O’Connell v. Donna E. Shalala, 79 F.3d 170 (1st Cir. 1996)

SELYA, Circuit Judge.

This is a petition for review and vacatur of a final rule promulgated by the Secretary of Health and Human Services (the Secretary) under the National Vaccine Injury Compensation Program, 42 U.S.C. §§ 300aa-10 through 300aa-34 (1994). We have jurisdiction under 42 U.S.C. § 300aa-32. In the pages that follow, we explore the pertinent statutory framework, recount the proceedings to date, and then examine the petitioners’ three-pronged challenge. When all is said and done, we deny the petition and leave the rule intact.

I. THE STATUTORY SCHEME

The administration of childhood vaccines, though critically important to public health, “is not always without risk.” Committee to Review the Adverse Consequences of Pertussis and Rubella Vaccines, Institute of Medicine, Adverse Effects of Pertussis and Rubella Vaccines 1 (1991) (IOM Report). Since vaccines generally contain either dead bacteria or live but weakened viruses, it is not surprising that they are capable of causing serious adverse effects. See id. Despite the infrequency of such episodes, Congress feared that the long shadow of tort liability cast by vaccine-related injuries would drive up prices and eventually force vaccine suppli

ers out of the market. See H.R.Rep. No. 908; 99th Cong., 2d Sess. 1, 4, 6-7 (1986), reprinted in 1986 U.S.C.C.A.N. 6287, 6344, 6345, 6347-48. Congress also worried that the vagaries of litigation, coupled with the cost, might leave many deserving victims of vaccine-related injuries undercompensated. See id.

To protect the supply of vaccines while ensuring fair, timely compensation for victims, Congress departed from the traditional tort system and wrote the National Childhood Vaccine Injury Act (the Act), Pub.L. No. 99-660, tit. III, 100 Stat. 3755 (1986). Among other things, the Act established a special tribunal (the Vaccine Court), and moved vaccine-injury cases partly outside the customary tort framework. See Schafer v. American Cyanamid Co., 20 F.3d 1, 2 (1st Cir.1994) (explaining the mechanics of the Act).1 In respect to cases brought before this tribunal, Congress eased the complainants’ burdens both by dispensing with the requirement of proving negligence and by greatly simplifying the requisite proof of causation. See 42 U.S.C. §§ 300aa-11. Of course, there are tradeoffs; for example, Congress limited the damages that a victim could obtain for vaccine-related injuries. See id. § 300aa-15.

In aid of the neoteric regulatory regime, the Act provides, in tabular format, a listing of vaccines and a parallel listing of medical conditions commonly associated with the use of each vaccine. See id. § 300aa-14(a). These listings, known collectively as the Vaccine Injury Table (the Table), are accompanied by, and are to be read in light of, the Qualifications and Aids to Interpretation (QAI). The QAI is a separate subsection that provides definitions and explanations for the terms used in the Table. See id. § 300aa-14(b). To receive compensation for a vaccine-related injury, a recipient must simply petition the Vaccine Court and show that, within a prescribed time span, she suffered one or more of the disorders listed in the Table as associated with the particular vaccine that she received. Thus, the content of the Table (a sample of which is excerpted in the Appendix) is critical: it is only when a vaccinated child suffers a listed condition within applicable temporal parameters that compensation will be forthcoming without the time, expense, proof requirements, and uncertainty of full-blown litigation.

The Table is not intended to be static. Congress gave the Secretary express power to promulgate regulations adding to or subtracting from the tabular list of conditions, and changing the delineated time periods. See 42 U.S.C. § 300aa-14(c)(3). This is a rather odd approach because it authorizes the Secretary, in effect, to amend the statutorily enacted Table by way of administrative rulemaking.2 This grant of power probably reflected a congressional consensus that the first iteration of the Table was not perfect. Driven by a sense of urgency to put something into place, the solons knowingly used incomplete data when forging the causal links between vaccines and associated medical conditions. Mindful of its haste, Congress directed the Secretary to have the Institute of Medicine (IOM) — an arm of the National Academy of Science — conduct an extensive review of all available information bearing on the relationship between vaccines and medical conditions, and thereafter to publish findings and revise the Table based on the IOM’s study. See Vaccine Act § 312, 100 Stat. at 3779.

