Testing the feasibility, acceptability, and preliminary efficacy of integrating accommodation strategies into an HIV prevention intervention for people who inject drugs with cognitive dysfunction

Abstract

Background:

Cognitive profiles of individuals with opioid use disorder (OUD) limit patients’ ability to learn, retain, and recall HIV prevention information. It also limits adherence to medications, such as pre-exposure prophylaxis (PrEP). Cognitive dysfunction accommodation strategies have shown promise at reducing HIV-related risk behaviors among individuals with OUD and increasing adherence to PrEP. This study investigated the feasibility, acceptability, and preliminary efficacy of integrating accommodation strategies into a behavioral HIV prevention intervention.

Methods:

This 2-arm single blind study provided 50 people who inject drugs (PWID) with OUD linkage to PrEP services and randomized them to a 4-week HIV prevention intervention condition. The active control condition received the HIV prevention intervention as treatment per usual, while the experimental condition received the enhanced HIV prevention intervention with added accommodation strategies. Participants completed acceptability ratings of intervention content and accommodation strategies post-intervention; feasibility was measured via participant recruitment and retention. HIV risk reduction information, motivation, and behavior (IMB) assessments and HIV risk reduction skills assessments were completed pre/post-intervention. Participants also completed weekly PrEP adherence assessments.

Results:

The intervention content received a high acceptability rating (89%). Intervention feasibility was deemed acceptable, with 80% of participants completing all study protocols. The accommodation strategies integrated into the HIV prevention sessions were also endorsed by 92% of participants. Participants in the experimental condition had significant increases in retention and recall of how to perform HIV risk reduction skills including how to properly clean a syringe (p=0.048) and how to accurately apply a female condom (p=0.025), compared to the control condition. Weekly PrEP adherence was reported by the three (7.5%) participants who indicated taking PrEP throughout the study. All three participants reported missing doses throughout each of the 4 weeks.

Conclusions:

Results from this study highlight the potential for integrating accommodation strategies into behavioral HIV prevention interventions to reduce the risk of HIV among PWID. Future research is needed to evaluate the use of such strategies by larger and diverse samples of PWID, as well as whether accommodation strategies enable the retention and recall of HIV prevention information and HIV prevention skills over longer periods of time.

Trial registration:

This trial has been retrospectively registered at ClinicalTrials.gov on June 12, 2023. (NCT05912374)

1. Introduction

The increase in injection drug use, associated with the opioid epidemic in the United States, has halted HIV prevention progress, as one in ten new HIV diagnosis are attributed to injection drug use (Centers for Disease Control and Prevention, 2022). Evidence-based HIV prevention interventions for people with opioid use disorder (OUD) generally consist of behavioral educational sessions to increase risk perception and HIV prevention knowledge, while reducing HIV risk behaviors (Mathers et al., 2010). As of 2013, a bio-behavioral approach including the promotion of pre-exposure prophylaxis (PrEP) is often integrated into these HIV prevention interventions, to maximize efficacy (Choopanya et al., 2013; Meyer et al., 2022; Mistler et al., 2020; Shrestha et al., 2019). These intervention sessions, by nature, require participants to have intact cognitive functioning to pay attention for long periods of time and retain the information presented to them. However, chronic use of opioids change the brain chemistry, leading to increased risk for cognitive dysfunction (Mistler et al., 2021).

In a recent systematic review, Winstanley et al. (2021) discussed how opioid-related overdose often causes cerebral hypoxia (oxygen deprivation in the brain) leading to a variety of cognitive deficits. The prevalence of cognitive dysfunction among persons receiving medication for OUD (MOUD), many of whom have experienced at least one overdose, is drastically high (Centers for Disease Control and Prevention, 2020). A recent study showed that 67% of people receiving MOUD experience cognitive dysfunction (Copenhaver et al., 2020). Cognitive dysfunction in people with OUD can lead to increased HIV risk behaviors, increased risk of acquiring HIV, and decreased adherence to medications (Buckingham et al., 2013; Huedo-Medina et al., 2016). While the prevalence of cognitive dysfunction among people with OUD is disproportionately higher than in general populations, this area of research is underdeveloped regarding HIV prevention among people with OUD (Bruijnen et al., 2019; Shrestha & Copenhaver, 2016). As over 80% of people who inject drugs (PWID) specifically inject opioids, innovative studies are needed to research effective methods on tailoring HIV prevention interventions to PWID with OUD (Degenhardt et al., 2017).

People with OUD often demonstrate cognitive dysfunction in specific domains, similar to people with attention deficit hyperactivity disorder (ADHD) and traumatic brain injury (TBI); these domains include deficits in executive functioning, attention, memory, and information processing (Gunstad et al., 2022; Mistler et al., 2021; Sanborn et al., 2020). Behavioral interventions for people with ADHD and TBI often include accommodation strategies that increase participants’ ability to pay attention, retain and recall information, and apply skills in daily activities (Mistler et al., 2021). These accommodation strategies include a structured session with a written agenda, a multi-modal presentation of information, hands-on activities, meditation, memory aids, and immediate feedback (Gunstad et al., 2022; Mistler et al., 2021; Sanborn et al., 2020). While effective, these accommodation strategies have not intentionally been integrated into behavioral interventions for people with OUD.

