Neurocognitive,Sociocultural,and Psychological Factors Impacting Medication Beliefs Among HIV-Seropositive Latinx Adults

Abstract

Among Latinx people living with HIV (PLWH), neurocognitive (NC) function, culture, and mental health impact medication adherence. Similarly, health beliefs and attitudes play a role in health care barriers and health behaviors. Research has not examined the effect that compromised neurocognition, sociocultural factors, and mental health have on health beliefs and attitudes. This is especially relevant for Latinx PLWH who are disproportionately impacted by HIV, given that sociocultural factors may uniquely impact HIV-related NC and psychological sequelae. This study investigated the associations between neurocognition, sociocultural factors, mental health, health beliefs, and health attitudes among Latinx HIV-seropositive adults. Within a sample of 100 Latinx PLWH, better verbal learning and executive functioning abilities were associated with more positive attitudes about the benefits of medications and memory for medications. In terms of sociocultural factors, higher English language competence was related to better self-reported memory for medications, and overall, higher US acculturation was associated with more positive attitudes toward health professionals. Depressive symptomatology was negatively associated with attitudes toward medications and health professionals, as well as with self-reported memory for medications. These findings highlight the important interplay between NC, sociocultural, psychological factors, and health beliefs among Latinx PLWH. Adherence intervention strategies and suggestions for dispensing medical information are presented for clinicians and health care practitioners.

Introduction

In the United States, HIV disproportionately impacts ethnic minority populations, particularly the Hispanic/Latinx community (herein referred to as “Latinx”). Given rapid growth within the Latinx community, the impact of HIV in this population is particularly concerning. Latinx individuals comprised 19% of the US population in 2020, and it is estimated that this figure will rise to 30% by 2060.¹ In 2019, HIV infection rates among Latinx were over three times as high as that of non-Hispanic Whites.² Moreover, worse health outcomes are observed in this population, including higher mortality rates. In 2019, Latinx individuals were over twice as likely to die of HIV/AIDS compared to non-Hispanic Whites.² Differences in medication adherence rates may impact these outcomes.

Suboptimal adherence (i.e., <85% to 95% of prescribed doses, depending on the medication) may result in HIV disease progression, worse health outcomes, and drug resistance to multiple classes of antiretroviral medications.^3,4 Latinx persons living with HIV (PLWH) face unique challenges with medication management and adherence.^5

–8 From a public health perspective, the increase in the Latinx population along with the impact that HIV/AIDS has on the Latinx community highlights the need for increased research on HIV/AIDS within this population. Examining potential barriers to medication adherence is particularly valuable. Health beliefs significantly contribute to medication nonadherence among PLWH.^9,10 However, barriers to adherence are not well understood among Latinx PLWH, and the impact of beliefs about medication and attitudes toward health professionals need to be examined. Furthermore, among Latinx PLWH, health beliefs may predict downstream outcomes such as neurocognitive (NC) functioning, cultural identity, and mental health.

Health beliefs about combined antiretroviral therapy (cART) and health professionals are important predictors of HIV medication adherence.¹¹ A qualitative examination of PLWH with optimal adherence to their medications¹² found that most patients with excellent adherence believed that 90–100% adherence was necessary to benefit from their medication. In a study examining health beliefs and self-efficacy,¹³ PLWH with limited self-efficacy and/or beliefs that their medication was ineffective were less likely to comply with complicated medication regimens¹⁰ Among PLWH in Brazil, lower medication-related self-efficacy was significantly associated with worse antiretroviral adherence.¹⁴ Additionally, PLWH may not adhere to their medication regimen because of a breakdown in communication with health providers.¹⁵ Communication is vital as health professionals often explain medical conditions, deliver medication instructions, stress the importance of adherence, and offer memory strategies to improve adherence.⁹ Providers may also question a patient's honesty about their adherence, particularly if their viral load is detectable, and resulting feelings of judgment and/or stigmatization may reduce trust in the patient–provider relationships. Lack of trust and poor patient–provider communication has been associated with worse HIV-related health outcomes, including adherence.¹⁶ These findings are congruent with the Health Belief Model, which hypothesizes that an individual's health beliefs will impact his/her actions to prevent or alleviate disease.¹⁷

NC functioning is also affected in HIV/AIDS; the virus enters the central nervous system during the earliest stages of infection, and HIV/AIDS specific NC deficits are consistent with dysfunction and pathology in frontostriatal circuitry (i.e., frontal lobe and basal ganglia) and white matter.^18

–22 Moreover, despite the availability of improved therapies (e.g., cART), the overall prevalence of observed NC impairment in PLWH remains high (up to 52%).²³

Given the importance of optimal treatment adherence and the NC impact of HIV, which may be subtle and may increase with age, it is necessary to examine how NC functioning may relate to factors associated with suboptimal adherence (e.g., health beliefs and attitudes). Learning, memory, and executive functioning are commonly associated with medication adherence among PLWH^13,24 and so are of particular interest in this study. For example, Hinkin et al.²⁵ found a relationship between cART regimen complexity and suboptimal medication adherence among PLWH with significant executive functioning deficits.

Both general and culture-specific health beliefs need to be considered among Latinx PLWH. Cultural identity, for instance, may be linked to health beliefs. This is especially relevant to Latinx PLHW in the United States.²⁶ In Latinx immigrants of Mexican descent, acculturation is related to perceptions of the physician–provider relationship, and beliefs about the utility of alternative medicines were associated with poorer medication adherence.²⁷ When focusing on Latinx PLWH, pilot research has suggested a positive correlation among US acculturation, bicultural self-efficacy, and medication adherence for Latinx PLWH.²⁸ Further, a study of Latinx PLWH residing in the Southeastern United States suggests a relationship between health attitudes and adherence.²⁹ Results of a qualitative study among PLWH in Tijuana suggested that receiving support from peer navigators, who were community-based and perceived as peers by study participants, had a positive impact on HIV-related knowledge and medication adherence.³⁰

Higher levels of mental illness are also observed among PLWH or AIDS,³¹ and depression, specifically, may be more common among PLWH.³² Psychological stressors impact medication adherence in PLWH.³³ Considering the social and economic disadvantages often experienced by Latinx PLWH, it's important to consider the impact of depressive symptoms on health beliefs and attitudes in this group.

