The Family and Medical Leave Act ("FMLA") entitles eligible employees to up to twelve weeks of medical leave each year "[b]ecause of a serious health condition that makes the employee unable to perform the functions of the position of such employee." 29 U.S.C. § 2612(a)(1)(D). The principal issue in this case is the meaning of that phrase, in light of regulations issued by the Department of Labor. The case also involves supplemental claims under state law for breach of implied promise and promissory estoppel.

Plaintiff-Appellant Mark Jones accepted a job as a telecommunications technician with the Denver Public Schools ("the District") on May 3, 1999. Although Mr. Jones was unaware of it, the written job description required that all technicians hold a valid driver's license. The month before, Mr. Jones's driver's license had been suspended for five years on account of multiple alcohol-related offenses. On his first day of work, the manager of the telecommunications department told Mr. Jones, "I think we can work around that." R. Vol. I, p. 68.

On May 26, 1999, Mr. Jones signed an employment agreement that included express language making him an at-will employee:

The School District and the Employee recognize that the Employee is an "at will" Employee, and that the Employee can terminate the employment relationship for any reason at any time, and the School District can also terminate the employment relationship for any reason at any time.

Id. at 153, ¶ 5. The District changed Mr. Jones's employment status in August 1999, making him a full-time classified employee rather than an hourly employee. Policy GDQD-R, which concerns the dismissal of classified District employees, entitles full-time employees facing discharge to two pre-termination hearings upon request, one before the Human Resources administrator and another before an impartial hearing officer, to determine "if there is any competent evidence to support the recommended dismissal." Id. at 156. Policy GDQD-R also states that those procedures "do not change the at-will status of classified employees." Id.

Until June 14, 2001, Mr. Jones worked under the supervision of his father. Co-workers complained that during this period, Mr. Jones abused the sick leave policy by taking more sick days than he was allotted. Some believed he was a hypochondriac or that he called in sick after nights of drinking. Also, his lack of a driver's license proved inconvenient to co-workers, who were required to drive Mr. Jones to and from work, as well as to and from job sites. Mr. Jones also reportedly refused to comply with the department's break policy, even after receiving a written reprimand. After his father's retirement, Mr. Jones came under the supervision of Defendant Jerry Brinkley. Shortly thereafter, Mr. Brinkley gave Mr. Jones a formal job evaluation rating his performance "fair" in fourteen categories and "poor" in six categories. The evaluation summarized Mr. Jones's deficiencies as follows:

Mark needs to work on absenteeism, negativism, anger management, listening skills and guidance acceptance.... Marks [sic] main strengths are quality and technical knowledge[.] His weakness lies in his personal judgments, prioritization and absenteeism.

Id. at 162-64.

Problems came to a head in early October 2001. On Monday, October 1, Mr. Jones claims that he fell at his home and aggravated a pre-existing back injury. On Monday and Tuesday, he called in sick, stayed at home, and took ibuprofin, but did not call a doctor. When he called in sick again on Wednesday, his supervisor told him he would need to bring a note from his doctor upon his return to work. On Thursday, October 4, Mr. Jones visited Dr. Ryan Kramer, who gave him a cortisone shot, indicated in his notes that Mr. Jones had "pain mid butt [to] mid calf," and wrote him a note that read "off work from 10-1-01 thru 10-5-01 due to wrenched back." Id. at 192. Dr. Kramer asked Mr. Jones to schedule a follow-up visit for roughly three weeks later.

By Sunday morning, October 7, Mr. Jones's back pain had subsided and he felt well enough to work. That afternoon, however, he became sick with the flu, and he therefore called in sick again on Monday and Tuesday.

Mr. Jones was terminated when he returned to work on Wednesday. A letter of dismissal, written by Mr. Brinkley and dated October 10, 2001, stated:

You are being dismissed due to your inability to conform with the job requirements of a Telecommunications Technician. Your unreliable attendance record, as evidenced from the last 7 days [sic] absence, is counterproductive to the efficient operation of the Department of Technology Services.

