Warren v. State

In the Supreme Court of Georgia

Decided: October 19, 2015

S15A0795. WARREN v. THE STATE.

NAHMIAS, Justice.

Jesse James Warren has been indicted on four counts of murder and many

additional charges in connection with a mass shooting at a Penske Trucking

Company location in Cobb County on January 12, 2010, in which four victims

were killed and a fifth victim was paralyzed. The State has given notice of its

intent to seek the death penalty. On March 4, 2013, Warren filed a special plea

of mental incompetence to stand trial. See OCGA § 17-7-130 (b) (2). That

same day, the trial court issued an order for Warren to be evaluated by the

Georgia Department of Behavioral Health and Developmental Disabilities. See

OCGA §§ 17-7-129, 17-7-130 (b) (1). On May 9, 2013, after receiving a report

on that evaluation from Dr. Brian Schief and Dr. Don Hughey, which indicated

that Warren was incompetent to stand trial but might benefit from treatment, the

trial court issued an order for Warren to be placed in the custody of the

Department in order to receive further psychological observation, evaluation,
and treatment. See OCGA § 17-7-130 (c). Warren has remained confined in a

state psychiatric hospital since that time. On November 18, 2013, the State filed

a motion requesting the authority to medicate Warren involuntarily in an attempt

to make him mentally competent to stand trial. On June 25, 2014, the trial court

held an evidentiary hearing, and on July 9, 2014, the court filed a short order

granting the State’s motion.1 Warren filed a notice of appeal to this Court.2 For

the reasons set forth below, we vacate the trial court’s order and remand the case

for further proceedings.

1
On August 7, 2014, the trial court filed a related order directing the Department to conduct
further evaluation and reporting in conjunction with the involuntary medication. See OCGA § 17-7-
130 (c) - (f). Warren has not appealed from this order.
2
The criminal case against Warren is still pending in the trial court, and he did not follow
the procedures required for an interlocutory appeal, see OCGA § 5-6-34 (b), but we hold that an
immediate appeal of the trial court’s order is authorized under the collateral order doctrine. Under
that doctrine, an order that does not resolve the entire case in the trial court may be appealed
immediately if it “(1) resolves an issue that is ‘substantially separate’ from the basic issues to be
decided at trial, (2) would result in the loss of an important right if review had to await final
judgment, and (3) completely and conclusively decides the issue on appeal such that nothing in the
underlying action can affect it.” Fulton County v. State, 282 Ga. 570, 571 (651 SE2d 679) (2007)
(citations omitted). The U.S. Supreme Court has held that a pretrial order to involuntarily medicate
a criminal defendant in an effort to render him competent to stand trial is immediately appealable
under the federal collateral order doctrine. See Sell v. United States, 539 U. S. 166, 175-177 (123
SCt 2174, 156 LE2d 197) (2003). We follow that Court’s lead, as we have in most of our previous
decisions on Georgia’s version of the doctrine. See Patterson v. State, 248 Ga. 875, 875-877 (287
SE2d 7) (1982) (looking to federal collateral order case law in developing the doctrine under Georgia
law); Scroggins v. Edmondson, 250 Ga. 430, 432 (297 SE2d 469) (1982) (same). See also Sosniak
v. State, 292 Ga. 35, 40 (734 SE2d 362) (2012) (overruling Georgia collateral order doctrine cases
that had “strayed” from the analogous U.S. Supreme Court precedents). But see Turner v. Giles, 264
Ga. 812, 813-814 (450 SE2d 421) (1994) (declining to follow the federal case law that allows
immediate appeals of pretrial rulings denying qualified immunity defenses in § 1983 actions).

2
 1. Introduction

In Sell v. United States, 539 U. S. 166 (123 SCt 2174, 156 LE2d 197)

(2003), the Supreme Court of the United States established a four-part test for

determining the “rare” instances when it is constitutionally permissible to

involuntarily medicate a mentally ill criminal defendant for the sole purpose of

making him competent to stand trial. Id. at 180. Under that test, the State must

demonstrate the following: (1) important governmental interests are at stake;

(2) involuntary medication will significantly further those governmental

interests; (3) involuntary medication is necessary to further those governmental

interests; and (4) the administration of the drugs to be used is medically

appropriate for the defendant. See id. at 180-181.

This Court has not previously applied the Sell test.3 We now hold, in

3
Sell was decided under the United States Constitution, and we follow it as a matter of
federal law. On appeal, Warren summarily cites three provisions of the Georgia Constitution, only
one of which he raised, also in summary fashion, in the trial court. Because the trial court did not
distinctly rule on the application of any of the provisions of the Georgia Constitution Warren cites
on appeal, we express no opinion on them. See Smith v. Baptiste, 287 Ga. 23, 30 (694 SE2d 83)
(2010) (“Because Appellees did not raise this constitutional issue in the trial court and obtain a
distinct ruling on it from that court, the issue cannot be considered for the first time in this Court.”).
See also Miller v. Deal, 295 Ga. 504, 511 (761 SE2d 274) (2014) (addressing the inadequacy of a
state constitutional argument made in summary fashion on appeal); Hall v. Terrell, 285 Ga. 448, 457
(679 SE2d 17) (2009) (holding under Supreme Court Rule 22 that claims raised on appeal without
“sufficient argument and citation to allow them to be meaningfully addressed” should be deemed
abandoned).

3
agreement with the majority of other courts that have addressed the issue, that

the first part of the test generally presents a legal question and thus should be

reviewed de novo on appeal, while the remaining three parts present primarily

factual questions and thus should be reviewed only for clear error by the trial

court. See, e.g., United States v. Dillon, 738 F3d 284, 291 (D.C. Cir. 2013)

(collecting cases); United States v. Diaz, 630 F3d 1314, 1330-1331 (11th Cir.

2011) (same).4 We also join the prevailing view and hold that the State should

bear the burden of proof on the factual questions involved under the clear and

convincing evidence standard. See, e.g., Dillon, 738 F3d at 291-292 (collecting

cases); Diaz, 630 F3d at 1331 (same). See also Addington v. Texas, 441 U. S.

418, 432-433 (99 SCt 1804, 60 LE2d 323) (1979) (holding that the clear and

convincing evidence standard is the lowest standard that due process permits for

cases involving civil commitment).

4
The points to be considered under the first part of the Sell test are often based on undisputed
facts, such as the maximum length of the defendant’s potential sentence, so the first part is generally
suitable for de novo review. However, to the extent that any circumstances in the case relevant to
the first part of the test necessitate findings of fact, those findings will be reviewed only for clear
error. See Dillon, 738 F3d at 291 (“To the extent that the District Court’s determination under the
first prong of Sell depends on findings of fact, we review those findings under a clear-error
standard.” (citation omitted)); United States v. Evans, 404 F3d 227, 236 (4th Cir. 2005) (“The
district court’s determination that the government’s interest is ‘important’ is a legal conclusion that
we review de novo, although we review any factual findings relevant to this legal determination for
clear error.” (citation omitted)). See also Sell, 539 U.S. at 180 (“Courts . . . must consider the facts
of the individual case in evaluating the Government’s interest in prosecution.”).

4
 In the divisions that follow, we first review the evidence presented to the

trial court and then address each of the four parts of the Sell test in detail.

Finding clear errors in the trial court’s rulings, we conclude with a discussion

of the appropriate remedy in this case.

2. The Evidence Presented to the Trial Court

The evidence presented by the State in support of its motion to have

Warren involuntarily medicated was comprised of testimony from the

psychiatrist and psychologist who had conducted his original competency

evaluation, Dr. Bryan Schief and Dr. Don Hughey. The evidence presented by

Warren included his medical records from before and after the shooting and

testimony from his treating psychiatrist and psychologist, Dr. Francis J. Kane,

Jr., and Dr. Norris Currence, and from a consulting pharmacologist, Dr.

Alexander Morton. According to the medical records and all five experts,

Warren was mentally incompetent to stand trial because of a delusional disorder

and was not currently taking any psychoactive medications. The evidence was

otherwise in conflict on important points, and disorganized and incomplete on

other significant points. Our review of this evidence is rather lengthy, but

necessary for the legal analysis that follows in Division 3.

