In the

    United States Court of Appeals
                 For the Seventh Circuit
                     ____________________
No. 18-3060
KEVIN CLANTON,
                                                  Plaintiff-Appellee,
                                 v.

UNITED STATES OF AMERICA,
                                              Defendant-Appellant.
                     ____________________

         Appeal from the United States District Court for the
                     Southern District of Illinois.
        No. 3:15-cv-124 — Nancy J. Rosenstengel, Chief Judge.
                     ____________________

 ARGUED SEPTEMBER 11, 2019 — DECIDED NOVEMBER 7, 2019
               ____________________

   Before RIPPLE, ROVNER, and BARRETT, Circuit Judges.
    BARRETT, Circuit Judge. For four years, nurse practitioner
Denise Jordan treated Kevin Clanton’s severe hypertension.
Jordan, an employee of the U.S. Public Health Service, failed
to properly educate Clanton about his disease or to monitor
its advancement. Clanton’s hypertension eventually devel-
oped into Stage V kidney disease requiring dialysis and a
transplant, and he sued the United States under the Federal
Tort Claims Act for Jordan’s negligent care. After a five-day
2                                                     No. 18-3060

bench trial, the district court found the United States liable.
The court determined that Clanton had not contributed at all
to his own injuries, noting that Clanton did not understand
why it was so important to take his medication and to attend
all appointments. The court awarded Clanton nearly $30 mil-
lion in damages. The government appeals the court’s compar-
ative-negligence determination and three aspects of the
court’s rulings on damages. We agree with the government
that the court erred in its analysis of comparative negligence,
so we vacate the judgment and remand for the court to apply
the proper legal standard. As for damages, however, we find
no reversible error.
                                I.
    Kevin Clanton first visited nurse practitioner Denise Jor-
dan for medical care after high blood pressure caused him to
fail a pre-employment physical exam. Jordan, an employee of
the Southern Illinois Healthcare Foundation, treated Clanton
at two health clinics in East St. Louis, Illinois. At his first ap-
pointment, Jordan ordered routine lab work and diagnosed
Clanton with obesity and hypertension. On Clanton’s second
visit, she gave him medication and asked him to come back in
one week. Clanton did not return in a week; instead, he failed
to return to see Jordan until he showed high blood pressure
at another employment-related physical exam two years later.
    During the two years that followed, Clanton returned to
Jordan for care ten times. Clanton often went long stretches
without returning to see Jordan, and he failed to take his pre-
scribed medicine if he was not feeling sick. For her part, Jor-
dan never explained why it was important for Clanton to take
his medications and to attend all appointments even if he was
feeling fine. In fact, Jordan never educated Clanton about
No. 18-3060                                                     3

hypertension, its associated risks, or the factors that increased
the risks for Clanton in particular.
    Three years after Clanton’s first visit, Jordan ordered new
lab tests, but she never reviewed the results. If Jordan had
seen the lab work, she would have noticed signs of kidney
disease and referred Clanton to a nephrologist. Instead, Clan-
ton did not learn for another year and a half that his hyper-
tension had caused serious damage to his kidneys. A doctor
finally diagnosed Clanton with end-stage renal disease, and
Clanton began hemodialysis treatment and joined the kidney-
transplant waiting list. Clanton eventually received a success-
ful kidney transplant, but he will probably need further
rounds of hemodialysis and one or more transplants in the
future. Medicare Part B paid for Clanton’s hemodialysis and
may cover future rounds of hemodialysis as well.
    Clanton sued the United States under the Federal Tort
Claims Act for Jordan’s negligence because Jordan and her
employer are employees of the U.S. Public Health Service. Af-
ter a five-day bench trial, the district court determined that
Jordan had deviated from the standard of care by failing to
adequately educate Clanton about the severity of his condi-
tion, failing to refer him to a specialist, and failing to consult
his lab reports. The court found no comparative negligence on
Clanton’s part and awarded him nearly $30 million in dam-
ages.
   In a motion to reconsider, the government challenged the
court’s damages rulings on three grounds. First, it argued that
the court erred in rejecting the government’s election to pay
some future damages in periodic installments. Second, the
government contended that the court should have conducted
a damages comparison when it calculated noneconomic
4                                                   No. 18-3060