To assist the Secretary in updating the Table, Congress created the Advisory Commission on Childhood Vaccines (ACCV) — a body composed of a cross-section of health

professionals, legal experts, interested citizens (including two who are parents of children victimized by vaccine-related injuries), and federal officials. See 42 U.S.C. § 300aa-19. Congress directed the Secretary to provide the ACCV with a copy of each contemplated regulation before formally proposing it, and then to await the expiration Of a ninety-day comment period before moving forward. See id. § 300aa-14(d).

II. THE COURSE OF EVENTS

In 1991, the IOM completed its study and, on August 27, issued the IOM Report. Among the many conclusions contained in this tome the IOM found a causal relation between DPT (diphtheria-pertussis-tetanus) vaccine, on one hand, and acute encephalopathy and hypotonic, hyporesponse episodes (HHE), on the other hand.3 See IOM Report at 118, 177. However, the IOM found insufficient evidence to indicate a causal relationship between DPT vaccine and residual seizure disorders (such as epilepsy). See id. at 118 n. 3. The project director gave the ACCV a full briefing on the IOM Report in September of 1991.

In anticipation of receiving the IOM Report, the Secretary formed a Public Health Service Task Force as a vehicle for revising the Table. She also enlisted yet another helpmate, the National Vaccine Advisory Committee (NVAC). Unlike the ACCV, which by statute counsels the Secretary in respect to the injury compensation program, see 42 U.S.C. § 300aa-19, the NVAC’s statutory responsibility is to advise the director of the separate national program for developing and administering the public health aspects of immunization policy, see id. § 300aa-5. The Secretary transmitted the IOM Report to the Task Force, which then recommended a number of changes to the Table (including the removal of encephalopathy, HHE, and residual seizure disorders as associated medical conditions vis-a-vis DPT vaccination).4 The NVAC concurred in these recommendations.

Despite the fact that the ACCV had not yet formally received the Task Force’s or the NVAC’s recommendations, it took up the substance of the proposed revisions at its December 1991 meeting. In lieu of the literal text of the suggested changes, the ACCV members received what has been referred to as a “matrix” — essentially, a table comparing a synthesis of Task Force and NVAC recommendations and summarizing the rationales advanced by those bodies. The ACCV discussed these recommendations at length and approved all but the one that suggested dropping encephalopathy from the Table. As a counter-proposal, the ACCV encouraged the Secretary to modify the QAI definition of encephalopathy in a way that would restrict its meaning to acute or chronic episodes of a type more likely to result in significant harm.

In due season, the Secretary published a Notice of Proposed Rulemaking (the Notice). See 57 Fed.Reg. 36,878 (proposed Aug. 14, 1992). The Notice included the required scientific findings and set forth regulations designed to revise the Table accordingly. These covered all the Task Force’s recommendations save only for the dropping of encephalopathy. On that point the Secretary acquiesced in the ACCVs view and proposed a new definition of the condition similar in most respects to the definition discussed at the ACCVs December 1991 meeting. See id. at 36,880. A comment period and public hearing ensued.

In 1993, the results of a ten-year study of acute childhood neurologic illnesses became available. Recognizing the potential importance of the study, the Secretary stayed her hand and requested the IOM to review the newly compiled material. In March 1994,

the IOM concluded that the “balance of the evidence is consistent with a causal relationship” between DPT vaccination and certain forms of chronic nervous system dysfunction suffered by children who experience an acute neurologic illness shortly after vaccination. Committee to Study New Research on Vaccines, Institute of Medicine, DPT Vaccine and Chronic Nervous System Dysfunction: A New Analysis 2-3 (1994). On March 24, 1994, the Secretary reopened the comment period for a limited time, restricting discussion to the question whether the previously proposed revisions should be modified in light of the supplemental report.