5
 (a) The State’s Two Expert Witnesses

(1) Dr. Schief

The State’s principal witness was Dr. Schief, a psychiatrist who works at

Georgia Regional Hospital and who had evaluated Warren multiple times and

had reviewed his medical records. Dr. Schief testified generally that

antipsychotic medications will make delusions go away completely in some

people, that other people will have a partial response, that others “don’t

improve,” and that there is no way to determine how a given person will respond

to the medications other than to administer them and observe the response. Dr.

Schief said, however, “Certainly most [people] get better enough to become

competent to stand trial,” adding, “Most of the time folks do very well with

these medications and you don’t have major side effects.”

Turning to Warren’s case, Dr. Schief testified that his delusions, including

his belief that he is an “emperor,” would render him incompetent to stand trial

because he would question the validity of his criminal proceedings and would

be “irrational about his approach to his defense and would [be] unable to assist

his attorney effectively.” When asked if Warren was refusing to be medicated,

Dr. Schief said that medications “haven’t been offered in quite some time, but

6
when they were last offered he refused to take them.” When asked if Warren

would be able to assist his defense counsel if medicated, Dr. Schief replied,

“[I]t’s hard to predict how [each individual is] going to respond, but there’s a

substantial probability he would be able to assist his attorney with medicine,”

adding that the medications “are unlikely to interfere with ability to assist in

court.” Regarding treatments other than medication, Dr. Schief said, “They

might have benefitted him a little bit in terms of organizing his thoughts and

confirming some of the things he needed to know about court, but they did not

come very close to getting him competent to stand trial.” Asked if administering

medication to Warren would be medically appropriate, Dr. Schief answered,

“Yes,” but offered no further explanation. Dr. Schief agreed with Warren’s

treating physician, Dr. Kane, that Warren suffers from a delusional disorder but

disagreed about “A, whether the medicine is likely to help him get a lot better

and get him competent to stand trial, and, B, whether the medicine is likely to

have problematic side effects for him.”

Dr. Schief then discussed one study conducted from 1965 to 1985, which

showed that, when medicated, “more than half of the patients recovered,

meaning were so much better that they were either symptom free or largely

7
symptom free,” that “28 percent achieved partial recovery,” and that “19 percent

did not improve.” He noted that another study published in 2006 showed that

“50 percent recovered, just about, and then another 40 percent showed

significant improvement.”

Dr. Schief next explained that there are three “generation[s]” of

antipsychotic medications. The first carries “a higher rate of movement

disorders over time”; the second carries a smaller risk of movement disorders

but “a little bit of a higher risk of increasing people’s weight, blood sugar and

cholesterol”; and the third is “very low in terms of likelihood of adding to the

risk of increased blood sugar, cholesterol and weight.” He testified that Warren

had taken a second generation drug, Geodon, for one day, but after Warren

suffered a decrease in blood pressure, Dr. Kane changed the medication to a first

generation drug, Haldol.5 Regarding the treatment with Haldol, Dr. Schief

testified:

[Warren] was prescribed that for, I think 20 – 19 days at four
milligrams. He complained of some tiredness, and then it was

5
According to his medical records, Warren took Geodon on May 21 and 22, 2013, and he
took Haldol for about six weeks beginning on May 29, 2013. The records indicate that Warren
“refused” to take the Haldol on three occasions, which were on June 17, July 9, and July 10, 2013.
The records do not show any other refusal to take prescribed medication.

8
 reduced to two milligrams. He took that for about 25 days and he
still complained of tiredness so the doctor stopped and didn’t try
any other medicine until after our second evaluation in late August
[2013], and . . . we suggested that another medicine be tried. Mr.
Warren agreed to us that he would try it, but then when the doctor
tried to prescribe it he refused. So really after that time, the 25-day
trial, he didn’t take anything more.

When asked how long it generally takes for antipsychotic medications to have

an effect on delusions, Dr. Schief said:

An adequate trial really depends on who you ask, but a minimum
probably of four to six weeks at an adequate dose. But probably
more realistically for someone that has had symptoms for as long as
Mr. Warren has and are as severe as Mr. Warren’s are would take
. . . a minimum of eight to twelve weeks for a trial.

As for side effects, Dr. Schief testified that “you can’t ever say I’m going

to ensure your safety,” but in Warren’s hospital setting, the staff would

“maximize [his] safety and make it so that any significantly bad outcome is

unlikely, or very unlikely.” Dr. Schief indicated that they would “monitor

[Warren’s] blood pressure at least daily and monitor his blood sugar regularly,

monitor his weight regularly, ask him about side effects, observe him, look at

his mental status.” At the conclusion of his direct examination by the State, Dr.

Schief said:

9
 If [Warren] was my patient and I was prescribing medicine I
wouldn’t have a problem in the right circumstances, either a court
order or he agreed with prescribing him medicine. Even had he –
even if – I did know that he had some side effects to other – to the
other two medicines, it doesn’t mean that he’s going to have side
effects to a different medication.

On cross-examination, Dr. Schief listed a number of first, second, and

third generation antipsychotic medications. He stated that Risperdal was the last

medication recommended to Warren but that “he refused to try it.” Dr. Schief

said that Warren’s drop in blood pressure when given Geodon was not

“dramatic” but that he would not criticize Dr. Kane’s decision to discontinue

that medication. Regarding the two studies he had cited, Dr. Schief

acknowledged that he was uncertain about what drugs had been involved,

including whether the studies included a drug called Pimozide, which he

admitted might not be available in the United States.

When asked directly if he was proposing any specific medication for

Warren, Dr. Schief answered, “No.” Dr. Schief testified that Warren was 64

years old and had a history of diabetes that was currently “diet controlled” but

had previously been controlled with medication. Asked whether an “atypical”

(third generation) antipsychotic drug could worsen Warren’s diabetes by raising

10
his blood sugar levels, Dr. Schief replied:

I don’t agree with that. I think there’s a few that – I don’t even
know if I would call it substantial, but there is some risk of that with
maybe the worst couple [of drugs]. Other than that, most of them
I wouldn’t call it substantial risk of making it significantly worse.
There’s some risk.

He could not quantify the risk, but he explained that he has “treated plenty of

people with diabetes” and “there are some [medications] that are known to be

neutral in this regard,” including “Latuda, Saphris, Fanapt, Geodon, and Abilify

is on the border.”

Dr. Schief acknowledged that Warren also had a history of hypertension

(high blood pressure), adding:

You have to be cautious. If they’re not contraindicated, and again,
there are some [drugs] that are fairly – are thought to be neutral in
this in terms of blood sugar, which can put people at higher risk of
heart disease, higher cholesterol, which can increase that risk, and
in terms of weight. Some of these medicines are neutral in this
regard. Again, for him I would probably want to avoid some of the
worst offenders, at least at first, in order to choose the ones that are
less likely to cause that problem. But the hospital can monitor very
closely things like his cholesterol, blood sugar, blood pressure and
weight, change the medicine, change the dose if something were to
happen. But even for people with heart disease and a history of
diabetes – or a history of heart disease, a history of diabetes, a
history of hypertension, these medicines aren’t contraindicated or
thought to be dangerous. They’re used all the time in this setting
with folks like that.

11
Without specifying which drugs he was referencing, Dr. Schief stated that some

medications are “worse offenders” that he would avoid but that “as a general

class, no, there isn’t a contraindication with those.”

Dr. Schief also agreed that someone like Warren who has suffered from

delusions for a long time is less likely to respond to medication, but he added,

“I don’t have statistics, but in my experience a pretty high percentage . . . get

better.” He continued:

[Y]ou can improve [Warren’s] behavior in the sense of working on
things he’s not working on like learning about his illness, engaging
in a relapse prevention plan so that he can think about how this
illness has affected his life and how he could some day reintegrate
into the community perhaps, or even into the smaller community of
the hospital in a way with a little more freedom, figuring out how
he can relate a little better to others that may be inside and outside
the hospital if he doesn’t have this belief and these sorts of things.
So in that sense, in a broader life functioning sense I think he could
do better. But in terms of his day-to-day interactions, you know,
we’ve heard he isn’t presenting any kind of problems, so that’s –
there’s really not much room for improvement in terms of
presenting a problem, to the hospital staff.

