v. Marston

The summaries of the Colorado Court of Appeals published opinions
constitute no part of the opinion of the division but have been prepared by
the division for the convenience of the reader. The summaries may not be
cited or relied upon as they are not the official language of the division.
Any discrepancy between the language in the summary and in the opinion
should be resolved in favor of the language in the opinion.

SUMMARY
August 6, 2020

2020COA121

No. 18CA1879, People v. Marston — Crimes — DWAI;
Evidence — Admissibility

A division of the court of appeals holds that a trial court may

admit evidence of the results of a horizontal gaze nystagmus (HGN)

test given to a motorist first without holding a Shreck hearing, if the

evidence is offered only as evidence of impairment and if the

witness testifying about the administration and the results of the

test is competent to give such testimony.
COLORADO COURT OF APPEALS 2020COA121

Court of Appeals No. 18CA1879
Jefferson County District Court No. 16CR2934
Honorable Tamara S. Russell, Judge

The People of the State of Colorado,

Plaintiff-Appellee,

v.

Shawn Patrick Marston,

Defendant-Appellant.

JUDGMENT AFFIRMED

Division V
Opinion by JUDGE J. JONES
Gomez, J., concurs
Welling, J., concurs in part and dissents in part

Announced August 6, 2020

Philip J. Weiser, Attorney General, Frank R. Lawson, Assistant Attorney
General, Denver, Colorado, for Plaintiff-Appellee

Megan A. Ring, Colorado State Public Defender, Meredith E. O’Harris, Deputy
State Public Defender, Denver, Colorado, for Defendant-Appellant
¶1 Defendant, Shawn Patrick Marston, appeals the judgment of

conviction entered on a jury verdict finding him guilty of driving

while ability impaired (DWAI). One of the issues he raises is

whether the district court was required to hold a Shreck hearing,

see People v. Shreck, 22 P.3d 68 (Colo. 2001), before allowing a

police officer to testify about the results of a horizontal gaze

nystagmus (HGN) test the officer administered to him immediately

before he was arrested. We hold that no such hearing was

required; the results of such a test are generally admissible, if

relevant, as evidence of impairment, so long as the person testifying

about the administration and results of the test is competent to give

such testimony. The officer in this case was, so the court didn’t err

by allowing the officer’s testimony. We also reject Marston’s other

challenges to the judgment and therefore affirm.

I. Background

¶2 Shortly before noon one day, J.P. was driving behind a red

truck when he saw the truck straddling lanes and speeding up and

slowing down erratically. He also saw the driver nodding off at the

wheel. J.P. called 911 and followed the truck to a 7-Eleven.

Marston got out of the truck and went into the 7-Eleven, at which

1
point J.P. approached a Jefferson County Sheriff’s deputy —

Deputy Aaron Fosler — who had just pulled into the parking lot.

J.P. told Deputy Fosler what he had seen. Deputy Fosler followed

Marston into the 7-Eleven.

¶3 Deputy Fosler asked Marston to step outside the convenience

store and answer some questions. Marston obliged. As Deputy

Fosler held the door open and Marston walked out, he saw that

Marston was “staggering, grabbing something to hold onto.”

Outside, Deputy Fosler questioned Marston about his driving and

whether he had been drinking. Marston told him that his driver’s

license was suspended, told the officer he had driven to the

7-Eleven but then said his girlfriend had driven him there, and told

the officer he had downed several “mixed drinks” the night before at

his girlfriend’s house. Unprompted, Marston asked Deputy Fosler if

there was any way to keep his truck from being towed if he was

taken to jail. Throughout this conversation, Deputy Fosler noticed

2
that Marston’s eyes were red and watery, he smelled of alcohol, and

his speech was “thick tongued.”1

¶4 Based on Marston’s statements and Deputy Fosler’s

observations, Deputy Fosler called for another officer, Deputy Kevin

Kehl, to administer roadside field sobriety tests. Kehl did so. One

such test was the HGN test. As discussed more fully below, that

test requires the subject to follow an object (for example, a pen) with

his eyes to the left and right. The person administering the test

must watch the subject’s eyes to detect any involuntary “jerking” of

his eyeballs.

¶5 At one point during the roadside tests, when Deputy Kehl was

demonstrating a walk-and-turn maneuver, Marston said, “I couldn’t

do that sober.” Deputy Kehl determined that Marston didn’t

perform as a sober person would on the HGN test and the other

roadside tests and arrested him. Marston refused to take a

chemical test.

1 Sometime during this conversation, a second deputy arrived and,
for part of the time, stood nearby. That officer also noticed
Marston’s signs of intoxication.