In sum, although cART has significantly improved the health and life longevity of PLWH, HIV-associated NC deficits are still prevalent, but may be subtle. Research shows that NC functions (including learning, memory, and executive functioning), mental health, acculturation, and health beliefs and attitudes are independently associated with nonadherence to ARTs. However, how NC, cultural, and mental health factors impact beliefs about medication adherence is not well understood. This is an important issue within the Latinx community given the significant HIV-related health disparities in this group compared with non-Hispanic Whites. Therefore, the aim of this study was to investigate the associations between NC, sociocultural, and mental health factors with health beliefs and attitudes among Latinx PLWH. This study represents the first investigation of the connection between neurocognition and medication beliefs, and will be one of the first studies to utilize the Beliefs Related to Medication Adherence (BERMA)⁹ among PLWH.

Methods

Participants

A total of 100 Latinx PLWH enrolled in a larger parent study (Medication Adherence Study, MÁS; PI: Monica Rivera Mindt, PhD; K23MH079718) were included in these analyses. Primary recruitment was via community outreach in New York City, particularly Spanish Harlem, through self-referral, and through clinics and related studies at the Icahn School of Medicine at Mount Sinai (ISMMS) in New York City.

All participants were HIV-positive (confirmed by medical records), between 18 and 80 years old, fluent in English, and on stable antiretroviral therapy with a minimum of 12 weeks on the same regimen. Only individuals who self-identified as Latinx were included in the current study. Exclusion criteria for the study included self-reported history of any of the following conditions that could impact cognition: severe psychiatric disorder (e.g., schizophrenia, psychosis, and bipolar disorder) or significant HIV related comorbid neurologic or medical condition (e.g., epilepsy, brain cancer or tumor, traumatic brain injury with loss of consciousness >60 min, neurosurgery, lupus, multiple sclerosis, Parkinson's disease, and cerebrovascular accident).

Procedure

Participants completed comprehensive NC and sociocultural evaluations, described below. Viral loads and CD4 lymphocyte counts were collected via blood samples. All participants provided written informed consent. The Institutional Review Boards (IRB) of both ISMMS and Fordham University approved the study.

Neuropsychological evaluation

All participants completed a comprehensive NC battery that has been validated and empirically supported for use with participants with HIV/AIDS.²³ Table 1 summarizes the NC measures examined in this study, which included tests of learning, memory, and executive functioning. Raw scores on all tests were converted to standardized T-scores using demographically corrected normative data (Table 1). Higher demographically corrected NC T-scores reflect better NC functioning. Scores were examined continuously, but for context, T-scores of 40 or higher are considered unimpaired. Specifically, T-scores of 0–19 reflect severe impairment, 20–24 reflect moderate-to-severe impairment, 25–29 reflect moderate impairment, 30–34 reflect mild-to-moderate impairment, 35–39 reflect mild impairment, 40–44 are below average, 45–54 average, and 55 and higher are above average.³⁴

Table 1.

Neurocognitive Test Battery and Normative Data by Areas for Computation of Average T-Scores and Deficit Scores

Neuropsychological domain and test	Normative data sources
Executive functioning
Wisconsin Card Sorting Task-64 Item	Kongs et al.^59,a,b
Trail Making Test (Part B)	Heaton et al.^60,a,b,c,d
Learning
Hopkins Verbal Learning Test—Revised	Benedict et al.^62,a
Brief Visuospatial Memory Test—Revised	Benedict^63,a
Delayed recall
Hopkins Verbal Learning Test—Revised	Benedict et al.^62,a
Brief Visuospatial Memory Test—Revised	Benedict^63,a

Normative data corrects for the demographic characteristics indicated by superscript: ^aAge; ^bEducation; ^cGender; ^dEthnicity.

Abbreviated Multidimensional Acculturation Scale

All participants completed the Abbreviated Multidimensional Acculturation Scale (AMAS),³⁵ a self-report acculturation questionnaire. The brief survey assesses both US and Latinx (i.e., culture of origin) acculturation levels and is comprised of six subscales, for each acculturation domain. Responses to individual items on each subscale were then summed for a total raw score. The current study will focus specifically on the US acculturation subscale raw scores in Language Competence, Cultural Competence, and Cultural Identity, as well as the total US acculturation raw score. Higher scores represent higher levels of acculturation. The AMAS has excellent reliability with Cronbach alpha coefficients ranging from 0.90 to 0.97 with validity established in both US born and Latin American born Latinx.³⁵ The AMAS has been used extensively in research with Latinx HIV-positive populations.^26,36
–38

Depressive symptomology

All participants completed the Beck Depression Inventory II³⁹ (BDI-II) to assess depressive symptoms over the past 2 weeks. Possible scores range from 0 to 63, with higher total BDI-II scores reflecting greater levels of depressive symptomatology. For context, categorically, scores of 13 or less are considered “minimal,” 14–19 are “mild,” 20–28” are “moderate,” and scores of 29 or more are considered “severe” symptoms of depression. BDI-II total scores were examined continuously.

Beliefs Related to Medication Adherence

The BERMA survey focuses on three efficacy-related beliefs: (1) memory for medications; (2) ability to deal with health professionals; (3) attitudes about medications. Participants rate their perceptions on a Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree. Scores regarding anxiety are reverse scored, that is, higher scores indicated lower anxiety. Hence, higher scores represent better memory for medications, better ability to deal with health professionals, and more positive attitudes toward medications.⁹ The BERMA survey has a Cronbach's alphas of 0.94 and has been utilized with PLWH in previous research.⁴⁰ The BERMA survey provides a thorough, yet simple assessment of numerous components impacting medication adherence.

