Abstract
Importance
Alcohol use disorder (AUD) has been associated with reduced adherence to and discontinuation of HIV preexposure prophylaxis (PrEP), potentially compromising its effectiveness.
Objective
This study examines the relationship between AUD and PrEP adherence and continuation.
Design
Retrospective cohort study using MarketScan Commercial Claims data.
Participants
We included individuals aged 16–64 who initiated PrEP between January 1, 2014, and December 31, 2021, and had continuous insurance coverage.
Main Measures
AUD diagnosis was identified within six months before PrEP initiation. Adherence was measured using the proportion of days covered (PDC) over 180 days and categorized by clinically relevant thresholds (≥ 85%, ≥ 80%, ≥ 57%). Continuation was assessed based on uninterrupted PrEP supply without a ≥ 30-day gap. Statistical analyses included propensity-score matching and regression modeling.
Key Results
Among 43,913 eligible individuals, 1,245 (2.84%) had an AUD diagnosis prior to PrEP initiation. In a matched sample of 1,153 individuals, those with AUD had lower mean PDC (59.54% [34.14] vs. 65.85% [33.18]; p < 0.001) and fewer mean days of continuous PrEP use (107 [67.20] days vs. 119.90 [65.21] days; p < 0.001) compared to the Without AUD group. Regression analyses showed individuals with AUD had 6.31% lower mean PDC (95% CI: -9.05% to -3.57%; p < 0.001) and 12.93 fewer days of PrEP continuity (95% CI: -18.34 to -7.52 days; p < 0.001) compared to the Without AUD group. Findings may not be generalizable beyond commercially insured individuals, and we may have not captured all factors that influence PrEP adherence.
Conclusion
Individuals with AUD before PrEP initiation exhibited lower adherence and continuation compared to those without AUD. Targeted interventions may be required to enhance PrEP adherence and continuity in this population.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1007/s11606-025-10114-7.
INTRODUCTION
New diagnoses of HIV in the US have remained stable between 2017 (38,433) and 2023 (38,850) and remain far higher than the national target of 3,000 new cases by 2030 [1–3]. These trends highlight the importance of expanding access to primary prevention strategies, such as HIV pre-exposure prophylaxis (PrEP). PrEP—HIV medications that have been approved for prevention—have been recommended by the Centers for Disease Control and Prevention (CDC) and the US Preventive Services Task Force for adults and adolescents who face substantial risk of HIV [4, 5]. The effectiveness of PrEP is contingent on adherence (taking medication as advised or indicated) and continuation or persistence (uninterrupted medication use or supply). Consistent use of PrEP can reduce the risk of HIV from sex by 99% and from injection drug use by 74% [6, 7]. However, nonadherence can reduce or eliminate PrEP’s protective effect. A CDC analysis of trial data found that when oral PrEP adherence fell to approximately 40% or lower, its association with HIV risk reduction was no longer statistically significant (RR 0.93, 95% CI: 0.72–1.20) [6].
Alcohol use, particularly unhealthy consumption, is one of several factors that affect PrEP adherence and continuation—two steps in the PrEP care continuum [8]. For example, men who have sex with men (MSM) in Toronto who engaged in unhealthy alcohol use, defined as having an Alcohol Use Disorders Identification Test (AUDIT) score of ≥ 16, had nearly 7 times higher odds of nonadherence (95% CI: 1.49–30.33; p = 0.013) compared to low-risk drinkers [9]. In Kenya, a study found that alcohol use disorder (AUD), a category of unhealthy alcohol use [10], was associated with discontinuation among MSM and transgender women who initiated PrEP [11]. Other studies, however, have found that alcohol use does not affect PrEP use, suggesting that individuals who use alcohol can use PrEP effectively [12–16].
To build on this literature, we sought to assess the effect of AUD diagnoses on PrEP adherence and continuation using large, multi-year data from a real-world (i.e., non-trial) patient population. While most prior studies on this topic have focused on self-reported alcohol use or AUD symptoms (measured via AUDIT), we focused on clinical AUD diagnoses, which represent an important and necessary step in treating AUD [17]. Findings from this study can inform PrEP care guidelines and identify the need for interventions that address AUD and PrEP use concurrently.
METHODS
Data Source
This retrospective cohort study followed the STROBE reporting guideline [18]. We used Merative MarketScan® data, which contains a convenience sample of > 25 million individuals per year with employer-sponsored commercial health insurance and their dependents. The data includes demographic information, diagnoses in International Classification of Diseases, Ninth and Tenth Revision (ICD-9 and ICD-10) codes, and records for inpatient and outpatient services and outpatient prescriptions. Because MarketScan data are de-identified, this study was exempted from full review by The University of Texas Institutional Review Board (STUDY00006393).
Study Population
We included individuals aged 16–64 years at the time of PrEP initiation from January 1, 2014, to December 31, 2021. PrEP initiation was determined through an algorithm that has been adopted in previous studies involving real-world data [19]. First, we identified pharmacy claims for Food and Drug Administration (FDA)-approved medications for PrEP: 2 oral combination medications—tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) and tenofovir alafenamide/emtricitabine—and a long-acting injectable (cabotegravir). These medications were identified using National Drug Codes for the formulations used in PrEP (eTable 1 in Supplement 1). Second, we excluded patients diagnosed with HIV or hepatitis B, which are other indications for TDF/FTC (eTable 2 in Supplement 1). Consistent with other studies that use real-world-data [20, 21], we further limited the sample to individuals that had 6 months of continuous insurance coverage before and after the date of their PrEP initiation (i.e., the index date).
We classified our sample based on their AUD diagnosis history, which we assessed using ICD-9 and ICD-10 codes in inpatient and outpatient service claims, as previous claims-based studies have also done (eTable 3 in Supplement 1) [22–25]. Individuals who had an AUD diagnosis in the 6 months before PrEP initiation were included in the “Pre-initiation AUD” group. Individuals without AUD diagnoses in the baseline and follow-up periods were included in the “Without AUD” group. Due to the underdiagnosis of AUD and the potential for misclassification of AUD status in claims data [26, 27], we also identified individuals who received an AUD diagnosis in the 6 months after PrEP initiation (“Post-initiation AUD” group) as a supplemental analysis. Previous research has found that individuals diagnosed with AUD after PrEP initiation have similar characteristics as those who were diagnosed before initiation [28], suggesting that these individuals may have been experiencing AUD symptoms before starting PrEP.
