In the United States, the publically supported national HIV medical care system is designed to provide HIV medical care to those who would otherwise not receive such care. Nevertheless, many HIV-infected persons are not receiving medical care. Limited information is available from HIV-infected persons not currently in care about the reasons they are not receiving care. From November 2006 to February 2007, we conducted five focus groups at community-based organizations and health departments in five U.S. cities to elicit qualitative information about barriers to entering HIV care. The 37 participants were mostly male (n = 29), over the age of 30 (n = 34), and all but one had not received HIV medical care in the previous 6 months. The focus group discussions revealed health belief-related barriers that have often been overlooked by studies of access to care. Three key themes emerged: avoidance and disbelief of HIV serostatus, conceptions of illness and appropriate health care, and negative experiences with, and distrust of, health care. Our findings point to the potentially important influence of these health-related beliefs on individual decisions about whether to access HIV medical care. We also discuss the implications of these beliefs for provider-patient communication, and suggest that providers frame their communications with patients such that they are attentive to the issues identified by our respondents, to better engage patients as partners in the treatment process.
Relocation from one’s birthplace may affect human immunodeficiency virus (HIV) outcomes, but national estimates of HIV outcomes among Hispanics/Latinos by place of birth are limited. We analyzed Medical Monitoring Project data collected in 2015–2018 from 2564 HIV-positive Hispanic/Latino adults and compared clinical outcomes between mainland US-born (referent group), Puerto Rican (PR-born), and those born outside the United States (non-US-born). We reported weighted percentages of characteristics and used logistic regression with predicted marginal means to examine differences between groups (p < 0.05). PR-born Hispanics/Latinos were more likely to be prescribed antiretroviral therapy (ART) (94%) and retained in care (94%) than mainland-US-born (79% and 77%, respectively) and non-US-born (91% and 87%, respectively) Hispanics/Latinos. PR-born Hispanics/Latinos were more likely to have sustained viral suppression (75%) than mainland-US-born Hispanics/Latinos (57%). Non-US-born Hispanics/Latinos were more likely to be prescribed ART (91% vs. 79%), retained in care (87% vs. 77%), and have sustained viral suppression (74% vs. 57%) than mainland-US-born Hispanics/Latinos. Greater Ryan White HIV/AIDS-funded facility usage among PR-born, better mental health among non-US-born, and less drug use among PR-born and non-US-born Hispanics/Latinos may have contributed to better HIV outcomes. Expanding programs with comprehensive HIV/AIDS services, including for mental health and substance use, may reduce HIV outcome disparities among Hispanics/Latinos.
We examined factors associated with antiretroviral therapy (ART) adherence among transgender women living with HIV (TWLWH).We used combined data from the 2009 to 2013 cycles of Medical Monitoring Project, an HIV surveillance system designed to produce nationally representative estimates of the characteristics of HIV-infected adults receiving HIV medical care in the United States. Rao-Scott chi-square tests and multivariable logistic regression were used to identify factors associated with dose adherence (defined as taking 100% of prescribed ART doses in the past 3 days).Among TWLWH who reported current ART use, an estimated 80.5% self-reported dose adherence. Multivariable factors independently associated with lower (<100%) dose adherence were younger age (30-39 vs. 40 and over), not having health insurance coverage, depression, lower self-efficacy to take medication as prescribed, and having greater than one daily ART dose.Our findings suggest several ways to potentially improve ART adherence of TWLWH including tailoring efforts to address the needs of TWLWH under age 40, increasing access to health insurance coverage, addressing mental health morbidities, building skills to improve medication adherence self-efficacy, and simplifying ART regimens when possible.
People with HIV at highest risk of anal cancer include gay, bisexual, and other men who have sex with men and transgender women aged 35 years or older as well as other people with HIV aged 45 years or older. Identifying and treating precancerous lesions can reduce anal cancer incidence in these groups. We assessed the prevalence of anal cytology and access to high-resolution anoscopy among people with HIV overall and in those individuals at highest risk.
Objective. Clinical interventions that lengthen life after HIV infection and significantly reduce transmission could have greater impact if more HIV-diagnosed people received HIV care. We tested a surveillance-based approach to investigating reasons for delayed entry to care. Methods. Health department staff in three states and two cities contacted eligible adults diagnosed with HIV four to 24 months previously who had no reported CD4+ lymphocyte (CD4) or viral load (VL) tests. The staff conducted interviews, performed CD4 and VL testing, and provided referrals to HIV medical care. Reported CD4 and VL tests were prospectively monitored to determine if respondents had entered care after the interview. Results. Surveillance-based follow-up uncovered problems with reporting CD4 and VL tests, resulting in surveillance improvements. However, reporting problems led to misspent effort locating people who were already in care Follow-up proved difficult because contact information in surveillance case records was often outdated or incorrect. Of those reached, 37% were in care and 29% refused participation. Information from 132 people interviewed generated ideas for service improvements, such as emphasizing the benefits of early initiation of HIV care, providing coverage eligibility information soon after diagnosis, and leveraging other medical appointments to provide assistance with linkage to HIV care. Conclusions. Surveillance-based follow-up of HIV-diagnosed individuals not linked to care provided information to improve both surveillance and linkage services, but was inefficient because of difficulties identifying, locating, and recruiting eligible people. Inefficiencies attributable to missing, incomplete, or inaccurate surveillance records are likely to diminish as data quality is improved through ongoing use.