Medication event monitoring system

HIV antiretroviral medication adherence was assessed electronically using the microchipped medication event monitoring system (MEMS) bottle and cap system. MEMS is a validated measure of adherence^41
–43 that records the date and time of each dose based on bottle opening. One medication [i.e., either the protease inhibitor, if taking, or the most frequent doses of non-PI antiretroviral (ARV) medication] was selected and placed in the MEMS bottle. Participants were given verbal and written instructions on MEMS cap use, following procedures described by Bova et al.⁴⁴ Participants used the MES cap for at least 30 days, and subsequently, 30-day medication adherences rates were obtained. Medication adherence was reported, below, as a percentage of doses taken over the course of 30 days, and participants were further categorized by optimal adherence (i.e., 95% adherence or better) or suboptimal adherence (less than 95% adherence).

Statistical analysis

The Statistical Package for the Social Sciences (SPSS) version 28 was used to analyze the results.⁴⁵ Pearson correlations, independent samples t-tests, and linear multiple regressions were computed. A p-level of 0.05 was used to determine statistical significance. Subscale scores for the BERMA survey were used as dependent variables. A series of linear regressions were computed to predict each BERMA subscale score. Variables with significant univariate relationships with BERMA subscale scores were entered into the model.

Results

Table 2 summarizes the demographic and clinical characteristics of the study sample. The majority of participants (67%) were born in the United States; the remainder was born in Latin American countries. Of note, most participants identified as Puerto Rican (75%). The sample consists of 68% males, overall mean age was 46.27 [standard deviation (SD) = 7.24] years, and mean years of education were 12.08 (SD = 2.47). In terms of HIV-related characteristics, more than half of the sample had an undetectable viral load (54%), and the sample had a median CD4 count of 444 (interquartile range = 267.50, 689.00). Participants reported an average duration of HIV infection of nearly 15 years. Of note, length of HIV duration was significantly correlated with HVLT-R Delayed T-score (r = −0.34, p = 0.03), but was not significantly associated with any other outcome variables (e.g., BERMA scores, clinical, or sociocultural characteristics). Participants' adherence rate over the last 30 days, as measured electronically by MEMS, was about 80%, and only about a third of participants exhibited optimal adherence (i.e., 95% adherence or better). Higher CD4 count (r = 0.32, p = 0.04) was associated with higher BERMA Attitudes Toward Health Professionals subscale scores, and lower HIV viral load (r = −0.39, p < 0.01) was associated with higher BERMA Memory for Medications subscale scores. No other demographic or virological characteristics were significantly associated with BERMA subscale scores.

Table 2.

Sample Demographic and Clinical Characteristics (N = 100)

Demographic characteristic	M or %	SD or IQR
Age (years)	46.27	7.24
Gender (% male)	68%
Ethnicity (% Puerto Rican descent)	75%
Born in the United States	67%
Education (years)	12.08	2.47

Clinical characteristics	M or %	SD
CD4 count^a	444	421.50
HIV viral load^b	2.31	1.34
HIV viral load (% undetectable)	54%
Duration of HIV infection (years)	14.79	5.83
MEMS medication adherence (mean percent)	81%	24%
Optimal adherence (% with ≥95% adherence)	32%

Median/IQR, ^blog₁₀ transformed.

IQR, interquartile range; MEMS, medication event monitoring system; SD, standard deviation.

Table 3 illustrates participant average T-scores on the NC tests, as well as scores on the BERMA, the AMAS, and the BDI-II. In terms of the BERMA results, the subscale means, where higher scores indicate more positive attitudes, suggested that the sample has mostly positive attitudes toward medications (M = 37.73, SD = 6.40) and health professionals (M = 80.98, SD = 16.46), endorsing items, such as “I believe the treatment plan my doctor has prescribed for me will effectively manage my symptoms” and “I trust the medical advice my doctor gives me.” Additionally, a high mean on the Memory for Medications subscale (M = 94.77, SD = 15.94) implies the sample has “good” memory for taking medications (e.g., “I am good at remembering to take my medication” and “I can remember my medication regimen as well as always”). An AMAS Total US Acculturation score of 23.71 out of a possible 28 indicates that participants, on average, reported a high level of US acculturation. The sample mean for the BDI-II (M = 10.40, SD = 9.77) suggests that, overall, the study participants were not experiencing a significant amount of depressive symptoms.

Table 3.

Descriptive Statistics for Neurocognitive Variables, Beliefs Related to Medication Adherence Scores, and Abbreviated Multidimensional Acculturation Scale Scales (N = 100)

Scores	M	SD
NC average T-scores learning
HVLT-R total recall	31.95	13.26
BVMT-R total recall	36.44	12.80
Memory
HVLT-R delayed recall	30.77	14.22
BVMT-R delayed recall	39.72	13.35
Executive functioning
Trails B	45.38	11.95
WCST-64 total errors	36.90	10.46
BERMA subscale scores
Attitude about medication	37.73	6.40
Memory for medications	94.77	15.94
Attitude toward health professionals	80.98	16.46
BDI-II total scores	10.40	9.77
AMAS subscale and total scores
Cultural identity—US	21.24	3.97
Language competence—English	32.12	4.36
Cultural competence—US	17.77	3.91
Total US acculturation	23.71	3.29

Interpretative notes: Higher NC T-scores reflect better functioning (NC T-scores of 30–34 = mild-to-moderate impairment, 35–39 = mild impairment, 40–44 = below average, 45–54 = average, and 45+ = above average); higher BERMA scores reflect more positive attitudes; higher BDI-II scores reflect greater depressive symptomatology, BDI-II <13 = minimal symptoms; higher AMAS scores reflect greater acculturation.

