Neurocognitive, psychiatric, and substance use characteristics in a diverse sample of persons with OUD who are starting methadone or buprenorphine/naloxone in opioid treatment programs

Scott

etal. Addict Sci Clin Pract (2021) 16:64

https://doi.org/10.1186/s13722-021-00272-4

RESEARCH

Neurocognitive, psychiatric, andsubstance

use characteristics inadiverse sample

ofpersons withOUD who are starting

methadone orbuprenorphine/naloxone

inopioid treatment programs

Travis M. Scott

1,2*

, Julia Arnsten

, James Patrick Olsen

, Franchesca Arias

5,6

, Chinazo O. Cunningham

and

Monica Rivera Mindt

7,8

Abstract

Background: Medications for opioid use disorder such as opioid agonist treatment (OAT, including methadone,

buprenorphine) are the gold standard intervention for opioid use disorder (OUD). Persons with OUD have high rates

of neurocognitive impairment and psychiatric and substance use disorders, but few studies have examined these

characteristics in diverse patients initiating OAT in opioid treatment programs (OTPs). Additionally, in these individuals,

poor neurocognitive functioning and psychiatric/other substance use disorders are associated with poor OUD treat-

ment outcomes. Given rapid changes in the opioid epidemic, we sought to replicate ﬁndings from our pilot study by

examining these characteristics in a large diverse sample of persons with OUD starting OTP-based OAT.

Methods: Ninety-seven adults with OUD (M age = 42.2 years [SD = 10.3]; M education = 11.4 years [SD = 2.3]; 27%

female; 22% non-Hispanic white) were enrolled in a randomized longitudinal trial evaluating methadone versus

buprenorphine/naloxone on neurocognitive functioning. All participants completed a comprehensive neurocogni-

tive, psychiatric, and substance use evaluation within one week of initiating OAT.

Results: Most of the sample met criteria for learning (79%) or memory (69%) impairment. Half exhibited symptoms

of current depression, and comorbid substance use was highly prevalent. Lifetime cannabis and cocaine use disorders

were associated with better neurocognitive functioning, while depression was associated with worse neurocognitive

functioning.

Conclusions: Learning and memory impairment are highly prevalent in persons with OUD starting treatment with

either methadone or buprenorphine/naloxone in OTPs. Depression and comorbid substance use are prevalent

among these individuals, but neither impact learning or memory. However, depression is associated with neurocog-

nitive impairment in other domains. These ﬁndings might allow clinicians to help persons with OUD starting OAT to

develop compensatory strategies for learning and memory, while providing adjunctive treatment for depression.

Trial Registration NCT, NCT01733693. Registered November 4, 2012, https:// clini caltr ials. gov/ ct2/ show/ NCT01 733693.

permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the

original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or

other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line

to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory

regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this

licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco

mmons. org/ publi cdoma in/ zero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Open Access

Addiction Science &

Clinical Practice

*Correspondence: [email protected]

VA Palo Alto Health Care System Sierra Paciﬁc Mental Illness Research

Education Clinical Center, 3801 Miranda Ave, Palo Alto, CA 94304, USA

Full list of author information is available at the end of the article

Page 2 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

Background

Medication treatment with opioid agonist treat-

ment (OAT, including methadone, buprenorphine, or

buprenorphine/naloxone) is the gold standard interven

tion for opioid use disorder (OUD; [1]). Methadone is

most often provided in physician monitored, long-term

outpatient opioid treatment programs (OTPs), and

buprenorphine/naloxone is becoming increasingly com

mon in the United States (U.S.) among individuals seek-

ing treatment for OUD [2]. Reporting on important

characteristics of persons with OUD initiating OAT in

OTPs, such as neurocognitive abilities, psychiatric con

ditions, and comorbid substance use, is vital to under-

standing and improving treatment outcomes in these

individuals.

Poor neurocognitive functioning is associated with

poor substance use treatment outcomes, such as higher

relapse and lower substance abstinence rates [3, 4].

Individual studies and systematic reviews indicate that

chronic opioid use is associated with impaired learn

ing, memory, attention/working memory, and executive

functioning [5–11]. More recent studies have character

ized cognitive impairments in persons with OUD already

engaged in OAT with methadone [12] or buprenorphine

[13] and have found similar domains of cognitive impair

ment (i.e., attention/working memory and executive

functioning). While these studies demonstrate adverse

neurocognitive eﬀects of chronic opioid use and describe

cognitive characteristics of those already engaged in opi

oid agonist treatment, only two studies to our knowledge

have characterized the neurocognitive proﬁles of persons

with OUD who are starting OAT [5, 14]. While these

studies found impairments in several neurocognitive

domains including learning, memory, executive function

ing, and motor skills, both were limited by inadequate or

outdated neurocognitive batteries and small sample sizes,

suggesting that more research is needed to understand

the neurocognitive characteristics of persons with OUD

who initiate OAT in OTPs.

Psychiatric disorders (e.g., depression) and other, non-

opioid substance use are associated with a host of fac

tors that can complicate OAT (e.g., poor health, high

rates of criminal activities; [15, 16]). Both depression

[17] and additional substance use [18] are more preva

lent in persons with an OUD than in the general popu-

lation. While studies have examined the psychiatric and/

or substance use characteristics of patients starting (or

initiating) OAT [5, 19, 20], only one study to date has

examined associations between psychiatric disorders,

substance use, and neurocognitive function in persons

with OUD who are initiating OAT. is study found high

rates of lifetime major depressive disorder (31%), moder

ate to severe current depressive symptomatology (28%),

and high rates (i.e., > 30%) of lifetime and current alcohol,

cannabis, and cocaine use among OUD patients [5]. Arias

and colleagues [5] also found that persons with OUD

who had a lifetime history of alcohol dependence had

worse global neurocognitive functioning than those with

out, and that persons with OUD who had a lifetime his-

tory of cocaine dependence had worse attention/working

memory and motor functioning than those without, and

concluded that there may be a synergistic eﬀect of multi

ple substance use disorders contributing to neurocogni-

tive impairment in persons with OUD.