To inform the current study, following the ADAPT-ITT Model (Wingood & DiClemente, 2008) of adapting interventions, we previously conducted focus groups with key stakeholders to gain insights on how to adapt an evidence-based HIV prevention intervention: Community-Based Community-friendly Health Recovery Program: Biobehavioral (CHRP-BB) (Mistler et al., 2022; Shrestha et al., 2019). Drug treatment counselors and people receiving MOUD provided insights and endorsed cognitive dysfunction accommodation strategies to further test among people with OUD. These strategies were then implemented into a two-session, 2-arm study with participants on medication for OUD; participants in the experimental condition (that received the accommodation strategies) demonstrated greater increases in retention and recall of HIV prevention knowledge and skills (Mistler & Copenhaver, 2022). These promising results informed our current study and supported the use of integrating cognitive dysfunction accommodation strategies into behavioral interventions to reduce HIV risk and HIV transmission among this population (Mistler & Copenhaver, 2022).

For the current study, we integrated accommodation strategies into a 4-week behavioral HIV prevention session to test the feasibility, acceptability, and preliminary efficacy of these strategies among PWID with cognitive dysfunction. We hypothesize that participants randomized to the experimental condition will demonstrate greater increases in HIV risk reduction skills, information, motivation, and behaviors pre/post-intervention compared to the control condition.

2. Materials and Methods

2.1. Recruitment and Sample

The study recruited 50 people who inject opioids via purposive sampling in Hartford, CT, through a needle exchange van and a social service non-profit organization. In addition to recruitment flyers, participants recruited other eligible people through word of mouth. A sample size of n=50 is appropriate for HIV pilot trials (Biello et al., 2019; Mayer et al., 2017; Shrestha et al., 2018). Screening consisted of a 2-step process. Initial screening criteria required participants to: a) be at least 18 years or older; b) have injected drugs in the past 30 days; c) have engaged in drug- and/or sex-related HIV risk behaviors; d) self-report being HIV negative, e) be willing to initiate PrEP use; f) be able to communicate (read, write and speak) in English. The study excluded participants if they reported: a) being unable to attend all four intervention sessions; and b) being actively suicidal and/or homicidal. Rolling recruitment ensured that the study randomized at least 6 participants to each condition and administered the sessions in a group format.

Once individuals completed an initial screening form to determine preliminary inclusion criteria, they completed the Brief Inventory of Neurocognitive Impairment (BINI) to determine if they experienced mild/moderate cognitive dysfunction. If a participant had a BINI score indicating mild/moderate cognitive dysfunction, they were invited to be a participant in the study. Participants received a $5 gift card if they completed the BINI form during recruitment. Linkage to PrEP services was provided for participants throughout the study duration; PrEP prescription was not required for participants to be enrolled in the trial.

2.2. Study Design

A 2-arm HIV prevention intervention determined if cognitive dysfunction accommodation strategies have the efficacy to increase retention of knowledge and application of skills over time. The study blindly assigned participants to one of two different HIV prevention intervention conditions. The active control condition received the HIV prevention intervention content via treatment as usual (lecture-style), while the experimental condition received the enhanced HIV prevention intervention content with added accommodation strategies.

During the initial intake session, participants completed self-report surveys to collect data on their demographics, drug- and sex-related HIV risk behaviors, and HIV risk reduction information, motivation, and behaviors skills (IMB) measures using Qualtrics survey software (Qualtrics, 2021). Participants also completed a risk reduction skills checklist at the intake session. Participants were then randomized to one of the two conditions; age, gender, ethnic-racial identity, and BINI scores were considered when randomizing participants to maintain homogeneity between the two conditions. Participants who had initiated PrEP completed weekly self-report PrEP adherence assessments after weekly intervention sessions, throughout the 4-week duration of the study. Following the pre/post study design, participants completed a second round of skills assessments and HIV-related IMB measures of risk reduction surveys at the end of the 4-week intervention. Participants also completed an intervention acceptability survey and accommodation strategy acceptability survey post-intervention.

2.3. Intervention Overview

The intervention was delivered in group format, in-person by trained research assistants, at a social service non-profit organization in Hartford, CT. Following an adapted version of the CHRP-BB intervention, the same intervention content was presented in both conditions through four weekly 45-minute sessions (Copenhaver et al., 2013; Shrestha et al., 2019). Session 1 focused on Navigating the Healthcare System and Personal Advocacy, Session 2 focused on Reducing Drug Risk and Taking PrEP, Session 3 focused on Sex Risk Reduction Strategies, Session 4 focused on Harm Reduction and Overdose Prevention. The conditions only varied by the presentation style of the content. The control condition received the intervention content by the facilitator reading through the intervention manual, in a lecture style manner. Accommodation strategies were included in the experimental condition to present the content. These accommodation strategies integrated into the experimental condition included a written agenda, mindfulness meditation, use of a whiteboard, projected presentation, handouts, hands-on demonstrations, videos, group quizzes, scenario-based activities, and formal closure.

2.4. Demographics

The study asked participants to self-report their age, gender, racial-ethnic identity, sexual identity, education level, yearly income, housing status, incarceration history, health insurance status, having a primary care doctor, psychiatric visitation in the past 12 months, length of substance use, and if they were on medication for opioid use disorder. They were also asked to report their overdose history and experience administrating Naloxone.