3
¶6 After Marston’s girlfriend picked up the truck from the

7-Eleven, she found bottles of vodka in the back of the truck. She

testified at trial that, contrary to what Marston had told Deputy

Fosler, she wasn’t with Marston the night before the arrest and in

fact hadn’t seen him in several weeks.

¶7 The People charged Marston with driving under the influence

(DUI) and driving under restraint. Marston went to trial on the DUI

charge. The jury ultimately convicted him of the lesser included

DWAI offense. The court then determined that Marston had at least

three prior alcohol-related driving convictions and sentenced him

for felony DWAI. See § 42-4-1301(1)(b), C.R.S. 2019.

II. Discussion

¶8 Marston contends that the district court erred by (1) denying

his motion to suppress his statements to Deputy Fosler at the

scene; (2) denying his request for a Shreck hearing on (a) the

reliability of the HGN test and (b) Deputy Kehl’s expertise; and (3)

determining that he had three prior alcohol-related driving

convictions by a preponderance of the evidence rather than having

the jury determine those convictions as elements that must be

4
proven beyond a reasonable doubt. We address and reject each

contention in turn.

A. Marston’s Statements to Police

¶9 First, Marston contends that his statements to Deputy Fosler

at the scene should have been suppressed because they were

involuntary.2 We disagree.

1. Standard of Review

¶ 10 “A trial court’s suppression ruling presents a mixed question

of fact and law.” People v. Ramadon, 2013 CO 68, ¶ 21. We won’t

overturn the trial court’s factual findings if they are supported by

competent evidence in the record; however, we review the legal

effect of those facts de novo. Id.; Effland v. People, 240 P.3d 868,

878 (Colo. 2010) (“[T]he ultimate determination of whether a

statement is voluntary is a legal question and is reviewed de novo.”).

And we review any error under the constitutional harmless error

standard; that is, we reverse unless the People show that the error

2Marston’s motion didn’t specify what statements he wanted
suppressed. At the suppression hearing, Marston’s attorney said
she was challenging “everything at the scene.” But on appeal,
Marston only challenges the voluntariness of his statements to
Deputy Fosler.

5
was harmless beyond a reasonable doubt. Hagos v. People, 2012

CO 63, ¶ 11.

2. Applicable Law

¶ 11 “[A] defendant’s statements must be voluntary to be

admissible as evidence.” Ramadon, ¶ 18. In determining whether a

defendant’s statements were voluntary, we “must consider the

totality of the circumstances ‘to determine whether the accused’s

will was actually overborne by coercive police conduct.’” People v.

Coke, 2020 CO 28, ¶ 18 (quoting Sanchez v. People, 2014 CO 56,

¶ 11). To do so, we engage in a two-step inquiry: we first look to

whether the police conduct was coercive; if so, we then look to

whether that conduct “played a significant role in inducing the

statements.” Ramadon, ¶ 20. The statements “must not be the

product of any direct or implied promises, nor obtained by exerting

an improper influence.” People v. Medina, 25 P.3d 1216, 1222

(Colo. 2001).

¶ 12 To determine whether the police conduct was coercive, we may

consider, among other things, the following factors:

(1) whether the defendant was in custody;
(2) whether the defendant was free to leave;

6
(3) whether the defendant was aware of the
situation;
(4) whether the police read Miranda rights to
the defendant;
(5) whether the defendant understood and
waived Miranda rights;
(6) whether the defendant had an opportunity
to confer with counsel or anyone else prior
to or during the interrogation;
(7) whether the statement was made during
the interrogation or volunteered later;
(8) whether the police threatened [the]
defendant or promised anything directly or
impliedly;
(9) the method or style of the interrogation;
(10) the defendant’s mental and physical
condition just prior to the interrogation;
(11) the length of the interrogation;
(12) the location of the interrogation; and
(13) the physical conditions of the location
where the interrogation occurred.

Ramadon, ¶ 20 (quoting Medina, 25 P.3d at 1222-23).

3. Analysis

¶ 13 We agree with the district court’s determination that Marston’s

will wasn’t overborne by coercive police conduct and that his

statements were therefore voluntary. The following facts, almost all

of which are undisputed, lead us to this conclusion:

• Deputy Fosler asked Marston if he wouldn’t mind

stepping outside and answering some questions; he

didn’t order him to do so. Marston agreed.

7
• No officer made any threats or promises to Marston.

• The encounter occurred outside, in a public place.

• Marston seemed be aware of the situation. (He had been

in this situation several times before.) For example,

unprompted, he asked if there was any way he could

prevent his truck from being towed if he was taken to jail.

• Marston appeared to understand Deputy Fosler’s

questions and gave responsive answers.

• Marston wasn’t restrained in any way and hadn’t yet

been taken into custody.

• Deputy Fosler used a conversational tone.