AMAS, Abbreviated Multidimensional Acculturation Scale; BDI-II, Beck Depression Inventory-Second edition; BERMA, Beliefs Related to Medication Adherence; BVMT-R, Brief Visuospatial Memory Test-Revised; HVLT-R, Hopkins Verbal Learning Test-Revised; NC, neurocognitive; WCST-64, Wisconsin Card Sorting Test-64 Item Version.

A series of bivariate correlations were computed to examine the associations between executive functioning, learning, and memory (respectively) with the BERMA subscales. As illustrated in Table 4, there was a significant positive relationship between performance on the Wisconsin Card Sorting Test-64 item version (i.e., WCST-64; Total Errors T-score; a measure of executive functioning) and the Attitudes Toward Medication subscale (r = 0.28, p < 0.01) as well as the Memory for Medications subscale (r = 0.23, p = 0.03). No significant relationships were found between Trails B and the BERMA subscales (all ps > 0.10). In terms of learning and memory, the HVLT-R Total Recall T-score (a measure of verbal learning) was significantly positively correlated with the BERMA Memory for Medications subscale (r = 0.30, p < 0.01). The HVLT-R Delayed Recall T-score (a measure of verbal memory) was also significantly correlated with the Memory for Medications subscale (r = 0.26, p = 0.01). There were no significant relationships between BVMT-R T-scores (i.e., visual learning and memory) and the BERMA subscales (all p's > 0.10).

Table 4.

Correlations Between Beliefs Related to Medication Adherence Subscales and Neurocognition, Depression, and Acculturation (N = 100)

	Attitudes toward medications	Memory for medications	Attitudes for health professionals
Learning
HVLT-R total recall	0.09	0.30^**	0.14
BVMT-R total recall	0.10	0.06	0.06
Memory
HVLT-R delayed recall	0.17	0.26^*	0.05
BVMT-R delayed recall	0.07	0.06	−0.01
Executive functioning
Trails B	0.11	0.05	0.11
WCST-64 total errors	0.28^**	0.23^*	0.11
BDI-II	−0.33^**	−0.42^***	−0.45^***
AMAS
Cultural identity—US	0.11	−0.03	0.15
Language competence—English	0.05	0.22^*	0.20^t
Cultural competence—US	0.13	0.07	0.16
Total US acculturation	0.11	0.12	0.21^*

Interpretive note: Higher scores NC T-scores reflect better neurocognitive functioning; higher BDI-II scores reflect greater depressive symptomatology; higher AMAS scores reflect greater acculturation.

p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^t Trend.

p Values in bold are statistically significant.

A series of bivariate correlations were computed to examine the associations between the AMAS subscales with the BERMA. As seen in Table 4, AMAS English Language Competence subscale was significantly positively correlated with the BERMA Memory for Medications subscale (r = 0.22, p = 0.03). There was also a trend between the AMAS English Language Competence subscale and the BERMA Attitudes Toward Health Professionals subscale (r = 0.20, p = 0.05). The AMAS acculturation summary score, Total US Acculturation, was significantly correlated with BERMA Attitudes toward Health Professionals (r = 0.21, p = 0.04). To examine the associations between depressive symptomology and the BERMA, a series of bivariate correlations were computed. As illustrated in Table 4, the BDI-II was negatively correlated with all BERMA subscales: BERMA Attitudes Toward Medications (r = −0.33, p < 0.01); BERMA Memory for Medications (r = −0.42, p < 0.01); and BERMA Attitudes Toward Health Professionals (r = −0.45, p < 0.01).

Next, independent sample t-tests were computed to examine relationships between medication adherence and both BERMA scores and NC T-scores. A trend was observed such that participants with optimal medication adherence (i.e., 95% or better adherence) had somewhat higher scores on the BERMA Attitudes Toward Health Professionals scale (M = 85.31, SD = 14.49) compared to scores among those with less than optimal adherence (M = 79.17, SD = 17.12; t = −1.67, p < 0.10). No significant group differences were observed on the other two BERMA subscales or on any NC T-scores (all p's > 0.10).

Last, a series of linear regressions were computed to examine the relative contribution of each significant predictor of BERMA subscale scores (i.e., Attitudes about Medication, Memory for Medications, and Attitudes Toward Health Professionals). For models predicting each individual BERMA subscale, variables with significant univariate relationships with BERMA scores were entered in Step 1. Table 5 describes the omnibus models and each predictor variable. Briefly, both BDI-II scores and WCST-64 total errors significantly predicted about 17% of the variance in BERMA Attitude about Medication scores. BDI-II scores and HIV viral load levels significantly predicted about 33% of the variance in BERMA Memory for Medications scores. HVLT-R Total and Delay T-scores were not significant predictors. Finally, BDI-II scores significantly predicted about 24% of the variance in BERMA Attitude Toward Health Professionals scores. CD4 count and AMAS Total US Acculturation scores were not significant predictors.

Table 5.

Linear Regressions Predicting Beliefs Related to Medication Adherence Subscale Scores (N = 100)

BERMA score and model predictors	R²	B	SE B	Statistic	p
BERMA attitudes about medication	0.17			F = 8.47	<0.001
BDI-II total		−0.21	0.07	t = −2.97	<0.01
WCST-64 total errors		0.15	0.06	t = 2.48	0.02
BERMA memory for medications	0.33			F = 5.95	<0.001
HIV viral load^a		−4.86	1.34	t = −3.64	<0.001
BDI-II total		−0.54	0.18	t = −2.92	<0.01
HVLT total T-score		−0.08	0.19	t = −0.43	0.67
HVLT delay T-score		0.14	0.18	t = 0.80	0.43
WCST-64 total errors		0.14	0.17	t = 0.82	0.42
AMAS language competence—English		0.27	0.41	t = 0.65	0.52
BERMA attitude toward health professionals	0.24			F = 8.91	<0.001
CD4		0.01	<0.01	t = 1.84	0.07
BDI-II total		−0.67	0.17	t = −3.87	<0.001
AMAS total US acculturation		−0.67	0.48	t = 1.40	0.17

Log₁₀ transformed.