2.5. Brief inventory of neurocognitive impairment (BINI)

The BINI is a 57-item self-report measure designed to assess neurocognitive dysfunction among high-risk drug users enrolled in treatment (Copenhaver et al., 2016; Copenhaver et al., 2020). The nine-factor measure includes a diverse set of factors with excellent to good reliability (i.e., F1 α = 0.97 to F9 α = 0.73) ranging from generalized neurocognitive symptoms (Global Impairment) to more specific forms of impairment (Learning-related; Language-related; Memory-related; Psychomotor/Physical; Psychomotor/Perceptual; Anger-related; Pain-associated; Traumatic Head Injury-related). Given its ease of administration, sound psychometric properties, and straightforward interpretation, the BINI is designed to serve as an abbreviated instrument to screen for neurocognitive dysfunction among patients entering or enrolled in addiction treatment and for monitoring symptoms of dysfunction over time (Copenhaver et al., 2016; Copenhaver et al., 2020). A 5-item Likert scale was used for each of the 57 questions, with 1=not at all, 2=slightly, 3=somewhat, 4=moderately, and 5=extremely. A total score was calculated from the BINI by adding up the output from the 1–5 Likert Scale options; a cutoff score is set at 155 to determine presence of dysfunction; a higher total score indicates greater levels of cognitive dysfunction (Shrestha et al., 2019). This assessment can be found as Supplementary Material 1.

2.6. Sex- and Drug- Related HIV Risk Behaviors

The HIV risk assessment, adapted from NIDA’s Risk Behavior was used to measure several aspects of HIV risk behaviors in the past 3 months, including measurement of “any” high-risk behavior (sexual or drug-related) as well as measurements of event-level (i.e., partner-by-partner) behaviors (Meade et al., 2010). These categorical measures included questions such as “In the past 3 months, did you use a needle or syringe that had previously been used by someone else?” and “In the past 3 months, how often did you use a condom or other latex protection when you had oral, vaginal, and/or anal sex?”

2.7. Intervention Acceptability Survey

All participants who completed the 4-week study completed an 11-item survey to measure the acceptability of intervention content for both conditions, as intervention content was the same for both conditions (Mistler & Copenhaver, 2022). Participants indicated if they believed the intervention content was acceptable for people who engage in high-risk behaviors, if the intervention content seemed effective in reducing HIV risk and HIV risk behaviors, if participants would be willing to recommend this intervention content to a friend, if there were any negative consequences of the intervention content, if they liked this HIV prevention approach, if this intervention was a good way to help individuals with high-risk behaviors, and if overall this program would benefit high-risk drug users. Participants were also asked to indicate what topics they found most useful, and if they preferred to attend two shorter sessions a week compared to one longer session a week.

2.8. Accommodation Strategy Acceptability Survey

A 6-item survey measured the acceptability of the accommodation strategies in the experimental condition. Participants indicated which strategies were most useful in helping them remember, recall, and apply the information presented to them, and how they were useful in anticipating future behaviors.

2.9. HIV Risk Reduction Information, Motivation, and Behavior (IMB) Construct Measures

Categorical HIV risk reduction IMB model constructs (Fisher et al., 2003; Shrestha et al., 2017) comprised information, motivation, and behavioral skills variables (Supplementary Material 2). Informational variables included four questions such as “If an HIV+ person only has sex with another HIV+ person, they don’t need to use a condom”. Participants were asked to indicate if the four questions were true, false, or if they were unsure. This valid measurement tool recommends recoding information variables to combine responses of “not sure” with “false”; given these were reverse coded, participants who answered “not sure” were given a lesser score. To be specific, correct responses were coded as a 1 and incorrect or “not sure” responses were coded as a 0 for each specific question. Eight Likert Scale motivation variables included questions on readiness to change and intentions to change HIV risk behavior, such as “I plan not to have sex during the next 3 months”. A Likert scale scored participants responses to the motivation questions ranging from 5=very true, 4=somewhat true, 3=neither true nor untrue, 2=somewhat untrue, or 1=very untrue. Four behavioral skills variables included questions related to self-efficacy towards risk reduction skills, such as “How hard would it be for you to always use condoms or latex protection if you have oral, vaginal, or anal sex?” A Likert scale scored participants responses to the motivation questions ranging 1=very hard to do, 2=fairly hard to do, 3=neither hard nor easy to do, 4=somewhat easy to do, or 5=very easy to do. Some HIV risk reduction IMB variables were recoded to standardize the scales, using reverse coding to simplify results as a higher HIV risk reduction IMB score indicates greater HIV risk reduction information, motivation, and behavioral skills. Each HIV risk reduction IMB construct was scored separately adding up the scored responses for each individual construct (information, motivation, and behavioral skills). The study calculated total HIV risk reduction IMB scores by adding the total from all three IMB constructs together.

2.10. Skills Assessment

Th study assessed participants’ HIV risk reduction behavioral skills as in prior randomized controlled trials by having participants demonstrate the 16 steps necessary to properly clean a syringe, 17 steps necessary to properly use a male condom, 15 steps necessary to properly use a female condom (Copenhaver et al., 2013). This checklist can be found as Supplementary Material 3. Ratings of audio-taped demonstrations of these procedures have shown high inter-rater reliability (inter-rater reliability = 0.98) in similar prior trials (Margolin et al., 2003). Participants were also asked to demonstrate the 13 steps necessary to accurately administer nasal Naloxone. Participants received 1 point for accurately completing a step of the checklist, and 0 points if they inaccurately completed a step of the checklist, for each individual skill. The study calculated a total score for each skill separately, by totaling the correct steps completed; a higher total score indicated a greater ability to accurately engage in the risk reduction behavior.