damages. And third, the government argued that the court
should have deducted some of its damages obligations be-
cause Medicare had covered Clanton’s hemodialysis. The dis-
trict court agreed to conduct a damages comparison analysis,
but after doing so, it declined to change its damages award. It
refused to reconsider its rulings with respect to periodic pay-
ment and partial offset. The government appeals the district
court’s determination that Clanton was not comparatively
negligent and all three of its rulings on damages.
                               II.
    The government argues that the district court failed to ap-
ply the correct legal standard when it evaluated Clanton’s
comparative negligence. Because Jordan treated Clanton in Il-
linois, Illinois tort law applies. 28 U.S.C. § 1346(b)(1). We re-
view de novo a district court’s determination of state law in
an FTCA case. Del Raso v. United States, 

, 570 (7th
Cir. 2001).
     The parties agree on the proper standard to assess a plain-
tiff’s comparative negligence under Illinois law. Courts must
apply the familiar reasonable-person standard, an objective
test that asks “whether plaintiff … used that degree of care
which an ordinarily careful person would have used … under
like circumstances.” McCarthy v. Kunicki, 

,
1101 (Ill. App. Ct. 2005). But the district court here failed to
articulate that standard—or to cite to any legal authority at
all—in its discussion of Clanton’s comparative negligence.
   Clanton insists that the district court applied the correct
standard in practice, even if not in name. But we are not con-
vinced. The court focused its assessment of Clanton’s negli-
gence on his own limited understanding of his condition.
No. 18-3060                                                   5

Because Clanton didn’t understand the seriousness of his se-
vere hypertension, the court reasoned, he did not act negli-
gently when he missed medical appointments or when he
failed to take his medicine. But Clanton’s subjective under-
standing does not properly end the inquiry. Illinois law re-
quires the court to take the additional step of comparing Clan-
ton’s understanding of his condition to that of a reasonable
person in his situation. 

in the position of a
person whose caregiver had failed to provide information
about the severity of his condition. Yet he also had a few ex-
ternal clues that he was seriously unwell—for example, two
employment-related physicals showed that he had danger-
ously high blood pressure. The district court must determine
how a reasonable person in the same position would have
acted and compare Clanton’s behavior to that objective stand-
ard of care.
     We vacate the judgment and remand to the district court
so that it can assess Clanton’s comparative negligence under
Illinois’s reasonable-person standard.
                              III.
    The government argues that on remand the district court
should also revisit some of its rulings on damages. The gov-
ernment first focuses on the court’s rejection of the govern-
ment’s election to pay Clanton some of the damages award in
periodic installments. When the court calculated Clanton’s
damages, a since-repealed Illinois statute offered a medical-
malpractice defendant that option. 735 ILCS 5/2-1705 to -1709
(repealed Aug. 16, 2019). But over the government’s objection,
the court held that the statute permitted it to reject the elec-
tion—which the court did.
6                                                      No. 18-3060

    The government contends that the court misread the stat-
ute. According to the government, the statute entitled it to
elect a periodic-payment schedule, and in rejecting that elec-
tion, the district court asserted discretion that the statute did
not give it. Maybe so. But Illinois has now repealed the peri-
odic-payment statute, and if periodic payment is no longer an
option, it doesn’t matter whether the district court made a
mistake. Before we go any further, therefore, we must deter-
mine whether Illinois law makes the repeal of this statute ret-
roactive.
     This question is controlled by Illinois’s long-standing rule
that the repeal of special remedial statutes applies retroac-
tively to pending cases. A special remedial statute is one that
creates a new remedy for one particular subject. See People ex
rel. Eitel v. Lindheimer, 

, 320–21 (Ill. 1939). In Shel-
ton v. City of Chicago, for example, the Supreme Court of Illi-
nois held that a law providing a damages remedy for victims
of mob violence qualified as a special remedial statute. 

