Faced with the arduous demands of legislating for an increasingly complex society, Congress often leaves interstitial details to selected administrative agencies. Congress followed this praxis when it enacted the Family and Medical Leave Act of 1993 (FMLA), 29 U.S.C. §§ 2601-2654, delegating implementation to the Secretary of Labor (the Secretary). See id. § 2654.

Responding to this directive, the Secretary promulgated extensive regulations. See 29 C.F.R. §§ 825.100-825.800. At one point in the process, however, she caught the nearest way; in lieu of tailoring the definition of terms such as "impairment," "major life activities," and "substantially limits" to suit the peculiar needs of the FMLA, the Secretary simply co-opted existing definitions designed by a different agency -- the Equal Employment Opportunity Commission (EEOC) for use in connection with a different statute -- the Americans with Disabilities Act (ADA), 42 U.S.C. §§ 12101-12213. See 29 C.F.R. § 825.113(c)(2). Some perplexing difficulties lurk in the shadows cast by this cross-reference, including questions about the extent to which the EEOC's informal interpretations of the borrowed definitions are binding in the FMLA context.

This appeal brings those difficulties into sharp focus. It requires us to explore terra incognita -- to date, no other court of appeals has grappled with the meaning of the term "disability" under the FMLA and set the parameters of a mother's right to take an unpaid leave of absence in order to care for her seriously ill adult child. The able district judge, considering himself bound to defer unhesitatingly to an EEOC interpretive guidance devised with the ADA in mind, found that the mother had no such entitlement in the circumstances of this case and, accordingly, granted the employer's motion for summary judgment. Navarro Pomares v. Pfizer Corp., 97 F. Supp. 2d 208, 214 (D.P.R. 2000). We think that the court below acquiesced too readily in this interpretive guidance. For FMLA purposes, the guidance neither merits Chevron deference, see Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43(1984), nor possesses persuasive force. The objectives and structure of the FMLA, and the scope of the relief that it provides, require us to give effect instead to the regulation as written. Doing so, we reverse and remand for further proceedings.

Because the district court determined this case on summary judgment, we recount the essential facts in the light most favorable to the summary judgment loser. Suarez v. Pueblo Int'l, Inc., 229 F.3d 49, 53 (1st Cir. 2000).

Plaintiff-appellant Gladys Navarro Pomares (Navarro) began working for Pfizer Corporation as a secretary in 1994. On October 14, 1997, she requested an unpaid leave of absence until January 5, 1998; her plan was to travel to Germany so that she might minister to her adult daughter (Gladys Hernandez) and her two grandchildren. At the time she made this request, the appellant provided Pfizer with a note from her daughter's attending physician which reported that "Mrs. Hernandez is pregnant in 36th week. Because of high blood pressure bed rest is recommended to carry the baby to full term. So she cannot watch her other children."

Pfizer denied the appellant's request. She implored the company to reconsider. On October 25, having received no further response from her employer, the appellant departed for Germany. On November 6, she received correspondence from Pfizer directing her to return to work forthwith. The appellant remained at her daughter's bedside and Pfizer terminated her employment within the week.

Eleven months later, the appellant sued.1 She asserted that Pfizer had denied her leave to which she was entitled under the FMLA and then had added insult to injury by cashiering her for attempting to exercise her rights. When, thereafter, Pfizer moved for brevis disposition, the district court determined that the appellant was not entitled to FMLA leave and granted the motion. Navarro Pomares, 97 F. Supp. 2d at 214.

On appeal, we consider the appellant's asseveration that she raised a trial worthy issue anent her entitlement to FMLA leave. Because she has not renewed her retaliation charge, we deem that claim abandoned. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir. 1990).

We review orders granting or denying summary judgment de novo. Suarez, 229 F.3d at 53. The decisional path is well-trodden, so we borrow an earlier description of how the operative rule, Federal Rule of Civil Procedure 56, functions:

Once a properly documented motion has engaged the gears of Rule 56, the party to whom the motion is directed can shut down the machinery only by showing that a trial worthy issue exists. As to issues on which the summary judgment target bears the ultimate burden of proof, she cannot rely on an absence of competent evidence, but must affirmatively point to specific facts that demonstrate the existence of an authentic dispute. Not every factual dispute is sufficient to thwart summary judgment; the contested fact must be "material" and the dispute over it must be "genuine." In this regard, "material" means that a contested fact has the potential to change the outcome of the suit under the governing law if the dispute over it is resolved favorably to the nonmovant. By like token, "genuine" means that the evidence about the fact is such that a reasonable jury could resolve the point in favor of the nonmoving party.

McCarthy v. Northwest Airlines, Inc., 56 F.3d 313, 315 (1st Cir. 1995) (citations and some internal punctuation omitted).

Applying these tenets in a given case requires the court to scrutinize the summary judgment record "in the light most hospitable to the party opposing summary judgment, indulging all reasonable inferences in that party's favor." Griggs-Ryan v. Smith, 904 F.2d 112, 115 (1st Cir. 1990). If no genuine issue of material fact emerges, then the case may be ripe for summary adjudication.

The FMLA applies to private sector concerns that employ fifty or more persons. 29 U.S.C. § 2611(4). Congress enacted it as a means of alleviating the tension that so often exists between the demands of earning a living and the obligations of family life. See Hodgens v. Gen. Dynamics Corp., 144 F.3d 151, 159 (1st Cir. 1998); Price v. City of Fort Wayne, 117 F.3d 1022, 1024 (7th Cir. 1997). To achieve this objective, the FMLA seeks to balance authentic family needs and legitimate employer interests. See 29 U.S.C. § 2601(b)(1), (3). This accommodation entails a set of entitlements for employees and a matched set of rules for employers.

An employee becomes eligible for FMLA leave if he or she has been employed by a covered employer for no less than a year and has worked at least 1250 hours during the preceding twelve months. Id. § 2611(2)(A). Once eligible, an employee may take reasonable periods of unpaid leave for medical reasons, for childbirth or adoption, or for the care of a spouse, parent, or child who suffers from a serious health condition. Id. § 2601(b)(2). Leave periods are circumscribed: an eligible employee may take a maximum of twelve workweeks of FMLA leave in any twelve-month span. Id. § 2612(a)(1). Following such a leave, an employee is entitled to reclaim his or her former job (or some other position with equivalent pay, benefits, and conditions of employment). Id. § 2614(a)(1).

Ministering to sick children falls within a section of the FMLA that permits a period of leave "[i]n order to care for the . . . son [or] daughter of the employee, if such . . . son [or] daughter . . . has a serious health condition." Id. § 2612(a)(1)(C). In providing this protection, the FMLA differentiates between children under eighteen years of age and children eighteen years of age and older, defining a son or daughter as

a biological, adopted, or foster child, a stepchild, a legal ward, or a child of a person standing in loco parentis, who is --

(A) under 18 years of age; or

(B) 18 years of age or older and incapable of self-care because of a mental or physical disability.