At its June 1994 meeting the ACCV discussed the new information as it concerned the proposed definition of encephalopathy. This time, the ACCV did not achieve consensus on the subject, and simply transmitted the minutes of its meeting to the Secretary. On February 8, 1995, the Secretary promulgated a final rule (under attack in this proceeding) that removed HHE and residual seizure disorders from the Table and changed the definition of encephalopathy in a manner very similar (but not identical) to the manner originally suggested by the ACCV and proposed in the Notice. See 60 Fed.Reg. 7678 (1995).

Francis and Lisa O’Connell (parents and legal representatives of Kelliann O’Connell), joined by a parents’ advocacy group, now seek judicial review and vacatur of the final rule.5

III. THE PETITIONERS’ CHALLENGE

The petitioners raise three objections to the Secretary’s action. First, the petitioners assert that the Act does not empower the Secretary to change the definitions included in the QAI, but, rather, only authorizes the Secretary to add and subtract entries (i.e., vaccines and associated medical conditions) and change time periods specified in the Table proper. Second, the petitioners contend that, even if the Secretary otherwise had authority to effectuate the' contested change, she failed to follow the procedures mandated by the Act. Finally, the petitioners insist that a decision to remove a medical condition from the Table must be based on definitive evidence refuting the existence of a causal relationship between the vaccine in question and the condition, and that the Secretary eliminated HHE and residual seizure disorders from the Table notwithstanding the absence of such an evidentiary predicate. We address each remonstrance separately.

A. Authority to Revise the QAI.

The petitioners argue that the Secretary’s attempt to change the definition of encephalopathy provided in the QAI is impuissant because it surpasses the authority granted to the Secretary by the Act. The Act states:

42 U.S.C. § 300aa-14(c)(3) (emphasis supplied). In the petitioners’ view, the underscored phrases limit the Secretary’s powers of alteration, and hence, because the QAI provision is distinct from the Table proper, the Secretary’s revisory authority does not extend to it. Ergo, changing the definition of encephalopathy contained in the QAI oversteps the Secretary’s bounds. The Secretary debunks this argument. She construes the statute more broadly, urging that it gives her authority to rewrite the QAI.

1. Chevron Deference. Before choosing between these competing views, we must address a preliminary issue. The Secretary, correctly observing that courts ordinarily defer to an agency’s plausible construction

of a silent or ambiguous statute as long as Congress has committed the statute to the agency for purposes of administration, see Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 2781-82, 81 L.Ed.2d 694 (1984); Strickland v. Commissioner, Me. Dep’t of Human Servs., 48 F.3d 12, 16 (1st Cir.), cert. denied, - U.S. -, 116 S.Ct. 145, 133 L.Ed.2d 91 (1995), asks that we defer to her construction of § 300aa-14(e). There may be more to this request than meets the eye. The petitioners’ objection is arguably not directed at a regulation that purports to apply a particular statutory directive which the Secretary is coneededly empowered to implement, but instead at a regulation that lies in an area as to which, the petitioners say, the statute grants the Secretary no rulemaking authority at all. In the current state of the law, it is unclear whether deference is appropriate under such circumstances.6

Discretion is sometimes the better part of valor. Because we decide, as a matter of original statutory construction, that the Act grants the Secretary authority to revise the QAI provision, see infra, we leave the question of deference unanswered.