Regarding how medication would affect the risk that Warren presents to the

hospital staff, Dr. Schief testified:

He would go from a small – a low risk, which he is now, to I
believe an even lower risk, because there would be less likely some

12
 chance that he could incorporate other people in the hospital into his
delusional system, which he’s obviously done in other places in his
life.

Dr. Schief said that he was uncertain whether medicating Warren would allow

him to reintegrate into society, but it would likely at least improve his ability to

communicate with and relate to others.

(2) Dr. Hughey

The State next presented testimony from Dr. Hughey, a psychologist at

Georgia Regional Hospital who had evaluated Warren multiple times and also

had reviewed his medical records. Dr. Hughey agreed with Dr. Schief that

Warren has a delusional disorder and that he is incompetent to stand trial

because, although he understood the nature of the proceedings, “he had certain

beliefs related to those charges that would interfere with his rationally assessing

his case” and would cause him difficulty in working with his counsel and

making decisions in his own best interest. Dr. Hughey testified that delusions

are usually treated with a combination of medication and “talk therapy” and that,

“[b]ased on the published literature, between 50 and 77 percent [of patients] can

have a reduction in their symptoms with treatment.” Dr. Hughey noted that

treating Warren at the hospital would allow for “frequent lab work and

13
monitoring daily by nursing staff.” Dr. Hughey opined that “with treatment

[Warren] could be restored to competency,” but said “[h]e is refusing

medications.” Regarding whether medication would interfere with Warren’s

ability to assist his counsel, Dr. Hughey answered generically, “Many people are

on medication and still proceeded to their trials.” He concluded his direct

testimony by stating, “For the past year at Central State [Hospital, Warren] has

received education on legal issues, he’s received individual and group therapies.

These have not affected his fixed beliefs. They are intact as they were when we

first saw him.”

On cross-examination, Dr. Hughey admitted that the 50 to 77 percent

improvement rate that he had referenced was from a general review of literature

regarding studies that included patients with problems other than delusions, that

longstanding delusions are more difficult to treat, and that Warren has suffered

delusions for at least 10 years. When asked if he was proposing a specific

treatment plan, Dr. Hughey replied, “[A]s a psychologist I do not prescribe

medications. That would be up to the treating physician at Central State.”

Regarding the decision by Warren’s treating physician, Dr. Kane, to lower the

dosage of Haldol and then to discontinue the medication completely, Dr.

14
Hughey said, “Well, [Warren] was complaining of side effects. Many times

people complain of side effects when they actually do not want to take the

medication.” Dr. Hughey acknowledged, however, that Dr. Kane had made a

medical decision to discontinue the Haldol. Dr. Hughey also acknowledged that

the earlier treatment with Geodon had caused a “dramatic” drop in Warren’s

blood pressure.

Dr. Hughey cited what appeared to be one particular study in which 17 of

either 22 or 32 patients with a delusional disorder “were restored to competency

with treatment” with a first generation antipsychotic medication. Dr. Hughey

agreed that there are “several side effects that can come with medications.

Movement disorders, called extrapyramidal symptoms would be one, they can

affect metabolic processes such as diabetes, weight, and can affect blood

pressure, attention, concentration, gait.” He also noted that age can be a risk

factor, depending on a patient’s health. Regarding the specific risks to Warren

based on his diabetes, hypertension, and possible history of a heart attack, Dr.

Hughey said, “These would be risk factors. You would need to ask a physician

for specific details of those, since as a psychologist I don’t prescribe

medication.”

15
 (b) Warren’s Three Expert Witnesses

(1) Dr. Kane

After the State rested its case, the defense presented testimony from Dr.

Kane, the psychiatrist who had served as Warren’s treating physician at Central

State Hospital since Warren was transferred there shortly after the trial court’s

May 9, 2013 order. Dr. Kane confirmed that Warren had been diagnosed as

having a delusional disorder from which he had suffered since at least 2005; the

delusions involve his beliefs about having made discoveries worth millions of

dollars that were taken from him or attempted to be taken from him by various

people. Dr. Kane testified that such longstanding delusions are more difficult

to treat; that delusions from a delusional disorder are more difficult to treat than

delusions from schizophrenia or bipolar disorder; that patients with a delusional

disorder seldom obtain complete relief from their delusions; and that he knew

of no one with a fixed delusion who had been cured, although some persons

with delusional disorders have improved to the point of being competent to

stand trial. Dr. Kane said that delusional patients may benefit from treatment

with medication even if they may not lose their psychotic symptoms, because

“they will become much less anxious and nervous” and “they may even be able

16
to work and live with their family and enjoy amenities of social life without

being unduly disturbed.”

Dr. Kane then discussed the “many risks” associated with antipsychotic

medications, including “tardive dyskinesia where [patients] receive permanent

neurological symptoms” such as contorted lips and bodies. When asked

whether treating Warren with antipsychotic medication would be medically

appropriate, Dr. Kane said:

Well, we tried medicine on two occasions with him, and he – on one
occasion [with Geodon] he – his blood pressure dropped, and of
course he has a history of – his EKG shows some suggestion he
may have had a previous heart attack, so we couldn’t continue that
medicine and have his blood pressure drop. We tried him on
another medicine [Haldol,] which was not risky for his diabetes, but
he became sedated on that. And of course we didn’t want him
falling down because he was drowsy, which sometimes happens.
So after I tried those two and since he was no clinical problem on
my unit, he was very well behaved, never mentioned his delusional
ideas to other people, unless we asked him, he was not a behavior
problem at all. Very cooperative, pleasant. I saw no urgency in
treating him for that reason.

Dr. Kane added that Warren had never been disruptive or injurious to anyone

else in the hospital and had always been compliant.

Asked again about Warren’s previous treatment with medications, Dr.

Kane explained that Warren had been given two drugs:

17
 Geodon is less likely to complicate the diabetes, but when he got
low blood pressure we couldn’t go there. And Haldol also does not
get in the way of his diabetes, but he became sedated and we
became concerned about him falling, so we stopped that.

Dr. Kane reiterated that the Haldol was discontinued based on a medical

decision, explaining:

It became clear to me that [Warren] was not tolerating treatment
with pharmac[ological] therapy very well, and since he was not a
big time behavior problem I felt – you know, I tried a couple of
times and, you know, it just – it was risky, and so I – that deterred
me from further experimentation.

Dr. Kane testified that he was “very doubtful” that antipsychotic

medication would have a positive effect on Warren because “the track record

and the scientific literature is such that it isn’t – they [patients with a delusional

disorder] are very difficult to treat.” When asked if he believed that Warren’s

competence to stand trial could be restored, Dr. Kane answered, “[T]hat

certainly hasn’t happened – occurred up till now.” As for the risks Warren faces

from medication due to his diabetes, Dr. Kane said:

There’s a certain percentage of patients who suffer for – the longer
the patient is on medicine the more likely it is that he will suffer
undesirable side effects. How serious those are, you know, we
seem to be getting better drugs in some regard for certain problems
than others, and so we keep hoping for that.

18
Dr. Kane noted that one drug, Pimozide, “worked pretty well [with fewer side

effects] but is unavailable in this country and the FDA would not approve it.”

He added that older people face a greater risk of suffering neurological side

effects from antipsychotic medications, and regarding Warren, “we tried and

he’s sensitive, and so that makes him not easy to treat.” When asked if Warren

had ever refused to take any medication, Dr. Kane replied, “He told me he didn’t

like them, he told me the side effects he was having, some of which – his blood

pressure dropped. No, he has never refused.”

On cross-examination, however, Dr. Kane said that Warren had refused

medication because he was suffering side effects and because he did not believe

that he was psychotic. When asked about the use of the alternative medications

Saphris, Latuda, and Fanapt, Dr. Kane stated that they were “me-too” drugs that

were “not significantly different than, say, Geodon.” He agreed that different

drugs could create different side effects, that it was impossible to know whether

they would improve Warren’s symptoms without actually trying them, that

Warren might “possibly” be restored to competence if given medication, and

that it could be “worth a try.” When asked if he was categorically against giving

Warren medication, Dr. Kane answered, “No,” but he added, “I am worried

19
about it, but I – I don’t want to go to jail,” apparently referring to the potential

for being jailed if he refused to comply with a court order to medicate Warren.