The rules for employers are straightforward. In writing the FMLA, Congress took pains to proscribe employers from "interfer[ing] with, restrain[ing], or deny[ing] the exercise of or the attempt to exercise, any right provided" by the law. Id. § 2615(a)(1). To this end, an employer may not discharge or otherwise unfairly discriminate against an individual for opposing practices made illegal by the FMLA. Id. § 2615(a)(2). An employer who flouts these rules can be held liable for compensatory damages and, unless the violation occurred in good faith, additional liquidated damages. Id. § 2617(a)(1)(A). Appropriate equitable relief, such as reinstatement, also may be available. Id. § 2617(a)(1)(B).

In this case, it is undisputed that Pfizer was a covered employer, that the appellant was an eligible employee, that Hernandez was over eighteen years of age, and that 29 U.S.C. § 2611(12)(B) governed the appellant's claim of entitlement to the requested leave. Hence, the appellant's case depends upon whether her daughter (1) had a serious health condition, (2) was incapable of self-care, and (3) was so incapacitated because of a mental or physical disability. For summary judgment purposes, the first two steps in this pavane have been satisfactorily executed, but the third is problematic.

We start with the existence vel non of a "serious health condition." This phrase can denote "an illness, injury, impairment, or physical or mental condition that involves . . . continuing treatment by a health care provider." Id. § 2611(11). The regulations promulgated for the FMLA by the Secretary supply further insight: one way to demonstrate a serious health condition based on continuing treatment by a health care professional is to show that the underlying condition involves a period of incapacity due to pregnancy or for prenatal care. See 29 C.F.R. § 825.114(a)(2)(ii).

The appellant's evidence suffices to create a genuine issue of material fact as to whether her adult daughter was in the throes of a serious health condition. The doctor's certification, which plainly indicates that Hernandez's incapacity was tied to her pregnancy, serves this purpose. It follows that the appellant has made a showing adequate to withstand summary judgment on the first of the three required inquiries. See, e.g., Pendarvis v. Xerox Corp., 3 F. Supp. 2d 53, 55-56 (D.D.C. 1998) (denying summary judgment to employer in an FMLA case on the ground that any pregnancy-related period of incapacity, including morning sickness, constitutes a serious health condition).

We turn next to the question of whether Hernandez was able to care for herself. An individual is incapable of self-care if she "requires active assistance or supervision to provide daily self-care in three or more of the 'activities of daily living' (ADLs) or 'instrumental activities of daily living' (IADLs)." 29 C.F.R. § 825.113(c)(1). The same regulation defines ADLs to encompass "adaptive activities such as caring appropriately for one's grooming and hygiene, bathing, dressing and eating." Id. IADLs "include cooking, cleaning, shopping, taking public transportation, paying bills, maintaining a residence, using telephones and directories, using a post office, etc." Id.

Considering the broad sweep of these definitions, the doctor's note appears sufficient to create a genuine issue of material fact as to Hernandez's capability to care for herself. After all, her physician confined her to bed for the remainder of her pregnancy. At a bare minimum, such a prescription would appear to signal the patient's need for active assistance or supervision in the performance of everyday activities such as cooking, cleaning, shopping, and doing housework. Cf. Bryant v. Delbar Prods., Inc., 18 F. Supp. 2d 799, 803 (M.D. Tenn. 1998) ("[I]t is only logical to conclude that [plaintiff's son] could not cook, clean, shop or take public transportation . . . while he was in the hospital."). It follows that the appellant has made a showing adequate to withstand summary judgment on the second of the three required inquiries.

The district court appropriately analyzed the case to this juncture. It then pondered the third inquiry and ruled that the appellant had alleged no facts sufficient to support a reasoned conclusion that her daughter's impairment qualified as a "disability" (and, therefore, that the appellant had failed to raise a genuine issue of material fact regarding her eligibility for FMLA leave). Navarro Pomares, 97 F. Supp. 2d at 214. This is the nub of the case, and it presents a question of first impression at the appellate level. We approach this question -- the meaning of the term "disability" under 29 U.S.C. § 2611(12) -- mindful that the crucial moment for determining if a particular condition qualifies as a disability for FMLA purposes is the time that leave is requested or taken. See, e.g., Bryant, 18 F. Supp. 2d at 804.

Congress left the task of defining "disability" to the Secretary, see 29 U.S.C. § 2654, who reasonably concluded that a disability is an "impairment that substantially limits one or more of the major life activities of an individual." 29 C.F.R. § 825.113(c)(2). In so doing, she abjured any independent effort to define the terms "impairment," "major life activities," and "substantially limits," instead referring the reader, without explanation, to regulations issued by the EEOC pursuant to the ADA. See 29 C.F.R. § 825.113(c)(2). Given the provenance of these borrowed definitions, we think it is useful to proceed by asking the same three questions as in an ADA inquiry: (1) Is there a physical impairment? (2) What, if any, major life activity is implicated? (3) Does the impairment substantially affect the identified major life activity? Bragdon v. Abbott, 524 U.S. 624, 631 (1998).

A. Impairment.

The EEOC's regulations state that an impairment can be:

Any physiological disorder, or condition, cosmetic disfigurement, or anatomical loss affecting one or more of the following body systems: neurological, musculoskeletal, special sense organs, respiratory (including speech organs), cardiovascular, reproductive, digestive, genito-urinary, hemic and lymphatic, skin, and endocrine.

29 C.F.R. § 1630.2(h)(1). High blood pressure is such an impairment. See Murphy v. United Parcel Servs., Inc., 527 U.S. 516, 523-24 (1999). Because the source of an impairment is irrelevant to a determination of whether that impairment constitutes a disability, cf. Cook v. R.I. Dep't of MHRH, 10 F.3d 17, 24 (1st Cir. 1993) (stating that the Rehabilitation Act -- a precursor of the ADA -- contains no language suggesting that its prophylaxis is linked to how an individual became impaired), it makes no difference that the appellant's daughter's high blood pressure was induced by her gravidity.

Even were the source of the impairment relevant, the result would be the same. While pregnancy itself may not be an impairment, the decided ADA cases tend to classify complications resulting from pregnancy as impairments. See, e.g., Gabriel v. City of Chicago, 9 F. Supp. 2d 974, 981-82 (N.D. Ill. 1998); Hernandez v. City of Hartford, 959 F. Supp. 125, 130 (D. Conn. 1997); Cerrato v. Durham, 941 F. Supp. 388, 393 (S.D.N.Y. 1996). But see Tsetseranos v. Tech Prototype, Inc., 893 F. Supp. 109, 119 (D.N.H. 1995) (finding that neither pregnancy nor pregnancy-related conditions are impairments under the ADA). We agree with the majority view.3

These cases indicate to us that there is at least a genuine issue of material fact as to whether the appellant's daughter's high blood pressure constitutes an impairment under the ADA. We discern no reason why we should not similarly consider it a possible impairment for FMLA purposes. Consequently, the appellant has made a sufficient showing, at the summary judgment stage, on the impairment prong.

