Sleep Disturbance is Associated with Missing PrEP Doses Among Young Black Sexual Minority Men in The N2 study

Abstract

PrEP uptake and adherence among young Black sexual minority men (YBSMM), has been sub-optimal. Multiple studies identified sleep as an important determinant of medication adherence, although it has not been examined with regard to PrEP among YBSMM. This study utilized data collected from HIV-negative cisgender YBSMM in the Neighborhoods and Networks (N2) study in Chicago using PrEP (N = 70). Sleep quality was measured using the PHQ-9 and PrEP adherence questions were adapted from Reynolds et al., 2004. Bivariate and multivariable regression analyses were used to estimate associations between sleep and missing PrEP doses, controlling for relevant demographic and behavioral factors. YBSMM who reported sleep disturbance a moderate amount of time (aOR 7.59 [1.05 to 54.57]) were more likely to miss taking PrEP because they had too many pills to take. Sleep quality is an overlooked determinant of medication adherence, and may negatively impact YBSMM’s ability to consistently take PrEP.

Introduction

Young Black sexual minority men (YBSMM) make up the largest proportion of new HIV diagnoses in the United States annually [1]. Information on the pre-exposure prophylaxis (PrEP) utilization practices of YBSMM are crucial because at risk YBSMM far underutilize PrEP compared to their White counterparts [2–5]. HIV rates among YBSMM have remained stagnant for a number of years, and all Getting to Zero efforts must consider factors associated with PrEP utilization, persistence, and adherence patterns among this key population [6–10]. Understanding this disparity is fundamental to efforts to end the HIV epidemic in the United States [11–14]. Multiple studies modeling the impact of PrEP demonstrate that this pharmacologic intervention has the potential to decrease the HIV epidemic in groups most vulnerable to disease such as YBSMM, but uptake has been disproportionate to need. In a study of 2,021 sexual minority men (SMM), only 55% of Black sexual minority men (SMM) were likely to start PrEP compared to 62% of Hispanic and 68% of White men, respectively [15]. Furthermore, Black SMM were less likely to have discussions about PrEP with their providers than White SMM [15]. Importantly, even when PrEP is prescribed, sub-optimal PrEP adherence is a critical health issue for YBSMM [9, 16]. PrEP adherence is key to HIV prevention among YBSMM as seroconversions happen most commonly when PrEP is discontinued without utilizing alternative methods for HIV prevention [17, 18]. Multiple studies point towards stigma, medical mistrust, and lack of culturally competent care as reasons for why YBSMM have low PrEP uptake and adherence, but sleep quality is not often mentioned [19–21].

The purpose of this study was to explore associations between sleep quality and missed PrEP doses among YBSMM, the first of its kind. There are many studies examining the ways in which sleep quality impacts health outcomes and health behaviors, but very few that center the role of sleep in pharmacologic HIV prevention through the use of PrEP, and none have focused on young YBSMM [22, 23]. Furthermore, a study by Caceres et al. revealed that Black sexual minority men had higher rates of very low sleep duration than Black heterosexual men, in part, perhaps due to Black SMM’s lack of stable housing and increased exposure to neighborhood violence [24, 25].

Given the relationship between sleep and HIV medication adherence and medication adherence more broadly, it is important that we examine the role of sleep in HIV prevention efforts for YBSMM [26]. To illustrate, a study in Peru among persons living with HIV (PLWH) demonstrated that participants with sleep disturbances as measured by the Pittsburgh Sleep Quality Index had difficulty with antiretroviral medication adherence [26]. Multiple studies have demonstrated the role of sleep in consolidation of cognition and memory, both of which are key factors in remembering to take routine medications such as PrEP [27, 28]. For instance, a study involving the subjective measure of sleep quality among PLWH showed that those who had worse sleep quality had worse memory and executive function [29]. We would expect that some of the same stressors and socioeconomic factors that impact sleep quality and medication adherence for Black men living with HIV, such as lower socioeconomic status, higher rates of homelessness, incarceration, and exposure to violence, would be the same for YBSMM taking PrEP given the parallel social circumstances, both childhood and adult, of both groups [30–32]. Furthermore, the YBSMM in our study may have been more susceptible to poor sleep quality given their potential for exposure to urban violence and housing instability, which has been shown to interfere with sleep quality and duration [33–35].