• The deputies didn’t use any subtle psychological

pressure to get Marston to talk.

¶ 14 Marston asserts that his statements were coerced because

more than one deputy was present, the two deputies stood close to

him, the encounter lasted twenty to thirty minutes, he wasn’t given

a Miranda advisement, and he wasn’t free to leave. But looking at

the totality of the circumstances, as we must, see Coke, ¶ 18, we

conclude that these facts don’t add up to coercion.

8
¶ 15 People v. Zadran, 2013 CO 69M, presents an instructive

comparison. In that case, the supreme court determined that the

police officer’s statements to the defendant that “I think it would be

in your best interest to talk to me[,]” “I think you are going to be

interested in some of the things that I already know[,]” “It is what it

is. You messed up. You know you messed up[,]” and “You want to

get ahead of this. You want to make things right. You want a

positive outcome from this. I’m trying to do the [least] invasive

thing that I can do here[,]” didn’t show police coercion. And this

even though the defendant was in custody and wasn’t free to leave.

Id. at ¶¶ 15-19. The court contrasted the facts before it with those

in cases in which the police exploited a defendant’s “particular set

of vulnerabilities.” Id. at ¶¶ 17-18.

¶ 16 In terms of showing coercion, the facts of this case don’t even

approach those in Zadran, in which the court found no coercion.

Deputy Fosler simply asked Marston, in a conversational tone,

questions about his driving and alcohol consumption. We see no

indication that Marston’s will was overborne by coercive police

conduct.

9
B. Denial of a Shreck Hearing

¶ 17 Next, Marston contends that the district court erred by

refusing to hold a Shreck hearing on the science, reliability, and

margin of error of the HGN test, as well as Deputy Kehl’s expertise

on those issues. We conclude that the district court didn’t err by

denying Marston a hearing, but that even if it did, any error was

harmless.

1. The District Court’s Ruling

¶ 18 After the prosecution endorsed Deputy Kehl as an expert in

HGN testing, Marston filed a motion requesting a Shreck hearing to

challenge the admissibility of the HGN test and Deputy Kehl’s

qualifications as an expert. In response, the prosecution argued

that a hearing wasn’t necessary, citing district court cases from

Colorado and appellate court cases from other jurisdictions in

which HGN testing had been found “reasonably reliable” and

accepted by the scientific community. At a motions hearing, the

district court heard argument from both sides as to why a Shreck

hearing was or wasn’t necessary. It ultimately found that “the

scientific community has accepted over many years the fact that the

HGN is a reliable and relevant tool to help police officers determine

10
if someone is under the influence of alcohol[,]” and concluded that

the science was reliable and that if the prosecution properly

qualified Deputy Kehl as an expert, his testimony would be useful

to the jury.

2. Standard of Review

¶ 19 We review a district court’s denial of a request for a Shreck

hearing for an abuse of discretion. See People v. Rector, 248 P.3d

1196, 1201 (Colo. 2011) (“Once a party requests a Shreck analysis,

a trial court is vested with the discretion to decide whether an

evidentiary hearing would aid the court in its Shreck analysis.”). “A

trial court abuses its discretion only if its decision is manifestly

arbitrary, unreasonable, [or] unfair, or is based on a

misunderstanding or misapplication of the law.” People v.

Thompson, 2017 COA 5, ¶ 91. And “we review any error in denying

a Shreck hearing under the nonconstitutional harmless error

standard.” People v. Wilson, 2013 COA 75, ¶ 24; cf. Campbell v.

People, 2019 CO 66, ¶ 34 (error in allowing officer to testify as a lay

witness regarding HGN reviewed for nonconstitutional harmless

error). We therefore reverse only if any error substantially

11
influenced the verdict or impaired the fairness of the trial. Wilson,

¶ 24.

3. Analysis

¶ 20 To decide whether the district court abused its discretion by

allowing Deputy Kehl to testify about Marston’s performance on the

HGN test without first holding a Shreck hearing, we proceed in the

following steps. First, we discuss the nature of the test and its use

in Colorado courts. Second, we discuss Shreck’s admissibility

framework. Third, we assess whether there is sufficient indication

that the HGN test satisfies Shreck’s admissibility test, or whether a

Shreck hearing was required.