MEASURES
Outcomes
Our primary outcomes were PrEP adherence and continuation. Adherence refers to the degree to which individuals follow the prescribed dosing regimen for their PrEP medication over time [29]. In this study, we measured adherence using proportion of days covered (PDC), calculated by dividing the number of days an individual has PrEP medications by the follow-up period (180 days) using pharmacy claims. We also created 3 binary adherence outcomes using clinically relevant PDC thresholds. The first was PDC ≥ 85%, representing an average of 6 doses per week for oral PrEP and about 4 doses of long-acting PrEP over 6 months [29]. An 80% PDC threshold is the most widely used cutoff in medication adherence studies and is recommended by the Pharmacy Quality Alliance [30]. Finally, we identified individuals who had a PDC ≥ 57%, which represents having 4 doses of oral PrEP per week (or 2 doses of long-acting PrEP)—the minimum threshold known to provide effective protection to cisgender males, who are the most common PrEP users [31]. This lower PDC threshold also captures individuals who follow “on-demand” or “event-driven” PrEP dosing, which involves taking 2 PrEP pills before a sexual encounter and 1 daily PrEP pill for the following 2 days [29].
We also assessed continuation or persistence, which refers to the consistency with which individuals remained on PrEP without a significant gap over a specified follow-up period [32]. We measured continuation by calculating the days individuals had PrEP from their index date to 180 days after initiation. Individuals were classified as having discontinued PrEP if they had a ≥ 30-day gap in sufficient PrEP coverage. For individuals who received long-acting PrEP, this represented missing 1 injectable dose following the first 2 loading doses.
Covariates
Sociodemographic characteristics, including age, sex assigned at birth, insurance plan, employment status, and geographic region, were measured on the index date. We also captured several clinical characteristics, including diagnoses of depression, anxiety, bipolar disorder, and substance use disorders (SUDs) related to the use of opioids, nicotine, cannabis, stimulants, sedatives, and other substances that have been associated with PrEP nonadherence in prior studies using ICD-9 and ICD-10 codes (eTables 4–5 in Supplement 1) [9, 33]. We also assessed diagnoses of STIs and use of STI screening and psychotherapy services (eTables 6–7 in Supplement 1).
To minimize the effect of measured confounders, we generated a matched cohort of individuals who use PrEP with and without AUD diagnoses using propensity scores. We first calculated each individual’s propensity score using multivariable logistic regression, including all sociodemographic and clinical characteristics as predictors of the probability of receiving an AUD diagnosis. Then, we matched individuals in the Pre-initiation AUD diagnosis and Without AUD diagnosis groups (the exposed and control groups) with a 1:1 ratio based on their propensity scores. We used greedy nearest neighbor matching without replacement and a 0.01 caliper width to ensure close matches. We also matched individuals in the Post-initiation AUD group with those in the Without AUD group. To evaluate the quality of the matching process, we assessed standardized mean differences (SMDs) of the covariates and used a threshold of SMD < 0.1 for covariate balance.
Statistical Analysis
We used descriptive statistics to summarize baseline characteristics separately for individuals who use PrEP with and without AUD diagnosis. Statistical differences between two groups were evaluated using Student’s t-test for continuous variables, Pearson’s chi-square test for binary variables, and Mantel–Haenszel chi-square test for nominal variables.
We also used descriptive statistics to summarize our outcomes. For adherence, we calculated the mean (standard deviation [SD]) PDC and reported the proportion of individuals that were considered adherent to PrEP based on 3 thresholds (85%, 80%, and 57%). For continuation, we calculated the mean (SD) days individuals had continuous or uninterrupted PrEP use.
To estimate the effect of AUD diagnosis on adherence and continuation, we conducted multivariable regression analyses using the propensity score-matched cohort of individuals who use PrEP. We used a logistic regression model to compare our binary adherence outcome among individuals with and without AUD diagnoses. We also used linear regression model to assess the association between AUD diagnoses and PDC and continuation among individuals who use PrEP.
RESULTS
Baseline Characteristics
Among 132,737 individuals who had claims for at least one FDA-approved PrEP, 43,913 met the full inclusion criteria. The most common type of PrEP received by individuals in the Pre-initiation AUD and No AUD group was TDF/FTC (87.71% and 86.36%, respectively), followed by TAF/FTC (12.29% and 13.64%, respectively; eFigure 1 in Supplement 1). We did not identify any individuals who used long-acting PrEP (cabotegravir).
1,245 (2.84%) individuals were included in the Pre-initiation AUD group and 37,639 (85.71%) individuals in the Without AUD group, who differed significantly in their baseline characteristics (Table 1). Compared to the Without AUD group, individuals in the Pre-initiation AUD group were younger, had a lower proportion of males assigned at birth 83.3% vs 90.3%, lower proportion of employed individuals, and a higher proportion with prior diagnoses of mental illness and SUDs. The matching process identified 1,153 matched pairs of individuals in the Pre-initiation AUD and Without AUD groups (Table 1; eFigure 2 in Supplement 1). Individuals in the Pre-initiation AUD group who were not matched (n = 92 or 7.4%) were younger, more likely to be male, and had more mental health comorbidities than individuals who were matched (eTable 9 in Supplement 1).
Table 1.
Baseline demographic and clinical characteristics of unmatched and matched individuals who use PrEP by AUD diagnosis status (2014–2021)
| Characteristics | Unmatched | Matched | |||||
|---|---|---|---|---|---|---|---|