, 123–24 (Ill. 1969). Shelton points the way here. Like
the statute in Shelton, the periodic-payment law created a new
remedy for one particular subject—it provided a method for
defendants to pay damages awards in medical-malpractice
cases. Under Shelton’s reasoning, then, the periodic-payment
statute qualifies as a special remedial statute.
    When a special remedial statute is repealed, the effect un-
der Illinois law is clear: “The unconditional repeal of a special
remedial statute without a saving clause stops all pending ac-
tions where the repeal finds them. If final relief has not been
granted before the repeal goes into effect it can’t be granted
afterwards.” 

. Since Illinois’s peri-
odic-payment statute was a special remedial statute with no
No. 18-3060                                                     7

saving clause, its repeal “stops all pending actions where [it]
finds them.” 

is no longer available to the
government, even if the district court erred in interpreting the
periodic-payment statute.
    Rather than addressing the rule about special remedial
statutes, the government invokes Illinois’s general rule about
repeals: those “that are procedural in nature may be applied
retroactively, while those that are substantive may not.”
Caveney v. Bower, 

, 602 (Ill. 2003). Because Illi-
nois’s periodic-payment statute had substantive compo-
nents—most notably, it entitled defendants to apply a 6% dis-
count rate to certain damages—the government insists that
the repeal does not retroactively eliminate the availability of
periodic payment.
    The Caveney rule, however, is inapposite. The substan-
tive/procedural distinction is the “clear legislative directive”
of the general saving clause of section 4 of Illinois’s Statute on
Statutes, 5 ILCS 70/4. 

. But the Su-
preme Court of Illinois has made clear that the general saving
clause “does not apply to express repeals of special stat[u]tory
remedies.” 

; see also People v. Glisson,

, 257 (Ill. 2002) (distinguishing repeals of spe-
cial statutory remedies). It follows that even those special re-
medial statutes that seem more substantive than procedural
are not applicable after repeal. Put differently, the rule regard-
ing special remedial statutes is a carve-out from the general
rule that the government invokes.
    The Supreme Court of Illinois reaffirmed the validity of
this carve-out as recently as 2018. In Perry v. Department of Fi-
nancial & Professional Regulation, the court held that the repeal
of a particular provision of Illinois law was substantive and
8                                                  No. 18-3060

therefore applied only prospectively. 

, 1034
(Ill. 2018). In explaining that conclusion, it emphasized that
the provisions were not procedural in nature; “[n]or can
[they] be said to be changes to special remedial statutes.” 

thus confirms that remedial statutes are a cat-
egory distinct from both substantive and procedural statutes.
Perry went on to explain that “courts can apply retroactively
statutory changes to procedural or remedial provisions.” 

782 N.E.2d at 257). The court thus reaffirmed
that repeals of remedial statutes, like procedural repeals, ap-
ply retroactively to make the repealed statute unavailable in
pending cases.
    Since periodic payment is no longer available to the gov-
ernment, the district court need not revisit its interpretation
of the periodic-payment statute on remand.
                              IV.
    The government’s second challenge to the damages award
addresses the district court’s calculation of noneconomic
damages. The district court initially awarded Clanton
$13,750,000 in noneconomic damages. In response to the gov-
ernment’s motion to reconsider, the court compared the dam-
ages in cases similar to Clanton’s. The court employed two
methods of comparison. The court first conducted a gross
award comparison to determine that the initial award was ap-
propriate. The court then confirmed that award using the ra-
tio method. The government takes issue with both methods.
We review the court’s approach to determining damages with
the recognition that the court’s new comparative-negligence
analysis may affect the actual damages values.
No. 18-3060                                                              9

                                    A.
    In its challenge to the gross award comparison, the gov-
ernment argues that the court improperly excluded two
cases. 1 To conduct a gross award comparison, a court com-
piles a list of cases from across districts that are factually sim-
ilar to the one at hand. The court then culls from the list those
cases that are sufficiently distinguishable to make them un-
helpful comparators. Finally, it compares the total noneco-
nomic damages awarded in those cases.
     Here, the district court began by creating a list of twenty
potential cases for comparison. In its culling process, the court
excluded two potential comparator cases that alleged torts oc-
curring in Hawaii: Campano v. United States, No. 15-439, 