a. The HGN Test

¶ 21 The HGN test measures the subject’s ability to maintain visual

fixation on an object as his eyes move from side to side. Nat’l

Traffic Law Ctr., Am. Prosecutors Research Inst., Horizontal Gaze

Nystagmus: The Science and the Law (n.d.),

https://perma.cc/UW93-GHPD. (“Nystagmus” is a rapid,

involuntary oscillation of the eyeballs.) An officer holds an object (a

pen, small flashlight, or finger) about twelve to fifteen inches in

front of a subject’s nose. The officer asks the subject to remove any

12
glasses, to stand still with his feet together and hands at his sides,

and to focus on the object. The officer then moves the object slowly

back and forth horizontally three times, observing whether each of

the subject’s eyes smoothly tracks the object. Continuing Legal

Education in Colorado, Inc., Colorado DUI Benchbook § 6.2.2

(2019-2020 ed.); see also Am. Prosecutors Research Inst.,

Admissibility of Horizontal Gaze Nystagmus Evidence 5 (2003),

https://perma.cc/R36W-CCXR. The officer looks for three testing

points for each eye (six total): lack of smooth pursuit, distinct

jerking of the eyes at maximum deviation, and jerking that occurs

before a forty-five-degree angle. Nat’l Highway Traffic Safety

Admin., DWI Detection and Standardized Field Sobriety Test (SFST)

Participant Manual Session 7, page 9 of 39 (rev. Feb. 2018),

https://perma.cc/4R6E-ZB3A; see State v. Baue, 607 N.W.2d 191,

201-02 (Neb. 2000) (describing the test). If four or more of these

clues are present, the subject’s blood alcohol content (BAC) is likely

at or above 0.08%. DWI Detection at Session 8, page 37 of 95;

Baue, 607 N.W.2d at 202.

¶ 22 The National Highway Traffic Safety Administration (NHTSA)

standardized the HGN test along with the walk-and-turn and

13
one-leg-stand tests in 1981, finding that, combined, the tests can

accurately determine whether a subject’s BAC is .10 or higher

eighty-three percent of the time. Horizontal Gaze Nystagmus: The

Science and the Law. Those tests are now the standard field

sobriety tests in use across the country.

¶ 23 In Colorado, “virtually every judge now takes judicial notice of

the scientific principles underlying HGN testing[,]” though

Colorado’s appellate courts haven’t yet addressed the test’s

admissibility. Colorado DUI Benchbook § 6.2.2.

b. Shreck

¶ 24 In Shreck, the Colorado Supreme Court adopted a “liberal,”

“totality of the circumstances” test, grounded in relevant rules of

evidence, for determining whether scientific evidence is admissible

through expert testimony. To get there, the court first rejected the

so-called Frye test (derived from Frye v. United States, 293 F. 1013

(D.C. Cir. 1923)), which Colorado appellate courts had applied,

albeit perhaps inconsistently, for several decades. The Frye test

requires that a scientific conclusion gain “general acceptance in the

particular field to which it belongs” as to both the underlying theory

supporting the conclusion and the techniques and experiments

14
employing it, before evidence based on that conclusion can be

introduced in court. Shreck, 22 P.3d at 73 (quoting Frye, 293 F. at

1014). The court determined that the Frye test is too rigid and is

inconsistent with the more flexible approach countenanced by the

Colorado Rules of Evidence. Id. at 76-77.

¶ 25 The Shreck court also discussed the United States Supreme

Court’s decision in Daubert v. Merrell Dow Pharmaceuticals, Inc.,

509 U.S. 579 (1993), which likewise rejected Frye, holding that its

“rigid general acceptance requirement [is] at odds with the ‘liberal

thrust’ of the Federal Rules [of Evidence] and their ‘general

approach of relaxing the traditional barriers to opinion testimony.’”

Id. at 588 (quoting Beech Aircraft Corp. v. Rainey, 488 U.S. 153, 169

(1988)). Rather than adopting the Daubert approach wholesale,

however, the Colorado Supreme Court went its own way.

¶ 26 Ultimately, the court adopted a rubric supported by CRE 403

and 702 “because their flexibility is consistent with a liberal

approach that considers a wide range of issues.” Shreck, 22 P.3d at

77. An admissibility analysis under Shreck requires the court to

determine whether “(1) the scientific principles underlying the

testimony are reasonably reliable; (2) the expert is qualified to opine

15
on such matters; (3) the expert testimony will be helpful to the jury;

and (4) the evidence satisfies CRE 403.” Rector, 248 P.3d at 1200;

see Shreck, 22 P.3d at 77-79. In making these determinations, the

court can consider a broad range of indicia “that may be pertinent

to the evidence at issue.” Shreck, 22 P.3d at 77. These may

include, but certainly aren’t limited to, the factors identified in

Daubert. Id. at 77-78.

¶ 27 Before applying the Shreck framework to the testimony at

issue in this case, we pause to address Marston’s assertion that

Shreck “is widely regarded as imposing a more rigorous ‘gatekeeper’

function on trial courts than Frye did.” (The point matters because,

as discussed below, some courts have applied the Frye test in

determining HGN test admissibility, and so understanding the

relative restrictiveness of the tests will prove informative.) To the

extent Marston intends to suggest that the Shreck test is more

limiting than the Frye test, he is wrong, for at least two reasons.