| Pre-initiation AUD group (N = 1,245) | Without AUD group (N = 37,639) | SMDa | Pre-initiation AUD group (N = 1,153) | Without AUD group (N = 1,153) | SMDa | ||
| Age in years, mean (SD) | 34.9 (11.4) | 35.9 (10.8) | 0.087 | 35.4 (11.2) | 35.3 (11.5) | 0.005 | |
| Sex assigned at birth, n (%) | |||||||
| Male | 1,037 (83.3) | 33,990 (90.3) | 0.087 | 981 (85.1) | 975 (84.6) | 0.002 | |
| Female | 208 (16.7) | 3,649 (9.7) | 172 (14.9) | 178 (15.4) | |||
| Insurance plan, n (%) | 0.063 | 0.035 | |||||
| HMO | 176 (14.1) | 5,015 (13.3) | 142 (12.3) | 166 (14.4) | |||
| POS | 231 (18.6) | 7,539 (20.0) | 230 (19.9) | 217 (18.8) | |||
| PPO | 584 (46,9) | 16,987 (45.1) | 543 (47.1) | 535 (46.4) | |||
| Others | 254 (20.4) | 8,098 (21.5) | 238 (20.6) | 235 (20.4) | |||
| Employed, n (%) | 986 (79.2) | 31,409 (83.5) | 0.109 | 907 (78.7) | 914 (79.3) | 0.016 | |
| Regionb, n (%) | N/A | N/A | |||||
| Northeast | 348 (27.9) | 9,531 (25.3) | 332 (28.9) | 315 (27.5) | |||
| Midwest | 175 (14.0) | 4,833 (12.8) | 149 (13.0) | 160 (13.9) | |||
| South | 428 (34.4) | 13,897 (36.9) | 401 (34.9) | 404 (35.2) | |||
| West | 288 (23.1) | 9,253 (24.6) | 266 (23.2) | 268 (23.4) | |||
| Unknown | 6 | 125 | 5 | 6 | |||
| Depression, n (%) | 389 (31.2) | 1996 (5.3) | 0.713 | 333 (28.9) | 316 (27.4) | 0.004 | |
| Anxiety, n (%) | 333 (26.8) | 2,488 (6.6) | 0.561 | 293 (25.4) | 280 (24.3) | 0.031 | |
| Bipolar, n (%) | 162 (13.0) | 590 (1.6) | 0.451 | 109 (9.5) | 118 (10.2) | 0.031 | |
| PTSD, n (%) | 94 (7.6%) | 738 (2.0%) | 0.265 | 0.021 | |||
| Schizophrenia, n(%) | 10 (0.8) | 28 (0.1) | 0.110 | 5 (0.4) | 9 (0.8) | 0.053 | |
| Substance use disorders | 0.888 | 0.023 | |||||
| Opioids, n (%) | 65 (5.2) | 90 (0.2) | 48 (4.1) | 18 (1.5) | |||
| Nicotine, n (%) | 164 (13.8) | 476 (1.3) | 125 (10.9) | 177 (15.3) | |||
| Cannabis, n (%) | 121 (9.7) | 116 (0.3) | 86 (7.4) | 68 (5.9) | |||
| Stimulants, n (%) | 100 (8.0) | 103 (0.3) | 82 (7.1) | 49 (4.2) | |||
| Sedatives, n (%) | 64 (5.1) | 29 (0.1) | 53 (4.6) | 14 (1.2) | |||
| Other, n (%) | 70 (5.6) | 99 (0.3) | 51 (4.4) | 42 (3.6) | |||
| STI n (%) | 132 (10.6) | 6,700 (17.8) | 0.207 | 115 (10.0) | 125 (10.8) | 0.025 | |
| STI screening n (%) | 189 (15.2) | 3,806 (10.1) | 0.153 | 170 (14.7) | 166 (14.4) | 0.010 | |
| Psychotherapy n (%) | 173 (13.9) | 2,084 (5.5) | 0.285 | 159 (13.8) | 150 (13.0) | 0.027 | |
aAn SMD < 0.1 shows adequate covariate balance
bRegion was excluded from the propensity score model because of insufficient sample size in the unknown group
Abbreviations: AUD, alcohol use disorder; HMO, health maintenance organization; PrEP, preexposure prophylaxis; POS, point of service; PPO, preferred provider organization; PTSD, posttraumatic stress disorder; SMD, standardized mean difference; STI, sexually transmitted infection
Adherence to PrEP
Using the matched sample, we found that the Pre-initiation AUD group had lower PrEP adherence compared to the Without AUD group (mean PDC [SD]: 59.54% [34.14] vs. 65.85% [33.18]; p < 0.001; Table 2). In regression analysis, we found that individuals in the Pre-initiation AUD group had 6.31% lower mean PDC (95% CI: -9.05% to -3.57%; p < 0.001; Table 3) compared to the Without AUD group.
Table 2.
Comparison of adherence and continuation to HIV preexposure prophylaxis by AUD diagnosis status
| Primary analysis | Supplemental analysis | |||||
|---|---|---|---|---|---|---|
| Pre-initiation AUD group (N = 1,153) | Without AUD group (N = 1,153) | P-value | Post-initiation AUD group (N = 4,303) | Without AUD group (N = 4,303) | P-value | |
| Adherence | ||||||
| Mean PDC (SD)% | 59.54 (34.14) | 65.85 (33.18) | < 0.001 | 63.39 (34.21) | 66.89 (32.30) | < 0.001 |
| PDC ≥ 85%, n (%) | 411 (35.65) | 501 (43.45) | < 0.001 | 1,775 (41.25) | 1,883 (43.76) | 0.019 |
| PDC ≥ 80%, n (%) | 473 (41.02) | 577 (50.04) | < 0.001 | 2,048 (47.59) | 2,183 (50.73) | 0.004 |
| PDC ≥ 57%, n (%) | 603 (52.30) | 700 (60.71) | < 0.001 | 2492 (57.91) | 2653 (61.65) | < 0.001 |
| Continuation, mean (SD) days | 107.00 (67.20) | 119.90 (65.21) | < 0.001 | 114.90 (67.42) | 121.90 (64.07) | < 0.001 |
This analysis used the propensity score-matched sample of individuals with and without AUD diagnoses
Abbreviations: AUD, alcohol use disorder; PDC, proportion of days covered; PrEP, preexposure prophylaxis
Table 3.
Results of regression analysis for adherence and continuation of HIV preexposure prophylaxis
| Outcome | Pre-initiation AUD vs. Without AUD groups (reference) | Post-initiation AUD vs. Without AUD groups (reference) | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P-value | OR | 95% CI | P-value | |
| Adherence | ||||||
| PDC ≥ 85% | 0.72 | 0.61 – 0.85 | < 0.001 | 0.90 | 0.83 – 0.98 | 0.019 |
| PDC ≥ 80% | 0.69 | 0.59—0.82 | < 0.001 | 0.88 | 0.81—0.96 | 0.004 |
| PDC ≥ 57% | 0.71 | 0.60 – 0.84 | < 0.001 | 0.86 | 0.79 – 0.93 | < 0.001 |
| Estimate | 95% CI | P-value | Estimate | 95% CI | P-value | |
| Adherence, mean PDC (%) | -6.31 | -9.05 to -3.57 | < 0.001 | -3.51 | -4.92 to -2.10 | < 0.001 |
| Continuation, mean days | -12.93 | -18.34 to -7.52 | < 0.001 | -6.94 | -9.72 to -4.16 | < 0.001 |
This analysis used the propensity score-matched sample of individuals with and without AUD diagnoses
Abbreviations: AUD, alcohol use disorder; OR, odds ratio; PDC, proportion of days covered
A lower proportion of individuals in the Pre-initiation AUD group met the 85% (35.65% vs. 43.45%; p < 0.001), 80% (41.02% vs. 50.04%; p < 0.001), and 57% (52.30% vs. 60.71%; p < 0.001) PDC thresholds for PrEP adherence (Table 2). Regression analyses found that the Pre-initiation AUD group had lower odds of meeting the 85% (OR 0.72; 95% CI: 0.61–0.85; p < 0.001), 80% (OR 0.69; 95% CI: 0.59–0.82; p < 0.001), or 57% (OR 0.71; 95% CI: 0.60–0.84; p < 0.001) PDC thresholds compared to the Without AUD group (Table 3).
Continuation of PrEP
The Pre-initiation AUD group had fewer days of continuous PrEP medication compared to the Without AUD group (107 [67.20] days vs. 119.90 [65.21] days; p < 0.001; Table 2). Regression analysis estimated the Pre-initiation AUD group had 12.93 fewer days of continuous PrEP medication (95% CI: -18.34 to -7.52 days; p < 0.001; Table 3) compared to the Without AUD group.