e opioid epidemic has rapidly shifted in the past sev

eral years from being initially fueled by heroin and opi-

oid analgesics to most recently synthetic opioids (e.g.,

fentanyl; [21]). As such, the objective of our study was

to replicate, expand, and update the ﬁndings of our pilot

study [5] by examining neurocognitive, psychiatric, and

substance use characteristics of a larger, more diverse

group of persons with OUD who were initiating OAT

with either methadone or buprenorphine/naloxone. To

achieve this objective, we analyzed baseline data from a

larger randomized trial examining how methadone and

buprenorphine/naloxone aﬀect neurocognitive function

ing. We expected high rates of neurocognitive impair-

ment (i.e., in domains of learning, memory, attention/

working memory, and executive functioning), depres

sion, and comorbid substance use. We hypothesized that

persons with OUD who had lifetime diagnoses of alcohol

or cocaine use disorder would have worse neurocogni

tive functioning than those without either condition.

Although Arias and colleagues [5] found that lifetime

major depression was not associated with neurocognitive

functioning in persons with OUD, based on previous lit

erature [22] we also hypothesized that depressed persons

with OUD would have worse neurocognitive functioning

than non-depressed persons with OUD.

Methods

Participants

e sample included 97 English-speaking adults with

opioid use disorder (OUD) starting opioid agonist treat

ment (OAT) with either methadone or buprenorphine/

naloxone. Participants were recruited from and assessed

Keywords: Opioid use disorder, Methadone, Buprenorphine/naloxone, Learning and memory, Depression, Comorbid

substance use

Page 3 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

in three clinics that together comprise the Einstein/Mon-

teﬁore Division of Substance Abuse (DoSA) in the Bronx,

New York. Recruitment was both active (e.g., approach

ing new patients at DoSA clinics) and passive (e.g.,

patients self-referred after seeing ﬂyers posted in DoSA

clinics, hearing about the study through word-of-mouth,

or seeing a study advertisement in a local newspaper).

Participants were enrolled in a randomized trial compar

ing neurocognitive outcomes among persons with OUD

who were initiating methadone versus buprenorphine/

naloxone; the analysis presented here is from the baseline

evaluation.

Eligibility criteria for the randomized trial included: (1)

diagnosis of an OUD without pharmacological treatment

for OUD within the previous 90days, (2) no current use

of street methadone or buprenorphine/naloxone and no

current prescription for either of these medications, (3)

English-speaking, (4) between the ages of 18 and 68, (5)

completed six or more years of education, and (6) able to

provide informed consent (e.g., not acutely intoxicated at

time of enrollment). We excluded persons with comor

bid illness likely to impact neurocognitive functioning,

including medical (i.e., liver disease, cardiovascular dis

ease, oxygen-requiring lung disease, or end stage renal

disease), neurological (i.e., history of head injury with

loss of consciousness > 24h, focal brain lesion, prior neu

rosurgery, non-alcohol related seizure disorder, or history

of non-HIV CNS opportunistic infection), or psychiatric

comorbidity other than major depression (i.e., schizo

phrenia or bipolar disorder). e trial was approved by

the Institutional Review Boards of both Albert Einstein

College of Medicine/Monteﬁore Medical Center and

Fordham University.

Measures

Neurocognitive functioning

Table 1 summarizes the comprehensive, standardized

neurocognitive test battery participants completed,

including tests in the following seven domains: executive

functioning, learning, memory, attention/working mem

ory, processing speed, motor abilities, and verbal ﬂuency.

Participants also completed the Wide Range Achieve

ment Test-Reading subtest, 3rd edition (WRAT-3; [23])

as a measure of premorbid intellectual functioning.

e battery was administered and scored by trained

psychometricians, supervised by a board-certiﬁed neu

ropsychologist and following standardized procedures,

and all measures have excellent reliability and validity

[24]. Raw test scores were ﬁrst converted to T-scores

based on the best available demographically corrected

normative data [25–30]. Next, T-scores for each individ

ual test were averaged to create mean domain T-scores.

en, all individual test T-scores were averaged to

create a mean global T-score [31]. Consistent with prior

research, we considered global neurocognitive function

ing and neurocognitive domain T-scores < 40 as impaired

(i.e., 35–39 mildly impaired, 30–34 mild-moderately

impaired, 25–29 moderately impaired, 20–24 moderate-

severely impaired, and < 20 severely impaired), T-scores

between 40 and 44 as “below average,” and T-scores

between 45 and 54 as “average” [31].

Depression

We used the Beck Depression Inventory–II (BDI-II)

to assess for current depressive symptomatology. e

BDI-II is a 21 item self-report scale assessing symptoms

of depression over the past 2 weeks [32]. Each of the

21 items contains at least four statements about spe

ciﬁc symptoms of depression (e.g., sadness, self-dislike,

worthlessness, loss of energy), listed in order of severity

from 0 to 3 with a total score ranging from 0 to 63. Sever

ity was deﬁned as: 0–13 for minimal depression, 14–19

for mild depression, 20–28 for moderate depression, and

29–63 for severe depression. We used the computerized

Composite International Diagnostic Interview (CIDI)

Table 1 Neuropsychological battery and normative data by

seven major neurocognitive domains

Wechsler Adult Intelligence Scales (WAIS); Paced Auditory Serial Arithmetic Test

(PASAT); Normative data corrects for the demographic characteristics indicated

by superscript:

Age;

Education;

Gender;

Ethnicity

Neuropsychological Domain and Test Normative

Data

Sources

Executive functioning

Wisconsin Card Sorting Task-64 Item [29]

a,b

Trail Making Test (Part B) [28]

a,b,c,d

Learning

Hopkins Verbal Learning Test—Revised [26]

Brief Visuospatial Memory Test—Revised [25]