Methods

This study analyzed data from HIV-negative cisgender YBSMM in the Neighborhoods and Networks (N2) Cohort Study in Chicago from January 19, 2018 to December 11, 2019, which included two hundred fifty-five self-identified HIV-negative cisgender Black men to recruit YBSMM from their own social networks through the use of respondent driven sampling [36, 37]. Briefly, the N2 Study aimed to examine the social, contextual, network, and geospatial contexts of HIV prevention and care among YBSMM [36]. Study participants were eligible for participation if they were sixteen to thirty-four years old, identified as African-American or Black, assigned male at birth, reported a sexual encounter with a man or transgender woman in the past twelve months, lived in the Chicago metropolitan area, and were willing to provide informed consent at the time of the study visit [36]. Of those two hundred fifty-five participants, only seventy self-reported current PrEP use, and only those who stated they were currently taking PrEP were included in the analytic sample.

Measures

All N2 participants were tested for HIV as part of study procedures. HIV-negative participants were included if they were currently taking PrEP [36]. The sleep-related questions were measured using the Patient Health Questionnaire-9 (PHQ-9), which is a widely utilized tool in health screening visits [38]. Participants were asked, “During the past week, you had trouble falling or staying asleep, or sleeping too much” to which participants answered in categorical format consisting of “rarely or none of the time (less than 1 day)”, “some or a little of the time (1 to 2 days)”, “occasionally or a moderate amount of time (3 to 4 days)”, “all of the time (5 to 7 days)”.

Reasons for missing or forgetting PrEP doses were measured using five items adapted from Reynolds et al. [39]. The questions selected included: “In the past month, how often have you missed taking your medications because you were busy with other things?”; “In the past month, how often have you missed taking your medications because you simply forgot?”; “In the past month, how often have you missed taking your medications because you had a too many pills to take?”; “In the past month, how often have you missed taking your medications because you fell asleep or slept through your dose time?”; “In the past month, how often have you missed taking your medications because you felt depressed or overwhelmed?” To which participants responded: “never”, “rarely”, “sometimes”, or “often”.

Analyses

Descriptive statistics were used to characterize the study sample. Percentages, counts, medians, and interquartile ranges were used to examine discrete variables. An exploratory analysis using multiple ordinal regressions were conducted to examine the relationship between sleep as reported in the PHQ-9 and the frequency of missing PrEP. We assumed a proportional odds ratio between each category. We controlled a priori for the following factors: age, sexual orientation, income, and education in our multivariable model [40]. Transgender women were excluded from the sample as there were only 3 of them, and we consider this population to be different from cisgender men socially, emotionally, and behaviorally [41]. All analyses were conducted using Stata 15, College Station, Texas.

Results

Demographics

Seventy (100%) of the men of the analytic sample were Black and (10%) also identified as Hispanic. Forty-two (60%) completed high school or some college. Forty (57%) identified as gay and twenty-five (36%) identified as “other” (Table 1). Thirty-three (47.1%) of the men sampled had some or moderate sleep disturbance while 6 (8.6%) had sleep disturbance all of the time.

Table 1.