Supplemental Analysis
5,029 out of 43,913 (11.45%) individuals who use PrEP received an AUD diagnosis within 6 months after initiation and were included in the Post-initiation AUD group (eFigure 1 in Supplement 1). Individuals in the Post-initiation AUD group were younger, had a lower proportion of males assigned at birth, a lower proportion of employed individuals, and a higher proportion with prior diagnoses of mental illness and SUDs compared to the Without AUD group (eTable 10 in Supplement 1).
The matching process identified 4,303 matched pairs of individuals from the Post-initiation AUD and Without AUD groups (eFigure 3 and eTable 10 in Supplement 1). A comparison of PDCs showed that the Post-initiation AUD group had lower PrEP adherence (mean PDC [SD]: 63.39% [34.21] vs. 66.89% [32.30]; p < 0.001; Table 2). The proportion of individuals who met the thresholds for adherence was also higher in the Without AUD group compared to the Post-initiation AUD group (Table 2). In regression analysis, we found that individuals in the Post-initiation AUD group had 3.51% lower mean PDC (95% CI: -4.92% to -2.10%; p < 0.001; Table 3) compared to the Without AUD group. The Post-initiation AUD group also had fewer days with continuous PrEP claims compared to the Without AUD group (mean [SD] 114.90 [67.42] days vs. 121.90 [64.07] days; p < 0.001; Table 2).
Regression analysis found that the Post-initiation AUD group had lower odds of meeting the 85% (OR 0.90; 95% CI: 0.83–0.98, p = 0.019), 80% (OR 0.88; 95% CI: 0.81–0.96, p = 0.004), and 57% PDC thresholds (OR 0.86; 95% CI: 0.79–0.93, p < 0.001) and had 6.94 (95% CI: -9.72 to -4.16 days, p < 0.001) fewer days of continuous PrEP use compared to the Without AUD group (Table 3).
DISCUSSION
This retrospective cohort study used a matched sample of individuals who initiated PrEP and found suboptimal PrEP use among those who were previously diagnosed with AUD. In our entire sample, the mean PDC was less than 70% across the AUD and non-AUD groups. However, individuals who were diagnosed with AUD before initiation had 7% lower mean PDC than their counterparts without AUD. Individuals who were diagnosed with AUD also had fewer days of continuous or “gap-free” PrEP supply. We found similar associations between AUD and PrEP adherence and continuation among those who were diagnosed with AUD after PrEP initiation.
Alcohol use has been repeatedly found to negatively affect medication adherence [34–36], likely by impairing prospective memory (i.e., “forgetfulness”) and increasing avoidance behaviors, particularly in individuals who believe alcohol interacts with their medications [37]. In HIV treatment, alcohol and substance use disorders have been found to reduce adherence to antiretroviral therapy (ART) [38–40].
Mirroring findings in ART studies, alcohol use is known to affect PrEP use. For example, in a Kaiser Permanente study (N = 2,995), individuals with AUD who were eligible for PrEP were 23% more likely to be linked with a PrEP provider within a year of referral but were 15% less likely to be adherent to daily PrEP use, defined as a medication possession ratio of ≥ 80% [41]. Despite similarities in study aims and approaches, our study improves on this previous analysis in several ways. First, we defined PrEP adherence using PDC, a measure that is less prone to overestimation than medication possession ratio [42]. Second, we employed propensity-score matching of individuals with and without AUD diagnoses and used data through 2021, which allowed us to include individuals who initiated TDF/FTC (approved in 2013) and TAF/FTC (approved in 2019). Finally, our use of real-world data allowed us to include a diverse cohort of individuals from all US regions who receive care from different healthcare settings.
Prior research, however, has also found null or positive associations between unhealthy alcohol use and PrEP outcomes [43, 44]. A study with a national sample of Black and Latino MSM found that higher AUDIT scores were associated with 6% lower odds of PrEP awareness but 6% higher odds of self-reported PrEP adherence, which can overestimate actual adherence [16]. In South Africa, a study with periconceptional women found that laboratory-confirmed adherence was more likely among those who reported any past-year drinking [13, 45]. These results suggest that individuals who use alcohol and may be at higher risk for HIV [46], making them ideal candidates for PrEP, can take the medication effectively and achieve optimal protection. If these findings hold, additional research should aim to understand the conditions or factors that are associated with PrEP adherence among people who use alcohol and other substances.
We also found that individuals with AUD experienced ≥ 30-day gaps in their PrEP supply about 12 days sooner than those without an AUD diagnosis. Like the literature on PrEP adherence, prior studies have yielded mixed results on the association between PrEP (dis)continuation and alcohol use [47, 48]. While gaps in PrEP supply can lower its effectiveness, other factors not related to PrEP use may impact a person’s HIV risk. For example, individuals may discontinue PrEP because of changes in their risk profiles, such as starting a monogamous relationship or adopting other prevention measures (e.g., condoms) [49, 50]. We used a 6-month follow-up to focus on the initial period of PrEP use, where significant changes in HIV risk are less likely to occur, and to identify early drop-offs that require re-engagement in PrEP care [29].
Contradicting findings on the association between alcohol use and PrEP may be influenced by several factors, such as the population (e.g., MSM vs. women), geography or location, length of follow-up, and presence of co-occurring conditions. How unhealthy alcohol use is measured or defined may also be a contributor. Studies that analyzed primary data often relied on self-reported alcohol use or responses to AUDIT, a validated screening tool, to identify individuals who have unhealthy alcohol use or AUD. While these studies may have been more sensitive to identifying individuals with AUD, self-reported alcohol use is still subject to social desirability bias [51]. By contrast, claims-based analyses like our study used AUD diagnosis codes to identify individuals with AUD [22–25]. While this approach may underestimate the burden of AUD due to coding errors and underdiagnosis, our study highlights needed improvements in the continuum of care among individuals at the intersection of HIV prevention and addiction, even if they may have fewer healthcare barriers (e.g., continuous insurance coverage).
While efforts are needed to improve PrEP use generally [52–54], our study underscores the importance of targeted efforts to address the challenges faced by people with AUD. Unfortunately, interventions developed to address unhealthy alcohol use and promote PrEP adherence and continuation have not been effective [55–58]. Most of these interventions utilized motivational interviewing approaches, web-based applications, or brief behavioral interventions. For example, Starks et al. (2024) and Karnik et al. (2023) both evaluated MI-based and brief interventions but found no statistically significant differences in alcohol use or PrEP-related outcomes between intervention and control groups. Similarly, Wray et al. (2024) tested a digital personalized feedback web app and found no improvements in PrEP adherence or alcohol outcomes compared to standard education. Common issues reported in these studies include high dropout rates, particularly in trials conducted during COVID-19, low participant engagement, and challenges in delivering interventions as intended.