Id. at 165. Upon learning that Mr. Jones was entitled to pre-termination hearings under Policy GDQD-R, Mr. Brinkley sent Mr. Jones a second letter on October 18, offering additional reasons for the dismissal. The second letter cited the department's shrinking workload, as well as Mr. Jones's unreliable attendance record, violations of the break policy, and lack of a driver's license, in support of the decision.

Mr. Jones visited Dr. Kramer again on October 24. The doctor's notes from the second visit include a passing reference to Mr. Jones's back pain: "Reviewed back — improving." Id. at 191. Mr. Jones apparently has not sought medical treatment for his back condition since October 2001.

Pursuant to Policy GDQD-R, Mr. Jones requested and received a pre-termination hearing before the Human Resources director, as well as an appeal before an impartial hearing officer. Both proceedings upheld the dismissal, which became effective November 6, 2001.

Mr. Jones brought this action in September 2002 against the District, Mr. Brinkley, and two other school officials in their individual capacities, raising three claims: (1) interference under the FMLA; (2) breach of implied contract under Colorado law; and (3) promissory estoppel under Colorado law. In an oral ruling, the district court granted the defendants' motion for summary judgment as to all claims. The district court granted summary judgment on the FMLA claim on the ground that a jury could not reasonably infer that the reason for Mr. Jones's termination was his five-day absence from work on account of his back pain. Tr. 331. The court noted that "he was also fired because he doesn't have a driver's license," which was a job requirement. Id. at 332. The court rejected his common law claims summarily:

I'm granting the summary judgment. The promissory estoppel doesn't make it. There's an express contract, and that — you can't rely on somebody telling you we can work around that to make that an enforceable promise. And, there's no implied contract. Judgment will enter for the defendants.

Id. at 333.

On appeal, Mr. Jones presses all three claims. We review the grant of summary judgment de novo, Wells v. Colo. Dep't of Transp., 325 F.3d 1205, 1209 (10th Cir.2003), and affirm only if the record, considered in the light most favorable to the plaintiff, establishes no genuine issue of material fact, Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248-50, 106 S. Ct. 2505, 91 L. Ed. 2d 202 (1986).

The FMLA makes it unlawful for any covered employer "to interfere with, restrain, or deny the exercise of or the attempt to exercise, any right provided in this subchapter." 29 U.S.C. § 2615(a)(1). To make out a prima facie claim for FMLA interference, a plaintiff must establish (1) that he was entitled to FMLA leave, (2) that some adverse action by the employer interfered with his right to take FMLA leave, and (3) that the employer's action was related to the exercise or attempted exercise of his FMLA rights. See Bones v. Honeywell Int'l, Inc., 366 F.3d 869, 877 (10th Cir.2004).

The district court granted summary judgment on the basis of the third element: whether the District's decision to terminate Mr. Jones was related to his absence from work during that week. The court held that because of the multiple reasons listed by Mr. Brinkley in his second letter of termination-particularly, his lack of a valid driver's licence, which was a requirement of the job-no reasonable jury could infer that he was terminated on account of his absence. Plaintiff argues that, viewing the evidence in the light most favorable to himself, as the non-moving party, there are disputed issues of material fact precluding that holding. Specifically, he argues that only the first letter of termination, which focused on his absence, is worthy of credibility, and that a jury could infer that the second letter was pretextual. We need not reach this issue, however, because we conclude that Mr. Jones's absence from work during the week of October 1 was not caused by a "serious health condition" as defined in the statute and regulations, and thus was not protected by the FMLA.