¶ 28 First, Marston’s assertion is based on a quote from an Alaska

Supreme Court case, State v. Coon, 974 P.2d 386, 390 (Alaska

1999), abrogated by State v. Sharpe, 435 P.3d 887 (Alaska 2019),

that is taken out of context. That case compared the Daubert test

16
(not the Shreck test, which the Colorado Supreme Court adopted

two years later) to the Frye test. And in saying that the Daubert test

imposes a more rigorous gatekeeper function, the Alaska court

merely observed that, under Daubert, admissibility depends on

consideration of several factors, while Frye gives dispositive weight

to but one — general acceptance.

¶ 29 Second, Shreck itself repeatedly contrasts Frye’s “rigid”

(indeed, too rigid) approach to the more “liberal” approach

contemplated by the Colorado Rules of Evidence. See Shreck, 22

P.3d at 76 (rejecting Frye because it “restricts the admissibility of

reliable evidence that may not yet qualify as ‘generally accepted’”)

(citation omitted).3 And the court emphasized that “[a]ny concerns

that invalid scientific assertions will be admitted under this liberal

standard are assuaged by Rule 702’s overarching mandate of

reliability and relevance. . . . Such concerns are also mitigated by

‘[v]igorous cross-examination, presentation of contrary evidence,

3Indeed, the court in Shreck concluded that certain scientific
evidence that may not have been generally accepted was
nonetheless admissible. People v. Shreck, 22 P.3d 68, 80-81 (Colo.
2001).

17
and careful instruction on the burden of proof.’” Id. at 78 (quoting

Daubert, 509 U.S. at 596).

c. Application

i. HGN Test Evidence is Generally Admissible

¶ 30 A number of relevant considerations identified in Shreck

support the admissibility of HGN test results as evidence of

impairment:

1. The technique has been tested and subjected to peer

review and publication. See State v. Ruthardt, 680 A.2d

349, 357 (Del. Super. Ct. 1996) (citing some of the

relevant literature); State v. Dahood, 814 A.2d 159, 166

(N.H. 2002) (recognizing “extensive scientific literature

dating back to the 1950s that thoroughly examines and

critiques the HGN test and the theory underlying that

test-that there is a strong correlation between the

amount of alcohol a person consumes and the onset of

nystagmus”); see, e.g., Marcelline Burns & Ellen W.

Anderson, Nat’l Highway Traffic Safety Admin., A

Colorado Validation Study of the Standardized Field

Sobriety Test (SFST) Battery (Nov. 1995),

18
https://perma.cc/5MMA-NR8W; Gregory W. Good & Arol

R. Augsburger, Use of Horizontal Gaze Nystagmus as a

Part of Roadside Sobriety Testing, 63 Am. J. Optometry &

Physiological Optics 467 (1986); V. Tharp, M. Burns, &

H. Moskowitz, Nat’l Highway Traffic Safety Admin.,

Development and Field Test of Psychophysical Tests for

DWI Arrest (Mar. 1981), https://perma.cc/5HA3-S3KV.

2. Studies have produced reliability rates indicating a

significant correlation between test results and

impairment. See, e.g., Horizontal Gaze Nystagmus: The

Science and the Law (combined, the HGN test, the

walk-and-turn test, and the one-leg-stand test accurately

determine whether a subject’s BAC was .10 or higher

eighty-three percent of the time); Jack Stuster &

Marcelline Burns, Validation of the Standardized Field

Sobriety Test Battery at BACs Below .10 Percent (Aug.

1998), https://perma.cc/E5BL-PGFH (study showed that

the HGN test could often accurately predict whether a

subject’s BAC is at or above 0.08).

19
3. Standards have been developed to ensure consistency in

the application of the test and the assessment of

responses. See DWI Detection and Standardized Field

Sobriety Test (SFST) Participant Manual at Session 4, page

35 of 36, Session 7, pages 8-11 of 39 (providing

instructions for application and evaluation of the HGN

test nationwide).

4. There is substantial specialized literature dealing with

the technique. See Dahood, 814 A.2d at 166 (reviewing

some of the literature); see, e.g., Stephanie E. Busloff,

Comment, Can Your Eyes Be Used Against You — The

Use of the Horizontal Gaze Nystagmus Test in the

Courtroom, 84 J. Crim. L. & Criminology 203 (Spring

1993).

5. The evidence has been offered in many other cases in

Colorado as evidence of alcohol impairment. See, e.g.,

Campbell, ¶¶ 7, 13-15 (HGN test results admitted at

trial); People v. Haack, 2019 CO 52, ¶¶ 1, 8 (trial court

20
found that HGN test results would have been admissible

but for an unrelated constitutional violation).4

See Shreck, 22 P.3d at 77-78 (identifying these considerations as

relevant).