2.11. PrEP interests, intentions, uptake and adherence

The study provided all participants linkage to PrEP services and referred to local PrEP providers for a PrEP prescription. Participants were asked if they received a PrEP prescription and if they picked up their PrEP prescription from the pharmacy after being prescribed it. Participants indicated “yes” to picking up their PrEP prescription before the first study session or “no” to picking up their PrEP prescription before the first study session. For those who indicated being prescribed it the study measured self-reported PrEP adherence at the end of each of the 4 weekly sessions,. These questions asked participants, “Have you taken PrEP since your last interview?” (no/yes) and “Since your last interview, how many days did you miss at least one dose of PrEP?” (never, sometimes, about half the time, most of the time, always). During the post-intervention data collection session, participants indicated if they intended to take PrEP for this study with a 5-point Likert scale response: definitely not, probably not, maybe, probably yes, and definitely yes. Using the same 5 Likert-scale responses, participants also indicated if they were more interested in taking PrEP after they had been more informed about it.

2.12. Data Analysis

All data analyses were conducted using SPSS v. 28 (IBM Corporation, 2017). Frequencies and percentages of the acceptability ratings were calculated and presented to determine the extent to which the participants supported the intervention content and the use of the accommodation strategies to improve their ability to learn intervention content. T-tests and Chi-squared tests determined if there were any significant differences between conditions for all demographic variables. Independent sample t-tests compared results between the two conditions on skills assessments and HIV risk reduction IMB measures. Chi-square tests compared results between the two conditions on PrEP adherence variables and HIV risk behaviors. Skip patterns were used if participants reported “no” to any variables to reduce redundancy. All comparisons met normality assumptions. If participants dropped out of the study at any point, research assistants attempted to contact them to complete the post-intervention assessments. Following a pre/post study design, the study removed any participants who did not complete both pre and post assessments from data analysis.

2.13. Ethics

The Investigational Review Board (IRB) at the University of Connecticut approved the study protocol. All participants received a verbal and written description of the study and signed an informed consent form prior to participation. All sessions and data collection were audio-recorded. Participants received $25 for each of the 5 sessions (intake and 4-intervention sessions).

3. Results

3.1. Retention

The study randomized 50 participants at baseline, following their completion of the initial intake session (Figure 1). Seven participants dropped out of the study after the intake session (four from the experimental condition and three from the control condition). Three participants dropped out after the first intervention session (one from the experimental condition, and two from the control condition). Therefore, 40 participants (80%) completed the 4-week pre/post-intervention study. Retention was homogeneous between the two conditions, with 20 participants in each condition completing the 4-week intervention.

Figure 1.

CONSORT diagram of pilot trial study procedures.

3.2. Demographics

The average age of participants enrolled in the trial was 43 years old (SD=11.72), and the average length of injection drug use was 3.9 years (SD=2.09). Participants average BINI score was 175.3, which is above the prespecified cutoff of 155 indicating presence of mild/moderate cognitive dysfunction. A majority of participants identified as male (72.5%). Those identifying as Caucasian made up 40% of the sample population; participants identifying as African American made up another 40%, while 15% identified as Hispanic/Latinx and 5% identified as Native American. Most of the participants identified as heterosexual and having less than $10,000 annual income (92.5%). Most of the sample reported being without secure housing (77.5%) and having ever been incarcerated (80%). No significant differences existed between the two conditions on any demographic variable (Table 1).

Table 1.