Memory

Hopkins Verbal Learning Test—Revised [26]

Brief Visuospatial Memory Test—Revised [25]

Attention/Working Memory

WAIS-IV Letter Number Sequencing [30]

a,b,c,d

PASAT-50 Total Correct [27]

a,b,d

Speed of Information Processing

WAIS-IV Coding [30]

a,b,c,d

WAIS-IV Symbol Search [30]

a,b,c,d

[28]

a,b,c,d

Trail Making Test (Part A)

Motor Skills

Grooved Pegboard Time [28]

b,c,d

Verbal Fluency

Controlled Oral Word Association Test (FAS) [28]

a,b,c,d

Semantic (Animal) Fluency [28]

a,b,c,d

Page 4 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

Version 2.1 to assess for lifetime (i.e., current or past)

major depressive disorder. e CIDI provides diagnostic

information based on Diagnostic and Statistical Manual

of Mental Disorders (DSM)-IV criteria [33].

Substance use

We also used the CIDI Version 2.1 to assess for life-

time history (i.e., current or past) of substance use dis-

orders [33]. To provide updated DSM-5 nosology about

substance use disorders, we deﬁned “use disorder” as

meeting criteria for either substance abuse or substance

dependence. We used the Addiction Severity Index (ASI;

[34]) to assess substance use during the 30 days prior

to the baseline evaluation. Speciﬁcally, patients were

asked how many of the past 30days they used heroin,

other opioids, cannabis, cocaine, amphetamines, alco

hol (including use to intoxication), and sedatives, and

patients who reported at least one day of use were con

sidered self-reported users. Additionally, urine was

collected (unobserved) to assess for use of opioids, can

nabinoids, cocaine, amphetamines, and benzodiazepines

on the day of the neurocognitive evaluation. ese urine

tests employed an Enzyme Multiplied Immunoassay

Technique (EMIT) analyzed at a commercial laboratory.

Statistical analyses

Statistical Package for the Social Sciences (SPSS) Ver-

sion 22.0 was used to analyze all results [35]. Descriptive

statistics were calculated to provide information means,

standard deviations, percentages, and ranges of relevant

demographic data along with neurocognitive, psychiatric,

and substance use data. Pearson correlations were used

to examine the relationship between current depres

sive symptomatology and neurocognitive outcomes. A

series of independent sample t-tests were then computed

to examine diﬀerences in neurocognitive functioning

between participants based on depression and substance

use disorder categories (e.g., cannabis, cocaine).

Results

Demographic characteristics

Table2 summarizes the sample’s demographic and clini-

cal characteristics. Almost one-third of the sample was

female, and the mean age was 42.2years (SD = 10.3) with

a mean of 11.4years of education (SD = 2.3). e major

ity (52%) was Latinx. e majority of Latinx participants

were of Caribbean heritage (Puerto Rican and Domini

can). e remainder of the sample was non-Hispanic

Black/African American (24%), non-Hispanic white

(22%), or of another background (2%).

Neurocognitive characteristics

Table3 summarizes neurocognitive characteristics of the

sample and prevalence rates of impairment. Estimated

premorbid intelligence for the overall sample was below

average (SS = 87.2, SD = 13.2). Consistent with this,

Table 2 Participant demographic and selected clinical

characteristics (N = 97)

n = 85. SR = Self-report use from Addiction Severity Index; UT = Urine

toxicology at baseline visit; BDI-II = Beck Depression Inventory – II;

CIDI = Composite International Diagnostic Interview–Version 2.1

M (SD) or % (n) Range

Demographic characteristics

Age, years 42.2 (10.3) 21–63

Education, years 11.4 (2.3) 6–18

Female 27% (26)

HIV + 6% (6)

Hepatitis C diagnosis

14% (14)

Race/ethnicity

Non-Hispanic White 22% (21)

Non-Hispanic Black/African-American 24% (23)

Latinx 52% (50)

Other 2% (3)

Opioid use characteristics

SR heroin in last 30 days 87% (79)

SR other opiates in last 30 days 33% (30)

UT opiates 98% (92)

Lifetime opioid use disorder 100% (97)

Cannabis use characteristics

SR cannabis in last 30 days 44% (40)

UT cannabis 25% (24)

Lifetime cannabis use disorder 44% (42)

Stimulant use characteristics

SR cocaine in last 30 days 40% (36)

UT cocaine 28% (27)

Lifetime cocaine use disorder 59% (57)

SR amphetamines in last 30 days 2% (2)

UT amphetamines 1% (1)

Sedative use characteristics

SR alcohol in last 30 days 41% (37)

SR alcohol to intoxication in last 30 days 24% (22)

Lifetime alcohol use disorder 54% (52)

SR sedatives in last 30 days 23% (21)

UT benzodiazepines 11% (11)

Lifetime sedative use disorder 28% (27)

Psychological characteristics

BDI-II Total Score 15.4 (10.6) 0–45

Minimal total score ≤ 13 50% (45)

Mild total score 14 and 19 17% (15)

Moderate total score 20 and 28 23% (22)

Severe total score 29 and 63 10% (9)

CIDI lifetime major depressive disorder 37% (35)

Page 5 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

global neurocognitive functioning T-score was below

average (M = 41.6, SD = 6.4). Individual neurocogni

tive domain T-scores ranged from mild-to-moderately

impaired for learning (M = 34.3, SD = 8.3) to average

performance in processing speed (M = 48.0, SD = 8.8).

Although all neurocognitive domain T-scores fell within

normal limits with the exception of for learning and

memory, neurocognitive impairments were still observed

in each domain. Over two-thirds of the sample scored

in the impaired range for learning (79% impaired) and

memory (69% impaired). e remaining prevalence of

neurocognitive impairment varied from 18% in speed of

information processing, to 42% in motor skills.

Psychiatric andsubstance use comorbidity

Approximately half of the sample endorsed depressive

symptomatology, with one-third of the sample endorsing

moderate to severe symptoms of depression, and 37% of

the sample meeting lifetime criteria for a major depres

sive disorder.