2. Interpreting the Statute. We turn now to the disputed statute. While one can focus with Cyclopean intensity on the words singled out by the petitioners and perhaps construct a coherent argument that those words restrict the Secretary’s revisory authority to the Table proper, courts are bound to afford statutes a practical, commonsense reading. See King v. St. Vincent’s Hosp., 502 U.S. 215, 221, 112 S.Ct. 570, 574, 116 L.Ed.2d 578 (1991). Instead of culling selected words from a statute’s text and inspecting them in an antiseptic laboratory setting, a court engaged in the task of statutory interpretation must examine the statute as a whole, giving due weight to design, structure, and purpose as well as to aggregate language. See National R.R. Passenger Corp. v. Boston & Me. Corp., 503 U.S. 407, 417, 112 S.Ct. 1394, 1401-02, 118 L.Ed.2d 52 (1992); Dole v. United Steelworkers of Am., 494 U.S. 26, 36, 110 S.Ct. 929, 934-35, 108 L.Ed.2d 23 (1990); K mart Corp. v. Cartier, Inc., 486 U.S. 281, 291, 108 S.Ct. 1811, 1817-18, 100 L.Ed.2d 313 (1988); Riva v. Massachusetts, 61 F.3d 1003, 1007 (1st Cir.1995).

The petitioners’ reading of the Act cannot survive the application of this global standard. Reading the statute as a whole, we are satisfied that Congress gave the Secretary authority to revise the QAI. In the absence of such authority the system for updating the Act is virtually unworkable. For instance, when the Secretary exercises her -undeniable power to include an emergent condition in the Table, she must be able to amend the QAI to reflect the addition. Surely Congress did not intend either to leave added conditions unexplained or itself to edit the QAI every time the Secretary saw fit to alter the Table. In short, the power to revise the QAI is a necessary adjunct of the power to revise the Table itself.7 Elsewise, the tension that would be created within the structure of the Act would be intolerable and would contravene the salutary principle that statutes should, whenever possible, be con

strued sensibly. See American Tobacco Co. v. Patterson, 456 U.S. 63, 71, 102 S.Ct. 1534, 1538-39, 71 L.Ed.2d 748 (1982); Riva, 61 F.3d at 1008; United States v. Meyer, 808 F.2d 912, 919 (1st Cir.1987); see also Norman J. Singer, Sutherland Statutory Construction § 45.12, at 61 (5th ed. 1992).

We add, moreover, that the petitioners’ proffered reading of the statute is excessively formalistic. If the Secretary could not change the definition of encephalopathy directly, she could certainly accomplish the same result indirectly. She need simply delete encephalopathy from the Table, thus rendering its definition nugatory, and then immediately add encephalopathy, redefined, to the Table.

This reality is lethal to the petitioners’ position. We cannot imagine that Congress intended to force the Secretary to go round and round the mulberry bush in order to revise the Table and its accompanying explanations. The shortest distance between two points is a straight line, and we will not lightly presume that Congress lost sight of so abecedarian a principle.8 See Singer, supra, § 45.12, at 61 (advocating the baseline assumption that an enacted statute should be construed to “achieve[ ] an effective and operative result”). To cinch matters, we note that the statutory grant of a greater power typically includes the grant of a lesser power, see, e.g., United States v. O’Neil, 11 F.3d 292, 296 (1st Cir.1993) (describing this principle as “a bit of common sense that has been recognized in virtually every legal code from time immemorial”), and the overall structure of the Vaccine Act confirms its applicability here: the brute power to subtract listed medical conditions from the Table encompasses the more modest power to trim the definitions associated with listed medical conditions.

We have said enough on this score. We hold that the Act grants the Secretary the authority to revise the Qualifications and Aids to Interpretation that accompany the Vaccine Injury Table. Consequently, the petitioners’ initial remonstrance fails.

B. Notification.

The petitioners accuse the Secretary of failing to observe the required notification procedures when she promulgated the regulation. The accusation is unfounded.

The relevant statute provides:

42 U.S.C. § 300aa-14(d). The petitioners contend that the Secretary neglected to follow these procedures twice over, by failing to provide the ACCV with (1) a copy of the Notice before publishing it, and (2) a copy of the final rule before issuing it. We examine each contention.

