United States v. Ruiz-Gaxiola

FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

UNITED STATES OF AMERICA,  No. 08-10378
Plaintiff-Appellant,
D.C. No.
v.
 4:06-cr-01289-FRZ-
VICENTE RUIZ-GAXIOLA, AKA HCE
Inocente G. Ruiz,,
OPINION
Defendant-Appellant.

Appeal from the United States District Court
for the District of Arizona
Frank R. Zapata, District Judge, Presiding

Argued and Submitted
August 31, 2009—Pasadena, California

Filed September 24, 2010

Before: Alex Kozinski, Chief Judge, Stephen Reinhardt,
Circuit Judge, and Robert J. Timlin,* District Judge.

Opinion by Judge Reinhardt

*The Honorable Robert J. Timlin, United States District Judge for the
Central District of California, sitting by designation.

16275
16280 UNITED STATES v. RUIZ-GAXIOLA
COUNSEL

Deirdre M. Mokos (argued), Saul M. Huerta, Jr., Assistant
Federal Public Defenders, Tucson, Arizona, for the defendant-
appellant.

G. Micah Schmidt (argued), Monte Cress Clausen, Assistant
U.S. Attorneys, Tucson, Arizona, for the plaintiff-appellee.

OPINION

REINHARDT, Circuit Judge:

The government is allowed to medicate a defendant invol-
untarily for the purpose of rendering him competent to stand
trial only in rare circumstances. The question is whether this
case presents those rare circumstances. Our answer is that it
does not.

On July 26, 2006, defendant Vicente Ruiz-Gaxiola was
charged with illegal reentry in violation of 8 U.S.C. § 1326.
He was subsequently diagnosed with Delusional Disorder,
grandiose type, and was found incompetent to stand trial. An
administrative hearing conducted pursuant to Washington v.
Harper, 494 U.S. 210 (1990), resulted in a determination that
Ruiz was not a danger to himself or others in the institutional
context and that he did not suffer from a grave disability justi-
fying involuntary medication. The government then sought a
court order authorizing it to medicate Ruiz involuntarily for
the sole purpose of restoring him to competency for trial. A
court may not grant such a request unless the government
shows (1) “that important governmental interests are at stake”
in prosecuting the defendant for the charged offense; (2) “that
involuntary medication will significantly further those con-
comitant state interests,” i.e., it is substantially likely to
restore the defendant to competency and substantially
 UNITED STATES v. RUIZ-GAXIOLA 16281
unlikely to cause side effects that would impair significantly
his ability to assist in his defense at trial; (3) “that involuntary
medication is necessary to further those interests,” i.e., there
are no less intrusive treatments that are likely to achieve sub-
stantially the same results; and (4) “that administration of the
drugs is medically appropriate, i.e., in the patient’s best medi-
cal interest in light of his medical condition.” Sell v. United
States, 539 U.S. 166, 180-81 (2003). Orders authorizing
involuntary medication pursuant to this standard are “disfa-
vored.” United States v. Rivera-Guerrero, 426 F.3d 1130,
1137 (9th Cir. 2005).

After receiving evaluation reports and hearing testimony
from experts for the government and the defense, the magis-
trate judge concluded that the government had proved each of
the Sell requirements by clear and convincing evidence. The
district court adopted the magistrate judge’s findings of fact
and conclusions of law in their entirety, and entered an order
authorizing the government to administer antipsychotic medi-
cation forcibly, under the conditions set forth in the magistrate
judge’s Report and Recommendation. Ruiz now appeals that
ruling. We have jurisdiction under the collateral-order excep-
tion to 28 U.S.C. § 1291. Sell, 539 U.S. at 177.

We conclude that the district court erred in finding that the
Sell factors were met without affording the question the “thor-
ough consideration and justification” and “especially careful
scrutiny” required. United States v. Williams, 356 F.3d 1045,
1055 (9th Cir. 2004). Further, we hold that the district court
clearly erred in finding that the government proved by clear
and convincing evidence that the proposed regime of involun-
tary medication is substantially likely to restore Ruiz to com-
petency, as required under the second Sell factor. We also
conclude that the district court clearly erred in finding that the
government proved by clear and convincing evidence that the
proposed treatment is medically appropriate, as required
under the fourth Sell factor. Our holding under either the sec-
16282 UNITED STATES v. RUIZ-GAXIOLA
ond or the fourth factor would be sufficient to require rever-
sal. Accordingly, we reverse.

FACTUAL AND PROCEDURAL BACKGROUND

Ruiz, a citizen of Mexico, has been in federal custody since
his arrest on June 27, 2006. The government alleges that on
or about that date, he illegally re-entered the United States
after having been deported, in violation of 8 U.S.C. § 1326.
He has an extensive criminal history that dates back to 1984
and consists primarily of drug-related offenses. If he were
convicted of the offense with which he is currently charged,
he would face a Sentencing Guidelines range of 100 to 125
months of imprisonment.

Shortly after his arrest, Ruiz’s counsel moved for an evalu-
ation of his competency to stand trial. He was diagnosed with
Delusional Disorder, grandiose type,1 an extremely rare men-
tal illness that rendered him unable to understand the nature
and consequences of the proceedings against him or to assist
properly in his own defense. The magistrate judge thus found
him incompetent to stand trial and ordered him committed to
the custody of the Attorney General for evaluation of his
potential for restoration to competency. Ruiz was transferred
to the Federal Medical Center in Butner, North Carolina
(“FMC-Butner”).

The FMC-Butner staff members who evaluated Ruiz
agreed with the initial diagnosis of Delusional Disorder, gran-
1
The record reflects that, although there is a range of mental illnesses
that may be broadly described as delusional disorders, the diagnosis of
Delusional Disorder refers to a specific mental illness. The 2006 Merck
Manual of Medicine describes Delusional Disorder as a distinct disorder
“characterized by non-bizarre delusions (false beliefs) that persist for at
least 1 [month], without other symptoms of schizophrenia.” The Manual
describes the disorder as uncommon, and with respect to the grandiose
subtype, notes that “the patient believes he has a great talent or has made
an important discovery.”
 UNITED STATES v. RUIZ-GAXIOLA 16283
diose type, and determined that he remained incompetent to
stand trial. They encouraged him to take psychotropic medi-
cation, but he repeatedly refused to do so. In a written report
(the “FMC-Butner Evaluation”), the evaluators requested a
judicial order authorizing them to forcibly medicate Ruiz for
the sole purpose of restoring him to competency for trial. The
evaluators recognized that their request was subject to the
requirements set forth in Sell v. United States, 539 U.S. 166
(2003). The FMC-Butner Evaluation set forth the evaluators’
reasons for concluding that the Sell requirements were satis-
fied.

Due in part to our admonition that “Sell orders are disfa-
vored” and that courts should therefore “explore other proce-
dures . . . before considering involuntary medication orders
under Sell,” Rivera-Guerrero, 426 F.3d at 1137, the magis-
trate judge did not immediately hold a Sell hearing in
response to the FMC-Butner evaluators’ request. Instead, the
court first ordered the government to conduct an administra-
tive hearing pursuant to Harper, in order to evaluate whether
involuntarily medicating Ruiz was justified on the alternative
basis that his mental illness rendered him gravely disabled or
dangerous to himself or others. 494 U.S. 210. The govern-
ment held the Harper hearing and determined that Ruiz did
not meet either of those criteria.