An unexplored but promising avenue involves integrating alcohol AUD pharmacotherapy with PrEP adherence interventions. To date, no studies have evaluated the combined delivery of AUD pharmacotherapies (i.e., naltrexone, acamprosate, and disulfiram) alongside routine PrEP care. Newer long-acting PrEP regimens, like twice-yearly lenacapavir injections [59], may also reduce the burden on individuals to take daily pills and offer a promising alternative for individuals with AUD who face adherence issues. However, long-acting PrEP may not be a panacea. In HIV therapy, people with HIV still face adherence challenges when using long-acting ART. Studies have found that while some patients with substance use and housing instability viewed long-acting ART as a way to improve adherence, other challenges, such as missed appointments, transportation difficulties, and lifestyle mismatches associated with the resource-intensive nature of long-acting ART use, can easily undermine its effectiveness [60, 61]. Thus, long-acting formulations may not eliminate all barriers to medication adherence. Future research should examine how long-acting formulations mitigate PrEP adherence challenges among individuals with AUD.
Limitations
Our study has several limitations. We relied on diagnosis codes to identify patients with AUD, which is subject to misclassification and underreporting [26]. Best practices for AUD diagnosis involve structured assessments and the use of validated tools like AUDIT; however, claims data such as MarketScan do not include details on the screening or diagnosis procedures followed. Some studies have sought to address this limitation by including > 1 AUD diagnoses (particularly in outpatient settings) to classify an individual as having AUD [62], but most studies have relied on the presence of 1 diagnosis of AUD in claims data [22–25]. Similarly, we measured adherence and continuation indirectly using pharmacy claims, and we could not ascertain whether patients were taking their PrEP medications. In previous research, adherence was assessed through other means, including self-reported behavior [63, 64] or pharmacological exposure measures like TFV concentrations in blood plasma, dried blood spots, and hair samples [13, 16, 65, 66]. Adherence monitoring using TFV diphosphate is often preferred because its blood concentration detects cumulative adherence over several weeks [13, 67] and is less prone to “white-coat compliance” and other forms of observer bias [65, 68]. However, while biomarker-based monitoring can confirm or validate self-reported PrEP adherence and continuation, these techniques can be costly and have little utility in real-world settings. By contrast, measuring adherence in PDC using pharmacy claims offers a low-cost alternative with high sensitivity, but low specificity, when used to assess adherence over a limited period [29].
Claims data exclude diagnoses or healthcare services received without insurance or outside traditional healthcare settings. This may be especially relevant for adolescents whose healthcare coverage is through their primary caregivers and may choose to receive PrEP and other related services (e.g., STI screening) through community access programs to avoid disclosure [69, 70]. Related to this, our algorithm for identifying individuals who take PrEP, while highly sensitive and specific [19] and routinely used for PrEP surveillance in the US [71], may have missed or misclassified individuals. MarketScan only includes individuals with employer-sponsored commercial insurance and leaves out other populations, limiting our study’s generalizability. MarketScan also lacks information on key variables such as gender, sexual orientation, race, and ethnicity, which are important determinants of both AUD and PrEP use. We also did not account for other barriers to PrEP adherence and continuation, such as medical and non-medical financial constraints, medication concerns, and other care-access barriers, which may explain the associations we found [72–74]. The absence of these variables may have introduced bias in our matching approach, which only controlled for measured confounders in estimating the association—and not the causal link—between AUD and PrEP adherence and continuation. Our matching procedure also failed to match 7.4% of our exposed individuals whose characteristics differed from those who were matched with an individual in the control group. While our strict matching criteria ensured covariate balance in our final matched sample, this approach may have contributed to lower precision (i.e., higher variance) and selection bias that future studies can address using other matching algorithms (e.g., many-to-one matching) [75, 76].
CONCLUSION
This retrospective cohort study found that individuals who use PrEP and were diagnosed with AUD before or after PrEP initiation had lower adherence and continuation compared to those without AUD diagnoses. These findings conflict with previous research, suggesting that AUD may affect PrEP adherence and continuation in specific contexts and populations that future research should explore. Interventions, including longer-acting PrEP formulations, may be needed to help individuals with AUD to take PrEP as recommended.
Supplementary Information
Below is the link to the electronic supplementary material.
Author Contributions
Dr. Avanceña had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Avanceña, Norwood, Schnarrs, Barner.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Avanceña, Okoye, Yokananth.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Okoye, Barner, Avanceña.
Obtained funding: Avanceña, Norwood, Schnarrs.
Administrative, technical, or material support: All authors.
Supervision: Avanceña, Norwood, Schnarrs, Barner.
Funding
This study was supported by a pilot grant from the Institute for Community Health Research (previously the Texas Institute for Sexual and Gender Minority Health Research; awarded to AA, AN, and PS) and the Dell Medical School’s Office of Research and the Office of Health Equity through the Health Equity Pilot Project Award (awarded to AA). AA was supported by a Translational Science Mentored Research Career Development (K12) Award funded by the National Center for Advancing Translational Sciences of the National Institutes of Health (K12TR004529).
Data Availability
MarketScan is a licensed database of Merative and therefore not publicly available. Statistical code is available upon request to the corresponding author (AA). All inquiries should be directed to the corresponding author.
Declarations
Human Ethics and Consent to Participate Declarations
We used de-identified healthcare claims and administrative data in this study. This study was exempt from full ethics review by The University of Texas Institutional Review Board (STUDY00006393).
Role of the Funder/Sponsor
The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funders.
Conflict of Interest Disclosures
Drs. Schnarrs and Norwood have received funding from Merck Sharp & Dohme Corp. Investigator Studies Program to study preferences for longer-acting preexposure prophylaxis among gay, bisexual, and other men who have sex with men (MISP 10080) and transgender and gender-expansive populations (MISP 60761).