1. The Proposed Rule. We rehearse the relevant facts. Shortly after publication of the IOM Report, the Task Force made its initial recommendations for changing the Table. The NVAC substantially concurred in those recommendations. The Secretary’s proposal was then circulated at the ACCVs December 1991 meeting in the form of a matrix detailing the various recommendations. During the ensuing discussion, the 'principal objection was to the Secretary’s proposal, reflected in the matrix, for removing encephalopathy from the Table. The ACCV urged instead that encephalopathy should be retained in the Table but that its definition should be modified in the QAI. The Secretary accepted the ACCVs unanimous recommendation and, in August 1992,

published a Notice that implemented this recommendation.9 See 57 Fed.Reg. at 36,880 (accepting ACCV recommendation to retain encephalopathy in the Table with a new definition).

The petitioners argue that the rule ultimately promulgated is invalid because the matrix distributed and discussed in December of 1991 was not literally a “copy of the proposed regulation” as required by § 300aa-14(d). Although the matrix may not have been produced in the typical format for a proposed administrative regulation, we think that for all intents and purposes it was a “copy” of the regulation that the Secretary planned to propose. The matrix contained the substance of all the proposed changes to the Table. The only real difference appears to have been in manner of presentation. The statute requires the Secretary to deliver all the meat of a planned rule to the Secretaxy without regard to how it is arranged on the platter. Thus, as long as the Secretary transmits the entire substance of her proposed regulation to the ACCV in the appropriate time frame, the form of the transmission is immaterial. This principle possesses particular force where, as here, the unorthodox format does not obfuscate or mislead. Indeed, the matrix’s tabular format increased comprehension by allowing the ACCV to discuss the revisions in a manner corresponding to the format of the Vaccine Injury Table itself. We hold, therefore, that the Secretary fulfilled her statutory pre-publication duty in regard to the proposed regulation.

If there were any room for doubt about the adequacy of the transmittal — and we do not believe that there is — the ACCVs actions would dispel it. We are particularly impressed by two things. First, the transcript of the ACCVs December 1991 meeting makes very clear that ACCV members thought they were discussing the Secretary’s proposed revisions to the Table. Second, after the Secretary published the proposed regulation, the ACCV did not cry “foul” or otherwise complain that it had been bypassed. These facts plainly show that the ACCV understood the matrix to be a “copy of the proposed regulation” and acted upon it as such.

2. The Final Rule. The Secretary’s issuance of the final rule poses a somewhat closer question. The Secretary openly admits that she did not provide the ACCV with the text of the final rule prior to its promulgation. She argues, however, that the Act does not oblige her to do so, or that, if it does, she substantially complied with that obligation.

Section 300aa-14(d) provides in material part that “the Secretary may not propose a regulation ... or any revision thereof’ without first furnishing the ACCV with a copy of “the regulation or revision,” requesting comment, and marking time for ninety days. This aspect of the controversy between the parties arises from the phrase “any revision thereof.” In context, this phrase is susceptible to at least two reasonable meanings. Under one interpretation, favored by the Secre-taxy, the phrase refers to a revision of a regulation, so that, while the Secretary may not propose either a new regulation or a revision to an existing regulation without advance notice to the ACCV, she may proceed to revise a proposed regulation -without resubmitting it to the ACCV. Under the second interpretation, favored by the petitioners, the phrase “any revision thereof’ refers to proposed regulations, so that the Secretary may propose neither a new regulation nor any later revision to that proposed regulation without first informing the ACCV.

Once again our analysis begins with a nod in the direction of Chevron. The rule of deference traditionally applies when the

agency’s interpretation is a “product of delegated authority for rulemaking,” Stinson v. United States, 508 U.S. 36, 43-44, 113 S.Ct. 1913, 1918, 123 L.Ed.2d 598 (1993), a sphere that ordinarily encompasses legislative rules and agency adjudications. Here, the Secretary’s interpretation of the law is not embodied in a legislative rule or an adjudication. The evidence of her view about how § 300aa-14(d) is supposed to operate comes exclusively from two sources: the refusal of her subordinates to send the ACCV a pre-publication copy of the final rule, and the tactical position adopted by her counsel.