The FMLA entitles a qualified employee to up to twelve weeks of leave during any twelve month period "[b]ecause of a serious health condition that makes the employee unable to perform the functions of the position of such employee." 29 U.S.C. § 2612(a)(1)(D). The statute defines "serious health condition" to mean "an illness, injury, impairment, or physical or mental condition that involves — (A) inpatient care in a hospital, hospice, or residential medical care facility; or (B) continuing treatment by a health care provider." Id. § 2611(11). Under regulations promulgated by the Department of Labor, "continuing treatment by a health care provider" includes:

A period of incapacity (i.e., inability to work, attend school or perform other regular daily activities due to the serious health condition, treatment therefor, or recovery therefrom) of more than three consecutive calendar days, and any subsequent treatment or period of incapacity relating to the same condition, that also involves:

(A) Treatment two or more times by a health care provider...; or

(B) Treatment by a health care provider on at least one occasion which results in a regimen of continuing treatment under the supervision of the health care provider.

29 C.F.R. § 825.114(a)(2)(i). For purposes of the regulations, "treatment" includes both "examinations to determine if a serious health condition exists and evaluations of the condition," while a "regimen of continuing treatment" includes "a course of prescription medication (e.g., an antibiotic) or therapy requiring special equipment ... (e.g., oxygen)" but excludes regimens that consist solely of over-the-counter medications, bed rest, fluids, or exercise. Id. § 825.114(b). Neither Mr. Jones nor the defendants challenge the agency's interpretation of the underlying statute.

We assume, for the sake of argument, that Mr. Jones's testimony and Dr. Kramer's notes create a genuine issue of material fact as to whether Mr. Jones suffered a "period of incapacity" from October 1 through the morning of October 7, 2001, a period in excess of three consecutive calendar days. The more difficult question is whether Mr. Jones's two visits to Dr. Kramer, first on October 4 and again on October 24, satisfy the requirement of "[t]reatment two or more times by a health care provider" under § 825.114(a)(2)(i)(A).

Both visits amount to "treatment" under the agency's definition. However cursory Dr. Kramer's examinations, his actions fit the two descriptions of treatment in § 825.114(b). At the first visit, he examined Mr. Jones to determine "if a serious health condition exist[ed]," and his note excusing Mr. Jones from work documents his findings. At the second, he performed an "evaluation[ ] of the condition," as evidenced by his handwritten note that Mr. Jones's back was "improving."

The timing of the two visits, however, presents a separate question under the regulations. "Continuing treatment" requires that a plaintiff experience "[a] period of incapacity ... of more than three consecutive calendar days, and any subsequent treatment or period of incapacity relating to the same condition, that also involves" either two or more treatments, or a single treatment followed by a regimen of continuing treatment. 29 C.F.R. § 825.114(a)(2)(i). Several district courts have held that "the regulation by its plain language merely requires two or more treatments, without distinguishing between treatments occurring during or after the initial period of incapacity." Summerville v. Esco Co. Ltd. P'ship, 52 F. Supp. 2d 804, 810 (W.D.Mich.1999); see also Jones v. Willow Gardens Ctr., No. C98-0007, 2000 WL 201170 (N.D.Iowa Jan.28, 2000); George v. Associated Stationers, 932 F. Supp. 1012, 1015 (N.D.Ohio 1996) ("The Regulations do not specify a time period during which the minimum two examinations must take place."). Because Mr. Jones's second treatment took place more than two weeks after his back pain and "period of incapacity" subsided, he urges us to adopt an interpretation that indefinitely extends the timeframe for "treatment two or more times."

Yet this "indefinite timeframe" reading defies the text, runs contrary to Congress's intent, and places employers in a position of grave uncertainty in complying with their obligations under the FMLA. We disagree that the plain language of the regulation imposes no time limit on the requisite "two or more" treatments. According to the regulation, a serious health condition that involves continuing treatment includes "[a] period of incapacity..., and any subsequent treatment or period of incapacity relating to the same condition, that also involves... [t]reatment two or more times by a health care provider." 29 C.F.R. § 825.114(a)(2)(i) (emphasis added). The question is whether the italicized phrase "that also involves" modifies "period of incapacity" or whether it modifies the entire phrase "period of incapacity ..., and any subsequent treatment or period of incapacity relating to the same condition." Id. If the two or more treatments must take place during the period of incapacity, Mr. Jones cannot prevail under this definition. If the two or more treatments may take place either during the period of incapacity or during any "subsequent treatment," however, his second session with Dr. Kramer would make him qualify.