B. Major Life Activity.

We next must identify the impacted major life activity. See Bragdon, 524 U.S. at 631. The appellant has made a prima facie showing that her daughter was substantially limited in at least three such pursuits: working, caring for herself, and reproduction. For ADA purposes, the EEOC specifically acknowledges that both working and caring for oneself are major life activities, see 29 C.F.R. § 1630.2(i), and the Supreme Court has held that reproduction so qualifies, see Bragdon, 524 U.S. at 637-39. We see no reason why this taxonomy should not hold true under the FMLA as well. As a taxonomic matter, then, the appellant has made a sufficient showing, for summary judgment purposes, on the "major life activity" prong.

C. Substantially Limiting.

We now reach the crux of the parties' dispute: whether the specified impairment substantially limits the identified major life activity. See 29 C.F.R. § 1630.2(g)(1). We put the major life activities of reproduction and working to one side4 and focus on what we perceive to be the appellant's strongest case: her emphasis on the major life activity of caring for oneself.

According to the regulations, "substantially limits" means that an individual is:

(i) Unable to perform a major life activity that the average person in the general population can perform; or

(ii) Significantly restricted as to the condition, manner or duration under which an individual can perform a particular major life activity as compared to the condition, manner, or duration under which the average person in the general population can perform that same major life activity.

Id. § 1630.2(j)(1). For ADA purposes, the factors to be evaluated in assessing whether an individual is substantially limited in a major life activity include (1) the nature and severity of the impairment, (2) the duration or expected duration of the impairment, and (3) the permanent or long-term impact, or the expected permanent or long-term impact, of or resulting from the impairment. Id. § 1630.2(j)(2).5

The appellant's position is that a fact finder, drawing reasonable inferences in her favor from the doctor's note, could conclude, consistent with the borrowed material, that Hernandez's high blood pressure constituted an impairment that substantially limited her in major life activities (including the ability to care for herself). For summary judgment purposes, the employer does not contest the facts that undergird this claim, but, rather, posits that so fleeting an impairment -- one that may last no more than a matter of weeks -- cannot substantially limit a major life activity (and, therefore, cannot constitute a covered disability). Accepting this reasoning, the district court ruled that the appellant's daughter's condition was "a temporary, non-chronic impairment [] of short duration" and that, therefore, it did not amount to a disability. Navarro Pomares, 97 F. Supp. 2d at 214. Though plausible at first blush, this ruling cannot survive close scrutiny.

In holding that a "temporary, non-chronic impairment" did not constitute a disability, the lower court relied entirely on an EEOC interpretive guidance, 29 C.F.R. Pt. 1630, App § 1630(h), at 396, thereby implicitly if not explicitly granting Chevron deference to the EEOC's interpretation of its own rules. This was error: although Chevron deference to an agency's interpretive guidance is generally appropriate when a regulation is ambiguous and the agency's resolution of the ambiguity is a permissible construction of the regulation, Christensen v. Harris County, 529 U.S. 576, 587 (2000), the Supreme Court recently has clarified that not all agency interpretations merit Chevron deference. See United States v. Mead Corp., ___ U.S. ___,121 S. Ct. 2164, 2171 (2001). Pertinently, the Mead Court warned that "where statutory circumstances indicate no intent to delegate general authority to make rules with force of law, or where such authority was not invoked," a court must review agency interpretations under a less tolerant standard. Id. at 2177 (directing, in such circumstances, resort to the rule of Skidmore v. Swift & Co., 323 U.S. 134 (1944)). In the last analysis, then, such an agency interpretation is entitled to respect only to the extent that the interpretation has the power to persuade. Mayburg v. Sec'y of HHS, 740 F.2d 100, 106 (1st. Cir. 1984) ("[U]nder Skidmore the agency ultimately must depend upon the persuasive power of its argument. The simple fact that the agency has a position, in and of itself, is of only marginal significance.") (Breyer, J.).

This is significant here because an EEOC interpretive guidance issued pursuant to the ADA simply is not entitled to Chevron deference when applied in the FMLA context. The EEOC never had any authority to promulgate regulations pursuant to the FMLA. To the contrary, Congress explicitly delegated to the Secretary of Labor the sole authority to promulgate such regulations. Even if the Secretary adopts certain EEOC rules as her own (as happened here), she does not automatically adopt the EEOC's informal interpretations of those rules. Moreover, the EEOC itself has been granted no rulemaking power under the FMLA, and therefore its interpretive guidance is certainly not entitled to deference. Indeed, it borders on the Kafkaesque to suggest that the EEOC, acting some three years before Congress passed the FMLA, had invoked the authority delegated to the Secretary of Labor and written interpretations to govern an as-yet-unenacted statute. Accordingly, we decline to grant Chevron deference to the EEOC's interpretive guidance and instead apply the Skidmore standard.

Despite the concerns of our dissenting brother, this seems to us a bedrock principle of administrative law. After all, a court cannot blindly defer to the interpretations of an administrative agency simply because that agency has expertise in a field that bears some relation to the statute at issue. To warrant Chevron deference, Congress must actually delegate authority to that agency, and the agency must invoke that authority.

Where, as here, an agency's pronouncement (in this instance, the EEOC's interpretive guidance) fails to meet these criteria, an inquiring court must scrutinize that pronouncement and question whether it is in harmony with the statute and the regulations. See Joy Techs., Inc. v. Sec'y of Labor, 99 F.3d 991, 996 (10th Cir. 1996); (explaining that regulations should be construed to mesh with the objectives of the statute that they implement); Dunn v. Sec'y of USDA, 921 F.2d 365, 367 (1st Cir. 1990) (similar); see also Martinez v. R.I. Hous. & Mortg. Fin. Corp., 738 F.2d 21, 26 (1st Cir. 1984) (noting that "a rule out of harmony with the statute is a mere nullity"). The results of this inquiry will, in turn, determine the persuasive force of the interpretive guidance. We turn, then, to the question of whether the interpretive guidance passes Skidmore muster when applied in an FMLA context.

Under Skidmore, we are constrained to weigh the "thoroughness evident in [the guidance's] consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control." Skidmore, 323 U.S. at 140. The EEOC's guidance does not fare well when measured against these benchmarks.

We can find no thoroughness evident in the consideration of the guidance. For one thing, an interpretive guidance, much like "interpretations contained in policy statements, agency manuals, and enforcement guidelines," is not the product of notice-and-comment rulemaking or formal adjudication. Christensen, 529 U.S. at 587. For another thing, this guidance simply was not meant to apply in the FMLA context; the EEOC promulgated it well before the FMLA was anything more than a gleam in its sponsors' eyes. By like token, the guidance is idiosyncratic; it has little consistency with other EEOC pronouncements on the FMLA as the EEOC has made no such pronouncements.