Having determined that involuntary medication was not
justified on any basis other than to render Ruiz competent for
trial, the magistrate judge held two hearings to address the
Sell requirements. On March 6, 2008, the court held a short
hearing to discuss the first Sell factor, the strength of the gov-
ernment’s interest in prosecuting Ruiz, which the court
described as a purely “legal matter.” On March 14, 2008, the
court held a more extensive hearing at which the parties pres-
ented evidence relevant to the remaining Sell factors: whether
involuntary medication would be substantially likely to
restore Ruiz to competency and substantially unlikely to cause
side effects that would impair significantly his ability to assist
16284 UNITED STATES v. RUIZ-GAXIOLA
in his defense at trial; whether there were any less intrusive
treatments that would be likely to achieve the same results;
and whether involuntary medication was in his best medical
interest in light of his medical condition. See Sell, 539 U.S.
at 180-81. The hearing focused on those criteria in the context
of the government’s proposed use of a particular antipsychotic
medication, Haldol Decanoate.2

Two of the co-authors of the FMC-Butner Evaluation testi-
fied as expert witnesses for the government at the Sell hearing.3
At the time of the hearing, Dr. Mark Cheltenham had been a
staff psychiatrist at FMC-Butner for 17 months. He graduated
from medical school in 2002, completed his residency in
2006, and worked as a local physician for four months before
beginning work at FMC-Butner in October 2006. He was not
board certified in psychiatry. Dr. Cheltenham met with Ruiz
four or five times, for a total of about three hours.

The government’s other expert was Carlton Pyant, a foren-
sic psychologist at FMC-Butner. He received a master’s
degree in psychology in 1984 and a Ph.D. in psychology in
1989. He worked for the Army for six years before beginning
work for the Bureau of Prisons, where he had been employed
for fifteen years prior to the Sell hearing. He met with Ruiz
“at least seven” times. Because Pyant is not a medical doctor,
the government relied primarily on Dr. Cheltenham’s testi-
mony regarding medical issues.

The defense expert was Dr. Robert Cloninger, a psychiatrist
who graduated from medical school in 1970, completed his
residency in 1973, and has been board certified in psychiatry
since 1975. He is a professor of psychiatry and of genetics
2
Throughout the record, the drug is also referred to as “Haldol,” “Halo-
dol,” or “Haloperidol.” These variations in name all refer to the same
drug.
3
The third co-author of the FMC-Butner Evaluation was a Predoctoral
Psychology Intern. He did not testify at the Sell hearing.
 UNITED STATES v. RUIZ-GAXIOLA 16285
and psychology at Washington University, where he has been
on the faculty since 1973. He also has a psychiatry practice
in which he treats patients for approximately 15 hours per
week. He has published over 350 articles and 8 books about
psychiatry and has served as the editor or associate editor of
several journals. He met with Ruiz by video teleconference
for two hours and ten minutes, and also reviewed a number
of documents including Ruiz’s initial competency report, the
FMC-Butner Evaluation, and the Harper hearing report.

Although all of the experts agreed that Ruiz suffered from
Delusional Disorder and remained incompetent to stand trial,
the two sides disagreed as to whether he should be involuntar-
ily medicated. Testifying for the government, Dr. Cheltenham
opined that involuntary medication would be substantially
likely to restore Ruiz to competency; that intolerable side
effects would be unlikely to manifest themselves during the
relatively brief time frame of the proposed treatment; that
Ruiz’s lack of cooperation and unwillingness to acknowledge
his mental illness precluded the possibility of less intrusive
treatments; and that the proposed treatment was medically
appropriate. Testifying for the defense, Dr. Cloninger opined
that involuntary medication would likely exacerbate rather
than improve Ruiz’s delusional thinking; that he would imme-
diately face a risk of serious and potentially irreversible side
effects; that a treatment approach focused on establishing a
trusting therapeutic alliance would be a less intrusive means
of restoring him to competency; and that the proposed treat-
ment was not medically appropriate.

On June 3, 2008, the magistrate judge issued a Report and
Recommendation in which he concluded that the government
had met its burden of proving each of the Sell requirements
by clear and convincing evidence. He recommended that the
district court grant the government’s request to medicate Ruiz
involuntarily, subject to specified limitations on medication,
dosage, and duration of treatment.
16286 UNITED STATES v. RUIZ-GAXIOLA
Ruiz filed objections with the district court, and requested
a stay in the event of an adverse decision. On August 19,
2008, the district court issued an order denying Ruiz’s objec-
tions and adopting the magistrate judge’s findings of fact and
conclusions of law in their entirety. The court authorized the
government to medicate Ruiz involuntarily, under the condi-
tions outlined in the magistrate judge’s Report and Recom-
mendation. The court stayed the involuntarily medication
order for a period of thirty days so as to allow Ruiz to seek
appellate review.

Ruiz submitted a timely Notice of Interlocutory Appeal of
the district court’s order. On September 17, 2008, we granted
his emergency motion to stay that order pending the resolu-
tion of this appeal.

DISCUSSION

“The Supreme Court has thrice recognized a liberty interest
in freedom from unwanted antipsychotic drugs.” Williams,
356 F.3d at 1053 (internal quotation marks omitted); see Sell,
539 U.S. 166; Riggins v. Nevada, 504 U.S. 127 (1992); Har-
per, 494 U.S. 210. Antipsychotic medications are designed to
cause a personality change that, “if unwanted, interferes with
a person’s self-autonomy, and can impair his or her ability to
function in particular contexts.” Williams, 356 F.3d at 1054
(citing Riggins, 504 U.S. at 137). In addition to the intended
changes in cognition and behavior, the drugs “can have seri-
ous, even fatal, side effects.” Id. (quoting Harper, 494 U.S. at
229-30). Accordingly, the Court has “refus[ed] to permit
involuntary medication except in highly-specific factual and
medical circumstances.” Rivera-Guerrero, 426 F.3d at 1136.
Together, “Harper, Riggins, and Sell demonstrate the Court’s
reluctance to permit involuntary medication except in rare cir-
cumstances.” Id. at 1138.

[1] As we stated earlier, under Sell, the government cannot
involuntarily medicate a mentally ill defendant for the pur-
 UNITED STATES v. RUIZ-GAXIOLA 16287
pose of rendering him competent to stand trial unless it proves
(1) “that important governmental interests are at stake;” (2)
“that involuntary medication will significantly further those
concomitant state interests;” (3) “that involuntary medication
is necessary to further those interests;” and (4) “that adminis-
tration of the drugs is medically appropriate.” Sell, 539 U.S.
at 180-81. The Sell factors do not represent a balancing test,
but a set of independent requirements, each of which must be
found to be true before the forcible administration of psy-
chotropic drugs may be considered constitutionally permissi-
ble. See United States v. Hernandez-Vasquez, 513 F.3d 908,
913 (9th Cir. 2008). Here, Ruiz contends that the government
failed to prove any of the four Sell requirements, and that the
district court’s order authorizing involuntary medication must
therefore be reversed.

I.

[2] In order to address Ruiz’s claim that the government
did not meet the burden imposed by Sell, we must first deter-
mine what burden of proof the government was required to
satisfy. The Supreme Court did not explicitly state the burden
of proof applicable to the government’s request for a Sell
order, and our court has not previously addressed that issue.
A review of the relevant case law, however, demonstrates that
a preponderance of the evidence standard would provide inad-
equate protection for the important constitutional interests at
stake.

The Supreme Court has repeatedly emphasized the impor-
tance of the liberty interests affected by an involuntary medi-
cation order. “The forcible injection of medication into a
nonconsenting person’s body represents a substantial interfer-
ence with that person’s liberty.” Harper, 494 U.S. at 229.
That interference is “particularly severe” in the case of invol-
untary medication with antipsychotic drugs. Riggins, 504 U.S.
at 134. Because an individual “possesses a significant liberty
interest in avoiding the unwanted administration of antipsy-
16288 UNITED STATES v. RUIZ-GAXIOLA
chotic drugs under the Due Process Clause of the Fourteenth
Amendment,” Harper, 494 U.S. at 221-22, the Court has
demanded “a finding of overriding justification and a determi-
nation of medical appropriateness” before an involuntary
medication order may be issued. Riggins, 504 U.S. at 135.

These significant liberty interests call for equally signifi-
cant procedural safeguards. Because “an order compelling a
person to take antipsychotic medication is an especially grave
infringement of liberty,” it requires “thorough consideration
and justification” and “especially careful scrutiny,” and must
be based on “a medically-informed record.” Williams, 356
F.3d at 1055-56. Applying those principles in United States v.
Rivera-Guerrero, 426 F.3d 1130, 1138 (9th Cir. 2005), we
held that the district court abused its discretion in denying the
defendant’s request for a continuance of his Sell hearing. We
reasoned that “[t]he importance of the defendant’s liberty
interest, the powerful and permanent effects of anti-psychotic
medications, and the strong possibility that a defendant’s trial
will be adversely affected by the drug’s side-effects all coun-
sel in favor of ensuring that an involuntary medication order
is issued only after both sides have had a fair opportunity to
present their case and develop a complete and reliable
record.” Rivera-Guerrero, 426 F.3d at 1138. Those consider-
ations also weigh in favor of requiring the government to sat-
isfy a heightened evidentiary standard.