The grammatical and logical structure of the regulation suggests that the former interpretation is more likely. First, to read the regulation as allowing one or both of the required treatments to take place during any "subsequent treatment" would eliminate any necessary temporal connection between the "period of incapacity" and the treatments. If that was Congress's intention, however, the regulation could have simply required that the underlying medical condition "involve" both a more-than-three-day period of incapacity and two or more treatments, with no connection between them. That the regulation instead frames the definition in terms of a "period of incapacity" that "involves" at least two treatments indicates that the timing of the treatments, and not just the need for treatments, is important. Indeed, structurally, the regulation consistently frames its alternative definitions as a set of conditions on a "period of incapacity." See 29 C.F.R. § 825.114(a)(2)(ii) ("[a]ny period of incapacity due to pregnancy, or for prenatal care"); id. § 825.114(a)(2)(iii) ("[a]ny period of incapacity or treatment for such incapacity due to a chronic serious health condition"); id. § 825.114(a)(2)(iv) ("[a] period of incapacity which is permanent or long-term due to a condition for which treatment may not be effective"). This emphasis on the "period of incapacity" reinforces the view that the necessary treatments must be temporally linked to that period.

Second, to read the phrase "that also involves" as modifying the combination of the period of incapacity and any subsequent treatment would collapse the distinction between 29 C.F.R. § 825.114(i)(A), which provides for "[t]reatment two or more times," and § 825.114(i)(B), which provides for "[t]reatment by a health care provider on at least one occasion which results in a regimen of continuing treatment under the supervision of the health care provider." The latter provision contemplates the situation where the worker obtains only one treatment during the period of incapacity, but where that treatment results in a continuing regimen of treatment which extends beyond the period of incapacity. Indefinitely extending the timeframe for the two treatments renders subpart (B) almost entirely redundant: the provisions would retain independent meaning only to the extent a "regimen of continuing treatment" never involves "treatment." It seems odd for the regulation to specify the need for a full "regimen of continuing treatment" after the period of incapacity if a single subsequent treatment suffices.

Third, the verb "involves" (rather than "involve") indicates that the subject is singular ("period of incapacity"), not plural ("period of incapacity ... and any subsequent treatment"). To treat the entire phrase "[a] period of incapacity ..., and any subsequent treatment or period of incapacity relating to the same condition" as a conjunctive subject would render it plural, and would necessitate use of the plural verb "involve." See id. § 825.114(a)(2)(i).

We therefore read the statutory phrase as equivalent to the following:

A serious health condition involving continuing treatment by a health care provider includes: [a] period of incapacity... that also involves [t]reatment two or more times by a health care provider, and any subsequent treatment or period of incapacity relating to the same condition.

Under this definition, to qualify for FMLA protection, the health condition must be sufficiently serious that it entails an absence of more than three consecutive calendar days during which the employee obtained treatment by a health care provider at least two times (or one time followed by a regimen of continuing treatment).

The function of the phrase, "and any subsequent treatment or period of incapacity relating to the same condition," under this reading, is not to extend the period during which the employee may obtain the two treatments, but to extend the temporal dimension of the serious medical condition so as to require the employer to grant leave for subsequent treatments or periods of impairment even if they would not independently qualify. For example, imagine that an employee misses a week of work and sees a doctor two times during his absence. It is clear that he is entitled to FMLA leave for that period. But imagine that a week later he has a flare-up that requires a day's absence. Even though a single day's absence would not, on its own, qualify as a "serious medical condition," the regulation under our reading would require leave because the "subsequent period of impairment" is included as part of the original "serious medical condition" when it "relat[es] to the same condition." The same would be true if the employee had to take a few hours off each week for follow-up medical treatments: those, too, would qualify for FMLA leave as part of the original "serious medical condition."