This interpretive guidance, moreover, cannot be reconciled with the fundamental premise that a balancing test should be pliant, the scale weighted differently in each case. The Supreme Court has cautioned that "in the context of a rule based on a multifactor weighing process[,] every consideration need not be equally applicable to each individual case." FCC v. Nat'l Citizens Comm. for Broad., 436 U.S. 775, 808 n.29 (1978). The regulation here at issue constructs just such a balancing test, and the Supreme Court's caveat conduces to the view that the regulation's list of factors should not be treated as some sort of mandatory checklist (even in the ADA context).6 The Court's heavy emphasis on the individualized nature of what constitutes a disability for purposes of the ADA, see Albertson's, Inc. v. Kirkingburg, 527 U.S. 555, 566 (1999); Sutton v. United Air Lines, Inc., 527 U.S. 471, 483 (1999); Bragdon, 524 U.S. at 641-42, reinforces the desirability of a flexible case-by-case approach.

Indeed, the Court's ADA jurisprudence strongly suggests that the three factors contained in the borrowed regulatory definition of "substantially limits" should not be given equal weight. When considering the definition of "disability" under the ADA, the Justices have maintained a steady focus on the present state of an individual's impairment. The Sutton Court observed that "[b]ecause the phrase 'substantially limits' appears in the Act in the present indicative verb form, we think that language is properly read as requiring that a person be presently -- not potentially or hypothetically -- substantially limited in order to demonstrate a disability." Sutton, 527 U.S. at 482. This keen attention to the statute's verb tense is persuasive evidence that an individual's present, actual state (rather than a hypothetical, projected state) is paramount in determining whether he or she suffers from a disability. In turn, this designated point of reference militates against according talismanic effect to factors such as duration and long-term impact, which may require the fact finder to hypothesize as to the future course of the impairment. AccordKatz v. City Metal Co., 87 F.3d 26, 30-32 (1st Cir. 1996) (reversing directed verdict for employer on ADA claim even though plaintiff had presented almost no evidence as to the duration of his impairment); EEOC Compliance Manual § 902.4(a), at 5311 (1999) (calling the "extent to which an impairment restricts one or more of an individual's major life activities" the hallmark of a disability under the ADA, and noting that the impairment's duration is no more than a "secondary factor that may affect the analysis") (emphasis supplied). The mechanistic assumption that all the enumerated factors invariably must be present before an impairment can be termed "substantially limiting" in an ADA case is, therefore, unfounded. The argument against the assumption is even more cogent in an FMLA case.

Most importantly, the EEOC interpretive guidance cannot be applied to the FMLA because it clashes with the underlying purposes of the statute. The ADA and the FMLA have divergent aims, operate in different ways, and offer disparate relief. These dissimilarities argue convincingly that the trio of factors - particularly duration - must be treated somewhat differently in the FMLA context than in the ADA context. Cf. Chevron, 467 U.S. at 863-64 (finding definition of "source" to be flexible and approving EPA's varying interpretations of it in different contexts); Stowell v. Sec'y of HHS, 3 F.3d 539, 542 (1st Cir. 1993) (deeming it "apodictic that Congress may choose to give a single phrase different meanings in different parts of the same statute").

Two salient considerations fortify this conclusion. First, the concept of disability serves a much different function in the ADA than in the FMLA. Where the ADA is concerned, a finding of disability is the key that unlocks the storehouse of statutory protections. Title I of the ADA provides that a covered employer may not discriminate against a qualified individual with a disability because of that disability. 42 U.S.C. § 12112(a). This means that the employer must, inter alia, make "reasonable accommodations to the known physical or mental limitations of an otherwise qualified individual with a disability" as long as that disability persists, unless and until those accommodations impose an undue hardship on the employer. Id. § 12112(b)(5)(A). Such accommodations can take various forms, and the duty to accommodate is an ongoing responsibility that is not exhausted by a single effort on the employer's part. Garcia-Ayala v. Lederle Parenterals, Inc., 212 F.3d 638, 648 n.12 (1st Cir. 2000). Given the centrality of a finding of disability under the ADA and the panoply of rights and responsibilities that such a finding triggers, it makes sense to insist that, in most cases, an impairment have an extended duration before it will be deemed so limiting as to constitute a disability.

In contrast, the only time that the concept of disability becomes relevant under the FMLA is in the relatively rare instance in which an employee seeks FMLA leave to care for a seriously ill child over the age of eighteen. Even then, the existence vel non of a disability simply provides a partial answer to the question of whether the employee is entitled to leave. See 29 U.S.C. §§ 2612(a)(1)(C), 2611(12)(B). The minor role that the disability determination plays in the context of the FMLA one of three criteria to be met in respect to the infirmity of an adult child before a modest unpaid leave can be taken -- indicates that very little weight should be placed on the duration of an impairment. This is especially so since the duration of the impairment is not even likely to determine the precise term of an FMLA leave, which is far more apt to be measured by how long the child's serious health condition lasts or how long the child is incapable of self-care. See, e.g., Bryant, 18 F. Supp. 2d at 804 (reporting that while son's disabling kidney condition persisted over a period of years, plaintiff's FMLA leave was only part of one day).

The second consideration that leads us to believe that factors such as duration must be accorded reduced significance in the FMLA context is that the FMLA deals in much lower levels of employer engagement and employee rewards than does the ADA. For one thing, the FMLA implicates shorter time frames: an employee may qualify for FMLA leave to care for a child under eighteen merely by showing that the child suffers from a serious health condition, which, as defined, can be an illness that lasts as little as four days. See 29 U.S.C. § 2612(a)(1)(C); 29 C.F.R. § 825.114(a)(2)(i); see also Brannon v. Oshkosh B'Gosh, Inc., 897 F. Supp. 1028, 1037 (M.D. Tenn. 1995) (finding that a child who had a fever, was taken to a doctor, and stayed home from day care from Friday through Tuesday had a serious health condition within the purview of the FMLA). For another thing, the maximum annual benefit under the FMLA is twelve weeks of unpaid leave, see 29 U.S.C. § 2612(a)(1), whereas reasonable accommodations under the ADA can last for years on end. Finally, the obligatory interactive process that is a staple of the ADA, see, e.g., Criado v. IBM Corp., 145 F.3d 437, 444 (1st Cir. 1998), is entirely foreign to the FMLA. We mention these contrasting levels of engagement and reward because we think it highly unlikely that, with so much less at stake under the FMLA, Congress would have required FMLA litigants to make the same durational showing as ADA litigants.

Having established that the differences between the ADA and the FMLA render the durational factor less important under the latter statute, we turn to the purpose of the FMLA and the light that it sheds on that factor's proper role. A regulation must harmonize with the purpose of the statute it implements. See Grunbeck v. Dime Sav. Bank, 74 F.3d 311, 336 (1st Cir. 1996) ("[C]ourts will reject an agency interpretation which conflicts with congressional intent."). The FMLA's primary purposes are "to balance the demands of the workplace with the needs of families, to promote the stability and economic security of families, and to promote national interests in preserving family integrity." 29 U.S.C. § 2601(b)(1). Those objectives would be frustrated by reading the implementing regulations through the prism of the EEOC's interpretive guidance, for this would impose a rigid requirement that an employee must prove that an impairment is long-lasting before it can qualify as substantially limiting (and, thus, furnish the basis for FMLA leave).