As for the first source, agency positions that are pieced together from offhand conduct of bureaucratic fussbudgets are entitled to little weight on judicial review, principally because they do not reflect the kind of delegated authority for policymaking that underlies the Chevron presumption.10 See Stinson, 508 U.S. at 43-44, 113 S.Ct. at 1918; Martin v. OSHRC, 499 U.S. 144, 157, 111 S.Ct. 1171, 1179, 113 L.Ed.2d 117 (1991); Public Citizen v. United States Dep’t of Justice, 491 U.S. 440, 463 n. 12, 109 S.Ct. 2558, 2571, n. 12, 105 L.Ed.2d 377 (1989). As for the second source, courts customarily withhold Chevron deference from agencies’ litigating positions. See, e.g., Bowen v. Georgetown Univ. Hosp., 488 U.S. 204, 212, 109 S.Ct. 468, 473-74, 102 L.Ed.2d 493 (1988); United States v. 29 Cartons of an Article of Food, 987 F.2d 33, 38 n. 6 (1st Cir.1993); Director, OWCP v. General Dynamics Corp.,* 980 F.2d 74, 79 (1st Cir.1992). We see no reason to take a different tack in this instance. We therefore decline to defer to the Secretary’s construction of § 300aa-14(d).

Approaching the statutory question without the Secretary’s thumb on the scale, we believe that both suggested interpretations are plausible but imperfect renditions of problematic language. The petitioners’ interpretation means that every alteration to the text of a proposed rule — even a minor technical or grammatical alteration — would have to be rerouted through the ACCV, subject to a fresh notice-and-comment period. This extra step would be necessary even when the Secretary changes a proposed regulation in oc-cordance with the ACCV’s announced wishes or to correct a syntactical bevue. Such a construction would create a nearly endless circle and attenuate the rulemaking process without achieving any corresponding benefit. Because it is difficult to believe that Congress intended to prolong the revisory process by directing the Secretary to engage in a mindless minuet, the prospect of wasted motion cuts against the petitioners’ interpretation. See Alabama Power Co. v. Costle, 636 F.2d 323, 360 (D.C.Cir.1980) (stating the obvious proposition that courts should be reluctant to interpret the terms of a statute “to mandate pointless expenditures of effort”).

The Secretary’s construction likewise presents a problem in that it may render the phrase “any revision thereof’ superfluous to some extent. Since the Secretary would have to issue a new regulation in order to change an existing one, see 42 U.S.C. § 300aa-14(c), Congress probably did not need to add the requirement that a revision to an existing regulation must be reviewed by the ACCV. Because courts usually presume that every word and phrase in a statute is pregnant with meaning, see, e.g., United States v. Ven-Fuel, Inc., 758 F.2d 741, 751-52 (1st Cir.1985), the prospect of redundancy cuts against the Secretary’s interpretation.

Faced with no ideal choice, we conclude that Congress more likely intended the statute to be read as the Secretary urges. This interpretation is more plausible and better serves the ends that the legislature sought to achieve. Though it is possible that Congress could have accomplished its purpose without adding the disputed phrase (“any revision thereof’), it is not certain that it could have done so. A wily lawyer could perhaps have argued that the unembellished word “regulation” referred only to brand-new regulations, not to adjustments of preexisting regulations. Cf. Public Serv. Co. v. United States EPA, 682 F.2d 626, 633 (7th Cir.1982), cert. denied, 459 U.S. 1127, 103 S.Ct. 762, 74 L.Ed.2d 977 (1983). Even more likely, Congress might have thought that the Secretary would be able to style a revision to a preexisting regu