Demographics variables

Demographic	Total Sample (n=40)	% / (S.D.)	Experimental Condition (n=20)	% / (S.D.)	Standard Condition (n=20)	% / (S.D.)	p-value
Age	43.85	(11.72)	46.7	(11.99)	41.0	(11.01)	0.388
BINI score	175.3	(19.71)	174.1	(13.68)	176.4	(24.64)	0.717
Length of injection drug use (years)	3.90	(2.09)	4.55	(2.21)	3.55	(1.99)	0.141
Gender							0.077
Male	29	72.5%	17	85.0%	12	60.0%
Female	11	27.5%	3	15.0%	8	40.0%
Racial-Ethnic Identity							0.355
Caucasian	16	40.0%	6	30.0%	10	50.0%
African American	16	40.0%	9	45.0%	7	35.0%
Native American	2	5.0%	2	10.0%	0	0.0%
Hispanic or Latinx	6	15.0%	3	15.0%	3	15.0%
Sexual Identity							0.387
Heterosexual	37	92.5%	18	90.0%	19	95.0%
Homosexual	1	2.5%	1	5.0%	0	0.0%
Bisexual	1	2.5%	0	0.0%	1	5.0%
Prefer not to say	1	2.5%	1	5.0%	0	0.0%
Education level							0.101
Some high school, no degree	15	37.5%	10	50.0%	5	25.0%
High school degree	17	42.5%	5	25.0%	12	60.0%
2-year college degree	5	12.5%	2	10.0%	3	15.0%
4-year college degree	2	5.0%	2	10.0%	0	0.0%
Graduate degree	1	2.5%	1	5.0%	0	0.0%
Yearly income							0.220
less than $10,000	37	92.5%	18	90.0%	19	95.0%
$10,000 - $19,999	2	5.0%	2	10.0%	0	0.0%
$20,000 - $29,999	1	2.5%	0	0.0%	1	5.0%
Housing status							0.705
With secure housing	9	22.5%	5	25.0%	4	20.0%
Without secure housing	31	77.5%	15	75.0%	16	80.0%
Ever been incarcerated							1.000
No	8	20.0%	4	20.0%	4	20.0%
Yes	32	80.0%	16	80.0%	16	80.0%
Health insurance							0.220
No	1	2.5%	1	5.0%	0	0.0%
Yes	37	92.5%	19	95.0%	18	90.0%
Not sure	2	5.0%	0	0.0%	2	10.0%
Primary care doctor							0.123
No	21	52.5%	8	40.0%	13	65.0%
Yes	18	45.0%	12	60.0%	6	30.0%
Not sure	1	2.5%	0	0.0%	1	5.0%
Psychiatric visit in past 12 months							0.752
No	21	52.5%	11	55.0%	10	50.0%
Yes	19	47.5%	9	45.0%	10	50.0%
On medication for opioid use disorder							0.327
No	25	62.5%	11	55.0%	14	70.0%
Yes	15	37.5%	9	45.0%	6	30.0%
Ever experienced a drug overdose?							1.000
No	14	35.0%	7	35.0%	7	35.0%
Yes	26	65.0%	13	65.0%	13	65.0%
Ever administered Naloxone?							0.288
No	11	27.5%	7	35.0%	4	20.0%
Yes	29	72.5%	13	65.0%	16	80.0%

^*p < 0.05 is statistically significant; chi-squared tests were used for categorical variables (n,%); t-tests were used for numerical variables (mean, S.D.). Two-sided p-values were used.

3.3. Drug-related HIV risk behaviors

More than half of participants enrolled in the trial (55%) reported injecting drugs a few times a day, sharing syringes within the past 3 months (62.5%), and carrying new or bleach-cleaned syringes with them (70%). Most participants knew where to obtain new, unused syringes (87.5%). A needle exchange van was where most participants went to get new, unused syringes. More than half of the participants (57.5%) reported sharing drug equipment (cookers, cotton, etc.), but less than half (45%) of participants reported drawing up a drug solution from a community container shared with others in the past 3 months. No significant differences existed between the two conditions on any drug-related HIV risk behaviors (Table 2).

Table 2.

Drug-related HIV risk behaviors

HIV drug risk behaviors	Total Sample (n=40)	%	Experimental Condition (n=20)	%	Standard Condition (n=20)	%	p-value
In the past 3 months, how often would you say you injected illicit drugs?							0.878
Once a month	6	15.0%	4	20.0%	2	10.0%
A few times a month	3	7.5%	1	5.0%	2	10.0%
A few times a week	5	12.5%	2	10.0%	3	15.0%
Once a Day	4	10.0%	2	10.0%	2	10.0%
A few times a day	22	55.0%	11	55.0%	11	55.0%
In the past 3 months, did you use a needle or syringe that had previously been used by someone else?							0.102
No	15	37.5%	10	50.0%	5	25.0%
Yes	25	62.5%	10	50.0%	15	75.0%
In the past 3 months, how often did you use a needle or syringe that had previously been used by someone else?							0.353
Sometimes	20	80.0%	7	70.0%	13	86.7%
About half the time	1	4.0%	0	0.0%	1	6.7%
Most of the time	3	12.0%	2	20.0%	1	6.7%
Always	1	4.0%	1	10.0%	0	0.0%
In the past 3 months, when you injected with needles or syringes that had previously been used, how often did you clean the needle?							0.712
Never	3	12.0%	1	10.0%	2	13.3%
Sometimes	10	40.0%	3	30.0%	7	46.7%
About half the time	1	4.0%	1	10.0%	0	0.0%
Most of the time	9	36.0%	4	40.0%	5	33.3%
Always	2	8.0%	1	10.0%	1	6.7%
In the past 3 months, did you give, lend, sell, or rent a needle or syringe to someone else after you had already used it?							0.185
No	26	65.0%	15	75.0%	11	55.0%
Yes	14	35.0%	5	25.0%	9	45.0%
In the past 3 months, how often did you give, lend, sell, or rent a needle or syringe to someone else after you had already used it?							0.233
Sometimes	11	27.5%	4	20.0%	7	35.0%
About half the time	2	5.0%	0	0.0%	2	10.0%
Most of the time	1	2.5%	1	5.0%	0	0.0%
Do you know of any place or person from which you can obtain new, unused needles or syringes?							0.151
No	5	12.5%	4	20.0%	1	5.0%
Yes	35	87.5%	16	80.0%	19	95.0%
Where can you obtain new, unused needles and syringes?							0.637
Needle Exchange Clinic	8	22.9%	4	25.0%	4	21.1%
Needle Exchange Mobile Van	26	74.3%	12	75.0%	14	73.7%
Family member, friend, acquaintance	1	2.9%	0	0.0%	1	5.3%
In the past 3 months, did you carry new or bleach-cleaned needles and works with you, in case you needed them?							0.168
No	12	30.0%	4	20.0%	8	40.0%
Yes	28	70.0%	16	80.0%	12	60.0%
In the past 3 months, how often did you carry new or bleach-cleaned needles and works with you, in case you needed them?							0.765
Sometimes	14	50.0%	7	43.8%	7	58.3%
About half the time	1	3.6%	1	6.3%	0	0.0%
Most of the time	5	17.9%	3	18.8%	2	16.7%
Always	8	28.6%	5	31.3%	3	25.0%
In the past 3 months, when you injected drugs, did you share a cooker/ vial/ container/ cotton/ filter, or rinse with someone else?							0.110
No	17	42.5%	11	55.0%	6	30.0%
Yes	23	57.5%	9	45.0%	14	70.0%
In the past 3 months, when you injected drugs, how often did you share a cooker/ vial/ container/ cotton/ filter, or rinse with someone else?							0.931
Sometimes	18	78.3%	7	77.8%	11	78.6%
About half the time	3	13.0%	1	11.1%	2	14.3%
Most of the time	2	8.7%	1	11.1%	1	7.1%
In the past 3 months, did you draw up your drug solution from a community container shared by others?							0.525
No	22	55.0%	12	60.0%	10	50.0%
Yes	18	45.0%		0.0%	180	900.0%
In the past 3 months, how often did you draw up your drug solution from a community container shared by others?							0.865
Sometimes	13	72.2%	5	62.5%	8	80.0%
About half the time	2	11.1%	1	12.5%	1	10.0%
Most of the time	2	11.1%	1	12.5%	1	10.0%
Always	1	5.6%	1	12.5%	0	0.0%