We illustrate this point with a practical example. A worker who seeks to take FMLA leave to care for a child often does so in response to a crisis situation. See, e.g., Caldwell v. Holland of Tex., Inc., 208 F.3d 671, 673 (8th Cir. 2000) (three-year-old son's sudden ear infection); Bryant, 18 F. Supp. 2d at 802 (adult son's unanticipated kidney failure). In many instances, the emergency will have abated by the time that the duration of the child's impairment can be ascertained. If a hard-and-fast durational requirement is enforced, an employee will be effectively prevented from taking family leave to care for an adult child until it can be established that the child's problem will have an adequate duration. By then, the crisis may well have passed.

Such a scenario would place an employee with a sick adult child between a rock and a hard place, forcing him or her to choose between employment demands and family needs. This would run at cross purposes with the FMLA's goal of reassuring workers that "[w]hen a family emergency arises . . . they will not be asked to choose between continuing their employment, and meeting their personal and family obligations." 29 C.F.R. § 825.101. We do not believe that Congress intended to create so illusory a benefit.

The foregoing analysis of the purpose of the FMLA, its structure, and the relief it provides leads us to conclude that -- as the borrowed definition provides -- the duration of an impairment is one of several factors that should be considered in determining the existence of a disability under the FMLA. We also conclude, however, that Congress did not intend that the impairment always be shown to be long-lasting. This last conclusion comports with the major goals of the statute while at the same time respecting Congress's clear intent to set a higher bar for a parent's leave entitlement to care for a child eighteen years of age or older. See 29 U.S.C. § 2611(12). All that is required to ground a leave request vis-a-vis a younger child is that the child have a serious health condition, see id. §§ 2612(a)(1)(C), 2611(12)(A), which can be an illness lasting as little as four days, see 29 C.F.R. § 825.114(a)(2)(i). In comparison, to be eligible for leave to care for an older child, the child not only must have a serious health condition but also must lack the capacity to care for himself or herself due to a disability (which requires demonstrating an impairment, identifying a major life activity, and showing how the impairment substantially limits the major life activity). See 29 U.S.C. § 2611(12)(B); 29 C.F.R. § 825.113(c)(2). Thus, we do not "overlook[]," as our dissenting brother charges, that Congress intended a meaningful consequence to attach to its use of the word "disability" in 29 U.S.C. § 2911(12)(B). Rather, we recognize that although the regulation, 29 C.F.R. § 825.113(c)(2), properly interpreted (i.e., without regard to the EEOC's interpretive guidance), does not eliminate duration as relevant factor, it leaves room for an impairment of modest duration to be regarded, in some cases, as "substantially limiting" for FMLA purposes.7

We now return to the case at hand. At this point, we crystalize the insights derived from our investigation into the nuances of administrative law and the comparative jurisprudence of the ADA and the FMLA. We distill these insights into a tripartite rule: (1) Courts facing the question, under the FMLA, of whether an adult child's impairment substantially limits a major life activity should apply the Secretary's borrowed regulation, 29 C.F.R. § 825.113(c)(2), as written, ignore the EEOC's unpersuasive interpretive guidance (crafted for use in connection with a different statute), and consider (a) the nature and severity of the impairment, (b) its expected duration, (c) its anticipated long-term impact, and (d) any other relevant factors. (2) This assessment must be performed on a case-by-case basis, balancing all factors in light of the FMLA's purpose, structure, and provisions for relief. See O'Coneell v. Shalala, 79 F.3d 170, 176 (1st Cir. 1996) (emphasizing the need to afford statutes a practical, commonsense reading that gives due weight to design, structure, purpose, and overall language). (3) The requisite test is a balancing test: apart from the severity of the impairment, no one factor is indispensable to finding that a disability exists for FMLA purposes.

Applying this rule, we hold that the provisions of 29 U.S.C. § 2611(12)(B) may be satisfied by various combinations of factors. One such permissible combination entails, at least in certain circumstances, a showing that the employee's adult child is suffering from a severe impairment which has a modest projected duration and an as-yet-unquantified long-term impact.8 The case before us fits within the contours of that category. It follows that the court below improvidently granted summary judgment for Pfizer. We expound on this conclusion.

High blood pressure is, by its nature, a serious impairment. Given that the attending physician ordered Hernandez confined to bed, a fact finder reasonably could regard its manifestation as severe. As to duration, the appellant provided evidence that Hernandez's high blood pressure would last at least to the end of her pregnancy, an interval of several weeks. Long-term impact hardly seems relevant to the appellant's leave request, see supra note 6, but in all events, Hernandez's condition arguably might persist after childbirth and have a lasting impact. Crediting this evidence, as we must at the summary judgment stage, the record seems adequate to support a finding -- although it surely does not compel one -- that the appellant's daughter had a "disability" within the purview of the FMLA.

We summarize succinctly. Taking the evidence as it stands, drawing all reasonable inferences in the appellant's favor, and applying the appropriate legal standard, there is a jury question as to whether Hernandez's high blood pressure substantially limited her in the major life activity of self-care. From the evidence, a jury could find that, at the time the appellant requested leave, her bedridden daughter was "[s]ignificantly restricted as to the . . . manner" in which she could perform the major life activity of self-care "as compared to the . . . manner [in] which the average person in the general population [could] perform that same major life activity." 29 C.F.R. § 1630.2(j)(1)(ii). The jury also could find that Hernandez's condition, though not proven to be of extremely protracted duration, threatened to persist long enough to qualify her as disabled for purposes of the FMLA. Consequently, the lower court erred in granting summary judgment.

We need go no further. Reading the statutory text and the applicable regulation with an eye toward congressional purpose and practical consequences, and disregarding the EEOC's interpretive guidance, see Mead, 121 S. Ct. at 2177; Skidmore, 323 U.S. at 139-40, we hold that the district court's decision overstates the importance of the durational element to the determination of the existence of a disability under the FMLA. Since the summary judgment previously entered in the employer's favor hinged on that overstatement, it must be set aside.

Reversed and remanded.

CAMPBELL, Senior Circuit Judge, dissenting.

While my colleagues' result is humanly appealing, I cannot agree with it. It seems to me to run counter to proper standards of legal analysis and to substitute judicial discretion for that conferred upon the Secretary of Labor. I would affirm the district court.