SE, standard error.

Discussion

This study examined the role of NC functioning, sociocultural factors, and depressive symptoms in beliefs regarding medication adherence among Latinx PLWH. In sum, study results indicated that better executive functioning was related to more positive attitudes toward medications and better memory for medication, and better verbal learning and memory ability was related to better memory for medications. No other relationships between NC functioning and BERMA scores were observed. Similarly, participants with optimal electronically monitored medication adherence did not differ from those with suboptimal adherence on NC T-scores. However, those with optimal adherence had more positive attitudes toward health professionals, but only at the trend-level. This is consistent with our previous work, which showed that higher scores on the BERMA attitudes toward health professionals subscale were significantly associated with better medication adherence in a larger sample of both Latinx and non-Latinx White PLWH.⁴⁶ Regarding acculturation, English language competence was related to better memory for medications, and higher total US acculturation (i.e., the AMAS summary score) was significantly associated with more positive attitudes toward health professionals. Depressive symptomology was strongly associated with poorer attitudes toward health professionals and medications, as well as higher self-reported memory difficulties for medications. Finally, in a series of stepwise, linear regressions, depression symptoms and executive functioning significantly predicted BERMA attitudes about medications (about 17% of the variance), depression symptoms and HIV viral load significantly predicted BERMA Memory for Medications (about 33% of the variance), and depression symptoms significantly predicted BERMA Attitudes Toward Health Professionals (i.e., about 24% of the variance).

The current findings are generally consistent with prior research on neurocognition and HIV medication adherence; in univariate analyses, one of our two executive functioning measures (i.e., WCST-64, our more complex EF measure) was associated with more positive attitudes toward medications, and both verbal learning and memory were associated with better reported memory for medications.^25,47
–49 The relationship between objective learning/memory performance and self-reported memory for medications is not surprising nor is the relationship among better conceptual reasoning, decision-making, and set-shifting and positive attitudes toward medications.⁵⁰ For example, PLWH often face difficult life circumstances in combination with having to manage their illness; individuals with better executive functioning likely navigate these complex situations more successfully.⁵¹ In turn, this can lead to improved quality of life and health outcomes, which may be attributed to healthy decisions and the effects of medication.

In terms of attitudes toward health professionals, acculturation may be important. Latinx individuals with lower overall acculturation to US culture had significantly less positive attitudes toward health professionals. Confusion and lack of cultural connection may create ruptures in the patient–physician relationship, leading to poorer attitudes toward health care providers. Of the individual BERMA subscales, only language competence reached trend-level. It is possible that discomfort with language may manifest itself not only as a discomfort toward the health professional⁵² but also as acculturation to the dominant culture that more generally may impact these attitudes.

The relationship between memory for medications and verbal learning/memory and English proficiency can be interpreted within an information-processing model. Individuals with deficits in verbal learning and memory may have more difficulty understanding information from their providers regarding their medication regimens and instructions often associated with them (e.g., taking with food, medication must be refrigerated). Research has shown that learning/memory and information processing impairments have been associated with worse performance on medication management tasks in PLWH⁵³ Regarding English language, health professionals frequently give medical information in English or via an interpreter with limited medical knowledge. Poor English language proficiency not only interferes with learning these instructions but also hinders the ability to ask clarifying questions.⁵²

Verbal learning and memory difficulties may make it difficult for patients to recall the intricacies of their medications if they have not learned them in the first place. Additionally, since medical information is commonly given in English, this may pose problems for patients without adequate proficiency in the language.

Further, the demand for bilingual and bicultural Latinx doctors specializing in HIV is likely high considering demographic trends in HIV transmission. Given this imbalance, it is possible that Latinx patients, particularly Spanish-speaking and nonnative English-speaking patients, may experience more difficulty understanding their medical instructions. In combination, these findings demonstrate the importance of ensuring that patients understand medical directions, as it is potentially dangerous to assume that all instructions are understood and will be followed.

An important finding of the current study was the significant role of depressive symptoms on all BERMA subscales, even in a sample without severe depressive symptomatology. Specifically, the association between depressive symptoms with attitudes toward medications and health professionals, as well as self-reported memory for medications, all had medium effect sizes. Depression is known to negatively impact medication adherence across various diseases.^54
–56 The impact of depression on adherence, in combination with the findings of the current study, point to the importance of assessing for and treating depressive symptomology in an effort to support the patient–physician relationship and presumably improve health outcomes. It is also noteworthy that in a multivariate model accounting for depressive symptomology, executive functioning performance on a set shifting and problem-solving task remained a significant predictor of attitudes toward medications. Individuals with the capacity to successfully handle various stressors and situations may be transferring these skills to cope with symptoms of depression or general malaise. In turn, managing these feelings can improve attitudes overall, including those for medications, as seen in the present findings. In our overall multivariate models, better executive functioning and less depressive symptomatology predicted better reported attitudes toward medications. Less depressive symptomatology and lower viral load predicted better reported memory for medications. Less depressive symptomatology predicted more positive attitudes toward health professionals. Overall, these findings suggest that depressive symptomatology was strongly associated with beliefs about medications across all domains. Therefore, even in the absence of clinically significant levels of depression, these symptoms are important to consider among Latinx PLWH.