Lifetime substance use disorder prevalence rates were

as follows: 59% cocaine (n = 57), 54% alcohol (n = 52),

44% cannabis (n = 42), and 28% sedatives (n = 27). Dur

ing the 30days prior to the baseline visit, with the excep-

tion of high opioid use (e.g., 87% heroin use) and low

amphetamine use (i.e., 2%), self-reported substance use

ranged from 23% for sedatives to 44% for cannabis. Urine

toxicology results during the baseline visit revealed lower

current prevalence rates of these substances ranging

from 11% for benzodiazepines to 28% for cocaine (again

excluding opiates and amphetamines). Additionally,

50% of the sample tested positive for at least one non-

opiate substance (i.e., cannabis, cocaine, amphetamines,

benzodiazepines).

Associations betweenneurocognitive functioning

anddepressive symptoms

Current depressive symptomatology was signiﬁcantly

correlated with the global composite of the seven

domains (r = − 0.23, p = 0.03), while the correlations

with each of the seven individual domains ranged from

r = − 0.20, p = 0.05 for verbal ﬂuency to r = − 0.11,

p = 0.32 for memory.

Table 4 summarizes diﬀerences in the relationship

between lifetime major depressive disorder and neuro

cognitive functioning. Participants with a history of life-

time major depressive disorder had signiﬁcantly worse

functioning in two domains: attention/working memory

(t(94) = 2.72, p = 0.01) and motor skills (t(92) = 2.64,

p = 0.01) than those with no lifetime history of major

depressive disorder, with medium eﬀect sizes (Cohen’s

d’s = 0.57–58). Additionally, participants with a history of

lifetime major depressive disorder exhibited worse exec

utive functioning, processing speed, and overall global

neurocognitive functioning than those without a his

tory of lifetime major depressive disorder at trend levels

(p’s ≤ 0.10), with modest eﬀect sizes (Cohen’s d = 0.35 to

0.40).

Associations betweenneurocognitive functioning

andsubstance use disorders

Table 4 also summarizes diﬀerences in the relation-

ship between lifetime substance use disorder categories,

global neurocognitive functioning, and each of the seven

neurocognitive domains. Participants with a lifetime his

tory of cannabis use disorder had better executive func-

tioning t(94) = − 2.25, p = 0.03) than those without a

lifetime history of cannabis use disorder with a medium

eﬀect size (Cohen’s d = 0.46). Similarly, participants with

cannabis use disorder also performed better on atten

tion/working memory tasks t(94) = − 2.77, p = 0.007)

Table 3 Participant neurocognitive (NC) characteristics based on average T-scores and rates of impairment (N = 97)

T-scores between 45 and 54 = average; T-scores between 40 and 44 = below average; T-scores < 40 = overall impaired; T = 35–39 mildly impaired; T = 30–34 mild-

moderately impaired; T = 25–29 moderately impaired; T = 20–24 moderate-severely impaired; and T < 20 severely impaired; Mod = moderately

M (SD) Overall

impaired

% (n)

Mildly

impaired

% (n)

Mild-mod

impaired %

(n)

Moderately

impaired % (n)

Mod-severely

impaired % (n)

Severely

impaired

% (n)

WRAT-3 reading (standard score) 87.2 (13.2) – – – – – –

Global NC functioning 41.6 (6.4) 35% (34) 18% (17) 13% (13) 3% (3) 1% (1) 0% (0)

Executive functioning 43.6 (7.7) 35% (34) 24% (23) 9% (9) 2% (2) 0% (0) 0% (0)

Learning 34.3 (8.3) 79% (77) 34% (33) 19% (18) 12% (12) 7% (7) 7% (7)

Memory 35.5 (8.7) 69% (67) 22% (21) 21% (20) 14% (14) 7% (7) 5% (5)

Attention/working memory 43.0 (9.5) 41% (40) 24% (23) 11% (11) 3% (3) 3% (3) 0% (0)

Speed of information processing 48.0 (8.8) 18% (17) 9% (9) 6% (6) 1% (1) 0% (0) 1% (1)

Motor skills 42.0 (10.3) 42% (40) 17% (16) 14% (13) 5% (5) 5% (5) 1% (1)

Verbal ﬂuency 44.5 (9.9) 31% (30) 18% (17) 6% (6) 4% (4) 2% (2) 1% (1)

Page 6 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

Table 4 Mean comparisons based on lifetime cannabis use disorder, cocaine use disorder, and major depressive disorder with neurocognitive (NC) functioning (N = 97)

Mean and Standard Deviations (M, SD) based on average T scores; Diagnosis assessed using the Composite Diagnostic Interview (CIDI); T-scores between 45 and 54 = average; T-scores between 40 and 44 = below

average; T-scores < 40 = overall impaired; T = 35–39 mildly impaired; T = 30–34 mild-moderately impaired; T = 25–29 moderately impaired; T = 20–24 moderate-severely impaired; and T < 20 severely impaired

p < 0.10

p < 0.05

***

p < 0.01

Neurocognitive domain Diagnosis of cannabis use disorder Diagnosis of cocaine use disorder Diagnosis of major depressive disorder

Yes (n = 42) No (n = 54) Yes (n = 57) No (n = 39) Yes (n = 35) No (n = 61)

M (SD) M (SD) t-test Cohen’s d M (SD) M (SD) t-test Cohen’s d M (SD) M (SD) t-test Cohen’s d

Global NC functioning 42.7 (5.7) 40.7 (6.9) − 1.5 0.31 42.1 (6.7) 40.8 (6.1) − 1.0 0.20 39.9 (7.0) 42.5 (6.0) 1.9* 0.41

Executive functioning 45.6 (7.3) 42.1 (7.9) − 2.3** 0.46 44.9 (8.1) 41.7 (6.9) − 2.0* 0.42 41.6 (8.0) 44.7 (7.5) 1.9* 0.40