¶ 31 As well, a survey of case law from other jurisdictions shows

that most allow evidence of HGN test results as evidence of

impairment. Some of these decisions employ a Daubert or

Shreck-like analysis, but some employ the even more restrictive

Frye test. See, e.g., Ballard v. State, 955 P.2d 931, 940 (Alaska Ct.

App. 1998) (HGN evidence meets Frye test if results are offered to

show a person has consumed alcohol and is potentially impaired),

overruled on other grounds as recognized by Alvarez v. State, 249

P.3d 286 (Alaska 2011); State v. City Court, 799 P.2d 855, 859 (Ariz.

1990) (HGN test satisfies the Frye standard if offered only as

evidence of impairment); State v. Commins, 850 A.2d 1074, 1080-81

(Conn. App. Ct. 2004) (HGN test evidence satisfied Daubert test),

4 As noted above, the prosecution cited several trial court decisions
admitting evidence of HGN test results in responding to Marston’s
motion for a Shreck hearing. Marston has never disputed the point
that Colorado trial courts regularly admit such evidence.

21
aff’d on other grounds, 886 A.2d 824 (Conn. 2005); Ruthardt, 680

A.2d at 356-60 (applying Shreck-like test; HGN test results

admissible as evidence of impairment); Williams v. State, 710 So. 2d

24, 30-32 (Fla. Dist. Ct. App. 1998) (the HGN test is a reliable

indicator of the presence of alcohol in blood, and there is no need

for trial courts to reapply a Frye analysis to HGN); Hawkins v. State,

476 S.E.2d 803, 806-08 (Ga. Ct. App. 1996) (HGN test results

admissible without expert testimony regarding the scientific validity

of the test; applying a Shreck-like totality of the circumstances test);

State v. Gleason, 844 P.2d 691, 694-95 (Idaho 1992) (HGN

testimony admissible under Frye test as evidence of impairment);

State v. Taylor, 694 A.2d 907, 911-12 (Me. 1997) (applying Frye

test; HGN test results admissible as evidence of impairment);

Schultz v. State, 664 A.2d 60, 69-70 (Md. Ct. Spec. App. 1995)

(courts may take judicial notice of the results of an HGN test); State

v. Klawitter, 518 N.W.2d 577, 584-86 (Minn. 1994) (HGN test

results satisfied Frye test); State v. Hill, 865 S.W.2d 702, 703-04

(Mo. Ct. App. 1993) (HGN test satisfies Frye test if offered as

evidence of intoxication), overruled on other grounds by State v.

Carson, 941 S.W.2d 518 (Mo. 1997); Baue, 607 N.W.2d at 201-04

22
(HGN test results admissible as evidence of impairment under Frye

test); Dahood, 814 A.2d at 166-67 (HGN test results admissible

under Daubert test); State v. Aleman, 194 P.3d 110, 115-16 (N.M.

Ct. App. 2008) (HGN test results admissible under Daubert test);

City of Fargo v. McLaughlin, 512 N.W.2d 700, 703-08 (N.D. 1994)

(HGN test results admissible under Frye test if offered in

conjunction with other field sobriety tests); State v. O’Key, 899 P.2d

663, 689 (Or. 1995) (HGN test results admissible under Daubert

test); Emerson v. State, 880 S.W.2d 759, 763-69 (Tex. Crim. App.

1994) (HGN test results admissible under Texas Rule of Criminal

Evidence 702; applying Daubert-like test); see also United States v.

Horn, 185 F. Supp. 2d 530, 561 (D. Md. 2002) (collecting cases).

¶ 32 Marston cites a few cases rejecting the admissibility of HGN

test results under the facts before them. Ex Parte Malone, 575 So.

2d 106 (Ala. 1990); State v. Meador, 674 So. 2d 826 (Fla. Dist. Ct.

App. 1996); People v. McKown, 875 N.E.2d 1029 (Ill. 2007); State v.

Witte, 836 P.2d 1110 (Kan. 1992); State v. Lasworth, 42 P.3d 844

(N.M. Ct. App. 2001); Commonwealth v. Apollo, 603 A.2d 1023 (Pa.