Study limitations

While our findings provide valuable information on beliefs about medication and health professionals among Latinx PLWH, there are limitations as well. Approximately 65% of the sample was born in the United States and over 70% of the sample identified as Puerto Rican, limiting the generalizability for Latinx from other countries of origin. Relatedly, participants were highly acculturated to US culture and had high levels of English-language fluency, which may limit generalizability for less acculturated, less bilingual Latinx individuals. Moreover, this restricted range in our sociocultural variables, including acculturation scores, may impact our findings. Greater range in relevant sociocultural characteristics, and more in-depth sociocultural characterization of the sample, may increase our ability to detect relationships among these variables. Additionally, this study sought a sample of exclusively Latinx PLWH, yet HIV also disproportionately affects Black/African American individuals. Data with Latinx PLWH may not be generalizable to those from other racial/ethnic backgrounds. In addition, men comprised more than two-thirds of the sample, and about two-thirds of participants had undetectable plasma HIV viral load. Therefore, these results may not generalize to women, nonbinary individuals, and/or those with detectable viral loads. This study's sample size was relatively small, and informant report was not available. Finally, other factors that may affect both NC functioning and HIV antiretroviral adherence, such as HIV-related stigma, were not examined in this study but may affect these relationships.⁵⁷

Clinical implications

Our results have significant implications for how medical information is disseminated to patients. The relationship between verbal learning and memory, English language competence, and memory for medications highlights the importance of ensuring that patients for whom English is not their first language understand their medical information. Visual methods of conveying information (i.e., brochures, visual aids) may improve communication.⁵⁸ Assessment of information comprehension, to address lack of understanding, may also be beneficial. Assessment might include a brief questionnaire that the patient completes or that is read aloud to the patient regarding their specific medication regimen, which then the provider could review and provide any needed remediation for issues that were unclear for the patient. Motivational interviewing may also be valuable to assess comprehension whereby the patient reflects back to the provider their understanding of the instructions provided to illuminate the need for further clarification.

Considering the strong relationship between depressive symptomatology and beliefs regarding medication adherence, health professionals should be aware of their patients' psychological health, as this may have salient physical health implications. This is particularly relevant for PLWH who show higher levels of depression and overall mental illness than HIV-negative individuals.^31,32 Research has shown several links between HIV and mental health, including higher levels of mental illness among PLWH.³¹

The relationship between neurocognition, depressive symptoms, and beliefs about medication adherence suggests the need for targeted and synergistic interventions that address the unique NC, psychiatric, and attributional characteristics of this population to tailor and enhance medication adherence, and ultimately health outcomes, within this group. For example, an intervention combining cognitive remediation and cognitive-behavioral treatment, focusing on both HIV-related NC impairment (i.e., executive functioning, learning, and memory) and psychiatric comorbidities that impact health beliefs in this population (i.e., depression), may be beneficial. Such culturally tailored interventions may improve understanding of the importance of optimal medication adherence and provide the necessary tools to engage in improved health behaviors.

Future directions

Based on the current findings, future research should further explore relationships between neurocognition (particularly executive functioning), mental health, and attitudes toward medication to better understand health behaviors, decisions, and beliefs in PLWH. It is important for future research to examine the relationship between verbal learning and a patient's comprehension of a health professional's instructions. Creating and piloting assessments of patients' understanding of their medical instructions before leaving the medical facility may assist health professionals in ensuring that their patients understand their directions. Further exploration of this relationship can help tailor the way in which medical information is dispensed to individual patients. It would be valuable to replicate this project on a larger scale with a more diverse sample. Additionally, the role of depression in beliefs regarding medication adherence in the study should be explored more thoroughly, especially among samples with more severe levels of depression symptomatology. Lastly, developing a combined, culturally tailored intervention targeting the NC and psychological symptoms common among PLWH may improve medication adherence and health outcomes in this population.

Footnotes

Acknowledgments

The authors would like to thank Letty Mintz, A.N.P. and Rhonda Burgess, M.A.; our community partners through the Manhattan HIV Care Network, the Harlem Community Academic Partnership; and most importantly, all of the individuals who participated in this study.

Authors' Contributions

A.F.: conceptualization, methodology, formal analysis, investigation, writing—original draft, writing—reviewing and editing, and visualization; K.C., C.M., M.A.R., J.M., and A.R.: investigation, writing—original draft, and writing—reviewing and editing; D.B.: methodology, writing—reviewing and editing, and supervision; A.A.: investigation, project administration, writing—original draft, writing—reviewing, and editing; M.R.M.: methodology, resources, writing—original draft, writing—reviewing and editing, supervision, and funding acquisition.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by K23MH07971801 and an Early Career Development Award from the Northeast Consortium for Minority Faculty Development (to MRM).

References

U.S. Census Bureau. Supplementary Tables on Race and Hispanic Origin:. 2020 Census Redistricting Data (PL 94-171). 2021.

Centers for Disease Control and Prevention. HIV Surveillance Report: Diagnoses of HIV Infection in the United States and Dependent Areas, 2019. 2021.

Wilson

, Laws

, Safren

, et al. Provider-focused intervention increases adherence-related dialogue, but does not improve antiretroviral therapy adherence in persons with HIV. J Acquir Immune Defic Syndr (1999), 2010; 53(3):338.

de Los Rios

, Okoli

, Punekar

, et al. Prevalence, determinants, and impact of suboptimal adherence to HIV medication in 25 countries. Prev Med, 2020; 139:106182.

, Sarafian

, Silvestre

, et al. Evaluation of adherence and factors affecting adherence to combination antiretroviral therapy among White, Hispanic, and Black men in the MACS Cohort. J Acquir Immune Defic Syndr, 2009; 52(2):290–293.

Arnsten

, Li

, Mizuno

, et al. Factors associated with antiretroviral therapy adherence and medication errors among HIV-infected injection drug users. J Acquir Immune Defic Syndr, 2007; 46:S64–S71.