Learning 35.3 (9.1) 33.6 (7.3) − 1.0 0.21 33.6 (7.8) 35.4 (9.0) 1.0 0.22 33.2 (7.7) 35.0 (8.6) 1.0 0.22

Memory 36.4 (8.1) 34.7 (9.3) − 0.9 0.19 35.1 (8.5) 36.1 (9.3) 0.5 0.11 34.7 (8.7) 35.9 (8.9) 0.6 0.14

Attention/working memory 46.0 (9.9) 40.8 (8.6) − 2.8*** 0.57 45.3 (9.6) 39.8 (8.5) − 2.9*** 0.60 39.7 (9.2) 45.0 (9.2) 2.7*** 0.58

Speed of information processing 47.9 (7.4) 48.2 (10.0) 0.1 0.03 48.7 (9.3) 47.1 (8.3) − 0.9 0.18 46.0 (9.6) 49.2 (8.3) 1.7* 0.36

Motor skills 42.1 (9.5) 41.9 (11.0) − 0.1 0.02 41.2 (10.5) 43.2 (10.1) 0.9 0.19 38.4 (9.4) 44.1 (10.3) 2.6** 0.57

Verbal ﬂuency 45.7 (9.1) 43.8 (10.5) − 1.5 0.19 46.0 (10.7) 42.6 (8.3) − 1.7* 0.35 46.0 (10.9) 43.9 (9.3) − 1.0 0.21

Page 7 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

with a medium eﬀect size (Cohen’s d = 0.56). ere were

no other diﬀerences between participants with and with

out a history of cannabis use disorder globally or in any

other neurocognitive domain.

Participants with a lifetime history of cocaine use dis

order outperformed those with no lifetime history of

cocaine use disorder on attention/working memory

tasks t(94) = − 2.90, p = 0.005) with a medium eﬀect

size (Cohen’s d = 0.61). Additionally, participants with a

lifetime history of cocaine use disorder had better execu

tive functioning t(94) = -2.00, p = 0.06) and verbal ﬂuency

t(94) = − 1.69, p = 0.10) at trend levels with medium

eﬀect sizes (Cohen’s d = 0.36 to 0.41). ere were also no

diﬀerences between participants with and without a his

tory of cocaine use disorder globally or in any other neu-

rocognitive domain.

Participants with or without a lifetime history of alco

hol use disorder or sedative use disorder also did not dif-

fer on any neurocognitive domain.

Discussion

We found that the vast majority of our diverse sample of

OUD patients who were initiating OAT with methadone

or buprenorphine/naloxone were impaired in domains of

learning and memory. Major depressive disorder was also

common, and half the sample reported current depres

sive symptomatology. Lifetime prevalence of substance

use disorders were high with over half of the sample

reporting a lifetime history of alcohol or cocaine use dis

order, 44% reporting lifetime cannabis use disorder, and

28% reporting lifetime sedative use disorder. As expected,

current depressive symptoms and a lifetime history of

major depressive disorder were both negatively related

to speciﬁc domains of neurocognitive functioning (e.g.,

attention/working memory, motor skills). However, con

trary to our hypothesis, we did not ﬁnd a negative rela-

tionship between lifetime diagnoses of alcohol or cocaine

use disorder and neurocognitive functioning. Instead,

we found no diﬀerences in neurocognitive function

ing between those with and without a history of alcohol

use disorder. Moreover, we found that attention/working

memory was signiﬁcantly better in individuals with a his

tory of cocaine or cannabis use disorder, and executive

functioning was signiﬁcantly better in individuals with a

history of cannabis use disorder.

We found a similar prevalence of learning and mem

ory impairment and impairment in other neurocogni-

tive domains as previous studies [5–9, 36]. For instance,

compared to Arias and colleagues [5] we found the fol

lowing impairment rates: learning: 79% present study

vs. 73% Arias and colleagues [5], memory: 69% present

study vs. 68% Arias and colleagues [5], attention/working

memory: 41% present study vs. 36% Arias and colleagues

[5], and executive functioning 35% present study vs. 44%

Arias and colleagues [5]. e especially high prevalence

of learning and memory impairment in persons with

OUD who are initiating OAT is not surprising given that

chronic opioid use decreases temporal lobe gray matter

density and results in corresponding decreases in hip

pocampal neurogenesis [37, 38]. Additionally, unlike

other neurocognitive domains, learning and memory

impairment appears to be independent of common

comorbidities (i.e., depression and substance use disor

ders) in this sample.

Similar to previous studies, we found that depression

was common in OUD patients [5, 17, 22]. Similar to

Loeber and colleagues [22], our study found that depres

sion (i.e., lifetime major depressive disorder and current

depressive symptomatology) negatively impacted neu

rocognitive performance (i.e., globally and domains of

attention/working memory and motor skills), but other

studies have not found a relationship between current

or past depression and cognitive functioning [5, 12] in

OUD patients. In contrast to our study, Sanborn and

colleagues [12] investigated this relationship in patients

already taking methadone (rather than in patients start

ing OAT). Arias and colleagues [5] studied a small sam-

ple and may not have had suﬃcient power to detect an

impact of depression on cognitive functioning, though

they did report a negative but non-signiﬁcant relation

ship between motor skills and lifetime history of major

depressive disorder with a similar eﬀect size (i.e., Cohen’s

d = 0.70) as we found. Because we found a signiﬁcant

negative relationship for both current and lifetime

depression and neurocognitive outcomes in persons with

OUD who are starting OAT, our ﬁndings have increased

generalizability compared to past studies. However, given

the simple bivariate relationship between depression and

cognition in the present study, future research should

clarify the speciﬁc role of current and lifetime history of

depression in negatively impacting neurocognitive per

formance in OUD patients.