Dr. Kane reiterated that his concerns included the fact that Warren had proven

“sensitive” to two drugs that are known to have a minimal effect on diabetes, but

he agreed that other drugs could be tried. He also agreed that treating patients

in Warren’s condition with drugs was “common” and that such treatment could

benefit Warren by making him competent. Dr. Kane noted that some

medications could be used to treat Warren only if he were willing to cooperate,

because the drugs are taken orally.

On redirect examination, Dr. Kane was asked again whether Warren had

ever refused medication, and explained:

[I]t was always a persuasion for Mr. Warren to take medicine. He
did not think he needed it, and I told him, you know, what I tell all
patients, that the judge has sent them there for evaluation and
treatment and I’m obligated to treat him. And he tried the medicine
and he – you know, he never liked it and he didn’t think he needed
it. And that was always – you know, I didn’t write about that [in
Warren’s medical record] because that was understood between he
and I. So – but he did take the medicine, and when he started
complaining of sedation and my staff became concerned about this
then we decided that was another strikeout and that’s when I
decided to sort of sit on things.

When asked if third generation antipsychotic medications were significantly

20
different than the previous generations, Dr. Kane answered, “The drug

companies would like you to think so, but they aren’t.”

(2) Dr. Currence

Warren next presented the testimony of Dr. Currence, a psychologist at

Central State Hospital who had treated Warren there. Dr. Currence testified that

Warren’s behavior at the hospital “has been compliant and appropriate” and that

he has not required “intensive clinical treatment” because he has been

“behaviorally stable and emotionally stable.” Dr. Currence confirmed that

Warren had been diagnosed as having a delusional disorder, explaining that he

“currently has a fixed delusion involving his belief that he has been the victim

of a robbery, millions of dollars have been stolen from him, and [it] is kind of

a grandiose and paranoid type of delusional belief system.” Dr. Currence noted

that Warren’s “fixed delusions” have existed for a long time, that delusional

disorder is more difficult to treat than delusions related to schizophrenia or

bipolar disorder, and that Warren’s delusions were not reduced when he was

taking medications. Dr. Currence testified that Warren’s delusions could affect

his ability to assist his counsel and to understand the implications of testifying.

Dr. Currence acknowledged that he had no specific education or training on the

21
effect of medications and was not authorized to prescribe medications. When

asked if medication could improve fixed delusions, he replied, “Once again, I’d

rather not be specific regarding Mr. Warren. I can say that I have worked with

individuals who had fixed delusions in the past and . . . those delusions were

highly resistant to medical interventions.”

On cross-examination, Dr. Currence admitted that he also knew persons

with fixed delusions who showed improvement to the extent that they would be

able to assist their counsel in their own defense. Dr. Currence also testified that

“if [Warren] was absolutely free without any type of supervision, I would

definitely have some concerns about his ability to care for himself or his risk to

the public.”

On redirect examination, Dr. Currence noted that Warren “has never given

us any reason to believe that he would be oppositional or noncompliant with any

reasonable demand of the hospital.” Dr. Currence added that Warren had

indicated that he believed that “retaliating in a very violent way” would be

appropriate if someone were the victim of a theft, which occurs often at the

hospital, but that “he hasn’t demonstrated any type of physical indicators that

he would be aggressive towards anybody” there. Dr. Currence explained, “He

22
has a very high level of coping skills. So he’s demonstrated behavioral control

despite the delusional belief.” Nevertheless, Dr. Currence said that Warren’s

ability to assist his counsel “would be funneled through that filter of delusion.”

Dr. Currence explained that he had seen “talk therapy” offer some success in

changing the behavior of delusional persons, but he had seen little success

through treatment with medication.

(3) Dr. Morton

Finally, Warren presented the testimony of Dr. Morton, a pharmacologist

with a consulting business who had never treated Warren but had reviewed his

medical records. Dr. Morton explained that older antipsychotic drugs were less

effective and “had a number of side effects”; that newer “atypical”

antipsychotics have been touted as being more effective and as having fewer

side effects, but “the clinical trials have questioned that”; and that “both typical

and atypical [drugs] have been relatively ineffective in treating fixed delusions,”

particularly persecutorial delusions and longstanding delusions. Dr. Morton

advised that the side effects can include tiredness and apathy that can mask

rather than cure symptoms, metabolic and endocrine problems, abnormal muscle

tone and restlessness, symptoms similar to Parkinson’s disease, tardive

23
dyskinesia producing involuntary muscle movement in the face, changes to

blood pressure and heart rate, dizziness, gastrointestinal problems, affected

vision, problems with urination, problems with concentration, and neuroleptic

malignant syndrome where body temperature rises out of control. Dr. Morton

added that side effects cannot be stopped quickly, because a dose of medication

wears off slowly, that neuroleptic malignant syndrome occurs in “one in a

thousand to one in a hundred” cases, and that tardive dyskinesia occurs in 20 to

40 percent of cases but at a lower rate with atypical antipsychotic medications.

Dr. Morton testified that Warren’s past reactions to medications indicated

that his physician should “stay away from” Zyprexa and Geodon; that a 64-year-

old would be at greater risk of side effects and a diabetic could have his blood

sugar get out of control, especially with certain drugs such as Olanzapine and

Zyprexa; and that other drugs, such as Asenapine and Latuda, might affect blood

sugar less but had not been studied in that regard. Asked about the use of

antipsychotic medications for someone with a history of diabetes, hypertension,

and a possible past heart attack, Dr. Morton replied, “It would depend on the

benefits over the risks. I don’t think they are contraindicated. They would have

to be used with a lot of caution, because . . . you would have to decide is the

24
diabetes control of that going to outweigh the potential risk of psychiatric

symptoms.” He explained that the treating physician usually makes this

decision. He added that typical antipsychotic medications will almost always

raise the blood sugar of patients with diabetes and that a diabetic, like Warren,

who previously had his disease managed through diet might need to go on

medication for high blood sugar.

Dr. Morton said that a dystonic reaction with symptoms like a seizure

occurs in 10 to 15 percent of persons medicated; that the incidence of other side

effects such as Parkinsonism, akathisia, and akinesia vary among the various

antipsychotic medications; and that side effects affecting muscle control would

be more dangerous for Warren because his dentures and difficulty swallowing

would put him at risk of choking. Dr. Morton noted that Mallaril and Thorazine

have lower rates of dystonic reactions but are more likely to induce sleepiness

and lowered blood pressure. He stated that Geodon had lowered Warren’s blood

pressure so dramatically that he was dizzy and not able to stand, and this drop

in blood pressure would have created an increased heart rate and a resulting

increased risk for a heart attack. Dr. Morton also noted that Warren had a

history of falling, partly due to vision problems, and that Warren had elevated

25
cholesterol and excess body weight that put him at an increased risk for strokes

and heart attacks.

Dr. Morton explained that additional side effects of antipsychotic

medications can include enlarged breasts and/or the expression of milk in male

patients, a dry mouth, some sedation that can sometimes lead to confusion,

difficulty focusing the eyes, and decreased bowel and bladder function. Dr.

Morton said that Haldol had not caused a decrease in blood pressure in Warren

and therefore “would be something to consider” for any future treatment,

adding, “Haldol, you know he’s tolerated it, and that might be something you

would want to try at a higher dose for 12 weeks, because he’s only been on it for

six weeks.” Dr. Morton noted, however, that Haldol had a sedating effect on

Warren that made him sleepy. Finally, Dr. Morton asserted that newer drugs

such as Fanapt, Saphris, and Latuda have not “dramatically been better than

anything else that we use,” although they cost more.

On cross-examination, Dr. Morton acknowledged that “[t]here is a benefit

of [antipsychotic] drugs,” but noted that the benefit must be balanced against

other concerns. He insisted that Warren’s situation “is loaded with many more

complications than other people” and that “[i]f he has neuroleptic malignant

26
syndrome he’ll probably die.” Dr. Morton also testified that he considered

Warren’s risk of falling or choking to be unacceptable. Dr. Morton stated,

“Maybe his doctor needed to give him Haldol longer,” but “[i]f we give

him . . . any of these medicines on this list we’re putting him at grave risk.”