Appellant complains that her employer violated the Family Medical Leave Act (FMLA or the Act) by refusing to grant leave so that she could look after her pregnant daughter. In her thirty-sixth week of pregnancy, the daughter had been placed on bed rest because of a pregnancy-induced hypertension. Because the district court found the daughter was not "disabled" -- a threshold requirement for FMLA leave in order to care for an adult child, "disability" being defined in the Act's regulations in terms of ADA criteria -- the district court dismissed appellant's FMLA claim. My colleagues now reverse that judgment. They do not suggest that, under ADA standards, appellant's daughter was disabled, but rather they hold that the FMLA requires a more relaxed standard of disability than does the ADA -- one with little or no durational requirements. I think my colleagues both misread the FMLA and improperly override the authority given by Congress to the Secretary of Labor to prescribe regulations in this area.

To explain, I begin with the words of the statute. In enacting the FMLA, Congress specifically distinguished between entitlements of leave to care for minor children and of leave to care for adult children. The Act broadly grants leave rights to employees "[i]n order to care for the spouse, or a son, daughter, or parent, of the employee, if such spouse, son, daughter, or parent has a serious health condition," 29 U.S.C. § 2612(a)(1)(C). However, the Act then proceeds to provide a limiting definition of who may be considered to be a son or a daughter: "[S]on or daughter" is defined as the employee's child who is "(A) under 18 years of age; or (B) 18 years of age or older and incapable of self-care because of mental or physical disability." 29 U.S.C. § 2911(12)(emphasis supplied). The statute's clause (B) thus imposes a significant limitation upon the class of adult children for whose care parental leave is mandated. Leave is provided solely to care for those adult children fitting within the category of children who are "incapable of self-care because of ... disability." Notably, the statute imposes no such disability limitation in respect to leaves to care for minor children, spouses and parents. While the majority brushes aside as "weak" the above limitation pertaining to adult children only, the fact that it was crafted as part of the very definition of the class for whose care leaves may be granted accentuates its importance. Whether or not we like the limiting phrase in § 2911(12)(B), we cannot responsibly ignore or downplay it. See Massachusetts Ass'n of Health Maintenance Organization v. Ruthardt, 194 F.3d 176, 181 (1st Cir. 1999) ("[A]ll words and provisions of statutes are intended to have meaning and are to be given effect, and no construction should be adopted which would render statutory words or phrases meaningless, redundant or superfluous.")(internal quotation marks omitted).

The Senate Report relating the legislative history of the FMLA illuminates Congress's reasons for inserting this unique provision, which limits leave to parents to care for their seriously ill adult children to only those children "incapable of self-care because of ... disability." This legislative history deserves repeating, ante note 2, as it clarifies the critical issue in this appeal: the significance and meaning of the phrase "incapable of self care because of mental or physical disability." The majority simply describes this history as "amorphous" and thereafter treats the Senate Report as essentially without import. Nevertheless, the Senate Report states:

The bill [FMLA] thus recognizes that in special circumstances, where a child has a mental or physical disability, a child's need for parental care may not end when he or she reaches 18 years of age. In such circumstances, parents may continue to have an active role in caring for the son or daughter. An adult son or daughter who has a serious health condition and who is incapable of self-care because of a mental or physical disability presents the same compelling need for parental care as the child under 18 years of age with a serious heath condition.

S. Rep. No. 103-3, at 22 (1993), reprinted in 1993 U.S.C.C.A.N. 3, 24.

The most obvious and reasonable construction of the above quoted passage -- a reading that the Secretary of Labor's regulations faithfully mirror, infra -- indicates that Congress wanted to restrict leave benefits for parents to care for their adult children 18 and older to only those special cases where because of some mental or physical disability the adult child is or remains especially dependent on the parent in the same ways minor children typically are dependent.1 In other words, for the adult child to trigger leave rights, it is not enough that he or she be seriously ill; the child must also be "incapable of self-care because of ... disability," a factor Congress deemed essential to place the adult child within a special class deserving of parental care for FMLA leave purposes. The Senate Report explains that in such cases "a child's need for parental care may not end [as presumably happens otherwise] when he or she reaches 18 years of age. ... [And] parents may continue to have an active role in caring for the son or daughter." Id. It is clear from this passage that Congress contemplated an adult child who is especially dependent over some period of time on parental care for physical or mental reasons.2 This is in contrast to the typical scenario in which, at and after age 18, a child may be regarded as having achieved substantial independence and self-sufficiency so as to be able to live on her own, support herself, and be ministered to by others than her parents. By restricting parental leaves to unemancipated minor children and a restricted class of adult children who may still require some sort of on-going parental care, Congress was imbuing the word "disability" with a serious and severe consequence, one which, however, the majority simply overlooks in its assumption that the sole and overriding purpose of the FMLA is to provide a liberal leave policy in all instances.

For the above reasons, I believe that Congress's intentions along the lines indicated are amply signaled both in the statutory language and the Senate Report. The FMLA, however, leaves to the Secretary of Labor a major role in the interpretation of Congress's wishes, hence I turn next to the Secretary's regulations. The FMLA provides that the "Secretary of Labor shall prescribe such regulations as are necessary to carry out ... this chapter...." 29 U.S.C. § 2654. In promulgating regulations to enforce congressional intent, including the difference in the statute between leave granted to care for minor children and leave granted to care for adult children, the Secretary of Labor has defined the relevant terms "incapable of self-care" and "physical or mental disability." And, in so doing, the Secretary has construed the terms precisely in accord with the congressional intent one would glean from the construction of the statute and the Senate Report I have just set forth.

As the majority notes, with regard to the term "physical or mental disability," instead of defining from scratch the term "disability" for the purposes of the FMLA, the Secretary has borrowed the statutory ADA definition as further refined in regulations issued by the EEOC pursuant to the ADA. The Secretary states in her FMLA regulations that

physical or mental disability means a physical or mental impairment that substantially limits one or more of the major life activities of an individual. [The ADA definition.] Regulations at 29 C.F.R. § 1630.2(h),(i), and (j), issued by the Equal Employment Opportunity Commission under the Americans with Disabilities Act ... define these terms.

29 C.F.R. § 825.113(c)(2)(2000)(emphasis supplied). The relevant ADA provision and EEOC regulations are undoubtedly restrictive in their scope with regard to the kinds and durations of impairments that will qualify as disabilities.3 Nonetheless, as already discussed above, Congress's very purpose in using the term "disability" seems to have been to limit the class of adult children for whose benefit leave was mandated. The ADA and EEOC definitions achieve this task very reasonably. They effectuate congressional intent to differentiate between entitlement to leave to care for, on the one hand, a narrower class of adult children whose long-term afflictions limit major life activities, and, on the other hand, all seriously ill minor children whether or not disabled.

And in cross-referencing to the ADA and to the EEOC regulations, the Secretary achieves advantages that would be lost were she to have defined the term "disability" by new and separate regulations tailored solely to the FMLA. By cross-referencing, the Secretary makes use of interpretations developed and being developed in another relevant on-going regulatory scheme, thereby achieving more precise standards in what -- given the vagueness of the term "disability" -- could otherwise be a chaotic area of interpretation.