(D. Haw. Mar. 7, 2018), and Mamea v. United
States, 

(D. Haw. 2011). Hawaii caps the
amount of pain-and-suffering damages that a court can
award as part of the total noneconomic damages, but Illinois
does not. The district court thus determined that Campano and
Mamea were “not particularly useful” comparators because it
had no way of determining the total noneconomic damages
those plaintiffs would have won if the torts had occurred in
Illinois.
    The government argues that it was legal error to exclude
Campano and Mamea from the comparison. We review a dis-
trict court’s damages methodology de novo, Kreg

    1  While we have acknowledged that “the Supreme Court of Illinois
does not require or even encourage such comparisons,” we have held that
this rule does not bind a federal court in a case brought under the Federal
Tort Claims Act. Arpin v. United States, 

, 776 (7th Cir. 2008).
10                                                  No. 18-3060

Therapeutics, Inc. v. VitalGo, Inc., 

, 419 (7th Cir.
2019), but we review the application of that methodology for
abuse of discretion. The district court’s methodology was
gross award comparison; the court’s decision to exclude Cam-
pano and Mamea was a choice it made in calculating damages
using that methodology. We therefore review the exclusion of
Campano and Mamea for abuse of discretion.
   Campano and Mamea might have provided the court with
some information for comparison, but it is true that Hawaii’s
damages cap made them imperfect comparators. It was there-
fore within the district court’s authority to determine that dis-
similarities between those cases and Clanton’s outweighed
their usefulness to the gross award comparison. The court did
not abuse its discretion by excluding Campano and Mamea
from its damages comparison.
                               B.
    The government also objects to the court’s second method
of damages comparison: the ratio method. As with the gross
award comparison method, a court applying the ratio method
begins with a list of comparable cases. Yet rather than com-
paring the total noneconomic damages, the court compares
the ratio between economic and noneconomic damages in the
comparator cases. The court then works backward from the
more concrete economic damages in the present case to arrive
at a proportionate number for noneconomic damages.
   The government argues that the ratio method is inappro-
priate for calculating damages in a personal-injury case. We
need not address that argument, though, because the court’s
application of the ratio method in this case did not affect the
damages award. Cf. Cook v. Niedert, 

, 1016 (7th
No. 18-3060                                                    11

Cir. 1998) (declining to remand on a legal error of methodol-
ogy in the calculation of attorneys’ fees because the error
would not have affected the total fees awarded). The court
first calculated noneconomic damages and then relied on the
gross award comparison to determine how its damages deter-
mination compared to other similar cases. Only then did the
court apply the ratio method, characterizing it as an alterna-
tive comparison methodology that merely confirmed the cor-
rectness of its award. In other words, the court gave every in-
dication that it would have awarded the same damages if it
had not applied the ratio method. There is therefore no reason
for us to instruct the district court to do anything differently.
                               V.
     Finally, the government argues that the district court
should have deducted a portion of Clanton’s damages be-
cause Medicare Part B covered his hemodialysis. The collat-
eral-source doctrine allows a tort victim whose out-of-pocket
costs are compensated—for example, by insurance or a bene-
fits program—to still recover his full losses from the tortfeasor
if the compensation comes “from a source wholly independ-
ent of, and collateral to, the tortfeasor.” Wills v. Foster, 

, 1022 (Ill. 2008) (citations omitted). The Supreme
Court of Illinois has explained that “[t]he justification for [the
collateral-source] rule is that the wrongdoer should not bene-
fit from the expenditures made by the injured party or take
advantage of contracts or other relations that may exist be-
tween the injured party and third persons.” Wilson v. Hoffman
Grp., Inc., 

, 530 (Ill. 1989).
    The government contends that Clanton’s Medicare bene-
fits do not come from a “collateral” source because general
taxpayer revenues fully fund Clanton’s damages and
12                                                             No. 18-3060

partially fund Medicare Part B, which fully covered (and may
continue to cover) Clanton’s hemodialysis. As the govern-
ment sees it, Clanton will be double-dipping into the public
purse if he receives the full amount of his damages. To pre-
vent that, the government asked the district court to reduce
Clanton’s damages award by the fraction of his Medicare ben-
efits that is traceable to general taxpayer funding. The district
court refused to give the government this partial offset, char-
acterizing the request as “an idea that was brought up in the-
ory forty years ago, but, as best the Court can tell, has never
been explicitly approved of, much less implemented in prac-
tice.” 2 The government urges us to reverse this ruling.
    Whether a benefit is collateral to a damages payment is a
question of state law. Molzof v. United States, 

