingly, we will treat the request as one seeking only a list of cases.

B. Names of Respondents

As a preliminary matter, we note that Dr. McCutchen does not challenge the district court's determination that the requested information was "compiled for law enforcement purposes" and thus satisfies the threshold requirement of Exemption 7. Therefore, we do not address that issue.

1. Privacy Interest

While acknowledging that the protection of Exemption 7(C) is not limited to intimate, embarrassing information, the district court held that "the privacy interest in one's identity is not nearly so strong when one's professional activities, rather than matters concerning personal conduct, are at issue." Mem. op. at 8. This statement overlooks our decision in Carter v. Dep't of Commerce, 830 F.2d 388 (D.C. Cir. 1987), which involved facts similar to those of this case. In Carter, we held that the Patent and Trademark Office properly invoked Exemption 6 to withhold information that would enable the requester to identify attorneys who had once been the targets of misconduct investigations but against whom all charges had been dismissed. Id. at 394-95.

It is true that in Washington Post Co. v. Dep't of Justice, 863 F.2d 96 (D.C. Cir. 1988), on which the district court relied, we observed that "[i]nformation relating to business judgments and relationships does not qualify for exemption [under 7(C)]" and that "[t]his is so even if disclosure might tarnish someone's professional reputation." Id. at 100. A few sentences later, however, we noted that "[i]t may be that [information], if it accused individual employees of having committed a crime, would implicate the privacy interest of personal honor" and that "the protection accorded reputation [under Exemption 7(C)] would generally shield material when disclosure would show that an individual was the target of a law enforcement investigation." Id. at 100-01. We went on to find that the requested document — a report prepared by outside directors of a pharmaceutical company to determine whether the company had indemnification claims against its employees for negligence in the development and marketing of a defective drug — did not identify any employees as targets of a law enforcement investigation. Id. at 101.

Here, the "personal honor" identified in Washington Post and the stigma that hangs over the targets of law enforcement investigations are on the line. The allegations that may trigger an OSI inquiry include plagiarism, fabrication of research results, and similar breaches of academic integrity. While these may not be prosecutable offenses, they carry a stigma and can damage a career. An affidavit in the record suggests that mere association with such allegations, even if they are never proven, could reduce a scientist's chances of securing grants or obtaining tenure.

Dr. McCutchen nevertheless contends that scientists who are investigated by OSI have already surrendered much of their privacy by applying for public funding to support their research, citing Sims v. CIA, 642 F.2d 562, 575 (D.C. Cir. 1980) (names of those entering into contracts with CIA is matter of "clear public concern" and cannot be withheld under Exemption 6). Even if this is true, we think that recipients of government grants or contracts are entitled to at least as much privacy protection as government employees who have been the targets of investigations. Courts have declined to reveal the names of such workers, particularly when they have been exonerated.

For example, in Stern v. FBI, 737 F.2d 84 (D.C. Cir. 1984), we permitted the FBI to withhold the identities of two of its low-level employees whose negligence, according to an investigative report, had contributed to a cover-up of illegal surveillance activities by FBI agents, although we required the agency to disclose the name of a higher-level employee who had knowingly participated in the cover-up. We commented that while "the status of the individuals in this case as federal employees diminishes their privacy interests . . . because of the corresponding public interest in knowing how public employees are performing their jobs," id. at 92, "the risk that such employees could be linked to serious criminal wrongdoing when, in fact, they were totally cleared of any such acts, increases the potential invasion of privacy that Exemption 7(C) was designed to protect." Id. at 93. See also Dunkelberger v. Dep't of Justice, 906 F.2d 779, 781 (D.C. Cir. 1990) (upholding FBI's refusal to disclose agent's employment records and noting that "a government employee has at least a minimal privacy interest in his own employment record and evaluation history"); cf. Department of the Air Force v. Rose, 425 U.S. 352, 380, (1976) (FOIA "request for access to summaries [of Air Force Academy disciplinary records] `with personal references or other identifying information deleted' respected the confidentiality interests embodied in Exemption 6").

We conclude that the respondents, who have been investigated and exonerated, have a substantial privacy interest in remaining anonymous.

2. Public Interest

The district court found that disclosing the names of the respondents would serve the public interest because "[t]he name of the scientist links up directly with the scientific research which he worked on. . . . This piece of data therefore . . . identifies scientific work done at taxpayer expense whose credibility has been called into question." Mem. op. at 11. Obtaining the respondents' names would permit other scientists to check the comprehensiveness of OSI's investigations, the district court reasoned, "because they would be best able to tell from statistical data and the accompanying documentation whether the findings are true and accurate." Id. at 12. The court also cited newspaper articles as evidence of "growing concern" that OSI was mishandling scientific misconduct investigations. Id. at 10. Finally, it commented that "OSI's own procedures raise questions which only oversight can address" because letting outside research institutions assist in investigations of their own employees creates a potential conflict of interest. Id. at 11.