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Guilamo-Ramos V, Thimm-Kaiser M, Benzekri A. Is the USA on track to end the HIV epidemic? Lancet HIV. 2023;10(8):e552-e556. 10.1016/S2352-3018(23)00142-X [DOI] [PubMed] [Google Scholar]
- 2.Centers for Disease Control and Prevention. HIV diagnoses, deaths, and prevalence: 2025 update. HIV Data. April 29, 2025. Accessed October 17, 2025. https://www.cdc.gov/hiv-data/nhss/hiv-diagnoses-deaths-and-prevalence-2025.html
- 3.Centers for Disease Control and Prevention. Diagnoses, deaths, and prevalence of HIV in the United States and 6 territories and freely associated states, 2022. HIV Surveill Rep. 2024;35:1-177. [Google Scholar]
- 4.US Public Health Service. Preexposure prophylaxis for the prevention of HIV infection in the United States – 2021 update clinical practice guideline. Published online 2021. Accessed February 4, 2023. https://www.cdc.gov/hiv/pdf/risk/prep/cdc-hiv-prep-guidelines-2021.pdf
- 5.US Preventive Services Task Force, Barry MJ, Nicholson WK, et al. Preexposure prophylaxis to prevent acquisition of HIV: US Preventive Services Task Force recommendation statement. JAMA. 2023;330(8):736. 10.1001/jama.2023.14461 [DOI] [PubMed]
- 6.Chou R, Spencer H, Bougatsos C, Blazina I, Ahmed A, Selph S. Preexposure prophylaxis for the prevention of HIV: updated evidence report and systematic review for the US Preventive Services Task Force. JAMA. 2023;330(8):746. 10.1001/jama.2023.9865 [DOI] [PubMed] [Google Scholar]
- 7.Centers for Disease Control and Prevention. Effectiveness of prevention strategies to reduce the risk of acquiring or transmitting HIV. June 17, 2022. Accessed February 4, 2024. https://www.cdc.gov/hiv/risk/estimates/preventionstrategies.html
- 8.Nunn AS, Brinkley-Rubinstein L, Oldenburg CE, et al. Defining the HIV pre-exposure prophylaxis care continuum. AIDS Lond Engl. 2017;31(5):731-734. 10.1097/QAD.0000000000001385 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shuper PA, Joharchi N, Bogoch II, et al. Alcohol consumption, substance use, and depression in relation to HIV Pre-Exposure Prophylaxis (PrEP) nonadherence among gay, bisexual, and other men-who-have-sex-with-men. BMC Public Health. 2020;20(1):1782. 10.1186/s12889-020-09883-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Saitz R. Unhealthy Alcohol Use. N Engl J Med. 2005;352(6):596-607. 10.1056/NEJMcp042262 [DOI] [PubMed] [Google Scholar]
- 11.Wahome E, Boyd A, Thiong’o AN, et al. Stopping and restarting PrEP and loss to follow-up among PrEP-taking men who have sex with men and transgender women at risk of HIV-1 participating in a prospective cohort study in Kenya. HIV Med. 2022;23(7):750–763. 10.1111/hiv.13237 [DOI] [PMC free article] [PubMed]
- 12.Hoenigl M, Jain S, Moore D, et al. Substance use and adherence to HIV preexposure prophylaxis for men who have sex with men. Emerg Infect Dis. 2018;24(12):2292-2302. 10.3201/eid2412.180400 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Miller AP, Shoptaw S, Moucheraud C, et al. Recent alcohol use is associated with increased pre-exposure prophylaxis (PrEP) continuation and adherence among pregnant and postpartum women in South Africa. JAIDS J Acquir Immune Defic Syndr. 2023;92(3):204. 10.1097/QAI.0000000000003133 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Strong SH, Oldfield BJ, van den Berg JJ, et al. Perspectives on unhealthy alcohol use among men who have sex with men prescribed HIV pre-exposure prophylaxis: a qualitative study. Prev Med Rep. 2023;37:102553. 10.1016/j.pmedr.2023.102553 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hojilla JC, Hurley LB, Marcus JL, et al. Characterization of HIV preexposure prophylaxis use behaviors and HIV incidence among US adults in an integrated health care system. JAMA Netw Open. 2021;4(8):e2122692. 10.1001/jamanetworkopen.2021.22692 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Watson RJ, Morgan E, Collibee C, et al. Substance use and healthcare utilization across the pre-exposure prophylaxis (PrEP) care cascade among Black and Latino men who have sex with men. Subst Use Misuse. 2022;57(11):1698-1707. 10.1080/10826084.2022.2108059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mintz CM, Hartz SM, Fisher SL, et al. A cascade of care for alcohol use disorder: Using 2015–2019 National Survey on Drug Use and Health data to identify gaps in past 12-month care. Alcohol Clin Exp Res. 2021;45(6):1276-1286. 10.1111/acer.14609 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10):e296. 10.1371/journal.pmed.0040296 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Furukawa NW, Smith DK, Gonzalez CJ, et al. Evaluation of algorithms used for PrEP surveillance using a reference population from New York City, July 2016–June 2018. Public Health Rep. 2020;135(2):202-210. 10.1177/0033354920904085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shortreed SM, Johnson E, Rutter CM, Kamineni A, Wernli KJ, Chubak J. Cohort restriction based on prior enrollment: Examining potential biases in estimating cancer and mortality risk. Obs Stud. 2016;2(1):51-64. 10.1353/obs.2016.0002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Jensen ET, Cook SF, Allen JK, et al. Enrollment factors and bias of disease prevalence estimates in administrative claims data. Ann Epidemiol. 2015;25(7):519-525.e2. 10.1016/j.annepidem.2015.03.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Huskamp HA, Reif S, Greenfield SF, Normand SLT, Busch AB. Medication utilization for alcohol use disorder in a commercially insured population. J Gen Intern Med. 2020;35(11):3262-3270. 10.1007/s11606-020-06073-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Mintz CM, Presnall NJ, Xu KY, et al. An examination between treatment type and treatment retention in persons with opioid and co-occurring alcohol use disorders. Drug Alcohol Depend. 2021;226:108886. 10.1016/j.drugalcdep.2021.108886 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Rittenberg A, Hines AL, Alvanzo AAH, Chander G. Correlates of alcohol use disorder pharmacotherapy receipt in medically insured patients. Drug Alcohol Depend. 2020;214:108174. 10.1016/j.drugalcdep.2020.108174 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Le P, Rich JJ, Bernstein EY, et al. Disparities in Treatment for Alcohol Use Disorder Among All of Us Participants. Am J Psychiatry. 2024;181(11):973-987. 10.1176/appi.ajp.20230730 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mercurio L, Garcia A, Ruest S, Duffy SJ, Eickhoff C. One third of alcohol use disorder diagnoses are missed by ICD coding. Subst Use Addict J. Published online November 7, 2024:29767342241288112. 10.1177/29767342241288112 [DOI] [PMC free article] [PubMed]