Characteristics of HIV negative YBSMM using PrEP in the N2 study, Chicago 2018

Variables (N = 70)	N(%) or median (IQR)
Age	26 (23 to 28)
Race/ethnicity
Black	70 (100)
Hispanic	10 (14)
Education
<High school	16 (22.9)
High school	24 (34.3)
Some College	18 (25.7)
BS or MS	12 (17.1)
Sexual Orientation
Gay	40 (57.1)
Bisexual	4 (5.7)
Straight	1 (1.4)
Other	25 (35.7)
Income
< $5,000	13 (18.6)
$5,000 to $29,999	36 (51.4)
>$30,000	21 (30.0)
Sleep Disturbance
Rarely	31 (44.3)
Some	18 (25.7)
Moderate	15 (21.4)
All the time	6 (8.6)

*BS or MS: Bachelor’s of science or Master of science degree

*Other refers to participants who wrote in sexual orientations other than “gay”, “bisexual”, or “straight” which included 25 separate orientations including “No Label” or “Confused” among other descriptors

*Sleep disturbance as defined by self-report on PHQ-9

Associations between sleep disturbance and missing PrEP doses

Table 2 includes bivariate and multivariable models examining the association between sleep and reasons for missing PrEP. In adjusted models, men with moderate sleep disturbance were more likely to miss taking PrEP because they had too many pills to take compared to those who rarely had sleep disturbance (aOR 7.59 [1.05 to 54.57]). Men with sleep disturbance all of the time (OR 11.30 [1.19 to 107.53]) were more likely to miss their PrEP because they felt depressed or overwhelmed compared to those who rarely had sleep disturbance, but this finding did not persist in the adjusted analysis.

Table 2.

Associations between sleep quality and PrEP adherence from the N2 Study, Chicago, Illinois (N = 70)

	Reasons for missing or forgetting PrEP doses
Sleep disturbance		Busy OR / aOR (95% CI)		Forgot OR / aOR (95% CI)		Too many OR / aOR (95% CI)		Slept OR / aOR (95% CI)		Depressed OR / aOR (95% CI)
Rarely		-	-	-	-	-	-	-	-	-	-
Some		1.14 (0.35 to 3.72)	3.06 (0.65 to 14.44)	0.79 (0.24 to 2.56)	0.88 (0.23 to 3.42)	6.00 (0.57 to 62.69)	6.73 (0.73 to 61.84)	1.01 (0.31 to 3.32)	1.95 (0.49 to 7.71)	0.55 (0.02 to 14.21)	0.62 (0.03 to 13.01)
Moderate		1.99 (0.52 to 7.65)	2.83 (0.53 to 15.24)	1.26 (0.35 to 4.61)	1.09 (0.27 to 4.33)	10.91 (1.10 to 108.55)	7.59 (1.05 to 54.57)	1.81 (0.52 to 6.29)	1.52 (0.44 to 5.23)	2.10 (0.20 to 22.01)	1.52 (0.14 to 17.00)
All the time		1.44 (0.23 to 9.11)	1.89 (0.23 to 15.64)	3.16 (0.33 to 30.43)	2.54 (0.22 to 28.90)	6.00 (0.32 to 112.26)	6.17 (0.44 to 87.47)	1.58 (0.27 to 9.17)	1.56 (0.28 to 8.63)	11.30 (1.19 to 107.53)	8.91 (0.50 to 157.20)