Gwadz

, Cleland

, Freeman

, et al. Stopping, starting, and sustaining HIV antiretroviral therapy: A mixed-methods exploration among African American/Black and Latino long-term survivors of HIV in an urban context. BMC Public Health, 2021; 21(1):1–26.

Crim

, Tie

, Beer

, et al. Barriers to antiretroviral therapy adherence among HIV-positive Hispanic and Latino men who have sex with men—United States, 2015–2019. Morbid Mortal Wkly Rep, 2020; 69(40):1437.

Mcdonald-Miszczak

, Maris

, Fitzgibbon

, et al. A pilot study examining older adults' beliefs related to medication adherence: The BERMA survey. J Aging Health, 2004; 16(5):591–614.

10.

Mitzel

, Vanable

. Necessity and concerns beliefs and HIV medication adherence: A systematic review. J Behav Med, 2020; 43(1):1–15.

11.

Spire

, Duran

, Souville

, et al. Adherence to highly active antiretroviral therapies (HAART) in HIV-infected patients: From a predictive to a dynamic approach. Soc Sci Med, 2002; 54(10):1481–1496.

12.

Malcolm

, Ng

, Rosen

, et al. An examination of HIV/AIDS patients who have excellent adherence to HAART. AIDS Care, 2003; 15(2):251–261.

13.

Barclay

, Hinkin

, Castellon

, et al. Age-associated predictors of medication adherence in HIV-positive adults: Health beliefs, self-efficacy, and neurocognitive status. Health Psychol, 2007; 26(1):40–49.

14.

Bomfim

IGdO

, Santos

SDS

, Napoleão

. Adherence to antiretroviral therapy in people living with HIV/AIDS: A cross-sectional study. AIDS Patient Care STDs, 2022; 36(7):278–284.

15.

Sleath

, Roter

, Chewning

, et al. Asking questions about medication: Analysis of physician-patient interactions and physician perceptions. Med Care, 1999; 37(1):1169–1173.

16.

Zhang

, McMahon

, Leblanc

, et al. Association of medical mistrust and poor communication with HIV-related health outcomes and psychosocial wellbeing among heterosexual men living with HIV. AIDS Patient Care STDs, 2020; 34(1):27–37.

17.

Janz

, Becker

. The health belief model: A decade later. Health Educ Q, 1984; 11(1):1–47.

18.

Ellis

, Langford

, Masliah

. HIV and antiretroviral therapy in the brain: Neuronal injury and repair. Nat Rev Neurosci, 2007; 8(1):33–44.

19.

Ernst

, Itti

, et al. Changes in cerebral metabolism are detected prior to perfusion changes in early HIV-CMC: A coregistered 1H MRS and SPECT study. J Magnet Reson Imaging, 2000; 12(6):859–865.

20.

Stankoff

, Tourbah

, Suarez

, et al. Clinical and spectroscopic improvement in HIV-associated cognitive impairment. Neurology, 2001; 56(1):112–115.

21.

Saloner

, Cysique

. HIV-associated neurocognitive disorders: A global perspective. J Int Neuropsychol Soc, 2017; 23(9–10):860–869.

22.

Devlin

, Giovannetti

. Heterogeneity of neuropsychological impairment in HIV infection: Contributions from mild cognitive impairment. Neuropsychol Rev, 2017; 27:101–123.

23.

Heaton

, Clifford

, Franklin

, et al. HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy CHARTER Study. Neurology, 2010; 75(23):2087–2096.

24.

Heaton

, Marcotte

, Mindt

, et al. The impact of HIV-associated neuropsychological impairment on everyday functioning. J Int Neuropsychol Soc, 2004; 10(3):317–331.

25.

Hinkin

, Hardy

, Mason

, et al. Medication adherence in HIV-infected adults: Effect of patient age, cognitive status, and substance abuse. AIDS (London, England), 2004; 18(Suppl 1):S19.

26.

Bianchi

, Zea

, Poppen

, et al. Coping as a mediator of the impact of sociocultural factors on health behavior among HIV-positive Latino gay men. Psychol Health, 2004; 19(1):89–101.

27.

Villagran

, Hajek

, Zhao

, et al. Communication and culture: Predictors of treatment adherence among Mexican immigrant patients. J Health Psychol, 2012; 17(3):443–452.

28.

Robbins

, D'Aquila

, Morgello

, et al. Cultural influences on antiretroviral therapy adherence among HIV-infected Puerto Ricans. J Assoc Nurses AIDS Care, 2012; 23(6):531–538.

29.

Vissman

, Young

, Wilkin

, et al. Correlates of HAART adherence among immigrant Latinos in the southeastern United States. AIDS Care, 2013; 25(3):356–363.

30.

Rocha-Jiménez

, Pitpitan

, Cazares

, et al. “He is the same as me”: Key populations' acceptability and experience of a community-based peer navigator intervention to support engagement in HIV care in Tijuana, Mexico. AIDS Patient Care STDs, 2021; 35(11):449–456.

31.

Brandt

The mental health of people living with HIV/AIDS in Africa: A systematic review. Afr J AIDS Res, 2009; 8(2):123–133.

32.

Ciesla

, Roberts

. Meta-analysis of the relationship between HIV infection and risk for depressive disorders. Am J Psychiatry, 2001; 158(5):725–730.

33.

Martinez

, Hosek

, Carleton

. Screening and assessing violence and mental health disorders in a cohort of inner city HIV-positive youth between 1998–2006. AIDS Patient Care STDs, 2009; 23(6):469–475.

34.

Heaton

, Grant

, Matthews

Comprehensive Norms for an Expanded Halstead–Reitan Battery: Demographic Corrections, Research Findings, and Clinical Implications. Psychological Assessment Resources: Odessa, FL; 1991.

35.