, 464
(7th Cir. 1993). To decide whether a payment is collateral, Il-
linois courts consider both the “source” of the benefits pay-
ments and the payments’ “purpose and nature.” Laird v. Ill.
Cent. Gulf R.R. Co., 

, 955 (Ill. App. Ct. 1991). We
therefore look to the source and the purpose and nature of
Clanton’s benefits to determine whether Illinois courts would
grant the government the partial offset that it seeks.



     2The government has not pointed to a single example of an Illinois
court granting a defendant a partial offset. Other circuits applying the sub-
stantive law of other states suggested in some older cases a hypothetical
openness to the possibility of a partial offset—but none actually allowed
the remedy. See, e.g., Overton v. United States, 

, 1309 (8th Cir.
1980) (noting in an FTCA case where Medicare covered the plaintiff’s in-
jury that “[t]here may be cases in which the government is entitled to a
partial set-off ….”); Steckler v. United States, 

, 1379 (10th Cir.
1977) (expressing openness to a partial set-off in an FTCA case involving
Social Security and veterans’ benefits).
No. 18-3060                                                               13

    Laird v. Illinois Central Gulf Railroad Co. guides our analysis.
In that case, a railroad was found liable to one of its employ-
ees for injuries that the employee’s disability benefits had
compensated. Both the railroad and the employee had con-
tributed to the disability fund that covered the employee’s in-

. But the court decided that the de-
fendant’s contribution to the disability fund did not justify de-
ducting any damages when the plaintiff had also contributed
to the source of the benefits. 

on the purpose and
nature of the disability benefits: they were meant to compen-
sate employees for their service with the employer, and the
plaintiff could collect them regardless of the defendant’s lia-
bility. 

Thus, unlike a fund established to indemnify
an employer from liability, the purpose of the disability ben-
efits was independent of the obligation to pay damages to the
plaintiff, which meant that they were collateral to the dam-
ages award. 

court in Laird did not par-
tially offset the defendant’s damages to the extent of the de-
fendant’s contribution to the fund. Instead, the Laird court de-
clined to deduct any damages at all.
    Both the source and the purpose of Clanton’s benefits are
analogous to the plaintiff’s in Laird. Medicare Part B is funded
in part by general taxpayer revenues and in part by monthly
premiums paid by covered parties. So, as in Laird, the defend-
ant and the plaintiff each contributed to the fund that paid
Clanton’s benefits. 3 And as in Laird, the benefits here are not


    3 The government asks us to distinguish Clanton’s contribution to
Medicare from the plaintiff’s contribution to the disability fund in Laird.
The plaintiff in Laird could collect disability benefits only because he paid
into the disability fund. By contrast, the government characterizes
14                                                           No. 18-3060

intended to indemnify the government. On the contrary,
Clanton would have been able to collect his Medicare benefits
even if the government had not been liable for Jordan’s negli-
gence. We therefore agree with the district court that Clan-
ton’s benefits are collateral to his damages award under Illi-
nois law, and the government is not entitled to a partial offset.
                                     ***
    We VACATE the judgment and REMAND the case for
further proceedings consistent with this opinion.




Clanton’s benefits as a gratuity available to everyone diagnosed with end-
stage renal disease—even those who don’t contribute to Medicare. Thus,
the government argues, Clanton did not bargain for the benefit of his Med-
icare payments, notwithstanding his contributions. Since Laird, however,
the Supreme Court of Illinois has rejected a “benefit of the bargain” theory
that entitles plaintiffs to a full recovery of medical expenses only “where
the plaintiff has paid some consideration for the benefit of the write-off.”

. For that reason, we don’t think that Illinois
courts would find the distinction dispositive.