lation as something other than a new regulation (perhaps as a clarification), and thereby evade the statutory safeguards. See, e.g., Detroit Edison Co. v. United States EPA, 496 F.2d 244, 249 (6th Cir.1974) (rejecting EPA’s attempt to characterize agency action as mere “clarification” of regulation as opposed to revision); cf. United States v. LaBonte, 70 F.3d 1396, 1411 n. 13 (1st Cir.1995) (noting Sentencing Commission’s practice of styling certain pronouncements affecting the interpretation of extant sentencing guidelines as “clarifications” rather than amendments). Moreover, a belt-and-suspenders approach is not uncommon when the Legislative Branch cedes rulemaking power to the Executive Branch. Indeed, Congress has frequently employed the phrase “revision thereof’ to confirm that its procedural mandates apply both to original regulations and future revisions of such regulations. See, e.g., 15 U.S.C. § 2934(f); 16 U.S.C. § 410cc-32(e); 33 U.S.C. § 1314(c); 42 U.S.C. §§ 4916(a)(3)-(4), 7521(a)(2), 7571(b).

Here, the Secretary’s interpretation not only is consistent with common congressional statute-drafting practice but also ensures the ACCV’s input into the rulemaking process without inviting the wasteful circularity of proposal, notice, comment, changed proposal, re-notice, additional comment, and so on and so forth, ad infinitum. The petitioners’ suggested alternative, on the other hand, creates a perverse incentive. If the Secretary is forced to recommit a proposed regulation and twiddle her thumbs for an additional three months every time she responds agreeably to an ACCV suggestion, she may be less inclined to acquiesce in the first place. Hence, the interpretation that we adopt actually may increase the chance that the Secretary will pay attention to, and act upon, the ACCV’s advice.

In reaching the conclusion that the statute refers to regulations and revisions thereof (and not to revisions of proposed regulations), we necessarily override two other concerns anent the ACCV’s place in the scheme of things. First, the petitioners boast that the ACCV’s statutorily prescribed part in the process of revising the Table evinces Congress’s distrust of the Secretary and proves that the reference to “any revision thereof’ is intended to give the ACCV a more prominent presence in the rulemak-ing process. This distorts the statutory alignment by grossly underestimating the Secretary’s role and aggrandizing the ACCV’s importance. In crafting the Act, Congress delegated unusually great authority to the Secretary, including the power to rewrite the statute by updating one of its hallmark provisions. In contrast, Congress assigned the ACCV a purely advisory function. See 42 U.S.C. § 300aa-19; see also H.R.Rep. No. 908, supra, 1986 U.S.C.C.A.N. at 6365. Thus, far from bolstering the petitioners’ ease, a comparison of the relative responsibilities that Congress entrusted to the Secretary and the ACCV, respectively, undermines the petitioners’ argument.

Second, the Secretary’s construction of the statute does not permit her effectively to bypass the ACCV by proposing one regulation and then issuing something radically different as a final rule. The Administrative Procedure Act applies here, and it is axiomatic under that regime that a final rule must be a lineal descendant of, and in character with, the earlier proposed rule. See, e.g., Kooritzky v. Reich, 17 F.3d 1509, 1513 (D.C.Cir.1994); American Medical Ass’n v. United States, 887 F.2d 760, 767 (7th Cir.1989). Put another way, changes must flow logically from the prescribed notice and comment. See Natural Resources Defense Council, Inc. v. United States EPA, 824 F.2d 1258, 1283 (1st Cir.1987). If the final rule deviates substantially from the proposed rule, it amounts to a new proposal and must run the regulatory gauntlet afresh. Thus, the ACCV’s right to be consulted is not stunted by the reading of § 300aa-14(d) that we adopt today.

To recapitulate, we believe that Congress might reasonably have inserted the phrase “any revision thereof’ to close what it suspected were potential loopholes. We therefore accept the Secretary’s thesis that the phrase “any revision thereof,” as used in § 300aa-14(d), refers exclusively to revisions of existing regulations (not to revisions of proposed regulations). On this understanding, we hold that the Secretary complied with the statutory notice-and-comment requirement by providing a pre-publication copy of her proposed regulation to the ACCV in December of 1991.