- 27.Williams EC, Fletcher OV, Frost MC, Harris AHS, Washington DL, Hoggatt KJ. Comparison of substance use disorder diagnosis rates from electronic health record data with substance use disorder prevalence rates reported in surveys across sociodemographic groups in the Veterans Health Administration. JAMA Netw Open. 2022;5(6):e2219651. 10.1001/jamanetworkopen.2022.19651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Avanceña ALV, Okoye G, Yokananth R, Norwood A, Schnarrs PW. Alcohol Use Disorder Diagnoses Among Individuals Who Take HIV Preexposure Prophylaxis. JAMA Netw Open. 2025;8(4):e257295. 10.1001/jamanetworkopen.2025.7295 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Pyra M, Rusie L, Castro M, et al. A taxonomy of pragmatic measures of HIV preexposure prophylaxis use. AIDS. 2020;34(13):1951. 10.1097/QAD.0000000000002618 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Baumgartner PC, Haynes RB, Hersberger KE, Arnet I. A systematic review of medication adherence thresholds dependent of clinical outcomes. Front Pharmacol. 2018;9:1290. 10.3389/fphar.2018.01290 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Anderson PL, Glidden DV, Liu A, et al. Emtricitabine-tenofovir concentrations and pre-exposure prophylaxis efficacy in men who have sex with men. Sci Transl Med. 2012;4(151):151ra125–151ra125. 10.1126/scitranslmed.3004006 [DOI] [PMC free article] [PubMed]
- 32.Barner JC, Khoza S, Oladapo A. ADHD medication use, adherence, persistence and cost among Texas Medicaid children. Curr Med Res Opin. 2011;27(sup2):13-22. 10.1185/03007995.2011.603303 [DOI] [PubMed] [Google Scholar]
- 33.Shuper PA, Varatharajan T, Kinitz DJ, et al. Perceived influence of alcohol consumption, substance use, and mental health on PrEP adherence and condom use among PrEP-prescribed gay, bisexual, and other men-who-have-sex-with-men: a qualitative investigation. BMC Public Health. 2022;22:1875. 10.1186/s12889-022-14279-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Braithwaite RS, McGinnis KA, Conigliaro J, et al. A temporal and dose-response association between alcohol consumption and medication adherence among veterans in care. Alcohol Clin Exp Res. 2005;29(7):1190-1197. 10.1097/01.ALC.0000171937.87731.28 [DOI] [PubMed] [Google Scholar]
- 35.Bryson CL, Au DH, Sun H, Williams EC, Kivlahan DR, Bradley KA. Alcohol screening scores and medication nonadherence. Ann Intern Med. 2008;149(11):795-803. 10.7326/0003-4819-149-11-200812020-00004 [DOI] [PubMed] [Google Scholar]
- 36.Grodensky CA, Golin CE, Ochtera RD, Turner BJ. Systematic review: effect of alcohol intake on adherence to outpatient medication regimens for chronic diseases. J Stud Alcohol Drugs. 2012;73(6):899-910. 10.15288/jsad.2012.73.899 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Woolf-King SE, Sheinfil AZ, Ramos J, et al. A conceptual model of alcohol use and adherence to antiretroviral therapy: systematic review and theoretical implications for mechanisms of action. Health Psychol Rev. 2022;16(1):104-133. 10.1080/17437199.2020.1806722 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Zhang Y, Wilson TE, Adedimeji A, et al. The Impact of Substance Use on Adherence to Antiretroviral Therapy Among HIV-Infected Women in the United States. AIDS Behav. 2018;22(3):896-908. 10.1007/s10461-017-1808-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Przybyla S, Ashare RL, Cioffi L, et al. Substance Use and Adherence to Antiretroviral Therapy among People Living with HIV in the United States. Trop Med Infect Dis. 2022;7(11):349. 10.3390/tropicalmed7110349 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Socias ME, Milloy MJ. Substance Use and Adherence to Antiretroviral Therapy: What Is Known and What Is Unknown. Curr Infect Dis Rep. 2018;20(9):36. 10.1007/s11908-018-0636-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Bruxvoort KJ, Schumacher CM, Towner W, et al. Referral linkage to preexposure prophylaxis are and persistence on preexposure prophylaxis in an integrated health care system. J Acquir Immune Defic Syndr 1999. 2021;87(3):918–927. 10.1097/QAI.0000000000002668 [DOI] [PubMed]
- 42.Pfaar O, Devillier P, Schmitt J, et al. Adherence and persistence in allergen immunotherapy (APAIT): A reporting checklist for retrospective studies. Allergy. 2023;78(8):2277-2289. 10.1111/all.15723 [DOI] [PubMed] [Google Scholar]
- 43.Okafor CN, Hucks-Ortiz C, Hightow-Weidman LB, et al. Associations between self-reported substance use behaviors and PrEP acceptance and adherence among Black MSM in the HPTN 073 study. J Acquir Immune Defic Syndr 1999. 2020;85(1):23–29. 10.1097/QAI.0000000000002407 [DOI] [PMC free article] [PubMed]
- 44.Triplett NS, Roberts ST, Hartmann M, et al. Examining Associations Between Mental Health, IPV Exposure, HIV Risk Behaviors, and PrEP Use in South African Women: An Analysis of Data from the Charisma Study. AIDS Behav. 2023;27(6):2030-2040. 10.1007/s10461-022-03936-7 [DOI] [PubMed] [Google Scholar]
- 45.Matthews LT, Jaggernath M, Kriel Y, et al. Oral preexposure prophylaxis uptake, adherence, and persistence during periconception periods among women in South Africa. AIDS. 2024;38(9):1342-1354. 10.1097/QAD.0000000000003925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Baliunas D, Rehm J, Irving H, Shuper P. Alcohol consumption and risk of incident human immunodeficiency virus infection: a meta-analysis. Int J Public Health. 2010;55(3):159-166. 10.1007/s00038-009-0095-x [DOI] [PubMed] [Google Scholar]
- 47.Lichtwarck HO, Mbotwa CH, Kazaura MR, Moen K, Mmbaga EJ. Early disengagement from HIV pre-exposure prophylaxis services and associated factors among female sex workers in Dar es Salaam, Tanzania: a socioecological approach. BMJ Glob Health. 2023;8(12):e013662. 10.1136/bmjgh-2023-013662 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Khadka N, Gorbach PM, Nyemba DC, et al. Evaluating the use of oral pre-exposure prophylaxis among pregnant and postpartum adolescent girls and young women in Cape Town, South Africa. Front Reprod Health. 2023;5:1224474. 10.3389/frph.2023.1224474 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Vanbaelen T, Rotsaert A, Jacobs BKM, et al. Why do HIV pre-exposure prophylaxis users discontinue pre-exposure prophylaxis care? A mixed methods survey in a pre-exposure prophylaxis clinic in Belgium. AIDS Patient Care STDs. 2022;36(4):159-167. 10.1089/apc.2021.0197 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Garofoli N, Siguier M, Robineau O, et al. Incidence and factors associated with PrEP discontinuation in France. J Antimicrob Chemother. 2024;79(7):1555-1563. 10.1093/jac/dkae133 [DOI] [PubMed] [Google Scholar]
- 51.Schell C, Godinho A, Cunningham JA. To thine own self, be true: examining change in self-reported alcohol measures over time as related to socially desirable responding bias among people with unhealthy alcohol use. Subst Abuse. 2021;42(1):87-93. 10.1080/08897077.2019.1697998 [DOI] [PubMed] [Google Scholar]
- 52.Haines M, Vandyk A, Skidmore B, Orser L, O’Byrne P. A systematic review of oral pre-exposure prophylaxis HIV adherence interventions. J Assoc Nurses AIDS Care. 2024;35(4):309-324. 10.1097/JNC.0000000000000460 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Garrison LE, Haberer JE. Pre-exposure prophylaxis uptake, adherence, and persistence: a narrative review of interventions in the U.S. Am J Prev Med. 2021;61(5):S73-S86. 10.1016/j.amepre.2021.04.036 [DOI] [PubMed]