^*Busy: busy with other things

^*Forgot: forgot to take pills

^*Too many: too many pills

^*Slept: slept through dose

^*Depressed: felt depressed or overwhelmed

^*Covariates in adjusted analysis: age, sexual orientation, income, education

Discussion

In the first study to assess associations between sleep and PrEP adherence in YBSMM, we found that sleep disturbance a moderate amount of time was associated with missing PrEP due to having too many pills compared to those who reported rarely having sleep disturbance. Additionally, we found that nearly 56% of our sample admitted to having sleep disturbance. In a study of women aged 15–24 taking antiretroviral (ARV) medications for HIV, some of the reasons why they reported forgetting to take their ARV were increased pill burden and depressed mood [42]. Additionally, reporting sleep disturbance all of the time was associated with forgetting to take PrEP because participants felt depressed or overwhelmed. Xie et al. have demonstrated that poor sleep impacts working memory, and that both poor sleep quality and depressed mood were predictive of capacity for working memory among college students at University of California, Riverside [43]. This study helps to contextualize some of the findings our study demonstrates, and further underscores the impact of sleep quality on skills that require focus and attention, such as remembering to take pills, particularly if there are many of them. Further, depression can exacerbate sleep dysfunction resulting in insomnia, thus creating a loop of poor of executive function and forgetfulness, which might explain the relationship between having sleep disturbance all of the time and forgetting PrEP due to depressed mood [44]. This also represents another opportunity for utilization of the PHQ-9 in the assessment of not only depression screening, but for sleep disturbance as well. Though we did not find missing PrEP doses due to depression to be significant in multivariable analyses, this finding is still striking given that many YBSMM experience depression, and we hypothesize that perhaps in a larger sample, we might have found the same results in the multivariable analysis [45, 46].

This study has a few limitations worth mentioning. One is that there was only one sleep item utilized to assess sleep disturbance. This is not an exhaustive analysis of sleep quality; however, it may be practical and can be easily and inexpensively implemented in a clinical setting. The findings in this study may not be generalizable as the study population was a small cohort of Black sexual minority men in an urban setting. We also did not assess for substance use in this wave of the study, which has been shown to impact sleep quality and duration, so residual confounding is possible [47]. Further, our sleep variable examined sleep difficulties over the past week while the questions pertaining to missed PrEP doses applied to the past month so the time relationship between both items are not congruent. This likely underestimates the level of sleep disturbance in our sample given the short interval of consideration based on the confines of the PHQ-9. Additionally, our sample was quite small, thus likely underestimating other reasons why YBSMM might have missed PrEP doses. The reasons for missing PrEP examined in this analysis were not exhaustive and in-depth qualitative interviews may help to elucidate additional reasons for missing PrEP doses among YBSMM. Some key reasons why YBSMM may have issues with poor quality sleep are likely related to many sociodemographic factors which have been demonstrated in other studies such as housing instability, developing adolescent brains, greater likelihood of working at night, and potential exposure to neighborhood violence which might cause sleep disturbance, hypervigilance, or lack of safety and security [36, 48–52].

Conclusions

Our work underscores the importance of sleep in remembering to take PrEP for HIV prevention among YBSMM, who have the highest rates of HIV in the United States. It also provides an additional avenue for intervention to improve pharmacological HIV prevention among this key population. Sleep quality may play a role in the ability of YBSMM to consistently take PrEP, which is essential for prevention of HIV seroconversions in populations not using barrier methods of HIV prevention [53]. While it cannot be understated that many of the reasons for poor sleep quality among YBSMM are related to structural issues such as neighborhood safety and availability of reliable and consistent housing, there may be some steps which can be undertaken at the individual level to improve sleep in this population [25, 36, 47, 54].

Interventions for sleep quality improvement (e.g., smartphone apps, housing) may represent one way to increase PrEP adherence among YBSMM. Other strategies that might help with medication adherence regardless of sleep quality might be utilization of an electronic pillbox, a cell phone pill reminder, or a peer navigator for PrEP adherence [4, 55]. A study by Nam et al. demonstrated that Black persons with lower educational levels had dysfunctional beliefs about sleep, as defined by the Dysfunctional Beliefs and Attitudes about Sleep scale, which measures worry about sleep, consequences of insomnia, sleep expectations, and medication, and, therefore, worse sleep quality [56]. It is possible that interventions which aim to educate YBSMM to improve sleep hygiene or the use of cognitive behavior therapy (CBT) to increase adaptive behaviors to improve sleep could be helpful [57, 58]. While improvements in sleep hygiene may help YBSMM have more restful sleep, the social, neighborhood, and cultural contexts in which that sleep is obtained must also be considered, especially in the context of the barriers to PrEP adherence among this particularly vulnerable group.

Availability of Data and Material

The data underlying the results presented in the study are available upon request from Jade Pagkas-Bather: jpagkasbather@medicine.bsd.uchicago.edu.