We have noted that “Sell inquiries are disfavored in part
because the medical opinions required for a Sell order are
more multi-faceted, and thus more subject to error, than those
required for a Harper analysis.” Hernandez-Vasquez, 513
F.3d at 915. Accordingly, although the Supreme Court
rejected the contention that the “clear and convincing evi-
dence” standard was required for Harper hearings conducted
by medical professionals, Harper, 494 U.S. at 235, the “more
error-prone analysis” involved in a Sell hearing conducted by
judicial officers, Hernandez-Vasquez, 513 F.3d at 915, calls
for a more stringent burden of proof.
 UNITED STATES v. RUIZ-GAXIOLA 16289
[3] Because of the importance of the liberty interests impli-
cated by a Sell order and the high risk of error, every circuit
to address the issue has concluded that the government must
bear the burden of proving the relevant facts by clear and con-
vincing evidence. See United States v. Bush, 585 F.3d 806,
814 (4th Cir. 2009) (“A higher standard . . . minimizes the
risk of erroneous decisions in this important context . . . .
[We] conclude that the government must satisfy the Sell fac-
tors by clear and convincing evidence.”); United States v.
Grape, 549 F.3d 591, 598 (3d Cir. 2008) (“[A]ll courts of
appeals addressing this issue have held that the Government
bears the burden of proof on factual questions by clear and
convincing evidence.”); United States v. Payne, 539 F.3d 505,
508-09 (6th Cir. 2008) (“[T]he risk of error and possible harm
involved in deciding whether to forcibly medicate for the pur-
pose of restoring competency are so substantial as to require
the government to prove its case by clear and convincing evi-
dence.” (internal quotation marks omitted)); United States v.
Bradley, 417 F.3d 1107, 1114 (10th Cir. 2005) (“[The] Sell
factors [that] depend upon factual findings . . . ought to be
proved by the government by clear and convincing evidence
. . . . [because of] the vital constitutional liberty interest at
stake . . . .” ); United States v. Gomes, 387 F.3d 157, 160 (2d
Cir. 2004) (“[T]he relevant findings must be supported by
clear and convincing evidence.”). We agree with those other
circuits and hold that the government has the burden of estab-
lishing the facts necessary to allow it to prevail as to each Sell
factor by clear and convincing evidence.

The first Sell factor, the importance of the government’s
interest in prosecution, is primarily a legal question. In con-
trast, the remaining Sell factors require the trial court to
resolve disputed issues by weighing expert testimony and
evaluating other medical evidence, and thus involve questions
that are primarily factual in nature. Accordingly, we “review
the district court’s determinations with regard to the first Sell
factor de novo, and the remaining Sell factors for clear error.”4
Hernandez-Vasquez, 513 F.3d at 915-16.
4
That the district court here adopted in full the findings of the magistrate
judge does not alter this standard of review. “Findings of fact of a magis-
16290 UNITED STATES v. RUIZ-GAXIOLA
A trial court’s factual finding “is ‘clearly erroneous’ when,
although there is evidence to support it, the reviewing court
on the entire evidence is left with the definite and firm con-
viction that a mistake has been committed.” United States v.
Hinkson, 585 F.3d 1247, 1260 (9th Cir. 2009) (en banc)
(quoting United States v. U.S. Gypsum Co., 333 U.S. 364, 395
(1948)). Again, this year we reiterated that well-established
standard, stating once again that we find clear error only when
we have “a definite and firm conviction that a mistake has
been committed.” Rhoades v. Henry, 596 F.3d 1170, 1177
(9th Cir. 2010) (citation and internal quotation marks omit-
ted). After sixty years of judicial experience with that stan-
dard, it needs no further explication. We jurists know what it
means.

II.

Having established the appropriate burden of proof and
acknowledged the appropriate standard of review, we turn
now to the merits of Ruiz’s claim and examine whether his
case presents those “rare circumstances” in which forcible
medication is appropriate. We must determine whether the
district court erred in concluding that the government satisfied
the first of the four requirements for an involuntary medica-
tion order under Sell, and whether the district court clearly
erred in finding that the government proved the remaining
requirements by clear and convincing evidence. In doing so,
we will also examine whether the strict procedural safeguards
applicable to so significant a deprivation of a defendant’s lib-
erty interest have been complied with.

A.

[4] The first factor of the Sell test requires the government
to establish “that important governmental interests are at

trate judge adopted by the district court are reviewed under the clearly
erroneous standard.” Wildman v. Johnson, 261 F.3d 832, 836 (9th Cir.
2001).
 UNITED STATES v. RUIZ-GAXIOLA 16291
stake.” Sell, 539 U.S. at 180. In its explanation of this require-
ment, the Supreme Court stated that “[t]he Government’s
interest in bringing to trial an individual accused of a serious
crime is important.” Id. The Court also cautioned, however,
that courts “must consider the facts of the individual case in
evaluating the Government’s interest in prosecution” because
“[s]pecial circumstances may lessen the importance of that
interest.” Id. For example, a defendant who refuses medica-
tion may be subject to a lengthy period of civil commitment,
which would “diminish the risks that ordinarily attach to free-
ing without punishment one who has committed a serious
crime,” and thereby lessen the need for prosecution. Id. Simi-
larly, the government has a reduced interest in prosecuting a
defendant who “has already been confined for a significant
amount of time (for which he would receive credit toward any
sentence ultimately imposed).” Id. (internal citation omitted).

[5] Hernandez-Vasquez held that “the likely guideline
range is the appropriate starting point” in determining whether
a defendant’s crime is serious enough to satisfy the first prong
of the Sell test.5 Hernandez-Vasquez, 513 F.3d at 919. We rea-
soned that the guidelines calculation provides “the best avail-
able predictor of the length of a defendant’s incarceration”
and that, unlike the statutory maximum, it includes some con-
sideration of “the specific circumstances of individual defen-
dants.” Id. We recognized that the likely sentencing guideline
range is not “the only factor that should be considered,” how-
ever, because “the sentencing guidelines do not reflect the full
universe of relevant circumstances.” Id.
5
In Hernandez-Vasquez, we determined that the offense at issue in that
case, illegal reentry in violation of 8 U.S.C. § 1326, may “at least under
some circumstances . . . constitute a ‘serious’ crime sufficient to justify
involuntary medication under Sell.” 513 F.3d at 919. Accordingly,
although Hernandez-Vasquez does not stand for the proposition that illegal
reentry will always qualify as a “serious” crime for purposes of the Sell
test, it forecloses Ruiz’s argument that the offense can never so qualify.
16292 UNITED STATES v. RUIZ-GAXIOLA
[6] Applying the framework set forth in Hernandez-
Vasquez, we begin by considering the guideline range appli-
cable to Ruiz, which is 100 to 125 months, not taking into
consideration possible adjustments and departures under the
Guidelines. This potentially lengthy sentence, which results
primarily from Ruiz’s extensive criminal history, suggests
that the government has an important interest in prosecuting
him for the current offense. The fact that he was arrested for
this offense less than 14 months after he was last released
from incarceration supports that conclusion. See Hernandez-
Vasquez, 513 F.3d at 919 (approving the district court’s con-
sideration of “the closeness in time of the prior offenses to the
current prosecution” as a factor in the seriousness inquiry).