- 54.Kamitani E, Peng Y, Hopkins D, Higa DH, Mullins MM. A Community Guide systematic review: digital HIV pre-exposure prophylaxis interventions. Am J Prev Med. 2024;67(2):303-310. 10.1016/j.amepre.2024.02.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Starks TJ, Samrock S, Lopez D, Bradford-Rogers J, Marmo J, Cain D. Testing the effectiveness of a motivational interviewing intervention to reduce HIV risk and drug use in young sexual minority men in a community-based organization setting. AIDS Behav. 2024;28(1):26-42. 10.1007/s10461-023-04191-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Karnik NS, Kuhns LM, Hotton AL, et al. Findings from the Step Up, Test Up study of an electronic screening and brief intervention for alcohol misuse in adolescents and young adults presenting for HIV testing: randomized controlled efficacy trial. JMIR Ment Health. 2023;10:e43653. 10.2196/43653 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Wray TB, Chan PA, Kahler CW, Ocean EMS, Nittas V. Pilot randomized controlled trial of Game Plan for PrEP: a brief, web and text message intervention to help sexual minority men adhere to PrEP and reduce their alcohol use. AIDS Behav. 2024;28(4):1356-1369. 10.1007/s10461-023-04223-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Ichite A, Jean-Gilles M, Rosenberg R, Abbamonte J, Dévieux JG. Assessing the preliminary efficacy of a nonrandomized, clinic-based HIV risk reduction pilot intervention for PrEP-initiated, alcohol- and other drug-using women of color in Miami, FL. J Racial Ethn Health Disparities. 2023;10(6):3077-3094. 10.1007/s40615-022-01482-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Bekker LG, Das M, Abdool Karim Q, et al. Twice-Yearly Lenacapavir or Daily F/TAF for HIV Prevention in Cisgender Women. N Engl J Med. 2024;391(13):1179-1192. 10.1056/NEJMoa2407001 [DOI] [PubMed] [Google Scholar]
- 60.Fletcher L, Burrowes SAB, Khan GK, et al. Perspectives on long-acting injectable HIV antiretroviral therapy at an alternative care site: a qualitative study of people with HIV experiencing substance use and/or housing instability. Harm Reduct J. 2023;20(1):4. 10.1186/s12954-023-00730-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Fletcher L, Burrowes S, Sabin LL, et al. Long-Acting Injectable ART in Practice: A Mixed Methods Implementation Study Assessing the Feasibility of Using LAI ART in High Risk Populations and At Alternative Low Barrier Care Sites. AIDS Patient Care STDs. 2024;38(5):221-229. 10.1089/apc.2024.0048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Busch AB, Mehrotra A, Greenfield SF, Uscher-Pines L, Rose S, Huskamp HA. A cohort study examining changes in treatment patterns for alcohol use disorder among commercially insured adults in the United States during the COVID-19 pandemic. J Subst Abuse Treat. 2023;144:108920. 10.1016/j.jsat.2022.108920 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.De França FÁP, Silva PG de B, Fonteles MM de F. The Impact of Alcohol and Illicit Drug use on Adherence to HIV Pre-exposure Prophylaxis among Men who have Sex with Men. J Young Pharm. 2021;13(3):251–256.
- 64.Kawuma R, Ssemata AS, Bernays S, Seeley J. Women at high risk of HIV-infection in Kampala, Uganda, and their candidacy for PrEP. SSM - Popul Health. 2021;13:100746. 10.1016/j.ssmph.2021.100746 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Gandhi M, Glidden DV, Liu A, et al. Strong correlation between concentrations of tenofovir (TFV) emtricitabine (FTC) in hair and TFV diphosphate and FTC triphosphate in dried blood spots in the iPrEx open label extension: implications for pre-exposure prophylaxis adherence monitoring. J Infect Dis. 2015;212(9):1402-1406. 10.1093/infdis/jiv239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Jalil EM, Torres TS, Luz PM, et al. Low PrEP adherence despite high retention among transgender women in Brazil: the PrEParadas study. J Int AIDS Soc. 2022;25(3):e25896. 10.1002/jia2.25896 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Velloza J, Donnell D, Hosek S, et al. Alignment of PrEP adherence with periods of HIV risk among adolescent girls and young women in South Africa and Zimbabwe: a secondary analysis of the HPTN 082 randomised controlled trial. Lancet HIV. 2022;9(10):e680-e689. 10.1016/S2352-3018(22)00195-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Podsadecki TJ, Vrijens BC, Tousset EP, Rode RA, Hanna GJ. “White coat compliance” limits the reliability of therapeutic drug monitoring in HIV-1—infected patients. HIV Clin Trials. 2008;9(4):238-246. 10.1310/hct0904-238 [DOI] [PubMed] [Google Scholar]
- 69.Carnevale C, Zucker J, Womack JA, et al. Adolescent Preexposure Prophylaxis Administration: An Education Curriculum for Health Care Providers. J Pediatr Health Care. 2019;33(3):288-295. 10.1016/j.pedhc.2018.09.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Kay ES, Pinto RM. Is Insurance a Barrier to HIV Preexposure Prophylaxis? Clarifying the Issue. Am J Public Health. 2020;110(1):61-64. 10.2105/AJPH.2019.305389 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Sullivan PS, DuBose SN, Castel AD, et al. Equity of PrEP uptake by race, ethnicity, sex and region in the United States in the first decade of PrEP: a population-based analysis. Lancet Reg Health - Am. 2024;33:100738. 10.1016/j.lana.2024.100738 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 72.Anyasi H, Idemudia A, Badru T, et al. Discontinuation of HIV oral pre-exposure prophylaxis: findings from programmatic surveillance within two general population HIV programs in Nigeria. BMC Public Health. 2024;24(1):1325. 10.1186/s12889-024-18808-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 73.Guo Y, Westmoreland DA, D’Angelo A, et al. PrEP discontinuation in a US national cohort of sexual and gender minority populations, 2017–22. Health Aff (Millwood). 2024;43(3):443-451. 10.1377/hlthaff.2023.00867 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Unger ZD, Golub SA, Borges C, Edelstein ZR, Hedberg T, Myers J. Reasons for PrEP discontinuation after navigation at sexual health clinics: interactions among systemic barriers, behavioral relevance, and medication concerns. JAIDS J Acquir Immune Defic Syndr. 2022;90(3):316-324. 10.1097/QAI.0000000000002952 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 75.Golinelli D, Ridgeway G, Rhoades H, Tucker J, Wenzel S. Bias and variance trade-offs when combining propensity score weighting and regression: with an application to HIV status and homeless men. Health Serv Outcomes Res Methodol. 2012;12(2-3):104-118. 10.1007/s10742-012-0090-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76.Austin PC. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat. 2011;10(2):150-161. 10.1002/pst.433 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
MarketScan is a licensed database of Merative and therefore not publicly available. Statistical code is available upon request to the corresponding author (AA). All inquiries should be directed to the corresponding author.