We believe that this interpretation of the regulation best accords with congressional intent. FMLA protection is not intended for minor ailments and injuries. It is reserved for workers who have a "serious health condition." The Senate Report on the FMLA comments:

The term "serious health condition" is not intended to cover short-term conditions for which treatment and recovery are very brief. It is expected that such conditions will fall within even the most modest sick leave policies. Conditions or medical procedures that would not normally be covered by the legislation include minor illnesses which last only a few days....

S.Rep. No. 103-3, at 28, reprinted in 1993 U.S.C.C.A.N. 3, 30 (1993). See also 29 C.F.R. § 825.114(c) ("Ordinarily, unless complications arise, the common cold, the flu, ear aches, upset stomach, minor ulcers, headaches other than migraine, ... etc., are examples of conditions that do not meet the definition of a serious health condition and do not qualify for FMLA leave.").

The term "serious health condition" is statutorily defined as one that requires either inpatient care or "continuing treatment by a health care provider." 29 U.S.C. § 2611(11). The Senate Report comments that "back conditions" may satisfy that test if they "requir[e] extensive therapy or surgical procedures." S.Rep. No. 103-3, at 29. If the employee is able to recover and return to work after only one treatment (a cortisone shot, at that), it is doubtful that he satisfies the statutory requirement of "continuing treatment." The regulations generously construe "continuing treatment" as either two or more treatments or one treatment followed by a regimen of continuing treatment. We are loath to reduce this to a requirement of one treatment, with a follow-up check-up weeks after the employee has returned to work.

Moreover, as Mr. Jones's case aptly demonstrates, the indefinite timeframe reading would place employers in a position of uncertainty regarding their FMLA obligations. At the time he returned to work, Mr. Jones had undergone only one treatment in connection with his back injury, and on any reading had not yet qualified for FMLA leave. When the District announced his termination, therefore, it acted legally. According to Mr. Jones, however, his follow-up visit to Dr. Kramer two weeks later retroactively made the District's conduct unlawful. This makes it exceedingly difficult for the employer to know what it has to do to comply with the law. It also risks frustrating the exercise of certification rights granted to employers by the FMLA. See 29 U.S.C. § 2613 (entitling an employer to require certification by a health care provider, and to require that employees obtain a second opinion, in connection with requests for leave under § 2612(a)(1)(D)). At the same time, reading the regulation to allow an indefinite timeframe for the second doctor's visit would invite strategic behavior by plaintiffs, who could schedule a second visit to "determine if a serious health condition exists" long after all symptoms have subsided, solely to bolster their claim of entitlement to FMLA leave in anticipation of litigation. We find it difficult to believe that the Congress intended such a result.

Although no federal appellate court has considered the issue at length, our reading finds some support in dicta from the Eighth Circuit, which paraphrased the regulations as requiring that the plaintiff "must have had a `period of incapacity requiring absence from work,' ... and she must have had `continuing treatment by ... a health care provider' within that period." Thorson v. Gemini, Inc., 205 F.3d 370, 377 (8th Cir.2000) (emphasis added). That court has since reiterated that "the fact that an employee is `sufficiently ill to see a physician two times in a period of just a few days' is all that FMLA requires for `continuing treatment.'" Rankin v. Seagate Techs., Inc., 246 F.3d 1145, 1149 (8th Cir.2001) (quoting Thorson, 205 F.3d at 379).