When asked if Warren could be made competent to stand trial, Dr. Morton

answered, “I don’t think so. I think there is a possibility. There’s a possibility.

I can’t say that those medicines will absolutely not work, but there’s a very low

likelihood that they will work.”

On redirect examination, Dr. Morton explained that “[a] treatment plan is

usually developed by a psychiatrist . . . in conjunction with other treatment

professionals there, looking at the benefits and risks,” and such a plan is

committed to writing. He noted that Warren had no written treatment plan and

added that if there were one, including one that might include a renewed use of

Haldol, he would be able to review it and comment on it. Finally, Dr. Morton

opined that, because Warren has special medical risks and does not appear to

present a danger to the hospital staff, the only possible benefit to Warren from

an antipsychotic medication would be a reduction in delusional thoughts that

might allow him to work with his defense counsel.

27
 (c) The Trial Court’s Rulings

In closing argument at the evidentiary hearing, Warren’s counsel asserted

that the absence of a specific treatment plan for Warren made it impossible for

the State to satisfy the Sell test. Counsel noted that he had asked the State

months earlier to provide a treatment plan so that the defense could have it

reviewed by their experts, but was told that the State was going to rely on the

treating physician to determine how to effectuate any court order for involuntary

medication. In the State’s closing argument, the prosecutor asserted, “Quite

frankly we can’t fashion a treatment plan until this Court first gives us the

ability to force treatment on the defendant so that we can get to the restoration

process.” The prosecutor added,

By requiring some sort of treatment plan in advance of forced
medications in this case you’re essentially tying the hands of the
State and you’re tying the hands of his clinicians, his doctors, from
being adequately able to treat the mental disorder such that we can
restore competency and also account for his unique medical
condition.

The trial court then announced:

I find by clear and convincing evidence there is an important
governmental interest to bring this case to trial, that involuntary
administration of medication will substantially further such
governmental interest. I find that it is necessary by clear and

28
 convincing evidence. And the testimony of the witnesses supports
a finding by this Court, and I do so find that less intrusive means are
unlikely to achieve substantially the same result. I further find that
the administration of the drug – of drugs is medically appropriate
in the case. I am not going to grant the [defense] request for a
treatment program in advance or further proceedings, and [the
prosecutor] can [draft] an order to that effect. . . .

Eight days later, on July 3, 2014, the trial court signed an order presented

by the State granting the State’s motion to medicate Warren involuntarily. The

order was filed on July 9. The substance of the order was less than a page long.

3. Application of the Sell Test

We turn now to reviewing the trial court’s application of the four-part test

enunciated in Sell v. United States, 539 U. S. 166, 180-181 (123 SCt 2174, 156

LE2d 197) (2003), for identifying the “rare” instances in which the Constitution

permits involuntary administration of antipsychotic drugs to be ordered solely

for trial competence purposes.

(a) Governmental Interest

Sell first directs courts considering a motion for an order authorizing the

involuntary medication of a criminal defendant for the purpose of making him

competent to determine whether “important governmental interests are at stake.”

Id. at 180 (emphasis in original). Here, the trial court’s written order says only,

29
“[t]he State has important governmental interests in prosecuting this criminal

Defendant,” without identifying what those interests are, and the court’s oral

pronouncement said only, “there is an important governmental interest to bring

this case to trial.” However, because an appellate court reviews this part of the

Sell test de novo, as we explained in Division 1 above, we can review the record

and determine for ourselves if this part of the test is satisfied.

The Sell Court explained that “[t]he Government’s interest in bringing to

trial an individual accused of a serious crime is important. That is so whether

the offense is a serious crime against the person or a serious crime against

property.” 539 U. S. at 180. In addition “[t]he Government has a substantial

interest in timely prosecution” and “a concomitant, constitutionally essential

interest in assuring that the defendant’s trial is a fair one.” Id. Warren is

accused, among other crimes, of shooting five people, killing four of them and

paralyzing the fifth. These are crimes against persons of the most serious

magnitude. Moreover, nearly six years have already passed since the alleged

crimes, and even if the administration of medication ultimately succeeded in

making Warren competent to stand trial, his trial would not begin for quite some

time, particularly given the extensive pretrial proceedings associated with a

30
death penalty proceeding. Thus, if this exceptionally serious case is to be tried,

and tried fairly, the State has compelling interests in doing so with as little

further delay as possible.

These important governmental interests are not significantly undermined

by any “[s]pecial circumstances” present in this case. Id. Special circumstances

may include the amount of time that a defendant who refuses to take

psychoactive drugs voluntarily is likely to remain confined in a mental health

institution if he is not tried, because a “lengthy confinement in an institution for

the mentally ill . . . would diminish the risks that ordinarily attach to freeing

without punishment one who has committed a serious crime.” Id. Georgia law

allows for civil confinement where a person accused of a violent crime is found

mentally incompetent to stand trial, which may continue in annual increments

for up to “the maximum period for which the accused could have been

sentenced on the most serious violent offense charged.” OCGA § 17-7-130 (e)

(2) (B) (iv).

However, civil commitment is not “a substitute for a criminal trial,”

particularly where “it may be difficult or impossible to try a defendant who

regains competence after years of commitment during which memories may fade

31
and evidence may be lost.” Sell, 539 U. S. at 180. Thus, “[t]he potential for

future confinement affects, but does not totally undermine, the strength of the

need for prosecution.” Id. Here, the State is prosecuting charges that, if proved

at trial, would at a minimum result in Warren’s receiving multiple life sentences

and serving a minimum of 30 years in prison. See OCGA §§ 16-5-1 (e) (1), 17-

10-6.1 (c) (1). See also United States v. Breedlove, 756 F3d 1036, 1041 (7th

Cir. 2014).

Another “special circumstance” identified in Sell is whether the defendant

has “already been confined for a significant amount of time” for which he would

receive credit toward any sentence ultimately imposed. 539 U. S. at 180.

Although Warren has been detained almost six years already, that time again

pales in comparison to the life sentences he faces if convicted. This case also

involves a circumstance, beyond any period of confinement, that weighs in favor

of the State’s interest in bringing Warren to trial – the State’s valid interest in

seeking Warren’s execution upon proof that he committed multiple capital

crimes.

Thus, although the trial court’s findings as to the first step of the Sell test

were incomplete, the court did not err in its conclusion that the State

32
demonstrated important governmental interests in rendering Warren competent

to stand trial, and those interests are not offset by any special circumstances of

the case.6

(b) Furtherance of Governmental Interest

The second part of the Sell test requires the trial court to determine that

“involuntary medication will significantly further” the governmental interests

in bringing the defendant to trial. 539 U. S. at 181 (emphasis in original). This

inquiry has two components. The court must find that “administration of the

drugs is substantially likely to render the defendant competent to stand trial”

and, at the same time, 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.

On these points, the trial court’s order in this case said only the following:

Involuntary medication will substantially further those important
governmental interests. The administration of drugs is substantially

6
Given the charges and potential sentences involved in this case, we need not delve into the
nuances that have divided other courts in applying the first part of the Sell test. See, e.g., United
States v. Hernandez-Vasquez, 513 F3d 908, 918-919 (9th Cir. 2007) (amended opinion) (discussing
the federal circuit split as to whether the statutory maximum sentence or the recommended federal
sentencing guidelines should be used as the basis for comparisons).

33
 likely to render Defendant competent to stand trial and is
substantially unlikely to have side effects that will interfere
significantly with Defendant’s ability to assist counsel in
conducting a trial defense.

Although this passage recites the relevant language from Sell, the trial court’s

written findings, which were not supplemented by any additional oral findings,

are insufficient to withstand review.

The fundamental problem with the trial court’s ruling on this part of the

Sell test – and with its rulings on the third and fourth parts of the test, as

discussed below – is that the court has not specified what antipsychotic

medication or medications may be forcibly administered to Warren, in what

dosage or range of dosages, for what period of time, and with what oversight by

the court. This is a problem created by the State – which drafted the order the

trial court signed – because the State has not presented a specific treatment plan

for Warren supported by expert testimony, nor can any particular treatment plan

be readily discerned from the evidence offered at the hearing. And this is not

a problem that has been identified for the first time on appeal. At the hearing,

Warren’s counsel asked the trial court to require the State to provide a specific

treatment plan, but the State objected, and the court denied the request.