C. Adding and Subtracting Medical Conditions.

The petitioners’ final shot injects a new notion into the case: the idea that the Act does not authorize the Secretary to remove HHE and residual seizure disorders from the Table in the absence of “definitive information” attesting to the lack of any causal link between DPT vaccination and these medical conditions. Since both the IOM and the Secretary herself found only that there was “insufficient evidence to indicate a causal relation” between vaccination and the kind of permanent neurological damage reflected in HHE and residual seizure disorders over time, see IOM Report at 118; Notice, 57 Fed.Reg. at 36,879, they tell us that the deletions cannot survive.

The Secretary accepts the petitioners’ premise — the available evidence does not flatly disprove the causal relation — but she vigorously disputes the petitioners’ conclusion. In her view, the criteria for revising the Table simply do not include the requirement that the petitioners seek to impose. We agree with the Secretary.

We need not tarry. Nothing in the text of the Act prohibits the Secretary from eliminating a condition from the Table if the evidence of a causal relationship between the vaccination and that condition is equivocal. The only explicit constraints on the Secretary are procedural. See, e.g., 42 U.S.C. § 300aa-14(c) (requiring notice-and-comment period, including public hearing); id. § 300aa-14(d) (mandating referral to the ACCV). While some other constraints may be readily inferred from the terms, structure, and history of the Act, there is no principled basis for the added constraint that the petitioners would have us infer.

The petitioners’ construct rests solely on a suggestion in the committee report accompanying the Act to the effect that the Secretary may revise the Table when “more definitive information” is available. They read this terse reference as superimposing on the text of the statute a requirement that the Secretary must have definitive evidence rejecting a causal relation between vaccination and a medical condition before she may delete the condition from the Table.

The petitioners read the committee report through rose-colored glasses. The passage on which they rely is reproduced in its entirety in the margin.11 The passage as a whole makes it abundantly clear that, though Congress, straggling with a lack of information, itself used an initial presumption that the conditions listed in the Table were caused by vaccination so long as they occurred within the specified time period following vaccination, it did not intend to carve this presumption into stone. To the precise contrary, the authors of the committee report explicitly recognized that the Table, as originally devised, might in some cases go too far, and relied on the Secretary to reconstitute it in light of the “more definitive information” that would be available as a consequence of the review, in order to reflect more accurately the causal relations between vaccines and allegedly associated medical conditions.12

Had Congress intended the Secretary to revise the Table by removing a medical condition only after a causal link was definitely disproven, it could quite easily have said so. It said nothing of the sort. What it did say

is that the Secretary should update the Table in light of new and better information about causation. The Secretary, in pursuance of this directive, decided inter alia to remove HHE and residual seizure disorders from the Table because the medical evidence failed to establish a causal connection between DPT vaccines and these disorders. Since there is nothing in the record to suggest that this decision is arbitrary or capricious, it must stand. See, e.g., Strickland, 48 F.3d at 17-18; United States v. Members of the Estate of Luis Boothby, 16 F.3d 19, 21 (1st Cir.1994); see also 5 U.S.C. § 706(2)(A).

IV. CONCLUSION

We need go no further. The Secretary had authority to issue the regulation about which the petitioners complain, and she exercised that authority in a procedurally appropriate and substantively permissible manner. No more is exigible.

APPENDIX

VACCINE INJURY TABLE

I. DTP; P; DTP/Polio Combination; or Any Other Vaccine Containing Whole Cell Pertussis Bacteria, Extracted or Partial Cell Bacteria, or Specific Pertussis Antigen(s).

Illness, disability, injury, or condition covered: Time period for first symptom or manifestation of onset or of significant aggravation after vaccine administration:

A. Anaphylaxis or anaphylactic shock. 24 hours

B. Encephalopathy (or encephalitis). 3 days

C. Shock-collapse or hypotonic-hyporesponsive collapse. 3 days

D. Residual seizure disorder in accordance with subsection (b)(2). 3 days

E. Any acute complication or sequela (including death) of an illness, disability, injury, or condition referred to above which illness, disability, injury, or condition arose within the time period prescribed. Not applicable