[7] We must also examine, however, whether the facts of
Ruiz’s case present any “[s]pecial circumstances [that] may
lessen the importance” of the government’s interest in prose-
cution. Sell, 539 U.S. at 180. Ruiz has been in federal custody
since his arrest on June 27, 2006; thus, he has already been
confined for more than 47 months. The long duration of
Ruiz’s current confinement, which would be credited towards
any sentence ultimately imposed, weakens the government’s
interest in prosecution. Id. Another circumstance that would
in some instances lessen the government’s interest in prosecu-
tion is the possibility that the defendant’s refusal to take medi-
cation voluntarily could result in “lengthy confinement in an
institution for the mentally ill[, which] . . . would diminish the
risks that ordinarily attach to freeing without punishment one
who has committed a serious crime.” Id. Because nothing in
the record suggests that Ruiz is subject to civil commitment
as a danger to himself or others, however, that possibility is
very slight here, if indeed it exists at all.6
6
The district court erroneously concluded that the record affirmatively
establishes that Ruiz is not subject to civil commitment. It reached that
conclusion on the basis that Ruiz was found non-dangerous at a Harper
hearing. The result of the Harper hearing, however, establishes only that
Ruiz does not pose a danger to himself or others while confined in the
 UNITED STATES v. RUIZ-GAXIOLA 16293
[8] When the duration of his present confinement (47
months) is deducted from the applicable guidelines range (100
to 125 months), Ruiz remains subject to an additional 53 to
78 months of imprisonment. In light of that substantial period
of additional confinement and the slim possibility of Ruiz’s
future civil commitment, we conclude that the government
would ordinarily have an important interest in prosecuting
Ruiz for the offense with which he is currently charged, and
we would ordinarily hold that the district court did not err in
determining that the government satisfied the first of the four
Sell requirements by clear and convincing evidence.

[9] There is one other circumstance that we should con-
sider before reaching a final conclusion on this point. It seems
reasonable to conclude from the record that Ruiz was inclined
to commit this offense at least in part by his mental condition,
his Delusional Disorder, grandiose type. In some cases, that
the offense was the result of a mental disorder of this type
might well render it less important that the government prose-
cute the particular defendant. For example, it may be less
important where, as here, the crime is neither against persons
nor property.7 It is also important, however, that Ruiz may, as

institutional context. See United States v. Godinez-Ortiz, 563 F.3d 1022,
1026 (9th Cir. 2009). The Harper hearing did not address whether Ruiz
might pose a danger to himself or others if released, and therefore did not
rule out the possibility that he will be subject to civil commitment in the
future.
We note that we examine dangerousness here only as it relates to the
possibility of future civil commitment, and not as a consideration that
would provide direct and independent support for an involuntary medica-
tion order. As we cautioned in Hernandez-Vasquez, courts must “remain
mindful of the Supreme Court’s distinction between the purposes and
requirements of involuntary medication to restore competency,” on the
one hand, “and involuntary medication to reduce dangerousness,” on the
other hand. 513 F.3d at 919. The two inquiries should not be permitted “to
collapse into each other.” Id.
7
As we noted in Hernandez-Vasquez, the Sell test does not create any
categorical rule precluding courts from determining that a defendant’s
16294 UNITED STATES v. RUIZ-GAXIOLA
a result of his mental condition, repeat his offense if he con-
tinues to believe that God wants him to be in the United
States, as he apparently thought when he last entered unlaw-
fully. Whether prosecution of such an individual remains
important under these circumstances is a question that we hes-
itate to answer now. We will assume, however, for purposes
of Ruiz’s case that his prosecution would serve an important
governmental interest, notwithstanding that the principal
cause of his conduct was his mental disorder, and thus that the
district court did not err as a matter of law with respect to the
first Sell factor.

B.

[10] Under the second prong of the Sell test, the govern-
ment must establish that “involuntary medication will signifi-
cantly further” its interest in prosecuting the defendant for the
charged offense. Sell, 539 U.S. at 181. This factor requires the
government to prove two facts by clear and convincing evi-
dence: first, “that administration of the drugs is substantially
likely to render the defendant competent to stand trial”; and
second, “that administration of the drugs is substantially
unlikely to have side effects that will interfere significantly
with the defendant’s ability to assist counsel in conducting a
trial defense, thereby rendering the trial unfair.” Id.

[11] In his Report and Recommendation, the magistrate
judge set forth the conflicting opinions of the parties’ experts
on both sub-prongs of this Sell requirement, then issued a
very brief statement of his findings and conclusions:

The Court finds by clear and convincing evidence
that the medication regimen FMC-Butner proposes is

“non-property, non-violent” crime is a serious offense. 513 F.3d at 918.
But neither does it preclude courts from considering the nature of the
crime as one of many factors that may be relevant in a particular case.
Rather, courts must “consider the facts of individual cases in evaluating
the government’s interest in prosecution.” Id.
 UNITED STATES v. RUIZ-GAXIOLA 16295
designed to reduce Defendant’s delusions, restore
normal thought processes, improve cognitive func-
tioning in the courtroom and enable Defendant to
assist his attorney. Consequently, the medication is
substantially likely to render Defendant competent to
proceed to trial and substantially unlikely to produce
side effects that would interfere with Defendant’s
ability to assist his attorney or that would be harmful
to him.

(Emphasis added). The magistrate judge’s reasoning is clearly
flawed. It does not follow that because the use of a product
is designed to accomplish an end, it does so. Nor does it fol-
low that it is substantially likely that it will do so, let alone
substantially unlikely that it will have unintended adverse
effects. Because the second factor of the Sell test requires the
government to show by clear and convincing evidence what
it is substantially likely that the involuntary medication regi-
men will do (and what it is substantially likely that it will not
do), the government cannot satisfy its burden by showing
what the involuntary medication regimen is designed to do.
Accordingly, the magistrate judge’s stated reasoning does not
adequately support his conclusion with regard to the second
Sell factor.

Moreover, the magistrate judge failed to make any factual
findings relevant to the second prong of the Sell test. There is
a compelling need in cases such as this for the district court
to make factual findings so that the defendant may be assured
that the trial court has conducted the stringent review man-
dated in light of the substantial infringement on his liberty
interests, and so that upon review the appellate court may
determine whether the findings are supported by clear and
convincing evidence. Here, several factual issues were both
vigorously disputed by the parties and critically important in
determining whether involuntary medication is substantially
likely to restore Ruiz to competency to stand trial. Rather than
resolve those disputes, the magistrate judge simply set forth
16296 UNITED STATES v. RUIZ-GAXIOLA
the testimony offered by each side and relied solely on the
conclusion that we have quoted above.

The inadequate reasoning and complete absence of findings
that we have just described demonstrate that the magistrate
judge and the district court, in adopting the magistrate judge’s
report, failed to comply with the procedural safeguards that
are required before a court may authorize the “especially
grave infringement of liberty” represented by the unwanted
administration of antipsychotic drugs. Cf. Williams, 356 F.3d
at 1055. Ruiz was denied the “thorough consideration and jus-
tification” and “especially careful scrutiny” that he was due.
Id.

[12] The district court’s failure to comply with the neces-
sary procedural safeguards would alone prevent us from
upholding its involuntary medication order. Although in some
instances we might remand to allow the court to remedy its
failures, doing so here would be futile. There is no explana-
tion that the court could provide on remand and no findings
consistent with the record before us that would allow us to
conclude that the government has met its burden under the
second Sell factor by clear and convincing evidence.

The parties agree that Ruiz suffers from Delusional Disor-
der, that he does not believe that he is mentally ill, and that
he has never been treated with antipsychotic medication in the
past. They do not agree, however, whether forcible injections
of Haldol are substantially likely to restore Ruiz to compe-
tency for trial. As we have noted, the principal factual issues
relevant to this prong of the Sell test are vigorously disputed
by the parties. Among the contested questions are 1) whether
antipsychotic medication is the clinically accepted treatment
for Delusional Disorder; and 2) whether a 2007 study con-
ducted at FMC-Butner (the “Herbel Study”) establishes that
involuntary medication is successful in restoring detainees
with Delusional Disorder to competency.
 UNITED STATES v. RUIZ-GAXIOLA 16297
With regard to the first contested question, the government
evaluators opined in the FMC-Butner Evaluation that “antip-
sychotic medication is the accepted and appropriate treatment
for an individual with the diagnosis of Delusional Disorder,
Grandiose Type.” Defense expert Cloninger opined that this
statement was entirely inaccurate, and that in fact, there is no
clinical consensus as to whether Delusional Disorder should
be treated with antipsychotic medication.