^*p < 0.05 is statistically significant. Two-sided p-values were used.

3.4. Sex-related HIV risk behaviors

Most participants enrolled in the trail (60%) reported having only one sexual partner in the past 3 months, while 25% reported having 2–4 sexual partners in the past 3 months. Sixty percent of participants reported never using a condom in the past 3 months if they had sex. About one-third of the sample (35%) reported transactional sex, and 35% reported never carrying condoms on them. Almost one-half (45%) of participants reported never having sex under the influence of alcohol, while 27.5% reported never having sex under the influence of illicit drugs in the past 30 days. Significant differences were present between conditions on the outcome that measured condom use in the past 3 months (Table 3).

Table 3.

Sex-related HIV risk behaviors

HIV sex risk behaviors	Total Sample (n=40)	%	Experimental Condition (n=20)	%	Standard Condition (n=20)	%	p-value
In the past 3 months, did you have oral, anal, or vaginal sex?
Yes	40	100.0%	20	100.0%	20	100.0%
How many partners have you had any form of sexual contact with in the past 3 months?							0.272
1	24	60.0%	13	65.0%	11	55.0%
2–4	10	25.0%	6	30.0%	4	20.0%
5–9	3	7.5%	0	0.0%	3	15.0%
10+	3	7.5%	1	5.0%	2	10.0%
In the past 3 months, how often did you use a condom or other latex protection when you had oral, anal, or vaginal sex?							0.031*
Never	24	60.0%	10	50.0%	14	70.0%
Sometimes	6	15.0%	3	15.0%	3	15.0%
About half the time	1	2.5%	1	5.0%	0	0.0%
Most of the time	3	7.5%	0	0.0%	3	15.0%
Always	6	15.0%	6	30.0%	0	0.0%
In the past 3 months, did you have sexual contact with anyone who is HIV positive?							0.264
No	32	80.0%	16	80.0%	16	80.0%
Yes	2	5.0%	0	0.0%	2	10.0%
Not sure	6	15.0%	4	20.0%	2	10.0%
In the past 3 months, have you received or provided drugs, money, or other items or services in exchange for any forms of sexual contact?							0.507
No	26	65.0%	14	70.0%	12	60.0%
Yes	14	35.0%	6	30.0%	8	40.0%
In the past 3 months, how often did you have any form of sexual contact while under the influence of alcohol?							0.673
Never	18	45.0%	9	45.0%	9	45.0%
Sometimes	9	22.5%	6	30.0%	3	15.0%
About half the time	5	12.5%	2	10.0%	3	15.0%
Most of the time	7	17.5%	3	15.0%	4	20.0%
Always	1	2.5%	0	0.0%	1	5.0%
In the past 3 months, how often did you have any form of sexual contact while under the influence of any illicit drugs?							0.277
Never	11	27.5%	8	40.0%	3	15.0%
Sometimes	10	25.0%	5	25.0%	5	25.0%
About half the time	3	7.5%	2	10.0%	1	5.0%
Most of the time	8	20.0%	2	10.0%	6	30.0%
Always	8	20.0%	3	15.0%	5	25.0%
In the past 3 months, how often did you carry condoms or other latex protection in case you needed them?							0.355
Never	14	35.0%	5	25.0%	9	45.0%
Sometimes	13	32.5%	9	45.0%	4	20.0%
About half the time	4	10.0%	1	5.0%	3	15.0%
Most of the time	3	7.5%	2	10.0%	1	5.0%
Always	6	15.0%	3	15.0%	3	15.0%

^*p < 0.05 is statistically significant. Two-sided p-values were used.