HHS rejects the district court's view that the information would permit the identification of the particular research projects that had been the subjects of OSI investigations. For purposes of this appeal, however, we will assume this proposition is true. The fundamental flaw in the district court's analysis is that it found a public interest in disclosure of the respondents' names based on " allegations in the scientific and political communities that OSI's handling of these cases has been inadequate." Id. at 10 (emphasis added).

A mere desire to review how an agency is doing its job, coupled with allegations that it is not, does not create a public interest sufficient to override the privacy interests protected by Exemption 7(C). In Davis v. Dep't of Justice, 968 F.2d 1276 (D.C. Cir. 1992), we reversed a district court order directing the release of recordings made in the course of a criminal investigation despite the requester's suggestion that the tapes contained evidence of government "entrapment" and "overreaching";

[W]hen . . . governmental misconduct is alleged as the justification for disclosure, the public interest is insubstantial unless the requester puts forward compelling evidence that the agency denying the FOIA request is engaged in illegal activity and shows that the information sought is necessary in order to confirm or refute that evidence.

Id. at 1282 (emphasis added; internal quotation marks omitted); see also Safecard Services, Inc. v. SEC, 926 F.2d 1197, 1206 (D.C. Cir. 1991) ("unless access to the names and addresses of private individuals appearing in files within the ambit of Exemption 7(C) is necessary in order to confirm or refute compelling evidence that the agency is engaged in illegal activity, such information is exempt from disclosure").

Whatever one may suspect about OSI from media reports, there is no evidence that it shirked its duties. We express no opinion on whether HHS could withhold the names in response to a future request by Dr. McCutchen for OSI's reports in all closed cases if accompanied by "compelling evidence" of official misconduct.

The district court also erred in treating the sharing of investigative responsibility between OSI and grantee institutions as a ground for disclosure. See Mem. op. at 11. The court noted that OSI typically conducted its investigations based on information developed in the course of a prior investigation by the university or institution that employed the individual whose work was being questioned. The court concluded that the results were suspect because of the interest of those universities and institutions in avoiding a finding of misconduct. Id. Congress, however, specifically provided for investigations by entities receiving research grants, 42 U.S.C. § 289b(a)(1), and required them to "report to the Secretary [of HHS] any investigation of alleged scientific fraud which appears substantial." Id. § 289b(a)(2).

We owe deference to Congress's judgment that the scheme it has established is effective. See Jacobs v. Barr, 959 F.2d 313, 319 (D.C. Cir.), cert. denied, ___ U.S. ___, 113 S.Ct. 95, 121 L.Ed.2d 56 (1992) ("although courts should not defer to Congress on constitutional questions, we should defer — or give great weight — to Congress on empirical questions") (emphasis in original); Brotherhood of Locomotive Firemen Enginemen v. Chicago, R.I. Pac. R.R. Co., 393 U.S. 129, 138-39, 89 S.Ct. 323, 327-28, 21 L.Ed.2d 289 (1968) (noting that "the question of [railroad] safety . . . is essentially a matter of public policy" for the legislature, and that the district court cannot "resolve conflicts in the evidence against the legislature's conclusion or . . . reject the legislative judgment" on the ground that it lacked support in the record). In light of that deference and in the absence of compelling evidence that OSI was not performing effectively, we find very little public interest in disclosure of the respondents' names.

Weighting a negligible public interest in disclosure against the respondents' substantial privacy interests, we hold that FOIA Exemption 7(C) permits HHS to withhold the respondents' names. Accordingly, we need not address whether Exemption 6 also protects that information.

C. The Complainants' Names

We uphold the district court's ruling allowing HHS to withhold the names of the complainants. The complainants have a strong privacy interest in remaining anonymous because, as "whistle-blowers," they might face retaliation if their identities were revealed. There is evidence in the record that one well-qualified immunologist was unable to obtain employment in her field after making allegations of misconduct. Protection against such retaliation was one of the purposes that Exemption 7(C) was designed to serve. See Holy Spirit Ass'n for Unification of World Christianity v. FBI, 683 F.2d 562, 565 (D.C. Cir. 1982) (MacKinnon, J., concurring) ("Where a person's fear of reprisals from the subject of his communication is reasonable . . . privacy interests support the application of both Exemption 6 and Exemption 7(C)"); L C Marine Transport v. United States, 740 F.2d 919, 922-23 (11th Cir. 1984) (employer could not obtain names of employees interviewed by Occupational Safety Hazard Administration because "[t]he employee-witnesses . . . have a substantial privacy interest . . . as disclosure would lead to the type of harm, embarrassment and possible retaliation that 7(C) was created to prevent").

Our analysis of the public interest side of the balance with respect to the respondents applies with equal force here: Even if learning the identities of the complainants would enable Dr. McCutchen to check the thoroughness of OSI's work, we will not force the agency to turn over the names of the complainants simply because someone has accused it of wrongdoing. We conclude that the scale again tilts in favor of privacy and that HHS thus properly withheld the complainants' names under Exemption 7(C). That being the case, we do not decide whether Exemptions 6 or 7(D) also protect the complainants' names from disclosure.

III. CONCLUSION

For the foregoing reasons, the decision of the district court is reversed in part and affirmed in part. None of the requested names are to be disclosed.

So ordered.