The government evaluators’ statement regarding antipsy-
chotic medication as the standard treatment for Delusional
Disorder conflicts with the information set forth in the 2006
Merck Manual of Medicine, the only medical reference text
introduced into evidence, a text identified by the defense,
without contradiction, as the medical equivalent of Black’s
Law Dictionary. According to the Merck Manual, treatment
for Delusional Disorder “aims to establish an effective
physician-patient relationship and to manage complications.”
The manual states that “[i]nsufficient data is available to sup-
port the use of any particular drug [in treating Delusional Dis-
order], although antipsychotics sometimes suppress
symptoms.” (emphasis added). In contrast, the government
evaluators simply stated in the FMC-Butner Evaluation, with-
out reference to any published authority, that “the standard
and accepted treatment for anyone with the diagnosis of Delu-
sional Disorder would involve the prescription of antipsycho-
tic medication.” (Emphasis added).

[13] Similarly, government expert Cheltenham initially
testified at the Sell hearing that he viewed antipsychotic medi-
cation as “the next step in [Ruiz’s] treatment” because Delu-
sional Disorder “is classified as a psychotic disorder, and the
first line of treatment . . . for a psychotic illness is an antipsy-
chotic medication.” Subsequently, however, he was asked
whether there was a “scientifically established consensus on
how to treat Mr. Ruiz in a way which is likely to be benefi-
cial.” He responded that
16298 UNITED STATES v. RUIZ-GAXIOLA
there’s not an explicit consensus. There’s a paucity
of data . . . regarding the treatment of delusional dis-
order in terms of with medications . . . . And so no,
there is no explicit practice guideline or consensus
on — on how to treat patients with delusional disor-
der, and particularly as it regards medications.

In sum, the government’s original claim that “antipsychotic
medication is the accepted and appropriate treatment for an
individual with the diagnosis of Delusional Disorder, Grandi-
ose Type,” was thoroughly discredited by the defense and
ultimately abandoned by the government. Thus, the govern-
ment’s initial contention that antipsychotic medication is the
clinically accepted treatment for Delusional Disorder is with-
out support in the record. Accordingly, it provides no support
for the government’s assertion that the forced medication is
substantially likely to restore Ruiz to competency for trial.

With regard to the second contested question, the govern-
ment experts offered the Herbel Study as evidentiary support
for their opinion that involuntarily medicating Ruiz is sub-
stantially likely to restore him to competency. The Herbel
Study consisted of a retrospective review conducted in 2007
by Drs. Herbel and Stelmach, who examined the outcomes of
twenty-two patients diagnosed with Delusional Disorder and
treated with antipsychotic medication at the FMC-Butner
facility. Because the Herbel Study specifically addressed the
involuntary medication of incompetent, non-dangerous pre-
trial detainees with Delusional Disorder, Dr. Cheltenham
opined that it provided particularly relevant evidence as to the
probable outcome of involuntarily medicating Ruiz. He testi-
fied that the study documented a “77 percent restoration of
competency with antipsychotic treatment” for the twenty-two
patients whose outcomes were reviewed.

Defense expert Cloninger opined that as a retrospective
review without a randomized control group, the Herbel Study
lacked sufficient scientific validity to support any firm con-
 UNITED STATES v. RUIZ-GAXIOLA 16299
clusions about the efficacy of involuntary medication in treat-
ing detainees with Delusional Disorder. Dr. Cloninger
explained that “treatment studies that do not make compari-
sons with untreated controls can easily lead to the erroneous
belief in the efficacy of whatever was done,” and opined that
the absence of a control group in the Herbel Study meant that
it “actually d[id] not show that the use of the medications
ha[d] a cause and effect relationship to the outcomes.” He
cited other studies that had shown similar rates of success
without the use of antipsychotic medication.8 He opined that
the Herbel Study provided only “preliminary anecdotal sup-
port, which is not an adequate scientific basis for expert opin-
ion.”

Government expert Cheltenham agreed that retrospective
reviews like the Herbel Study are generally biased in favor of
finding that treatment produced favorable results. The Herbel
Study itself acknowledged its structural limitations, noting
that

some patients may have been misdiagnosed and
wrongly included or excluded from this study popu-
lation. Standard research methods to reduce bias . . .
were not possible in this study. Without these safe-
guards, the opinions of the forensic examiners may
have been biased in favor of finding a positive
response to treatment.

....
8
Specifically, a study published in 2000 found that 63% of patients with
Delusional Disorder improved with appropriate psychological support
only, “whether or not they [were] administered antipsychotic medication.”
The study was referenced in Dr. Cloninger’s evaluation report, which was
made available to the government experts prior to the Sell hearing. When
asked about the study at the Sell hearing, government expert Cheltenham
testified that he was not familiar with it.
16300 UNITED STATES v. RUIZ-GAXIOLA
Despite the limitations of this study, the results
provide mental health professionals some evidence
that most of the incompetent male defendants with a
diagnosis of delusional disorder, especially the per-
secution subtype, will respond favorably to involun-
tary treatment with standard doses of first- and
second-generation antipsychotic medications. Addi-
tional research is needed to confirm and expand on
these findings.

(Emphases added). By their own terms, the findings of the
Herbel Study are both limited and tentative. Certainly, they do
not constitute clear and convincing evidence that involuntarily
medicating Ruiz, who is not even a “persecution subtype,” is
substantially likely to restore him to competency.

In addition to contending that the government experts’ reli-
ance on the Herbel Study was unwarranted, defense expert
Cloninger opined that “involuntary treatment of Mr. Ruiz-
Gaxiola with antipsychotic medication is likely to worsen his
condition, not restore him to competency.” Dr. Cloninger
based that opinion on “the crucial role of inferiority feelings
and hypersensitivity to powerlessness in the development of
Delusional Disorder in general and in Mr. Ruiz-Gaxiola in
particular.” He testified that involuntary medication would
“basically make [Ruiz] fight back,” using his delusions to
reduce his sense of powerlessness, “and he’ll become more
grandiose in proportion to how much he is forced to do things
against his will.”

The government experts disagreed, opining that the pro-
posed regime of involuntary medication would reduce rather
than increase Ruiz’s delusional thinking. In support of that
opinion, the government experts primarily relied on the
effects of antipsychotic medication on delusional thought pro-
cesses generally, rather than evidence specific to the particu-
lar mental illness from which Ruiz suffers.9 The FMC-Butner
9
The exceptions, which we have already discussed, are the govern-
ment’s discredited and abandoned position that antipsychotic medication
 UNITED STATES v. RUIZ-GAXIOLA 16301
Evaluation noted that because Ruiz had never been treated
with antipsychotic medication, “there is no established pattern
of treatment response to review and upon which to base the
probability of his current response to medication.” Instead, the
government evaluators supported their assertions regarding
Ruiz’s likely response to the government’s proposed treat-
ment by reference to “general observations . . . regarding
treatment response to antipsychotic medications,” including
statistics published by the American Psychiatric Association
(APA) about the efficacy of antipsychotic medications in
treating patients with Schizophrenia.10 During the Sell hear-
ing, however, government expert Cheltenham acknowledged
that Delusional Disorder and Schizophrenia are “distinct dis-
orders.” He conceded that the FMC-Butner Evaluation pro-
vided data for Schizophrenia, rather than Delusional Disorder,
because there was no known treatment for Delusional Disor-
der that had been validated in randomized clinical trials.