23
Super. Ct. 1992).5 But those cases usually apply the Frye test,

which, as discussed above, is more restrictive than the “liberal”

Shreck test. And even then, they don’t reject such evidence

outright, but hold that the evidence before them wasn’t sufficient to

establish reliability and leave open the possibility that such

evidence could be presented. See also Commonwealth v. Sands,

675 N.E.2d 370, 371-73 (Mass. 1997) (trial court erred by allowing

officer to testify about HGN test results without being qualified as

an expert, but indicating that HGN test results would be admissible

if supported by expert testimony from the person administering the

test); State v. Helms, 504 S.E.2d 293, 294-96 (N.C. 1998) (same);

State v. Murphy, 953 S.W.2d 200, 202-03 (Tenn. 1997) (same); State

v. Cissne, 865 P.2d 564, 566-69 (Wash. Ct. App. 1994) (same).6

5 Marston also cites State v. Superior Court, 718 P.2d 171 (Ariz.
1986), in support of his position. But the court in that case
actually held that HGN test results are admissible as evidence of
impairment (but not as evidence of a specific BAC level). Id. at 182.
6 In at least one instance, the court reversed course after additional

evidence of reliability was presented. State v. Aleman, 194 P.3d
110, 115-16 (N.M. Ct. App. 2008) (backtracking on its earlier
holding in State v. Lasworth, 42 P.3d 844 (N.M. Ct. App. 2001)).

24
¶ 33 The weight of judicial authority therefore favors admissibility

of HGN test results without the need for additional evidence of

scientific reliability — at least if the evidence is offered only as

evidence of impairment and not a specific BAC level.7

¶ 34 We recognize that there isn’t unanimous agreement among

academics and other commentators concerning the reliability of the

HGN test. But much of the disagreement doesn’t relate to whether

alcohol can cause nystagmus — that point is widely accepted.

Rather, some have noted that there can be reasons other than

alcohol impairment that may explain test results in a particular

case, see, e.g., William A. Pangman, Horizontal Gaze Nystagmus:

Voodoo Science, 2 DWI J. 1, 2 (1987), or that police officers may not

be sufficiently trained in administering the test. And some claim a

higher incidence of false positives than reported by NHTSA and

7 A few courts have held that this type of evidence isn’t scientific
evidence at all, and therefore isn’t subject to restrictions on
scientific evidence. State v. Murphy, 451 N.W.2d 154, 156-58 (Iowa
1990); State v. Nagel, 506 N.E.2d 285, 286 (Ohio Ct. App. 1986);
Salt Lake City v. Garcia, 912 P.2d 997, 1000-01 (Utah Ct. App.
1996). We align ourselves, however, with the vast majority of courts
that have held that such evidence is based, at least in part, on
scientific principles.

25
other studies. Busloff, 84 J. Crim. L. & Criminology at 211 (noting

that NHTSA’s “experimental procedure has been further challenged

for its intentional screening out of those individuals highly likely to

be misclassified as false positives”). We believe such concerns go to

the weight of the evidence, not its admissibility. See Ruthardt, 680

A.2d at 359-60 (so holding); Dahood, 814 A.2d at 166-67 (same).

They can be addressed through the presentation of evidence by the

defense, cross-examination, and rigorous application of the rules

governing an expert’s qualifications. See Daubert, 509 U.S. at 596

(“Vigorous cross-examination, presentation of contrary evidence,

and careful instruction on the burden of proof are the traditional

and appropriate means of attacking shaky but admissible

evidence.”); Shreck, 22 P.3d at 78 (same).

¶ 35 We therefore conclude that evidence of HGN test results is

admissible as evidence of impairment if offered through a qualified

expert witness. Contrary to Marston’s assertion, such an expert

doesn’t have to be an expert in the science underlying the test. It is

enough that the witness is an expert in administering the test and

interpreting the subject’s responses. See, e.g., Ballard, 955 P.2d at

942 (trooper could testify about the results of the HGN test because

26
the court could find he was qualified to administer the test and

assess the results); People v. Leahy, 882 P.2d 321, 336 (Cal. 1994)

(once HGN testing is shown to be generally accepted, officers may

be deemed qualified to testify as to test results and the prosecution

won’t be required to submit expert testimony confirming a police

officer’s evaluation of an HGN test); Taylor, 694 A.2d at 912 (“A

proper foundation shall consist of evidence that the officer or

administrator of the HGN test is trained in the procedure and the

test was properly administered.”); State v. Torres, 976 P.2d 20,

34-35 (N.M. 1999) (the expert is qualified to testify if it is shown

that he had ability and training to administer the HGN test

properly, and that he did in fact administer the test properly); see

also Campbell, ¶¶ 26-31 (officer testifying as to HGN test results

must be qualified as an expert in administration and interpretation

of the test).