3.5. Intervention Acceptability

All participants who completed the 4-week study indicated high acceptability (89%) of intervention content and HIV prevention approach. A majority of participants (>90%) agreed or strongly agreed that: 1) this intervention content provided an acceptable program for individuals who engage in high-risk behaviors; 2) this intervention content should be effective in reducing the chances of HIV infection; 3) that they would be willing to recommend this intervention content to friends who may benefit from it; 4) that they liked this HIV prevention approach; 5) that this was a good way to help individuals with high-risk behaviors; and 6) that overall the intervention content would benefit high-risk drug users. Almost all (88%) of participants agreed or strongly agreed that this approach did not have any negative consequences to participants. In order of topics that participants found the most useful, participants identified Naloxone administration, syringe cleaning, local resources, HIV and Hepatitis C prevention, condom application, navigating healthcare, and partner negotiations as most useful, respectively. While 40% of participants indicated they could pay attention better if sessions were shorter, 42.5% noted that it would not make a difference. Similarly, 40% of participants preferred to attend two shorter sessions a week, compared to 30% of participants who preferred one longer session a week. Notably, 30% of participants did not have a preference. In terms of how realistic it would be for participants to attend two shorter sessions per week, 17.5% said it would be somewhat difficult, 25% reported it would be neither easy nor difficult, 32.5% noted it would be somewhat straightforward, and 25% said it would be extremely easy to do.

3.6. Accommodation Strategy Acceptability

Participants in the experimental condition highly endorsed (92%) the accommodation strategies integrated into the HIV prevention sessions. Participants highlighted that using mindfulness meditation, multimodal presentation of information, written agenda, scenario-based activities, peer crosstalk and feedback, and formal closure was helpful. Mindfulness meditation was noted to help participants focus on learning the material and calm their minds so they could learn. Using a written agenda kept the group organized, helped participants pay attention better, and helped them know what to expect throughout the session. Participants noted that case scenarios helped to plan for risky situations, learn from others, and consider handling future situations. Group discussions and peer feedback were identified to help participants understand the information more in-depth. Using a formal closure at the end of the session helped participants recall intervention content, keep the information organized, and pay attention to the content.

3.7. HIV risk reduction Information, Motivation, and Behavior (IMB) construct measures

There were mean increases in both conditions for HIV risk reduction motivation, HIV risk reduction self-efficacy behaviors, and total HIV risk reduction IMB scores. HIV risk reduction information decreased from pre to post in both conditions. There were no significant differences in mean changes between conditions (Table 4).

Table 4.

HIV risk reduction Information, Motivation, and Behavior outcomes

HIV risk reduction IMB scores	Condition	Mean change	S.D.	95% C.I.	p-value	Cohen’s d
Information
	experimental	−.300	1.53	(−1.07-.772)	0.774	1.440
	standard	−.150	1.35
Motivation
	experimental	.500	8.21	(−5.95–3.24)	0.555	7.170
	standard	1.85	5.95
Behavior
	experimental	.800	4.81	(−2.45–2.85)	0.879	4.136
	standard	.600	3.33
Total IMB score
	experimental	1.00	8.18	(−6.46–3.86)	0.613	8.060
	standard	2.39	7.93

Levene’s test for equality of variances were not significant for all variables. 95% C.I. = 95% confidence intervals of the difference. Two-sided p-values were used.

3.8. HIV Risk Reduction Skills Assessment

Participants in the experimental condition, who received an HIV prevention session with cognitive dysfunction accommodation strategies, demonstrated greater retention of intervention content on properly cleaning a syringe (p=0.048) and accurately applying a female condom (p=0.025). The mean increase in total scores for cleaning a syringe was 6.45 (S.D. = 3.30) in the experimental condition compared to 4.30 (S.D. = 3.34) in the control condition. The mean increase in total scores for applying a female condom was 2.75 (S.D.=1.89) in the experimental condition compared to 1.25 (S.D.=2.17) in the control condition. Participants in the experimental condition also had greater increases in mean total scores for applying a male condom and administering Naloxone between conditions, but these results were not significant (Table 5).

Table 5.

HIV Risk Reduction Skills Assessment Outcomes

Skills Assessment	Condition	Mean increase in total scores	S.D.	95% C.I.	p-value	Cohen’s d
Syringe cleaning
	experimental	6.45	3.30	0.02–4.28	0.048*	0.647
	standard	4.30	3.34
Female condom
	experimental	2.75	1.89	0.20–2.80	0.025*	0.737
	standard	1.25	2.17
Male condom
	experimental	2.50	1.85	−0.51–1.91	0.250	0.370
	standard	1.80	1.94
Naloxone
Administration	experimental	5.45	3.07	−0.38–3.39	0.116	0.509
	standard	3.95	2.82

^*p < 0.05 is statistically significant. Levene’s test for equality of variances were not significant for all variables. 95% C.I. = 95% confidence intervals of the difference. Two-sided p-values were used.

3.9. PrEP interest, intention, uptake, and adherence

When participants were asked if they intended to take PrEP, 25% reported definitely not, and 35% reported probably not. When participants were asked if they were more interested in taking PrEP after being more informed about it, 25% reported definitely not, 20% said probably not, and 30% said they were maybe interested in taking it. Only 3 participants (7.5%) reported being prescribed PrEP throughout the study duration. Weekly PrEP adherence was reported by the 3 participants who reported taking PrEP throughout the study. All 3 participants reported missing doses “sometimes”, “half of the time”, or “most of the time’” throughout each of the 4 weeks. There were no significant differences in PrEP adherence, uptake, or interest between the two conditions.