In addition to asserting the general proposition that antipsy-
chotic medication tends to reduce delusional thinking, Dr.
Cheltenham testified that he had treated two patients with
Delusional Disorder at a clinic outside of the prison system
for three months, and both patients “had improvement with
antipsychotic medication.” The voluntary nature of that treat-
ment, however, distinguishes it from the involuntary treat-
ment that Dr. Cloninger testified would trigger an increase in
Ruiz’s delusional thinking as a means of asserting control.
Moreover, Dr. Cheltenham admitted that he had never
“treated a patient with delusional disorder with antipsychotic
medication” within the federal prison system.

is the clinically accepted treatment for Delusional Disorder and its reliance
on the small-scale and structurally limited Herbel Study.
10
The FMC-Butner Evaluation noted that “[r]esponse to antipsychotic
medication is highly individual,” and explained that “[b]ecause it is diffi-
cult to predict an individual’s response to antipsychotic medication, [the
APA statistics] have been provided to indicate the likelihood of response
if an individual is treated with an antipsychotic medication.”
16302 UNITED STATES v. RUIZ-GAXIOLA
The dispute over the factual issues described above notably
parallels the government and defense witnesses’ relative
expertise and knowledge regarding Delusional Disorder and
the response to treatment with antipsychotic medication of
persons suffering from that “uncommon” and distinct disor-
der. See supra note 1. The expert reports and testimony dem-
onstrate that Dr. Cloninger had a far superior knowledge base
than the government’s two “experts.” At the time of the hear-
ing, Dr. Cloninger had been board certified in psychiatry for
more than 33 years and had published hundreds of books and
articles on psychiatry, while the government psychiatrist, Dr.
Cheltenham, was not yet board certified and had been practic-
ing psychiatry for less than two years. Dr. Cloninger referred
to multiple scientific studies supporting his views, and the
government experts provided no explanation as to why those
studies did not support his conclusions. When asked about
one particularly relevant study, Dr. Cheltenham frankly
acknowledged that he had not read it, even though it was
referred to in a report that had been furnished to him.

In contrast to the multiple scientific studies that supported
Dr. Cloninger’s views, all of which were described in his
evaluation report and none of which the government even
attempted to discredit, the government’s evaluation report
cited only two studies and a “Practice Guideline,” none of
which had any apparent connection to Delusional Disorder
and one of which was specifically limited to a different men-
tal illness. The government experts’ evaluation report also
unquestionably misstated the existence of a scientific consen-
sus for treating Delusional Disorder. In their testimony at the
Sell hearing, the government experts attempted to provide sci-
entific support for their position by citing the Herbel Study,
but that study does not purport to be anything more than a
small-scale and structurally limited retrospective review that,
according to its authors, “may have been biased” and required
“additional research” to confirm its findings. The weaknesses
in the Herbel Study were readily admitted by the government
experts when they were questioned about it by the defense. In
 UNITED STATES v. RUIZ-GAXIOLA 16303
short, unlike Dr. Cloninger, the government experts did not
base their opinion as to the substantial likelihood of involun-
tary medication restoring Ruiz to competency on the charac-
teristics of his particular mental illness or on scientific data
obtained through randomized clinical trials.

The government experts relied in large part on the general
proposition that antipsychotic medication reduces mentally ill
patients’ delusional thought processes, and not on any specific
studies or other information regarding persons suffering from
the distinct disease of Delusional Disorder. From that general
proposition, they drew the direct inference that the involun-
tary administration of antipsychotic medication would reduce
Ruiz’s delusional thought processes. This type of reasoning
has very little value to courts conducting the Sell inquiry, as
the Fourth Circuit has recognized:

Instead of analyzing [the defendant] as an individ-
ual, the [government] report simply sets up syllo-
gisms to explain its conclusions: (1) atypical
antipsychotic medications are generally effective,
. . . (2) [the defendant] will be given atypical antip-
sychotic medications, (3) therefore, atypical antipsy-
chotic medication will be effective . . . for [the
defendant]. To hold that this type of analysis satis-
fies Sell’s second . . . factor[ ] would be to find [that]
the government necessarily meets its burden in every
case it wishes to use atypical antipsychotic medica-
tion.

United States v. Evans, 404 F.3d 227, 241 (4th Cir. 2005).
Like the Fourth Circuit, “[w]e do not believe that Sell’s analy-
sis permits such deference.” Id.

[14] In Ruiz’s case, as in Evans’s, the government experts
rely on generalities and fail to apply their views to Ruiz’s
condition with specificity. They also have made multiple mis-
statements and introduced inadequate supporting authority.
16304 UNITED STATES v. RUIZ-GAXIOLA
This showing stands in stark contrast to the specific and well-
supported testimony of the more-experienced and knowledge-
able defense expert. We conclude that the generalized state-
ments and unsupported assertions of the government experts,
when contrasted with the specific and authoritative rebuttal
evidence presented by the defense, were plainly insufficient to
establish by clear and convincing evidence that the proposed
regime of involuntary medication is substantially likely to
restore Ruiz to competency.11 The district court’s contrary
finding leaves us with “a definite and firm conviction that a
mistake has been committed.” Accordingly, we hold that the
district court clearly erred in finding that the government
proved the second Sell factor by clear and convincing evidence.12

11
We note that other circuits have recognized the weakness of evidence
that antipsychotic medication is successful in treating Delusional Disorder,
and accordingly, have hesitated to approve Sell orders for detainees with
this particular mental illness. See United States v. Bush, 585 F.3d 806, 817
(4th Cir. 2009) (vacating a Sell order for a detainee with Delusional Disor-
der and remanding for application of the appropriate burden of proof, in
part because “all experts agreed that there is a dearth of medical evidence
about the success of medicating persons suffering from Delusional Disor-
der, Persecutory Type”); United States v. Ghane, 392 F.3d 317, 319 (8th
Cir. 2004) (reversing a Sell order for a detainee with Delusional Disorder,
based in part on findings that the illness “resists treatment by . . . antipsy-
chotic medication”). But see United States v. Gomes, 387 F.3d 157, 161-
62 (2d Cir. 2004) (affirming a Sell order for a detainee with Delusional
Disorder where the defense did not present any expert testimony and both
government experts opined that involuntary medication was substantially
likely to restore him to competency).
12
In light of our conclusion that the district court clearly erred in finding
that the government proved by clear and convincing evidence that invol-
untary medication is substantially likely to restore Ruiz to competency, we
need not at this point review the district court’s finding with regard to the
other prong of the second Sell factor, which requires the government to
prove by clear and convincing evidence that involuntary medication is
substantially unlikely to have side effects that would render Ruiz’s trial
unfair.
 UNITED STATES v. RUIZ-GAXIOLA 16305
C.

[15] Under the third prong of the Sell test, the government
must establish “that involuntary medication is necessary to
further” its interest in prosecuting the defendant for the
charged offense. Sell, 539 U.S. at 180-81. Of course, if a pro-
posed treatment will not further the government’s interest in
prosecution, as required under the second Sell factor, it could
not possibly provide a necessary means of doing so. This
third factor thus assumes that the government has established
the second factor, and it then requires it to prove by clear and
convincing evidence “that any alternative, less intrusive treat-
ments are unlikely to achieve substantially the same results.”
Id. at 181.

Here, as we have explained, the district court clearly erred
in finding that the government proved the second Sell factor
by clear and convincing evidence. Had the government suc-
ceeded in establishing the second factor, however, the district
court’s finding as to the third factor would not be clearly erro-
neous.

Defense expert Cloninger opined that a therapeutic alliance
with a private psychiatrist, unaffiliated with the prison system,
would be less intrusive than involuntarily medicating Ruiz
and would produce superior results. Dr. Cloninger explained
that there was “clear and compelling scientific evidence” that
establishing a therapeutic alliance was beneficial to patients
with Ruiz’s diagnosis and personality. In his Report and Rec-
ommendation, which the district judge adopted, the magistrate
judge found that this “proposed alternative and less intrusive
treatment [was] unlikely to achieve the same results as invol-
untarily administered antipsychotic medication.” He offered
several reasons for that conclusion.