¶ 36 Though a party may request a Shreck hearing on the

admissibility of the proposed testimony, a trial court isn’t required

to grant that request if it has “sufficient information to make

specific findings under CRE 403 and CRE 702 about the reliability

of the scientific principles involved, the expert’s qualification to

27
testify to such matters, the helpfulness to the jury, and potential

prejudice.” Rector, 248 P.3d at 1201. The district court in this case

had ample information before it from which it could determine the

admissibility of the HGN test results without granting Marston’s

request for a Shreck hearing. And Deputy Kehl was sufficiently

qualified to testify about the administration and interpretation of

the test. (Marston doesn’t contest this point.) Though the court

didn’t make extensive findings, it didn’t abuse its discretion.

ii. Alternatively, Any Error Was Harmless

¶ 37 Even if the district court did err, we agree with the People that

any error was harmless.

¶ 38 In Campbell, the supreme court determined that while the

district court erred by allowing a police officer to testify about the

results of an HGN test without being qualified as an expert, the

error was harmless. Campbell, ¶¶ 1-2, 35. The court discussed the

“overwhelming” evidence against Campbell aside from the HGN test,

including the following:

• “Campbell’s breath had an odor of alcohol.”

• “[H]is eyes were bloodshot.”

• “[H]is speech was slurred.”

28
• He was uncoordinated: “he dropped his wallet while

trying to retrieve his identification[,]” and, “when

Campbell tried to get out of his truck, the officer saw him

reach for the door handle twice without success before

grabbing it and opening the door.”

• Bottles of alcohol were found in his car.

• He failed two of the other field sobriety tests.

• He had admitted to consuming alcohol.

• He took two breath tests, which showed BAC levels of

0.07 and 0.086, respectively.

Id. at ¶¶ 36-40.

¶ 39 There is similar overwhelming evidence against Marston:

• Two of the three officers reported that Marston’s breath

smelled like alcohol.

• All three observed that he had bloodshot eyes.

• Two of the three noticed that his speech was slurred.

• Marston had trouble keeping his balance as he walked

out of the 7-Eleven into the parking lot, and two of the

29
three officers reported that he had to reach to keep his

balance.

• Bottles of alcohol were found in the truck.

• Marston didn’t complete the other two field sobriety tests

as a sober person would. Deputy Kehl observed three of

eight clues on the walk-and-turn test. On the

one-leg-stand test, Marston could only lift his leg for

approximately four seconds.

• Marston admitted to drinking alcohol the night before.

¶ 40 Unlike Campbell, Marston didn’t submit to a breath test. (Nor

did he submit to a blood test.) But a witness saw him driving

erratically and nodding off at the wheel. And Marston said, “I

couldn’t do that sober,” when Deputy Kehl demonstrated the

walk-and-turn maneuver.

¶ 41 Because of the overwhelming amount of evidence against

Marston independent of the HGN test results, we conclude that any

error in allowing Deputy Kehl to testify about those results without

first holding a Shreck hearing didn’t substantially influence the

verdict or impair the fairness of the trial. See Hagos, ¶ 12; Wilson,

¶ 24.

30
C. Prior Convictions

¶ 42 Lastly, Marston contends that the district court erred by

finding that he had at least three prior alcohol-related driving

convictions by a preponderance of the evidence rather than

submitting the issue to the jury for it to decide beyond a reasonable

doubt. He argues that the prior convictions are elements of the

crime. We disagree.

¶ 43 Several divisions of this court have determined that such prior

convictions operate as sentence enhancers, not as elements, and

therefore don’t have to be decided by a jury. See People v. Ambrose,

2020 COA 112, ¶ 40; People v. Jiron, 2020 COA 36, ¶ 14; People v.

Quezada-Caro, 2019 COA 155, ¶ 24; People v. Gwinn, 2018 COA

130, ¶ 49. We agree with those divisions and disagree with another

division’s holding to the contrary. See People v. Viburg, 2020 COA

8M, ¶ 28.

III. Conclusion

¶ 44 The judgment is affirmed.

JUDGE GOMEZ concurs.

JUDGE WELLING concurs in part and dissents in part.

31
JUDGE WELLING, concurring in part and dissenting in part.

¶ 45 I concur fully in Parts II.A and II.B of the majority opinion but

dissent as to Part II.C. I do so for the reasons set forth by Judge

Berger in People v. Viburg, 2020 COA 8M (petition for cert. filed Mar.

12, 2020), which I joined. See also People v. Schreiber, 226 P.3d

1221, 1225-27 (Colo. App. 2009) (Bernard, J., concurring in part

and dissenting in part). Accordingly, I would reverse Marston’s

felony DUI conviction and remand the case for further proceedings.

I wouldn’t reach Marston’s contention on appeal that double

jeopardy bars retrial on the felony DUI. Instead, as in Viburg, I

would order that “[i]f there is a retrial of the felony DUI charge and

[Marston] interposes a double jeopardy defense, the trial court must

rule on that defense.” Viburg, ¶ 32 (“express[ing] no opinion

regarding the merits of any such defense”).

32