We agree, and hold that under the regulations defining "continuing treatment by a health care provider," the "[t]reatment two or more times" described in § 825.114(a)(2)(i)(A) must take place during the "period of incapacity" required by § 825.114(a)(2)(i).1 Mr. Jones's period of incapacity involved only a single treatment on October 4, 2001. Accordingly, he was not entitled to FMLA leave and cannot make out a claim for FMLA interference.2

Mr. Jones also appeals the district court's decision granting summary judgment for the defendants on his claim for breach of implied contract. Colorado law presumes that an employee hired for an indefinite period "is an `at will employee,' whose employment may be terminated by either party without cause and without notice." Continental Air Lines, Inc. v. Keenan, 731 P.2d 708, 711 (Colo.1987). Nonetheless, "[w]hen a school board limits its power to discharge personnel by promulgating and issuing a handbook, an employee may rely on the provisions in the handbook to state a claim for relief for breach of implied contract." Smith v. Bd. of Educ. of Sch. Dist. Fremont RE-1, 83 P.3d 1157, 1163 (Colo.Ct.App.2003). In Smith, the Colorado Court of Appeals held that an employee handbook that provided detailed examples of "Just and Proper Cause" for disciplining an employee "could reasonably be read to require that termination be only for just cause." Id. at 1164. On the other hand, in George v. Ute Water Conservancy Dist., 950 P.2d 1195, 1198-1200 (Colo.Ct.App.1997), the same court emphasized that "summary judgment is appropriate if the employer has clearly and conspicuously disclaimed intent to enter into a contract limiting the right to discharge employees," and found handbook language promoting "fair and equitable standards for all employees" insufficient to alter employees' at-will status. Although Colorado courts do not consider a disclaimer "clear and conspicuous" if it is "inconspicuously placed in an appendix to the handbook," see Ferrera v. Nielsen, 799 P.2d 458, 461 (Colo.Ct.App.1990) (construing Cronk v. Intermountain Rural Elec. Ass'n, 765 P.2d 619 (Colo.Ct.App.1988)), they have found disclaimers clear and conspicuous when they appear on the first page of the handbook, see id., or "in close proximity to the provision setting forth ... required procedure on termination," George, 950 P.2d at 1199.

The gist of Mr. Jones's claim is that Policy GDQD-R, which entitles employees to two pre-termination hearings "strictly confined to whether there is any competent evidence to support the dismissal recommendation," R. Vol. I, p. 157, operated as an implied contract that prohibited the District from dismissing him without just cause. He contends that by reneging on its promise to "work around" his lack of a driver's license, faulting him for absences despite the fact that he still had sick days available, and enforcing the break policy against him but not against other employees, the District acted without just cause and thus breached the contract.

We conclude that under Colorado law, Policy GDQD-R did not create an implied contract whereby the District would not terminate Mr. Jones except for just cause. Mr. Jones's original employment agreement contains a clear and conspicuous disclaimer making him an at-will employee, and Policy GDQD-R itself states, in its opening paragraph and on the same page as the language describing the pre-termination hearings, that "[t]he procedures do not change the at-will status of classified employees." R. Vol. I, p. 156. Unlike the language in Smith, Policy GDQD-R does not require "just cause" for termination, but instead creates an essentially procedural guarantee of pre-termination hearings. Although Mr. Jones boldly states that requiring "competent evidence" in support of dismissal is the same as requiring "just cause," the policy in fact imposes no limits on the reasons, just or unjust, for which the District may dismiss an employee.

Further, even if the policy created an implied contract, Mr. Jones cannot survive summary judgment because no reasonable jury could find that the District breached the terms of that contract. The District fully complied with the procedures laid out in Policy GDQD-R, and both the director of Human Resources and an impartial hearing officer affirmed the District's decision. The record unquestionably contains at least some "competent evidence" to support each of the reasons cited by the District: the depositions of Mr. Jones's supervisors and co-workers confirm his excessive absences, the written reprimand documents his violations of the break policy, his own testimony concedes his lack of a driver's license, and documents from the fall of 2001 support a decline in the department's workload. Any of those reasons appears sufficient, under the terms of Policy GDQD-R, to "support" the dismissal.

Mr. Jones vigorously contests each of the District's reasons, but it is not the role of this Court to decide whether the District acted for sound reasons, or even whether it acted for its stated reasons. We limit our review to whether the District breached an implied contract that, at most, obligated the District to dismiss him only when "any competent evidence" supported the decision. No reasonable finder of fact could conclude that there is no competent evidence, on any ground, to support Mr. Jones's dismissal. We therefore affirm the district court's grant of summary judgment for the defendants on the breach of implied contract claim.