First, the report rejected Dr. Cloninger’s proposed alterna-
tive treatment because the “proposal [was] based on a two
hour ten minute long video teleconference interview of
16306 UNITED STATES v. RUIZ-GAXIOLA
Defendant, unlike the Government’s extended examinations
and observations.” (internal citations omitted). We find this
aspect of the court’s reasoning extremely troubling. Because
defendants are routinely detained at federal medical centers
prior to Sell hearings, it will almost always be the case that
the government experts who work at the federal medical cen-
ters will have had the opportunity to spend more time with the
defendant than a defense expert possibly could. If a court can
reject the opinion of a defense expert based solely on the
lengthier observation conducted by the government experts, it
will always be possible for the government to prevail in a Sell
hearing by having its experts spend a greater amount of time
with the defendant. Such an approach would create an unjusti-
fiable risk of erroneous results and deprive defendants of the
vitally important protections that a Sell hearing is designed to
provide.13

The report also found, however, that Ruiz “1) does not
believe he is mentally ill; 2) does not believe he needs medi-
cation; 3) does not believe he needs treatment of any kind; . . .
4) is vested and fixed in his delusional ideation”; and 5) “har-
bors elaborate delusions that involve the correctional system
generally, and his caretakers in particular, in conspiracies
against him.” The expert reports and testimony were substan-
tially in agreement with regard to those findings. Moreover,
government expert Cheltenham testified that the alternative
treatment proposed by the defense could not be provided
within the constraints of the prison environment, and defense
expert Cloninger conceded that those constraints would pres-
ent “a real challenge”:
13
In addition to our concern with this reasoning as a matter of legal prin-
ciple, we note that it finds little to no support in the facts of this case. The
government’s principal expert witness testified that he met with Ruiz “four
or five times,” for a total of “probably three” hours. The government’s
other expert witness testified that he “was probably in at least seven ses-
sions” with Ruiz, for a total of possibly three to four hours. This testimony
fails to establish that either of the government experts spent a significantly
longer amount of time with Ruiz than did the defense expert.
 UNITED STATES v. RUIZ-GAXIOLA 16307
[T]he reason that I think this is especially problem-
atic in a prison setting is that the forensic evaluators
do not regard themselves as having a doctor/patient
relationship . . . . . [T]hey are agents of the Court, not
his advocates, and he picks up on that. And he sees
that, you know, they’re not there for me; they’re
there for the Court, and he feels some adversarial
relationship between the Court that’s charging him
with a crime and himself.

. . . . And so, I think it’s very hard to treat him with
the current justice system the way it’s set up . . . .
[W]hen I really try to think about how could we do
this the way things are done now, I think it’s very
hard for the doctors at [the federal medical center]
. . . . [I]t really is a dilemma in our current situation.

Based on this evidence, the court could reasonably conclude
that aspects of Ruiz’s condition identified by the expert wit-
nesses, especially his resistance to treatment and his conspira-
torial delusions, made it unlikely that he would engage in the
voluntary therapy proposed by the defense while confined
within the federal prison system. Accordingly, the court did
not clearly err in finding that the government proved by clear
and convincing evidence that less intrusive treatments would
be unlikely to restore Ruiz to competency for trial.

The third Sell factor also requires courts to “consider less
intrusive means for administering the drugs, e.g., a court order
to the defendant backed by the contempt power, before con-
sidering more intrusive methods.” Sell, 539 U.S. at 181. Here,
the district court’s Sell order required the government “to
request of Defendant that he voluntarily take medication
orally before each and every scheduled administration of
medication by injection.” This requirement would ensure that
forcible administration of the drugs would not occur, or would
not continue to occur, if Ruiz consented to taking the medica-
tion. Because there does not appear to be any other means of
16308 UNITED STATES v. RUIZ-GAXIOLA
reducing the intrusiveness of the medication process, this
aspect of the order establishes that the court adequately con-
sidered less intrusive means of administering the drugs.

[16] Because the government failed to prove the second
Sell factor, it could not possibly have proved the third factor.
For the foregoing reasons, however, had the government
proved the second factor by clear and convincing evidence,
we would conclude that the court did not clearly err in finding
that the third Sell factor was satisfied by clear and convincing
evidence as well.

D.

[17] Under the fourth and final prong of the Sell test, a
court may authorize involuntary medication only if “adminis-
tration of the drugs is medically appropriate.” Id. This factor
requires that the government prove by clear and convincing
evidence that the proposed regime of involuntary medication
is “in the patient’s best medical interest in light of his medical
condition.” Id. (emphasis added).

The existence of factor four reflects the importance of the
liberty interests at stake in an involuntary medication order:
Even if forcible medication provides the only means of bring-
ing an incompetent defendant to trial for a serious crime, and
is substantially likely to restore him to competency without
causing side effects that would render the trial unfair, the
treatment is nonetheless impermissible unless it is also in the
defendant’s best medical interest. We find it noteworthy that
the Court used the word “patient” in its explanation of this
prong of the Sell test, rather than the word “defendant,” which
it used in its discussion of the other three factors. Sell, 539
U.S. at 180-81. The choice of words serves to emphasize that,
in analyzing this factor, courts must consider the long-term
medical interests of the individual rather than the short-term
institutional interests of the justice system.
 UNITED STATES v. RUIZ-GAXIOLA 16309
A proposed regime of involuntary medication will not ordi-
narily be medically appropriate unless it is, at the least, sub-
stantially likely to restore the defendant to competency, in
conformity with Sell factor two. Certainly that is true under
the facts of this case, and the magistrate judge explicitly based
his finding of medical appropriateness on his conclusion that
Ruiz was “restorable through monitored medication within a
matter of months.” Although the term “restorable” may be
ambiguous, it is clear from the context that the magistrate
judge meant “sufficiently restored to competency for the
period of the trial so as to be able to assist counsel in conduct-
ing trial defense” and no more. The magistrate judge also rec-
ognized other issues relevant to the medical appropriateness
inquiry, including the questions whether Ruiz would be “sub-
stantially harmed” by the medication or “experience long-
term effects from the medication.” He then stated that he had
already addressed those questions in his analysis under factor
two when he considered whether the proposed regime of
involuntary medication was substantially unlikely to have side
effects that would undermine the fairness of Ruiz’s trial. On
that basis, he incorporated the reasoning and conclusions of
the section of the Report and Recommendation that addressed
the second Sell factor in the section addressing the fourth Sell
factor, and did not engage in any further discussion of the
unlikelihood of the proposed treatment’s side effects or the
likelihood that the involuntary medication would restore Ruiz
to competency.

In relying on the portion of his report that addressed the
second Sell factor to satisfy the requirements of the fourth fac-
tor, the magistrate judge committed three serious errors. First,
the conclusions that the magistrate judge reached with regard
to the second Sell factor could not serve to inform his analysis
of the fourth Sell factor because the former section of the
report contained no findings, resolved none of the disputed
factual questions, and rested its conclusion on flawed reason-
ing. It included only a restatement of the various contentions
of the parties, followed immediately by the statement that
16310 UNITED STATES v. RUIZ-GAXIOLA
because the forced medication regime was designed to restore
Ruiz to competency and to avoid side effects, it was “substan-
tially likely to render Defendant competent to proceed to trial
and substantially unlikely to produce side effects that would
interfere with Defendant’s ability to assist his attorney or that
would be harmful to him.” We have already explained that
determining what a course of treatment is designed to do is
not sufficient when a court is charged with the responsibility
of determining what it actually does or is substantially likely
to do. See supra Part II.B.

Second, as discussed supra, the magistrate judge also
clearly erred in his consideration of the second factor by con-
cluding that the government proved by clear and convincing
evidence that the proposed regime of involuntary medication
is substantially likely to restore Ruiz to competency. Because
that erroneous conclusion was not only incorporated into the
magistrate judge’s analysis of the fourth factor, but formed an
integral part of that analysis, his conclusion as to medical
appropriateness was rendered erroneous as well.

Third, the magistrate judge apparently failed to appreciate
that the scope of the inquiry with respect to the fourth Sell
factor is far broader than with respect to the second Sell fac-
tor. The second prong of the Sell test requires courts to deter-
mine only whether the medication is “substantially likely to
render the defendant competent to stand trial” and “substan-
tially unlikely to have side effects that will interfere signifi-
cantly with [his] ability to assist counsel in conducting trial
defense.” 539 U.S. at 181. In contrast, the fourth Sell factor
requires the court to consider all of the medical consequences
of the proposed involuntary medication, including those con-
sequences that may not affect the defendant’s trial in any way,
but result in long term side effects.14 It also requires the court,
14
For example, if a proposed medication were known to cause side
effects that did not become apparent until years after the medication was
administered, those side effects would be relevant to the fourth prong of
the Sell test but not to the second.
 UNITED STATES v. RUIZ-GAXIOLA 16311
in making the decision, to consider the length of time the
treatment regime must be continued in order to provide the
desired medical benefit to the patient.