4. Discussion

This novel study investigated the feasibility and impact of integrating tailored cognitive dysfunction accommodation strategies into a 4-week HIV prevention behavioral intervention for PWID. Participants who received the experimental condition (with accommodation strategies) demonstrated greater recall and application of HIV prevention risk reduction skills compared to those in the control condition who received the intervention content presented in a lecture-style. While participants in both conditions received the same HIV prevention content across 4 weekly sessions, those who received the HIV prevention information by structured sessions with a written agenda, multi-modal presentation of information, hands-on activities, meditation, memory aids, and immediate feedback were able to better retain how to properly clean a syringe and how to apply a female condom. There were no significant differences in participants ability to apply a male condom or administer Naloxone pre/post, highlighting that these strategies may be more beneficial in improving new learning than reiterating skills that participants are already familiar with. For example, 72.5% of participants had administered Naloxone and had likely applied a male condom prior to engaging in the intervention. Therefore, these tailored accommodation strategies may be more efficacious when delivering intervention content unfamiliar to participants, providing rationale as to why we should implement strategies in situations that emphasize participants learning new content.

While results from HIV risk reduction skills assessments showed promise in supporting the use of tailored cognitive dysfunction accommodation strategies, there were no significant differences between conditions on participants’ HIV risk reduction information, motivation, or behaviors (IMB) pre/post-intervention. While the IMB Model suggests that one’s knowledge, motivation, and self-efficacy indicate HIV risk, using accommodation strategies did not significantly change participants’ HIV risk reduction IMB (Fisher et al., 2003). This may be explained by the subject matter of the 4 intervention sessions, which may have been unintentionally specific to the skills assessments. The use of hands-on demonstrations, which were also specific to the skills assessments, may have influenced these outcomes. While this is the first trial to investigate integrating cognitive dysfunction accommodation strategies into an HIV prevention intervention, results from this study are similar to other pilot trials that have attempted to promote PrEP uptake and adherence among PWID (Dumchev et al., 2024; Meyer et al., 2021).

This study adds to the limited research focused on PrEP interest, intentions, uptake, and adherence among PWID (Bazzi et al., 2023; Mistler et al., 2020). This study linked PWID to PrEP services and increased PrEP awareness and interest among PWID, a vital step in the PrEP care cascade. However, PrEP uptake and adherence was low in this sample, and majority of participants reported not intending to take PrEP for this study. Most of the participants reported insecure housing (77.5%) and having ever been incarcerated (80%), so the limited PrEP interest and uptake may have been attributed to HIV prevention being low on the priority list of basic needs. Nonetheless, our findings are consistent with the national average of PrEP uptake among PWID (Mistler et al., 2020). Daily adherence to PrEP has been noted as primary barrier to PrEP uptake in PWID. Injectable PrEP may serve as a solution to this barrier, although studies are limited (Shrestha et al., 2020). However, behavioral interventions will still be crucial to support health behavior change, and participants will continue to benefit from behavioral counseling to reduce their HIV risks. Tailoring behavioral interventions to the cognitive profiles of people with OUD is likely to increase treatment outcomes across a variety of prevention content, expanding beyond HIV prevention and harm reduction. Adapting behavioral interventions for people with OUD to accommodate for cognitive dysfunction may serve as a vital and innovative approach to decreasing health disparities and promoting health nationwide.

Recommendations from the Substance Abuse and Mental Health Services Administration (SAMHSA), suggest that reviewing patients’ psychosocial history and cognitive functioning should be implemented in all behavioral interventions for PWID (Substance Abuse and Mental Health Services Administration, 2012). The current study provides evidence to support the implementation of screening for cognitive dysfunction, and tailoring HIV prevention behavioral interventions to the cognitive profiles of PWID to reduce HIV risk. Future research is needed to update clinical guidelines and optimize training efforts to ensure PWID receive standard of care HIV prevention.

4.1. Limitations

The experimental research design of this study increases the validity of the results; however, there are some notable limitations. The sample size of this study was small compared to a large-scale randomized controlled trial but was appropriate for a small-scale HIV prevention pilot trial (Biello et al., 2019; Mayer et al., 2017; Shrestha et al., 2018). The skills assessments were measured objectively, increasing the validity of the significant results in this study. However, all other measures were self-report, which reduced the validity of those results, including eligibility measures such as HIV risk and cognitive dysfunction. This study was also conducted in a small city in the northeast region of the United States and may lack generalizability to different subpopulations and locations. Additionally, this study was single-blinded, as the participants did not know what condition they were randomly assigned to. Researchers held group sessions on different days to avoid cross-contamination between the two conditions.

4.2. Conclusions

This study provides greater insights into innovative approaches to address the HIV epidemic, amidst the opioid epidemic in the United States. Tailoring HIV prevention interventions to the cognitive profiles of PWID by accommodating cognitive dysfunction shows efficacy in increasing HIV risk reduction behaviors. Results from this study provide evidence to support a larger randomized trial to determine which combinations of accommodation strategies may be most influential in reducing HIV incidence in PWID in the United States.

Highlights:

• People who inject drugs display disproportionate rates of cognitive dysfunction

• Cognitive dysfunction can limit the ability to learn and retain information

• Accommodating for cognitive dysfunction to promote HIV prevention is feasible