34
 As recounted in Division 2 above – at considerable length, to give a sense

of the disorganized and incomplete way the evidence was presented – there was

agreement among the State’s two experts and Warren’s three experts that

antipsychotic medications offer some likelihood of making a delusional person

competent to stand trial. But there was no consensus, and the certainty of the

opinions offered by the State’s experts dissipated, when the questions focused

on the likelihood of success using the various “generations” and particular

names of drugs and when the discussion moved from delusional patients in

general to persons like Warren who suffer from a long-term delusional disorder

and fixed delusions. For example, the State’s principal expert, Dr. Schief, cited

only two scientific studies, and he did not know what drugs were used in the

studies and did not say whether the studies involved patients with a delusional

disorder like Warren’s. Likewise, the State’s other expert, Dr. Hughey,

mentioned his general review of scientific studies, but he admitted that they

included patients with different and easier-to-treat conditions than Warren’s, and

he did not say what drug or drugs had been used. Dr. Hughey also referred to

what was apparently one particular study using a first-generation drug, but he

was not sure if the success rate was 77 percent of patients (17 of 22) or just 53

35
percent (17 of 32) – a significant discrepancy. No written reports on these

studies were offered into evidence.

Similarly, there was agreement among the experts that antipsychotic

medications, and certain drugs in particular, have numerous side effects that can

be severe, especially for an older person like Warren who has diabetes,

hypertension, and a possible history of a heart attack. But the experts offered

different predictions about whether different medications would have particular

side effects for Warren – and they did so with little focus on side effects that

would interfere with his ability to assist his counsel at trial, rather than affect his

well-being in other ways (which is a consideration in the fourth part of the Sell

test). See Sell, 539 U. S. at 185 (noting the absence of expert testimony “about

trial-related side effects and risks” of the proposed medications). For example,

Dr. Schief testified that the three generations of antipsychotic medications carry

different risks of various side effects, acknowledged that Warren had shown

significant side effects when given Geodon (a second generation drug) and

Haldol (a first generation drug), and said that he did not criticize Dr. Kane’s

decision to discontinue the Geodon after only one day when Warren’s blood

pressure dropped. However, while opining that there was a “substantial

36
probability” that Warren would be rendered competent if medicated, Dr. Schief

never said what drugs should and would be used to do that; to the contrary, Dr.

Schief made it clear that he was not proposing any specific medication. His

testimony, and the court’s order, would not preclude giving Geodon to Warren

again – this time forcibly.

As the factfinder in this case, the trial court was entitled to sift through the

evidence and give what credit and weight it deemed appropriate to what was

often qualified, uncertain, and conflicting testimony (although the court made

no specific oral or written findings in this regard). And the trial court indicated

(orally) that it was properly considering the record using the clear and

convincing standard of proof and placing the burden of proof on the State. See

Division 1 above. Viewed in this way, the record might support findings that

one or several types of antipsychotic medication would be substantially likely

to render Warren competent to stand trial while also being substantially unlikely

to have side effects that would interfere significantly with his ability to assist his

counsel. However, the record does not support a finding that the involuntary

administration of any of the many medications discussed in the expert

testimony, in any dosages and for any periods of time, meets this standard, and

37
yet that was the scope of the finding made in the trial court’s order. See United

States v. Chavez, 734 F3d 1247, 1252 (10th Cir. 2013) (“Because different types

of antipsychotic drugs can produce different side effects and result in different

degrees of success, granting the government such unfettered discretion in

determining which drugs will be administered to a defendant does not conform

with the findings required by Sell.”). See also Sell, 539 U. S at 185 (explaining

that “[w]hether a particular drug will tend to sedate a defendant, interfere with

communication with counsel, prevent rapid reaction to trial developments, or

diminish the ability to express emotions are matters important in determining the

permissibility of medication to restore competence” (emphasis added)).

“[A]n individual has a significant constitutionally protected liberty interest

in avoiding the unwanted administration of antipsychotic drugs.” Sell, 539 U. S.

at 178 (citation and quotation marks omitted).

The interference is “particularly severe” when, as in this case, the
medication in question is an antipsychotic, for the use of such
medications threatens an individual’s “mental, as well as physical,
integrity.” On the physical side, there is the “violence inherent in
forcible medication,” compounded when it comes to antipsychotics
by the possibility of “serious, even fatal, side effects.” But it is the
invasion into a person’s mental state that truly distinguishes
antipsychotics, a class of medications expressly intended “to alter
the will and the mind of the subject.”

38
United States v. Watson, 793 F3d 416, 419 (4th Cir. 2015) (citations omitted).

Given “the severity of the intrusion and corresponding importance of the

constitutional issue,” Sell, 539 U. S. at 177, the judicial findings required to

authorize such an intrusion by the State must be made with care and

thoroughness, and with sufficient detail to allow meaningful review on appeal.

See Chavez, 734 F3d at 1252 (“While Sell does not explicitly identify what

level of specificity is required in a court’s order for involuntary medication . . . ,

the need for a high level of detail is plainly contemplated by the comprehensive

findings Sell requires.”).7

The State contends that it is impossible to establish a specific treatment

plan for Warren, because the only way to know how any given patient will

respond to antipsychotic medication is to administer various drugs and see what

happens through “trial and error.” But Sell does not require certainty as to

7
Cf. Higgenbottom v. State, 288 Ga. 429, 430-431 (704 SE2d 786) (2011) (explaining that
to allow adequate appellate review in cases implicating a defendant’s constitutional right to a speedy
trial, it is imperative that the trial court enter findings of fact and conclusions of law consistent with
the four-part test set forth in Barker v. Wingo, 407 U. S. 514 (92 SCt 2182, 33 LE2d 101) (1972));
Wang v. Liu, 292 Ga. 568, 570-571 (740 SE2d 136) (2013) (explaining, in the non-constitutional
context of the forum non conveniens doctrine, that “[w]hat is required to permit meaningful
appellate review is that the trial court set out upon the record the essential reasoning that forms the
basis for its exercise of discretion. . . . Without such a statement of the essential reasoning of the
trial court, we frequently cannot ascertain whether the decision of the trial court was a reasoned and
reasonable one in the light of the standard set out.”).

39
whether medication will make a defendant competent to stand trial or as to the

side effects, only factually supported predictions as to what results are

“substantially likely” from the treatment regime proposed by the State. And Sell

did not condone – nor will this Court allow – trial courts to cede oversight of

such a significant constitutional matter to the State, allowing its doctors to force

a mentally ill criminal defendant to take whatever medications in whatever

dosages and for whatever period of time they consider appropriate.

We would hope that the State’s physicians, as healthcare professionals,

would not misuse such unfettered authority, but history teaches that involuntary

medical treatment, especially of the poor, the outcast, and the incarcerated, is

worthy of close and independent oversight. See Watson, 793 F3d at 419

(emphasizing that “forcible medication under Sell is ‘a tool that must not be

casually deployed,’ and courts must be vigilant to ensure that such orders, which

‘carry an unsavory pedigree,’ do not become ‘routine’” (citation omitted)).

To approve of a treatment plan without knowing the proposed
medication and dose range would give prison medical staff carte
blanche to experiment with what might even be dangerous drugs or
dangerously high dosages of otherwise safe drugs and would not
give defense counsel and experts a meaningful ability to challenge
the propriety of the proposed treatment.

40
United States v. Evans, 404 F3d 227, 241 (4th Cir. 2005).

Sell indicates that its test should be applied in the context of a specific

proposed treatment plan for a specific defendant. See 539 U. S. at 183 (stating

that the ultimate question for the trial court is, “Has the Government, in light of

the efficacy, the side effects, the possible alternatives, and the medical

appropriateness of a particular course of antipsychotic drug treatment, shown

a need for that treatment sufficiently important to overcome the individual’s

protected interest in refusing it?” (emphasis added; citations omitted)).