Even if the district court had not made the serious errors
that we have just described, it would be clear error for the dis-
trict court to find on the basis of the record before us that the
government proved by clear and convincing evidence that the
proposed regime of involuntary medication is medically
appropriate for Ruiz, the patient, in light of his medical condi-
tion. No doctor would prescribe a medication regime unless
the likelihood and value of its potential benefits outweighed
the likelihood and severity of its potential harms over the
course of the treatment, as opposed to during the period in
which the government wishes to conduct a criminal trial.
Here, the government unquestionably failed to prove by clear
and convincing evidence that its proposed treatment satisfies
this risk-benefit analysis over the relevant period.

[18] As we made clear earlier, there is no medical consen-
sus that antipsychotic medication is the appropriate treatment
for Delusional Disorder and very little evidence that it would
be successful in reducing Ruiz’s type of delusions. See supra
Part II.B. Even if there were such evidence, however, indispu-
tably the government did not make a clear and convincing
showing of medical appropriateness over the course of the
necessary treatment — Ruiz’s lifetime. There is no evidence
that the antipsychotic drug does more, at best, than control the
symptoms of Delusional Disorder, and none that it cures the
mental disorder from which Ruiz suffers. Moreover, the value
of even controlling the symptoms of Delusional Disorder is
unclear. Government expert Cheltenham testified that “pa-
tients with delusional disorder are generally high functioning
unless you elicit a discussion regarding their delusion. Other
than that, they function normally in society.” The record also
suggests that Ruiz, in particular, is not currently experiencing
significant distress as a result of his condition. The FMC-
Butner Evaluation stated that his “overall adjustment to the
16312 UNITED STATES v. RUIZ-GAXIOLA
institution has been appropriate” and that “no disciplinary
charges have been filed against him,” although there is proba-
bly some risk that his delusion that God wishes him to be in
the United States may cause him to repeat his border crossing.
When speaking with defense expert Cloninger, Ruiz described
his mood as “happy” despite his confinement. Certainly, it is
not medically appropriate to reduce Ruiz’s delusions by invol-
untary medication for the period of the criminal trial if the
delusions will resume upon its completion and, due to the
medication, cause him to experience long-term side effects.

The need for lifelong treatment and the uncertainty of the
medical benefit of the antipsychotic drug as a control device,
even over the short run, must be viewed in light of the evi-
dence regarding the very real risk associated with the use of
the proposed medication. Both parties agreed that Haldol
could cause serious side effects, most notably tardive dyskine-
sia, which government expert Cheltenham described as “a
very disfiguring side effect that can affect muscles anywhere
in the body.” Dr. Cheltenham testified that tardive dyskinesia
can be reversed “in up to 50 percent of the patients, if it’s
detected early.” In other words, according to the evidence
presented by the government, tardive dyskinesia is irrevers-
ible in at least fifty percent of those patients who experience
it, even when it is detected early.

The government experts proposed to treat involuntarily by
medication the symptoms of Ruiz’s illness only during the
period of his trial, including trial preparation, and asserted
that over that three- or four-month period of time, the risk of
tardive dyskinesia as a side effect would be very small. It is
undisputed, however, that the medication will not have any
effect after the conclusion of the three- or four-month period
unless Ruiz continues to take it, and to do so for an indefinite
period of time. From a patient’s standpoint, the medical bene-
fit of becoming competent to stand trial for only a few months
(even if that outcome were likely) and then returning to his
prior state of Delusional Disorder could not outweigh even a
 UNITED STATES v. RUIZ-GAXIOLA 16313
miniscule risk of a disfiguring and potentially irreversible side
effect. Because the proposed course of medication could be
considered medically appropriate treatment only for those
patients who expect or hope to continue undergoing it indefi-
nitely (and clearly any patient who must be forced to take the
antipsychotic medication involuntarily is not in that category),
“the patient’s best medical interest” cannot be measured, as
the district court and the government did, by evaluating the
benefit and risk over the period of the trial only. There is no
medical value to a medication regime that alleviates the men-
tal disorder, if it does so at all, only for the short period of
time necessary for trial preparation and the trial itself, while
creating a risk of side-effects that would render the regime
inappropriate for a patient’s long-term treatment. If the invol-
untary antipsychotic medication is to be administered for so
short a period, it is clearly not in “the patient’s best medical
interest” to risk serious medical consequences for a benefit
that, if one results at all, is only a temporary alleviation of the
symptoms and not a long-term remedy for the mental illness.

Because a medical justification for an involuntary regime
of Haldol treatment would be conceivable only if the antipsy-
chotic medication were to be taken on a long-term basis, the
issue would become whether the likelihood and value of the
long-term benefits outweigh the likelihood and severity of the
long-term harms. With regard to the latter, government expert
Cheltenham testified that the medication carries a twenty per-
cent “lifetime prevalence or lifetime risk” of tardive dyskine-
sia. Even disregarding the testimony of the more-experienced
defense psychiatrist that the administration of Haldol would
be counter-productive in any event, the government did not
prove by clear and convincing evidence that a Haldol regime
would be in the patient’s best medical interest in light of his
medical condition, because it introduced no evidence that the
purported lifetime benefits to Ruiz would make the proposed
treatment medically appropriate in spite of the significant life-
time risk.15 In fact, the government offered no evidence at all
15
If anything, the government expert’s testimony that people with Delu-
sional Disorder tend to lead relatively normal lives, and thus gain a rela-
tively limited benefit from treatment, supports the opposite conclusion.
16314 UNITED STATES v. RUIZ-GAXIOLA
as to whether the Haldol treatment would render Ruiz compe-
tent or have any positive effect at all on his Delusional Disor-
der at any time after the three or four months during which
the trial preparations and the trial itself would be conducted.

[19] For the above reasons, the government failed to estab-
lish by clear and convincing evidence that the proposed treat-
ment regime of involuntary medication is medically
appropriate in light of Ruiz’s medical condition, and the dis-
trict court clearly erred in finding that it did. Thus, we are
compelled to reverse not only because of the government’s
failure to meet the requirements of Sell factor two, but also
because of its failure to do so with respect to factor four.

Conclusion

Ruiz suffers from a mental disorder that is extremely rare
and difficult to treat. The government proposes to administer
antipsychotic medication involuntarily to Ruiz in order to fur-
ther its interest in prosecuting him for a serious criminal
offense by rendering him competent to stand trial. Ruiz, how-
ever, like all others, “possesses a significant liberty interest in
avoiding the unwanted administration of antipsychotic drugs
under the Due Process Clause of the Fourteenth Amendment.”
Harper, 494 U.S. at 221-22. The Supreme Court has resolved
the conflicting interests by establishing “rare” circumstances
under which the government will be permitted to administer
antipsychotic drugs involuntarily. In Sell, 539 U.S. at 180, it
set forth the four conditions that the government must satisfy
in order to obtain an order authorizing it to involuntarily med-
icate a non-dangerous criminal defendant. A failure to meet
any of the four is fatal to the government’s request.

[20] Under Sell, an involuntary medication order by the
district court cannot be issued unless the government proves
1) “that important governmental interests are at stake”; 2)
“that involuntary medication will significantly further those
concomitant state interests”; 3) “that involuntary medication
 UNITED STATES v. RUIZ-GAXIOLA 16315
is necessary to further those interests”; and 4) “that adminis-
tration of the drugs is medically appropriate.” Id. at 180-81.
The government has the burden of establishing the facts nec-
essary to allow it to prevail on its request by clear and con-
vincing evidence. Here, the government fell far short of
meeting its burden with respect to at least two of the Sell fac-
tors. Because the district court clearly erred in finding that the
government proved it met the second and fourth Sell factors
by clear and convincing evidence, either of which failures in
itself would require reversal, we reverse the district court’s
order authorizing involuntary medication.

REVERSED. This court’s emergency stay order is
VACATED.