In contrast to Arias and colleagues [5], results of the

present study revealed that cocaine and cannabis use

disorder diagnoses were positively associated with neu

rocognitive functioning in speciﬁc domains. It is possible

that the diﬀerence in diagnostic criteria (DSM-IV in Arias

vs. DSM-5 in present study), along with the increased

power via larger sample size, might partially explain this

diﬀerence in ﬁndings. e positive relationship between

attention/working memory and cocaine use is some

what consistent with previous research [39]. Speciﬁcally,

Byrd and colleagues [39] found a positive association

between positive urine toxicology results for cocaine and

attention/working memory in a sample of HIV ± poly

substance users, including current opioid users. ese

Page 8 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

converging ﬁndings may be due to the “neuro-activat-

ing” properties of cocaine that could result in enhanced

attentional performance. Future research should explore

whether potential neuro-activating properties of cocaine

in polysubstance users persists overtime as in those with

a lifetime history of cocaine use disorder. e current

study’s ﬁnding regarding the positive association between

lifetime cannabis use disorder and cognitive functioning

is both consistent [40] and inconsistent [41] with limited

prior literature. For instance, Gruber and colleagues [40]

reported improved executive functioning performance

over time in medical marijuana users. Ultimately, our

results appear to indicate that prior cocaine or cannabis

use disorder do not confer a risk for poor neurocognitive

functioning in patients entering OAT.

is study has several important implications and

strengths. First, by characterizing the prevalence of neu

rocognitive impairment, depression, and other substance

use in persons with OUD who are starting OAT, our

results provide useful information to treatment providers.

is knowledge might allow clinicians to help patients

taking OAT develop compensatory strategies for poor

neurocognitive functioning (e.g., learning and memory),

while providing adjunctive treatment for depression. Sec

ond, our study was able to successfully replicate the prev-

alence of neurocognitive impairment, depression, and

comorbid substance use reported in our previous study

by Arias and colleagues [5]. We also were able to provide

the additional power necessary to detect a statistically

signiﬁcant association between depression and neuro

cognitive functioning, while uncovering that OAT-initiat-

ing persons with a lifetime history of comorbid substance

use disorders did not have greater neurocognitive impair

ments than those without. Although our ﬁndings must

still be replicated in diﬀerent patient populations and

in diﬀerent locations, we are conﬁdent in the validity of

these results. ird, we used recent and well-validated

measures of neurocognitive functioning, depression, and

substance use, which addresses important limitations of

some prior studies on this topic (e.g., [5, 42, 43]). Fourth

and ﬁnally, the high racial/ethnic diversity of the sample

is both a strength and a limitation. e present study ﬁlls

an important gap in the currently sparse literature on

characteristics of racial/ethnic minority patients enter

ing OAT. However, our predominantly Latinx sample of

Caribbean heritage (Puerto Rican and Dominican) may

not generalize to persons from other racial/ethnic groups

and/or regions of the U.S. [44, 45]. Future studies should

replicate these ﬁndings in other racial/ethnic groups (e.g.,

non-Hispanic white, Asian American, American Indian/

Alaska Native) and settings (e.g., rural, suburban).

Despite these strengths, our study has several limita

tions. e cross-sectional nature of our research design

limits the generalizability of our results. Speciﬁcally,

these ﬁndings only apply to patients early in OAT (i.e.,

within the ﬁrst 14days) and do not account for individu

als who will drop out of treatment. erefore, it remains

to be seen if these high prevalence rates of neurocogni

tive impairment, depression, and other substance use are

present in patients who remain in treatment long-term.

Future longitudinal studies should be conducted to pro

vide characteristics of OAT patients over time. Finally,

because we had no healthy control group, it is unknown

if our ﬁndings are applicable to OUD patients in general

or just those seeking OAT. Future studies might incor

porate a non-treated control group to determine if these

ﬁndings are directly related to initiating OAT.

Conclusions

We found high prevalence rates of neurocognitive

impairment, depression, and other substance use among

diverse persons with OUD who were starting OAT. e

vast majority of these patients exhibited impairments

in learning and memory, and current depression and

substance use, especially cannabis, cocaine, and alco

hol, were common. Depressed patients were especially

likely to have neurocognitive impairments, but patients

with a lifetime history of either cannabis or cocaine use

disorder had no worse neurocognitive functioning than

those without. Treatment providers should be aware that

patients starting OAT may present with neurocognitive

and psychiatric complications, and depression in these

patients might be associated with especially poor neu

rocognitive outcomes. is knowledge might allow clini-

cians to help patients taking OAT develop compensatory

strategies for poor neurocognitive functioning (e.g.,

learning and memory), while providing adjunctive treat

ment for depression.

Abbreviations

OAT: Opioid agonist treatment; OUD: Opioid use disorder; OTP: Opioid treat-

ment program; U.S.: United States; DoSA: Division of substance abuse; WRAT-3:

Wide Range Achievement Test-Reading subtest, 3rd edition; BDI-II: Beck

depression inventory–II; CIDI: Composite International Diagnostic Interview;

DSM: Diagnostic and Statistical Manual of Mental Disorders; ASI: Addiction

Severity Index; EMIT: Enzyme Multiplied Immunoassay Technique; SPSS: Statis-

tical Package for the Social Sciences.

Acknowledgements

The authors acknowledge and thank Yuming Ning and Aprille Mangalonzo for

expert technical assistance, and also the patients who generously gave their

time to this study.

Authors’ contributions

All authors contributed to the preparation of this manuscript. MRM, CO, and

JA contributed to study design and implementation. Data collection was

performed by FA, JO, and TMS. Data analysis was performed by TMS and MRM.

All authors have read and approved the ﬁnal manuscript.

Page 9 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

Funding

This research was funded by NIH R01DA032552; Co-PIs: M.R.M., PhD, ABPP-CN

& J.A., MD, MPH.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from

the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The trial was approved by the Institutional Review Boards of both Albert Ein-

stein College of Medicine/Monteﬁore Medical Center and Fordham University.

Informed consent was collected for all participants in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Author details

VA Palo Alto Health Care System Sierra Paciﬁc Mental Illness Research Educa-

tion Clinical Center, 3801 Miranda Ave, Palo Alto, CA 94304, USA.