Finally, Mr. Jones appeals from the district court's decision granting summary judgment on his claim for promissory estoppel, which under Colorado law consists of four elements: "(1) the promisor made a promise to the promisee; (2) the promisor should reasonably have expected that the promise would induce action or forbearance by the promisee; (3) the promisee in fact reasonably relied on the promise to the promisee's detriment; and (4) the promise must be enforced to prevent injustice." Patzer v. City of Loveland, 80 P.3d 908, 912 (Colo.Ct.App.2003). The promise, according to Mr. Jones, was the telecommunications department manager's assurance that the District could "work around" the suspension of Mr. Jones's driver's license. The District allegedly broke that promise when it cited the lack of a license as one of the reasons for his dismissal. See Bd. of County Comm'rs of Summit County v. DeLozier, 917 P.2d 714, 716 (Colo.1996) (describing promissory estoppel as "a modest extension of the basic contract principle that one who makes promises must be required to keep them").

The claim fails on several grounds. First, promissory estoppel requires a promise that (1) either discloses promissory intent or constitutes "a commitment by the employer," and (2) is "sufficiently specific so that the judiciary can understand the obligation assumed and enforce the promise according to its terms." Soderlun v. Pub. Serv. Co. of Colo., 944 P.2d 616, 620 (Colo.Ct.App.1997) (emphasis in original). As a result, "vague assurances" of continued employment cannot serve as the basis for a promissory estoppel claim. Id. (citing Hayes v. Eateries, Inc., 905 P.2d 778, 780-81 (Okla.1995), which found that statements assuring an employee he would have a job so long as his performance was adequate were too vague to function as a promise). The statement in this case meets neither requirement. The language itself is qualified — "I think we can work around that" — and Mr. Jones had reason to doubt that it represented a commitment by the District when he subsequently signed an express agreement and read written job requirements to the contrary. Even an unequivocal pledge to "work around" a job requirement would provide the judiciary with too little information about the nature and duration of the District's obligations.

Second, to the extent the statement is promissory, no reasonable jury could conclude that the District failed to do what it promised. Colorado courts do not apply promissory estoppel in cases where the employer has effectively kept its promises to employees. See Seeley v. Bd. of County Comm'rs for La Plata County, 771 P.2d 21, 22-23 (Colo.Ct.App.1989) (holding promissory estoppel inapplicable where the employer had effectively discharged its duty, under the employee handbook, to carry out a unilateral investigation into events precipitating a dismissal). Here, the District "work[ed] around" Mr. Jones's lack of a driver's license for more than two years, ordering other employees to drive Mr. Jones to and from work, and to and from job sites. The District's accommodation of Mr. Jones became so extensive that one employee complained to his supervisor that he felt like Mr. Jones's "personal chauffeur." R. Vol. I, p. 176. To conclude that the District broke its promise would require an unreasonable interpretation of the District's obligations to Mr. Jones.

Third, and most basically, the circumstances of Mr. Jones's case are not "such that injustice can be avoided only by enforcement of the promise." Nelson v. Elway, 908 P.2d 102, 110 (Colo.1995). This final element of promissory estoppel involves a discretionary decision for the court, and is not a question of fact for the jury. Hoffman v. Red Owl Stores, Inc., 26 Wis. 2d 683, 133 N.W.2d 267, 275 (1965), cited by Kiely v. St. Germain, 670 P.2d 764, 767 (Colo.1983) (describing promissory estoppel as grounded in principles "familiar to equity jurisprudence"). In this case, the District had well-documented independent reasons for terminating Mr. Jones: excessive absences, violations of the break policy, and a decline in the workload of the department. Judgment for Mr. Jones would not prevent injustice on these facts, and accordingly his action for promissory estoppel must fail.

We AFFIRM the decision of the district court granting summary judgment to the defendants on all claims.