Accordingly, as many other courts have held, we now hold that the Sell test can

be properly applied only in relation to an individualized treatment plan that

specifies, at a minimum, (1) the drug or drugs the treating physicians are

permitted to use on the defendant, (2) the maximum dosages that may be

administered, and (3) the duration the drugs may be used before the physicians

report back to the court.8

8
See, e.g., Breedlove, 756 F3d at 1043-1044 (“To satisfy its duty [to approve an
individualized treatment plan], the district court must indicate the medication or range of
medications to be administered, the dose range and the length of treatment,” which may be done by
“reference to the government’s detailed treatment plan”); Chavez, 734 F3d at 1253 (“[W]e hold that
an order to involuntarily medicate a non-dangerous defendant solely in order to render him
competent to stand trial must specify which medications might be administered and their maximum
dosages. Without this information, a court cannot ensure that the ‘administration of the drugs is
substantially unlikely to have side effects that will interfere significantly with the defendant’s ability

41
 The presentation and approval of such specific and individualized

treatment plans in numerous other cases demonstrates that what the State

contends is impossible is, in fact, not even impractical.9 This is particularly so

because, like the courts whose holdings we are joining,
we are mindful of the balance we must strike between the judicial
oversight necessary to protect defendants’ constitutional rights and
the need of prison medical staff to retain a degree of flexibility in
order to provide effective treatment. Therefore, so long as all drugs

to assist counsel in conducting a trial defense,’ as required by the second prong of Sell. Similarly,
without knowing which drugs the government might administer and at what range of doses, a court
cannot properly conclude that such a vague treatment plan is ‘medically appropriate, i.e., in the
patient’s best medical interest’ as the fourth part of Sell demands” (citations omitted)); Hernandez-
Vasquez, 513 F3d at 916-917 (“[W]e hold that a Sell order must provide at least some limitations
on the medications that may be administered and the maximum dosages and duration of treatment.
At a minimum, to pass muster under Sell, the district court’s order must identify: (1) the specific
medication or range of medications that the treating physicians are permitted to use in their treatment
of the defendant, (2) the maximum dosages that may be administered, and (3) the duration of time
that involuntary treatment of the defendant may continue before the treating physicians are required
to report back to the court on the defendant’s mental condition and progress.”); Evans, 404 F3d at
241 (“[F]or the district court even to assess whether involuntary medication is constitutionally
permissible under Sell’s second and fourth factors, the government must set forth the particular
medication, including the dose range, it proposes to administer to [the defendant] to restore his
competency.”). See also United States v. Green, 532 F3d 538, 556-557 (6th Cir. 2008) (approving
an order directing involuntary medication that did not itself specify a treatment plan for the defendant
but referenced the plan proposed by the State’s expert during the Sell hearing).
9
See, e.g., United States v. Ruark, 611 Fed. Appx. 591, 599 (11th Cir. 2015) (per curiam)
(“Here, the government has presented an individualized treatment plan that details the drugs to be
used and the relevant dosage ranges. . . . Dr. Sarrazin, further, proposed a detailed treatment plan
describing the procedure to be followed if a court orders Ruark to be involuntarily medicated.”);
Diaz, 630 F3d at 1324-1325, 1326-1327, 1330 (affirming a detailed trial court order that approved
the government’s very detailed treatment plan for Diaz, which was in turn supported by detailed and
specific expert testimony); Green, 532 F3d at 555-557 (affirming the trial court’s approval of the
government’s proposed treatment plan, individualized to the defendant, which “set[] forth the
specific medications, alternative means of injecting it, the specific dosage, and the potential side
effects [the defendant] could face”).

42
 that might be administered to a defendant and their maximum
dosages are specified, courts may properly approve treatment plans
identifying a range of medications that could be used if the first
drug or drugs administered prove unsatisfactory. We also note that
either the government or the defendant may move to revise the
court’s Sell order if circumstances change during a defendant’s
treatment.

Chavez, 734 F3d at 1254 (citations omitted).10

For these reasons, the trial court’s ruling with respect to the second part

of the Sell test was plainly insufficient. That alone is enough to vacate the trial

court’s order, but because the court may be called on to apply the Sell test again

on remand, we will proceed to discuss the remaining two parts of the test.

(c) Necessity

The third part of the Sell test requires the trial court to conclude that

“involuntary medication is necessary to further” the governmental interests in

proceeding with the defendant’s prosecution. Sell, 539 U. S. at 181 (emphasis

in original). To reach this conclusion, “[t]he court must find that any

alternative, less intrusive treatments are unlikely to achieve substantially the

10
See also Green, 532 F3d at 557 (“The fact that [the physician] offered alternatives
depending on Green’s reaction to forced medication only supports the individualized and
appropriately tailored nature of her treatment plan.”); Hernandez-Vasquez, 513 F3d at 917 (“[W]hile
the court may not simply delegate unrestricted authority to physicians, the restrictions it does impose
should be broad enough to give physicians a reasonable degree of flexibility in responding to
changes in the defendant’s condition.”).

43
same results,” and the court also “must consider less intrusive means for

administering the drugs, e.g., a court order to the defendant backed by the

contempt power, before considering more intrusive methods.” Id. (emphasis

added). As with its other rulings, the trial court’s order in this case simply

recites a portion of the relevant language from Sell: “Involuntary medication is

necessary to further those interests and any alternative less intrusive treatments

have been and are unlikely to achieve substantially the same results.” These

findings are again insufficient to allow proper appellate review, and the court

also failed to fully address this part of the Sell test.

To begin with, the absence of a specific treatment plan for Warren,

identifying what drug or drugs would be given to him in what doses, leaves us

unable to evaluate whether the record supports a finding that compelling Warren

to take the unidentified “medication” is likely to be effective in making him

competent to stand trial or to compare that likelihood to the potential

effectiveness of the alternative, non-drug therapy discussed by the expert

witnesses. Moreover, even if medication of some sort is more likely to be

effective than less intrusive treatments, the trial court’s written order does not

advert to consideration of any less intrusive means for administering the drugs,

44
despite the conflicting evidence in the record on this issue, and the court said

only conclusorily at the hearing – without identifying what medication(s) the

court was referencing – that “the testimony of the witnesses supports a finding

by this Court, and I do so find that less intrusive means are unlikely to achieve

substantially the same result.”

Thus, although the evidence in the record was mixed as to whether Warren

had literally refused to take either or both of the antipsychotic medications

previously prescribed for him, or rather whether the discontinuation of his

medications was instead based on the medical judgment of his treating physician

Dr. Kane (informed by Warren’s complaints about side effects), the trial court

made no clear finding as to whether Warren would refuse to take whatever

medication might now be prescribed for him. When a court considers the

constitutionality of involuntarily medicating a defendant, the relevant

involuntariness is that of the defendant, not his treating physician.11

Accordingly, an order for involuntary medication must address the defendant’s

11
For this reason, we disagree with Warren’s argument that ordering him to take medication
would be legally improper unless his current treating physician concurs with the trial court’s
judgment regarding the appropriateness of the proposed treatment, although we recognize that a
physician faced with a court’s medication order that is contrary to his own medical opinion may have
difficult professional and ethical issues to resolve.

45
refusal to voluntarily take the medication being considered, although the order

may do so in a manner that contemplates alternative approaches for situations

that may unfold during the defendant’s treatment, such as his subsequent or

intermittent refusal to take the medication. As this case now stands, it would be

hard to make a finding on this point (or to evaluate such a finding on appeal),

without knowing what specific drug or drugs the State’s doctors actually

propose to present to Warren (e.g., a drug like Geodon, to which he previously

had a serious negative reaction, or a third generation antipsychotic he has never

been given). There is little if any evidence in the record that Warren has

indicated his refusal to take any medication under any circumstances.