Department

of Psychiatry and Behavioral Sciences, Stanford School of Medicine, Stanford,

CA, USA.

Department of Medicine, Albert Einstein College of Medicine

and Monteﬁore Medical Center, Bronx, NY, USA.

Department of Neurology,

North Shore University Hospital, Manhasset, NY, USA.

The Aging Brain Center,

Hebrew SeniorLife, Boston, MA, USA.

Department of Cognitive Neurology,

Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA,

USA.

Department of Psychology and Latin America and Latino Studies

Institute, Fordham University, New York, NY, USA.

Department of Neurology,

Icahn School of Medicine at Mount Sinai, New York, NY, USA.

Received: 19 May 2021 Accepted: 12 October 2021

References

1. Mattick RP, Breen C, Kimber J, Davoli M. Methadone maintenance therapy

versus no opioid replacement therapy for opioid dependence. Cochrane

Database Syst Rev. 2009. https:// doi. org/ 10. 1002/ 14651 858. CD002 209.

pub2.

2. Ducharme LJ, Abraham AJ. State policy inﬂuence on the early diﬀusion

of buprenorphine in community treatment programs. Subst Abuse Treat

Prev Policy. 2008;3(1):17.

3. Domínguez-Salas S, Díaz-Batanero C, Lozano-Rojas OM, Verdejo-García A.

Impact of general cognition and executive function deﬁcits on addiction

treatment outcomes: systematic review and discussion of neurocognitive

pathways. Neurosci Biobehav Rev. 2016;71:772–801.

4. Passetti F, Clark L, Mehta MA, Joyce E, King M. Neuropsychological

predictors of clinical outcome in opiate addiction. Drug Alcohol Depend.

2008;94(1–3):82–91.

5. Arias F, Arnsten JH, Cunningham CO, Coulehan K, Batchelder A, Brisbane

M, et al. Neurocognitive, psychiatric, and substance use characteristics in

opioid dependent adults. Addict Behav. 2016;60:137–43.

6. Baldacchino A, Balfour DJK, Passetti F, Humphris G, Matthews K. Neu-

ropsychological consequences of chronic opioid use: a quantitative

review and meta-analysis. Neurosci Biobehav Rev. 2012;36(9):2056–68.

7. Ersche KD, Sahakian BJ. The neuropsychology of amphetamine and

opiate dependence: implications for treatment. Neuropsychol Rev.

2007;17(3):317–36.

8. Fernández-Serrano MJ, Pérez-García M, Verdejo-García A. What are the

speciﬁc vs. generalized eﬀects of drugs of abuse on neuropsychological

performance? Neurosci Biobehav Rev. 2011;35(3):377–406.

9. Gruber SA, Silveri MM, Yurgelun-Todd DA. Neuropsychological conse-

quences of opiate use. Neuropsychol Rev. 2007;17(3):299–315.

10. Lee TMC, Pau CWH. Impulse control diﬀerences between abstinent

heroin users and matched controls. Brain Inj. 2002;16(10):885–9.

11. Verdejo-Garcia A, Perez-Garcia M. Proﬁle of executive deﬁcits in cocaine

and heroin polysubstance users: common and diﬀerential eﬀects on sep-

arate executive components. Psychopharmacology. 2007;190(4):517–30.

12. Sanborn V, Gunstad J, Shrestha R, Mistler CB, Copenhaver MM. Cognitive

proﬁles in persons with opioid use disorder enrolled in methadone treat-

ment. Appl Neuropsychol Adult. 2020. https:// doi. org/ 10. 1080/ 23279 095.

2020. 17690 99.

13. Saroj R, Ghosh A, Subodh BN, Nehra R, Mahintamani T, Rana D, et al.

Neurocognitive functions in patients on buprenorphine maintenance

for opioid dependence: a comparative study with three matched control

groups. Asian J Psychiatr. 2020. https:// doi. org/ 10. 1016/j. ajp. 2020. 102181.

14. Mintzer MZ, Correia CJ, Strain EC. A dose-eﬀect study of repeated

administration of buprenorphine/naloxone on performance in opioid-

dependent volunteers. Drug Alcohol Depend. 2004;74(2):205–9.

15. Betts KS, Chan G, McIlwraith F, Dietze P, Whittaker E, Burns L, et al. Dif-

ferences in polysubstance use patterns and drug-related outcomes

between people who inject drugs receiving and not receiving opioid

substitution therapies. Addiction. 2016;111(7):1214–23.

16. Havard A, Teesson M, Darke S, Ross J. Depression among heroin users:

12-Month outcomes from the Australian Treatment Outcome Study

(ATOS). J Subst Abuse Treat. 2006;30(4):355–62.

17. Goldner EM, Lusted A, Roerecke M, Rehm J, Fischer B. Prevalence of Axis-1

psychiatric (with focus on depression and anxiety) disorder and symp-

tomatology among non-medical prescription opioid users in substance

use treatment: systematic review and meta-analyses. Addict Behav.

2014;39(3):520–31.

18. Agrawal A, Lynskey MT, Madden PAF, Bucholz KK, Heath AC. A latent

class analysis of illicit drug abuse/dependence: results from the national

epidemiological survey on alcohol and related conditions. Addiction.

2006;102(1):94–104.

19. Krausz M, Degkwitz P, Kühne A, Verthein U. Comorbidity of opiate

dependence and mental disorders. Addict Behav. 1998;23(6):767–83.

20. Griﬃn ML, Dodd DR, Potter JS, Rice LS, Dickinson W, Sparenborg S, et al.

Baseline characteristics and treatment outcomes in prescription opioid

dependent patients with and without co-occurring psychiatric disorder.

Am J Drug Alcohol Abuse. 2014;40(2):157–62.