Given what Congress was attempting to accomplish by creating in the FMLA a narrower, more needy class of adult children, I see nothing wrong or unreasonable with the Secretary's giving the FMLA term "disability" the exact same meaning as provided in the ADA and its interpretive regulations. The ADA pre-dates the FMLA and is perhaps the primary federal statute dealing with the subject of disability. Its incorporated concept of a relatively long-term physical or mental condition ties precisely into what the FMLA intended when separating out all adult children whose need for parental care has ended from those whose "parents may continue to have an active role" in their care. S. Rep. No. 103-3, at 22 (1993), reprinted in 1993 U.S.C.C.A.N. 3, 24. Over time, the ADA definition has been refined by regulation and administrative and judicial precedent so that by now its meaning in many (or even most) situations has become relatively clear. The Secretary's cross-reference to the ADA's definition of "disability" with its concomitant history and administrative and judicial guidance makes it possible for employers, employees and tribunals interpreting the FMLA to refer to well-established coherent principles and precedent, providing predictability and clarity to a term "disability" that, by itself, is anything but plain.

Viewed this way, the Secretary's borrowing of ADA and EEOC criteria to define "disability" under the FMLA makes eminent good sense. These criteria, it is true, will limit leave to parents whose sons or daughters suffer from more chronic, fairly long-term physical and mental handicaps. Borrowing and faithfully applying to the FMLA the ADA's disability definition means that parental leaves will not be available in all situations where leaves, from a purely compassionate point of view, may seem equally well-justified - as where, for example, an adult child needs but lacks care, yet falls short of having a "disability" within the definition of that term in the ADA and the applicable EEOC regulations. But, given both the plain statutory language and Congress's intent as explained in the Senate Report in inserting the limiting phrase "incapable of self-care because of mental of physical disability," I cannot see how one can contend that the limitation inherent in borrowing from the ADA and its precedent runs counter to the objectives of the FMLA.

I therefore believe that the district court's analysis and judgment is correct on the facts of this case. The pregnancy-related medical condition of appellant's daughter lacked sufficient duration to be a "disability" as that term is used within the ADA as further defined by the EEOC regulations. Indeed, my colleagues do not seem to contend otherwise. That should end the matter.

Instead, however, my colleagues insist that because this is an FMLA case, a different, more relaxed durational standard of their own invention needs to be read into the ADA and EEOC criteria. This they term a "balancing" process. But balancing usually means balancing the facts of a case against statutory and regulatory standards, not altering the latter at will. In any case, I see no contradiction between the intent of Congress when using the term "disability" in the FMLA context and the ADA "disability" definition and related EEOC criteria as used in an ADA context. Curiously, my colleagues do not suggest that borrowing the ADA definition of "disability" and related EEOC regulations constituted legal error by the Secretary. They accept the Secretary's borrowing from the ADA and EEOC but then say the same regulations should mean different things depending on whether used in an ADA case or in an FMLA case. Not only do I find this incomprehensible, but I can see no reason for attempting such an exercise given the close fit, see supra, between Congress's reasons for using the term "disability" in the FMLA and the meaning of that term as developed in ADA case law.

In taking a different view, my colleagues point only to the FMLA's broadly stated, and by no means self-explanatory, statutory purpose of balancing family needs with employer interests. But striking the balance so as to favor only persons disabled under ADA criteria appears to meet this very principle given the statutory language and the intent of Congress as set out in the Senate Report. Of course, one may still argue as a matter of personal preference that it would be better to strike the balance differently or more in favor of the family, but it is not our business as judges to choose among competing policies where the statutory text, the legislative history, and the Secretary's interpretation are all so plainly in accord.

In effect, my colleagues are instructing courts to turn their backs on the Secretary's entirely rational invocation of the ADA standards - standards which, if applied as construed in ADA cases, reasonably effectuate congressional purpose in using the term "disability" here. The result of the majority's opinion will be simply to destabilize the meaning of the FMLA in an area requiring clarification, not greater obscurity. The only future guidance the majority gives to litigants and the courts is to "balance" and presumably to follow the majority's preference for granting leaves liberally to all parents with sick adult children. This approach effectively reads the phrase "incapable of self-care because of ... disability," as applied solely to adult children, out of the FMLA. I see no proper legal justification for this position.

To be sure, in reversing the district court, the majority says that it does not dispute the Secretary's importation of the ADA and its concomitant EEOC regulations into the FMLA. The majority says it takes issue only with the district court's consideration of the EEOC's interpretive guidance in its analysis of the plaintiff's entitlement to FMLA leave. In particular, the majority contends that it was error for the district court to rely on the EEOC's interpretive guidance that a "temporary, non-chronic impairment" does not constitute a disability. 29 C.F.R. Pt. 1630, App § 1630(h), at 396. Such reliance was error, says the majority, because the EEOC's interpretive guidance, issued pursuant to its own regulations promulgated to enforce the ADA, merits no deference in the context of another, distinct statute, such as the FMLA. Ante at 98-99 (citing to classic administrative law jurisprudence, such as Chevron and Skidmore, as modified by recent Supreme Court cases, such as Mead and Christensen). In so holding, the majority seems to imply either that the EEOC's interpretive rule flies in the face of the FMLA or that absent reliance on the EEOC guidance, the result below would have been different. I see no basis for either position. As already suggested, and as further discussed, the ADA's durational concepts, as clarified in the EEOC's interpretive guidance, mesh well with the purpose of the term "disability" as used in this part of the FMLA. Moreover, I fail to see how one can reject the EEOC's interpretive guidance without also rejecting the ADA or the EEOC's regulation, both of which the Secretary expressly adopts and which my colleagues do not question.4 The result of the majority's opinion, even absent consideration of the EEOC interpretive guidance, is to read out of the Secretary's and the EEOC's regulations any requirement that the plaintiff-parent provide some evidence of the duration and long-term impact of their child's impairment. This, I think, we are not permitted to do. I explain briefly.

Let us assume arguendo, in line with my brethren's conclusion, that the EEOC interpretive guidance deserves no deference (despite my own belief that utilization of the provision is entirely sound, see supra note 4). The outcome, in my view, would still be to affirm the district court by relying solely on the ADA standards and EEOC regulations, the reasonableness of which no one -- not even my colleagues -- disputes.

In finding a genuine dispute of fact as to whether Navarro's daughter is disabled within the meaning of the FMLA (i.e., within the meaning of the ADA minus the EEOC interpretive guidance), the majority argues that the balancing required of the ADA and EEOC factors, see supra note 3, "should not be treated as some mandatory checklist." Ante at 99-100. From this, the majority explains that the Supreme Court has decided that each factor of the "substantially limits" prong as illuminated by the EEOC "should not be given equal weight." Ante at 100 (citing Sutton v. United Air Lines, Inc., 527 U.S. 471, 481-82 (1999)). In particular, the majority argues that the durational and long-term impact factors (numbers (ii) and (iii) at 29 C.F.R. § 1630.2(j)(2), see supra note 3) should not be accorded "talismanic effect". As this is so in the ADA context where the label of "disability" is the touchstone for any relief under the statute, it should be even more true, they say, in the FMLA context, given that the FMLA provides only short-term relief and that the term "disability" becomes relevant only with regards to relatively rare cases in which FMLA leave is requested to care for an adult child. A proper balancing on this record of the EEOC factors in light of the FMLA's distinct purpose (as compared with the ADA), leads the majority to conclude that a jury could find that Navarro's daughter is disabled despite the lack of evidence that her impairment would last beyond the three weeks remaining in her pregnancy.