Similarly, the record shows that some of the drugs that the experts

discussed must be administered by injection, while others are taken orally, a

consideration that bears on how intrusive the forcible administration of different

drugs may be, which again could not be properly evaluated without knowing

which specific drugs are proposed for use. See Sell, 539 U. S. at 181 (holding

that the court “must consider less intrusive means for administering the drugs”

(emphasis added)). Furthermore, the trial court’s order says nothing about

whether Warren would agree to take medication, even though he would prefer

46
not to, if he were ordered by the court to do so on penalty of contempt. Sell

indicates that this is an alternative that the trial court should consider before

issuing an order authorizing the State to physically force a defendant to take an

antipsychotic medication. See id. (giving, as the only specific example of the

less intrusive means for administering the drugs that the court must consider, “a

court order to the defendant backed by the contempt power”).12

Finally, the trial court’s order was inadequate because it failed to address

whether ordering involuntary medication solely for the purpose of making

Warren competent to stand trial was necessary in light of the potential

12
In the absence of clear evidence that Warren would refuse all antipsychotic medication,
it is worth comparing the unfettered discretion that the trial court’s order would give the State to
force drugs on him with the type of careful and stepwise treatment plans that have been approved
in other cases. See, e.g., Ruark, 611 Fed. Appx. at 596 (“Dr. Sarrazin’s written report details the
treatment plan that will be implemented should a court order that Ruark be involuntarily medicated.
The staff at [the hospital] will first present Ruark with a copy of the order and will try to convince
him to take an oral antipsychotic medication at the lowest effective dose. If Ruark is willing to
cooperate, he will be given Abilify, Geodon, Risperdal, or Haldol. If he suffers from any side effects
that are not relieved by adjunctive medications, he will be switched to another antipsychotic. If
Ruark is unwilling to cooperate and must be forcibly medicated, Dr. Sarrazin will begin by
administering a test dose of 5 milligrams of Haldol. If Ruark develops neuromuscular side effects
during his treatment, he will be given other medications to treat those adverse effects. If Ruark
becomes agitated or combative during the involuntary medication process, he will be given an
injection of Lorazepam, a sedative.”); Breedlove, 756 F3d at 1042 (“[T]he district court found that
the treatment method would follow the least intrusive course possible, only using forced injections
if Breedlove refused to take the medication orally.”); Diaz, 630 F3d at 1326-1327, 1330 (outlining
the detailed treatment plan and approving the district court’s order directing the government to
medicate Diaz forcibly in accordance with the plan, but only after “first seeking to obtain Mr. Diaz’s
voluntary participation with any treatment.”).

47
alternative of ordering involuntary medication “for a different purpose, such as

the purposes set out in Harper related to the individual’s dangerousness, or

purposes related to the individual’s own interests where the refusal to take drugs

puts his health gravely at risk.” Sell, 539 U. S. at 182 (emphasis in original)

(referring to Washington v. Harper, 494 U. S. 210 (110 SCt 1028, 108 LE2d

178) (1990)). The Sell Court explained that “the inquiry into whether

medication is permissible, say, to render an individual nondangerous is usually

more ‘objective and manageable’ than the inquiry into whether medication is

permissible to render a defendant competent,” and

[t]he medical experts may find it easier to provide an informed
opinion about whether, given the risk of side effects, particular
drugs are medically appropriate and necessary to control a patient’s
potentially dangerous behavior (or to avoid serious harm to the
patient himself) than to try to balance harms and benefits related to
the more quintessentially legal questions of trial fairness and
competence.

Id. at 182 (citation omitted).

The Supreme Court noted that these other grounds for involuntary medical

treatment are typically addressed by courts as a civil matter, with various

procedural protections for the patient, and added:

48
 Even if a court decides medication cannot be authorized on the
alternative grounds, the findings underlying such a decision will
help to inform expert opinion and judicial decisionmaking in
respect to a request to administer drugs for trial competence
purposes. At the least, they will facilitate direct medical and legal
focus upon such questions as: Why is it medically appropriate
forcibly to administer antipsychotic drugs to an individual who (1)
is not dangerous and (2) is competent to make up his own mind
about treatment? Can bringing such an individual to trial alone
justify in whole (or at least in significant part) administration of a
drug that may have adverse side effects, including side effects that
may to some extent impair a defense at trial?

Id. at 183 (emphasis in original). For these reasons, the Supreme Court

concluded that a trial court asked to approve forced administration of drugs to

render a defendant competent to stand trial “should ordinarily determine

whether the Government seeks, or has first sought, permission for forced

administration of drugs on these other Harper-type grounds; and, if not, why

not.” Id. See also United States v. Hernandez-Vasquez, 513 F3d 908, 915 (9th

Cir. 2007) (amended opinion) (stating that the “[trial] court may wish to inquire

further as to the Government’s reasons for not seeking involuntary medication

on the basis of dangerousness, and should note for the record its reasons for not

proceeding under Harper, before undertaking the Sell inquiry,” because “[a]

defendant’s liberty interest in avoiding unnecessary involuntary medication is

49
too important to allow for situations in which the court is asked to undertake the

more error-prone [Sell] analysis for what may be arbitrary or tactical reasons”).

In this case, nothing in the record indicates that the State has sought

permission to involuntarily medicate Warren for the alternative purpose of

preventing him from being a danger to himself or others, or that the trial court

asked why the State had not done so. This is especially concerning because

there was expert testimony that Warren would likely present a “risk to the

public” if released, that he believed that “retaliating in a very violent way”

would be an appropriate response to the thefts that are common in the mental

hospital, and that antipsychotic medication would cause him to go from a “low

risk” to hospital staff to “an even lower risk, because there would be less likely

some chance that he could incorporate other people in the hospital into his

delusional system.” In light of this evidence, and in the absence of a cogent

explanation from the State as to why it has not sought to medicate Warren under

Harper, it was error for the trial court to jump directly to the forcible

administration of medication for the sole purpose of rendering Warren

competent to stand trial.

50
 (d) Medical Appropriateness

The fourth and final part of the Sell test requires the trial court to

“conclude that administration of the drugs is medically appropriate, i.e., in the

patient’s best medical interest in light of his medical condition.” Sell, 539 U. S.

at 181 (emphasis in original). This part of the test is closely related to the

second part, and the trial court’s ruling on it suffers the same infirmities as

identified above as to the ruling on the second part. Again, the court’s order

simply summarily recites that “[t]he administration of medications is medically

appropriate and involuntary medication is in Defendant’s best interest in light

of his medical condition,” without identifying what medications in what dosages

for what durations the court was blessing.

The Sell Court’s explanation immediately following the language that the

trial court’s order tracked makes the inadequacy of the order apparent: “The

specific kinds of drugs at issue may matter here as elsewhere. Different kinds

of antipsychotic drugs may produce different side effects and enjoy different

levels of success.” Id. (emphasis added). As discussed above, the expert

testimony presented in this case referred to a variety of antipsychotic drugs and

generations of drugs, which the experts indicated could produce different side

51
effects, in general and for Warren in particular, and could have different chances

for success, in general and for Warren in particular, although the experts were

far from uniform in their opinions and the testimony was often imprecise in

linking specific predictions of success and specific possible side effects to

specific medications. The trial court’s order does nothing to sort out this

evidence, and thus this aspect of the order is also flawed. See Evans, 404 F3d

at 242 (holding that the government must not only present a specific treatment

plan, but must also “relate the proposed treatment plan to the individual

defendant’s particular medical condition” in order to satisfy the second and

fourth parts of the Sell test).

4. Conclusion

For the reasons discussed above, we conclude that the trial court’s order

was insufficient in numerous respects to justify Warren’s involuntary

medication for the sole purpose of making him mentally competent to stand trial

for the very serious crimes he is accused of committing. Accordingly, as the

U.S. Supreme Court did in Sell, we vacate the trial court’s order and remand the

case for further proceedings in light of this opinion. We note that the

evidentiary hearing was held more than a year ago, and it is possible that

52
Warren’s mental and physical condition has materially changed in the meantime

and also possible that new scientific information is available regarding

antipsychotic drugs and their likely effectiveness and side effects.

Consequently, if the State elects to pursue its motion for involuntary medication

on remand, the trial court should allow the parties to present additional evidence

to ensure that the court’s findings are based on current circumstances. See Sell,

539 U. S. at 186. See also Evans, 404 F3d at 242-243 (vacating and remanding

“with instructions to the district court to reassess the motion after affording the

parties the opportunity to supplement the record in a manner consistent with [the

Circuit Court’s] opinion”); Hernandez-Vasquez, 513 F3d at 919 (vacating and

remanding with similar instructions).

Judgment vacated and case remanded with direction. All the Justices

concur, except Melton, J., who concurs in the judgment only.

53