21. O’Donnell JK, Halpin J, Mattson CL, Goldberger BA, Gladden RM. Deaths

involving fentanyl, fentanyl analogs, and U-47700-10 states, July–Decem-

ber 2016. MMWR Morb Mortal Wkly Rep. 2017;66(43):1197–202.

22. Loeber S, Nakovics H, Kniest A, Kiefer F, Mann K, Croissant B. Factors

aﬀecting cognitive function of opiate-dependent patients. Drug Alcohol

Depend. 2012;120(1–3):81–7.

23. Wilkinson GS. Wide range achievement test 3—administration manual.

Wilimington: Jastak Associates Inc; 1993.

24. Strauss E, Sherman EMS, Spreen O. A compendium of neuropsychologi-

cal tests: administration, norms, and commentary. Washington: American

Chemical Society; 2006.

25. Benedict RHB, Schretlen D, Groninger L, Dobraski M, Shpritz B. Revision

of the brief visuospatial memory test: studies of normal performance,

reliability, and validity. Psychol Assess. 1996;8(2):145–53.

26. Benedict RHB, Schretlen D, Groninger L, Brandt J. Hopkins Verbal Learning

Test–revised: normative data and analysis of inter-form and test-retest

reliability. Clin Neuropsychol. 1998;12(1):43–55.

27. Diehr MC, Cherner M, Wolfson TJ, Miller SW, Grant I, Heaton RK. The 50

and 100-item short forms of the Paced Auditory Serial Addition Task

(PASAT): demographically corrected norms and comparisons with the

full PASAT in normal and clinical samples. J Clin Exp Neuropsychol.

2003;25(4):571–85.

28. Heaton RK, Miller SW, Taylor MJ, Grant I. Revised comprehensive norms

for an expanded Halstead-Reitan battery: demographically adjusted

neuropsychological norms for African American and Caucasian adults.

Lutz: Psychological Assessment Resources; 2004.

29. Kongs SK, Thompson LL, Iverson GL, Heaton RK. Wisconsin card sorting

test-64 card version professional manual. Odessa: Psychological Assess-

ment Resources; 2000.

30. Wechsler D. Advanced clinical solutions for the WAIS-IV and WMS-IV. San

Antonio: Pearson; 2008.

31. Woods SP, Rippeth JD, Frol AB, Levy JK, Ryan E, Soukup VM, et al. Interrater

reliability of clinical ratings and neurocognitive diagnoses in HIV. J Clin

Exp Neuropsychol. 2004;26(6):759–78.

Page 10 of 10

Scottetal. Addict Sci Clin Pract (2021) 16:64

•

fast, convenient online submission

•

thorough peer review by experienced researchers in your ﬁeld

•

rapid publication on acceptance

•

support for research data, including large and complex data types

•

gold Open Access which fosters wider collaboration and increased citations

maximum visibility for your research: over 100M website views per year

•

At BMC, research is always in progress.

Learn more biomedcentral.com/submissions

Ready to submit your research

? Choose BMC and benefit from:

32. Beck AT, Steer RA, Brown GK. Beck depression inventory-II. San Antonio:

Psychological Corporation; 1996.

33. World Health Organization. CIDI-Auto Version 2.1: administrator’s guide

and reference. Sydney: Training and reference centre for WHO CIDI; 1997.

34. McLellan AT, Kushner H, Metzger D, Peters R, Smith I, Grissom G, et al.

The ﬁfth edition of the Addiction Severity Index. J Subst Abuse Treat.

1992;9(3):199–213.

35. SPSS. IBM SPSS Statistics Version 21. Boston: Mass Int Bus Mach Corp;

2012.

36. Darke S, Ross J. Polydrug dependence and psychiatric comorbidity

among heroin injectors. Drug Alcohol Depend. 1997;48(2):135–41.

37. Eisch AJ, Harburg GC. Opiates, psychostimulants, and adult hippocampal

neurogenesis: insights for addiction and stem cell biology. Hippocampus.

2006;16(3):271–86.

38. Lyoo IK, Pollack MH, Silveri MM, Kyung HA, Diaz CI, Hwang J, et al. Prefron-

tal and temporal gray matter density decreases in opiate dependence.

Psychopharmacology. 2006;184(2):139–44.

39. Byrd DA, Robinson-Papp J, Mindt MR, Mintz L, Elliott K, Lighty Q, et al.

Isolating cognitive and neurologic HIV eﬀects in substance-dependent,

confounded cohorts: a pilot study. J Int Neuropsychol Soc. 2013. https://

doi. org/ 10. 1017/ S1355 61771 20016 34.

40. Gruber SA, Sagar KA, Dahlgren MK, Racine MT, Smith RT, Lukas SE. Splen-

dor in the grass? A pilot study assessing the impact of medical marijuana

on executive function. Front Pharmacol. 2016;7:355.

41. Crean RD, Crane NA, Mason BJ. An evidence based review of acute and

long-term eﬀects of cannabis use on executive cognitive functions. J

Addict Med. 2011;5(1):1.

42. Gruber SA, Tzilos GK, Silveri MM, Pollack M, Renshaw PF, Kaufman MJ, et al.

Methadone maintenance improves cognitive performance after two

months of treatment. Exp Clin Psychopharmacol. 2006;14(2):157–64.

43. Rapeli P, Fabritius C, Alho H, Salaspuro M, Wahlbeck K, Kalska H. Metha-

done vs. buprenorphine/naloxone during early opioid substitution

treatment: a naturalistic comparison of cognitive performance relative to

healthy controls. BMC Clin Pharmacol. 2007;7:5.

44. Jones CM, Logan J, Gladden RM, Bohm MK. Vital signs: demographic and

substance use trends among heroin users—United States, 2002–2013.

MMWR Morb Mortal Wkly Rep. 2015;64(26):1–7.

45. Substance Abuse and Mental Health Services Administration. Treatment

Episode Data Set (TEDS): 2004–2014. Rockville: National admissions to

substance abuse treatment services; 2016.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in pub-

lished maps and institutional aﬃliations.