The difficulty with this analysis is that it disregards the ADA's and the EEOC's requirement that some consideration be given both to duration and long-term or permanent impact. The plaintiff has pointed to no evidence that would support an inference that Navarro's daughter would not fully recover upon the birth of her child from her pregnancy-induced hypertension, an impairment that developed in her thirty-sixth week of pregnancy. As was the case, the record shows that plaintiff requested leave on October 14, 1997 to begin on October 25, 1997. Plaintiff's daughter gave birth on October 26, 1997. The only evidence offered by plaintiff in support of her contention that her daughter qualified as disabled under the FMLA was a physician's certificate wherein her daughter's doctor certified that "Navarro's daughter was in her thirty-sixth week of pregnancy, was suffering from high-blood pressure, and had been placed on bed rest so that she could bring her fetus to term, which made her incapable of caring for her two young children." As the district court stated "[p]laintiff[] does not allege that [her] daughter was suffering from high blood pressure throughout much of her pregnancy, or that her condition would have any long-term or permanent impact." Without more, there is simply not enough evidence to raise a genuine issue of material fact as to the existence of a disability as defined by the ADA, unless, of course, we do not consider duration or long-term impact at all, two of the three factors the ADA's regulations require a court to consider.

In disregarding these two factors, the majority contends it is merely "balancing" and, in so doing, according little weight to the duration and long-term impact prongs in view of the FMLA's purpose. With respect, I see no balancing whatsoever in light of the total absence of evidence, as described above, regarding any duration or long-term impact of Navarro's daughter's hypertension beyond the three weeks left in her pregnancy. Morever, I see no basis for the court to rely on the FMLA's broadly stated purposes (e.g., balancing "the demands of the workplace with the needs of families," 29 U.S.C. § 2601(b)(1)("Purposes [for the FMLA]")) as a reason to accord so little weight to two of the three EEOC factors (if any weight could in fact be given on this bare record). The majority is, in effect, using the FMLA's generally stated aims to overturn what are otherwise the specifics of the operation of the Secretary's regulation. As noted, however, Congress seems to have used the term "disability" deliberately and precisely to limit in certain respects the granting of parental leaves in situations involving adult children. It is circular reasoning to evade that deliberate limitation, as adduced by Congress and construed by the Secretary, by reference to nothing more than bland and by their nature imprecise statutory objectives.

On this record, applying only the ADA standards and its EEOC regulations and adducing no durational or long-term impact of plaintiff's daughter's impairment, no reasonable jury could conclude that Navarro's adult daughter had a "disability" and that Navarro was, therefore entitled to leave to care for her under the FMLA.

As Congress gives the parents of some seriously ill adult children a statutory right to leave (their children having a "disability"), and yet denies leave to other parents of adult children whose situations may be equally disturbing (their children not having a "disability" but being nonetheless seriously ill), any line-drawing in this area will obviously to some degree be discomfiting. Unhappiness over this dilemma seems to be the impetus behind the majority opinion.

As an antidote, my colleagues have created a legal cure that is, in my view, worse than the disease. They have rejected the durational aspects of the ADA definition of "disability", thus blurring the line drawn in the FMLA between minor children and adult children, a line that Congress itself inserted into the statute. The reasons they offer for doing so are that a strict application of the EEOC factors would, in their opinion, be out of harmony with the general aims of the FMLA and that, under the ADA, the Supreme Court has mandated "balancing." For the reasons stated in Parts I and II of this dissenting opinion, however, it is by no means obvious that this lack of harmony exists or that a proper balancing does not lead to an affirmance of the district court's judgment. Congress itself added the disability condition to leaves for parents to care for their adult children, while inserting no such condition limiting the granting of leaves to care for parents, minors or spouses. Compare 29 U.S.C. § 2611(12)(B)(children 18 years of age or older) with 29 U.S.C. § 2611(7)("Parent"),(12)(A)(children under 18 years of age) and (13)("Spouse"). Thus, by using the term "disability", Congress rather clearly intended to place special limits, not imposed for care of other family members, on leaves to provide parental care for children 18 years of age or older - otherwise why require that adult children, but not others, be "incapable of self care because of ... disability" in addition to having a serious health condition which in all other cases would alone justify a leave? See Senate Report, supra. By eviscerating the difference Congress clearly intended there be between leave policy for parents to care for their adult children and leave policy for parents to care for their minor children, my colleagues ignore the strong evidence of the congressional purpose behind the FMLA provision at issue. Consideration of the record in light of the statute and the Senate Report leads, in my mind, to the conclusion that summary judgment was properly granted.

In sum, without much clearer evidence of congressional purpose favoring my colleagues' position, I see no basis for rejecting a textual reading of the Secretary's directive -- adopting the ADA and the EEOC's regulations promulgated thereunder -- as the basis for determining whether or not a parent-employee may take leave under the FMLA to care for an adult child. In so saying, I do not wish to imply that, had I been in Congress when the FMLA was enacted, I would necessarily have favored the disability distinction that Congress inserted. As a matter of policy, I might well agree with my colleagues that the current disability yardstick is a rather arbitrary and clumsy way to separate out those adult children entitled to be cared for by their parents under the FMLA from those who are not. But, as judges, our own philosophies and policy-choices are not the issue. The questions here are what Congress wrote, how the Secretary of Labor has exercised her power under the statute, whether what she did was within her authority, and, finally, whether putting that all together, the district court construed the law properly. I am constrained to believe that the district court did construe the law with total propriety. Indeed, I find it hard to see how the district court could have read the statute and regulations differently. This is not an obscure or ambiguous statute as regards the provision in issue. That seems to me to end the matter, however any one of us might have acted as a member of Congress or the Secretary of Labor.

Unwonted activism in the present appeal not only upsets a district court judgment reached by application of the appropriate and conventional legal rules, it creates a precedent with the potential for serious mischief, since our decision will create confusion as to the relevant standard, while adherence to the Secretary's directive would not.5 If Congress were to be persuaded in the future that the Secretary's interpretation of the Act is too narrow, or that its own language needs enlargement, it can always amend the FMLA; and, of course, the Secretary, too, can rewrite her regulations. These well-established remedies would come too late, to be sure, to assist the present appellant, but I think they better serve the public than a judicial opinion that is sure to create more uncertainty than answers.

I would affirm the district court.