Zea

, Asner-Self

, Birman

, et al. The Abbreviated Multidimentional Acculturation Scale: Empirical validation with two Latino/Latina samples. Cult Divers Ethnic Minor Psychol, 2003; 9(2):107.

36.

Arentoft

, Byrd

, Robbins

, et al. Multidimensional effects of acculturation on English-language neuropsychological test performance among HIV+ Caribbean Latinas/os. J Clin Exp Neuropsychol, 2012; 34(8):814–825.

37.

Poppen

, Reisen

, Zea

, et al. Predictors of unprotected anal intercourse among HIV-positive Latino gay and bisexual men. AIDS Behav, 2004; 8:379–389.

38.

Zea

, Reisen

, Poppen

, et al. Disclosure of HIV-positive status to Latino gay men's social networks. Am J Commun Psychol, 2004; 33(1–2):107–116.

39.

Beck

, Steer

, Brown

. Beck Depression Inventory-II. San Antonio, 1996; 78(2):490–498.

40.

Woods

, Moran

, Carey

, et al. Prospective memory in HIV infection: Is “remembering to remember” a unique predictor of self-reported medication management?. Arch Clin Neuropsychol, 2008; 23(3):257–270.

41.

Arnsten

, Demas

, Farzadegan

, et al. Antiretroviral therapy adherence and viral suppression in HIV-infected drug users: Comparison of self-report and electronic monitoring. Clin Infect Dis, 2001; 33(8):1417–1423.

42.

Pearson

, Simoni

, Hoff

, et al. Assessing antiretroviral adherence via electronic drug monitoring and self-report: An examination of key methodological issues. AIDS Behav, 2007; 11:161–173.

43.

McNabb

, Nicolau

, Stoner

, et al. Patterns of adherence to antiretroviral medications: The value of electronic monitoring. AIDS, 2003; 17(12):1763–1767.

44.

Bova

, Fennie

, Knafl

, et al. Use of electronic monitoring devices to measure antiretroviral adherence: Practical considerations. AIDS Behav, 2005; 9:103–110.

45.

IBM SPSS Statistical Package for the Social Sciences, Version 28.0. IBM: Armonk, NY; 2021.

46.

Rivera Mindt

, Arentoft

, Tureson

, et al. Disparities in electronically monitored antiretroviral adherence and differential adherence predictors in Latinx and non-Latinx White persons living with HIV. AIDS Patient Care STDs, 2020; 34(8):344–355.

47.

Heaton

, Marcotte

, Mindt

, et al. The impact of HIV-associated neuropsychological impairment on everyday functioning. J Int Neuropsychol Soc, 2004; 10(03):317–331.

48.

Murphy

, Johnston Roberts

, et al. Barriers and successful strategies to antiretroviral adherence among HIV-infected monolingual Spanish-speaking patients. AIDS Care, 2003; 15(2):217–230.

49.

Watermeyer

, Penn

. “Tell me so I know you understand”: Pharmacists' verification of patients' comprehension of antiretroviral dosage instructions in a cross-cultural context. Patient Educ Counsel, 2009; 75(2):205–213.

50.

Lovejoy

, Suhr

. The relationship between neuropsychological functioning and HAART adherence in HIV-positive adults: A systematic review. J Behav Med, 2009; 32:389–405.

51.

González-Andrade

, García-Torres

, Pérez-García

, et al. Assessment of executive functions as a measure of impairments in everyday functioning in persons with HIV. Psychol Health Med, 2022; 27(7):1443–1449.

52.

Sudore

, Landefeld

, Perez-Stable

, et al. Unraveling the relationship between literacy, language proficiency, and patient–physician communication. Patient Educ Counsel, 2009; 75(3):398–402.

53.

Thames

, Arentoft

, Rivera-Mindt

, et al. Functional disability in medication management and driving among individuals with HIV: A 1-year follow-up study. J Clin Exp Neuropsychol, 2013; 35(1):49–58.

54.

Tang

H-Y

, Sayers

, Weissinger

, et al. The role of depression in medication adherence among heart failure patients. Clin Nurs Res, 2014; 23(3):231–244.

55.

Lin

, Katon

, Von Korff

, et al. Relationship of depression and diabetes self-care, medication adherence, and preventive care. Diabetes Care, 2004; 27(9):2154–2160.

56.

Catz

, Kelly

, Bogart

, et al. Patterns, correlates, and barriers to medication adherence among persons prescribed new treatments for HIV disease. Health Psychol, 2000; 19(2):124.

57.

Thompson

, Muhammad

, Adimora

, et al. Internalized HIV-related stigma and neurocognitive functioning among women living with HIV. AIDS Patient Care STDs, 2022; 36(9):336–342.

58.

Schubbe

, Scalia

, Yen

, et al. Using pictures to convey health information: A systematic review and meta-analysis of the effects on patient and consumer health behaviors and outcomes. Patient Educ Counsel, 2020; 103(10):1935–1960.

59.

Kongs

, Thompson

, Iverson

, et al. Wisconsin Card Sorting Test—64 Card Computerized Version. Psychological Assessment Resources: Odessa, FL; 2000.

60.

Heaton

, Miller

, Taylor

, et al. Revised Comprehensive Norms for an Expanded Halstead-Reitan Battery: Demographically Adjusted Neuropsychological Norms for African American and Caucasian Adults (HRB). Psychological Assessment Resources, Inc.: Lutz, FL; 2004.

61.

Bechara

IGT Professional Manual. Psychological Assessment Resources: Lutz, FL; 2007.

62.

Benedict

, Schretlen

, Groninger

, et al. Hopkins Verbal Learning Test–Revised: Normative data and analysis of inter-form and test-retest reliability. Clin Neuropsychol, 1998; 12(1):43–55.

63.

Benedict

Brief Visuospatial Memory Test-Revised Professional Manual. Psychological Assessment Resources, Inc.: Odessa, FL; 1997.