Objective: To estimate the proportion of US HIV-positive men who report a male HIV-negative/unknown status (HIV-discordant) sexual partner taking preexposure prophylaxis (PrEP), and the use of multiple HIV prevention strategies within partnerships. Design: The Medical Monitoring Project is a complex sample survey of US adults with diagnosed HIV. Methods: We used data collected during June 2016 to May 2018 among sexually active HIV-positive men who had at least one HIV-discordant male partner ( N = 1871) to estimate the weighted prevalence of reporting at least one partner taking PrEP. Among HIV-discordant partnerships ( N = 4029), we estimated PrEP use, viral suppression among HIV-positive partners, and condomless anal sex. We evaluated significant ( P < 0.05) differences between groups using prevalence ratios with predicted marginal means. Results: Twenty-eight percent of sexually active HIV-positive MSM reported at least one HIV-discordant male partner taking PrEP. Twenty percent of HIV-discordant partners were reported to be taking PrEP; 73% were taking PrEP or the HIV-positive partner was virally suppressed. PrEP use was lower among black and Hispanic partners compared with white partners (12% and 19% vs. 27%). Fewer black than white MSM were in partnerships in which PrEP was used or the HIV-positive partner had sustained viral suppression (69% vs. 77%). Condomless anal intercourse was more prevalent in partnerships involving PrEP use and in partnerships involving either PrEP use or sustained viral suppression among the HIV-positive partner. Conclusion: PrEP use was reported among one in five partners, with disparities between black and white partners. Increasing PrEP use and decreasing racial/ethnic disparities could reduce disparities in HIV incidence and help end the US HIV epidemic.
Recent advances in statistical software1 have enabled public health researchers to fit multilevel models to a variety of outcome variables. Multilevel models facilitate inferences regarding unexplained variability among randomly sampled clusters of units (e.g., hospitals) in outcomes of interest and identify covariates that explain the variance in a given outcome at each level of a particular data hierarchy (e.g., patients within hospitals).2,3 Models with random intercepts enable researchers to accommodate correlations within higher-level units resulting from longitudinal or clustered study designs, and models with random coefficients enable researchers to identify higher-level covariates that explain between-cluster variance in relationships of interest.2,3
Public-use survey data sets collected from large national samples, such as the National Health and Nutrition Examination Survey, also have become widely available.4 The samples underlying these data sets are often “complex” in nature for 2 reasons: (1) the use of stratified multistage cluster sampling to increase sampling and cost efficiency and (2) unequal probabilities of selection from target populations for sampled elements, often as a result of oversampling of key subgroups (leading to the need to use weights for generating unbiased population estimates). Secondary analysts can accommodate these design complexities statistically by using “design-based” analyses, which ensure that population inferences are unbiased with respect to the sample design.4 However, these design-based approaches generally do not enable the types of cluster-specific inferences afforded by multilevel models,2,3 and researchers are now considering multilevel models for complex sample survey data.
Multilevel modeling represents a “model-based” approach to survey data analysis, in which dependencies in the data introduced by complex sampling features are generally accounted for by sound specification of the underlying probability model.5,6 Advocates of this approach argue that any information contained in the sample design features should be accounted for in the model specification, making the sampling uninformative.5 However, analysts may not have access to covariates capturing all of this information. In this case, the use of weighted estimation when fitting multilevel models provides some protection against potential biases introduced by informative sampling.6 Informed by recent methodological and computational developments in this area,1–3,6,7 we show that changes in inferences are possible when fitting multilevel models to complex sample survey data and ignoring the sampling weights.
We analyzed data from the 2013 Medical Monitoring Project HIV Provider Survey, sponsored by the Centers for Disease Control and Prevention, for which a probability sample of HIV care providers was selected from outpatient HIV care facilities in 16 states and Puerto Rico.8,9 Briefly, the provider survey followed a 2-stage probability-proportionate-to-size sample design, first sampling states and territories and then HIV facilities and selecting all providers within a facility. Unbiased estimation of multilevel model parameters requires the use of weights at all levels of a given data hierarchy,7 so we used previously calculated sampling weights adjusted for nonresponse at the facility level and inverses of estimated response probabilities at the provider level.
We focus on only facilities with multiple responding providers and include covariates that are both theoretically relevant for the dependent variables described later in this article and related to the sampling weights (e.g., an indicator of the provider serving more than 200 patients). Details about computation of the Medical Monitoring Project sampling weights for both providers and facilities are available on request.10 We scaled the final provider-level weights to sum to the sample sizes within each facility. A failure to do this would overstate actual sample sizes within each higher-level unit (facility), possibly resulting in biased estimates of model parameters.2,3,7
We fit multilevel logistic regression models to 2 binary dependent variables, indicating whether the responding provider delivered adequate drug use risk reduction and sexual risk reduction services to patients (defined as delivering approximately 70% of recommended risk reduction services to most or all of the patients). The models included random intercepts to capture between-facility variation in each proportion, in addition to fixed effects of several provider- and facility-level covariates of interest. We fit these models with the new GLIMMIX command11 in SAS/STAT version 13.1 (SAS Institute, Cary, NC), which can fit multilevel models to complex sample survey data. Identical results can be obtained with the new svy: melogit command in Stata version 14 (StataCorp LP, College Station, TX).
We did not test whether the parameter differences in the weighted and unweighted models were significant,12 but we did observe several shifts in inference when using weighted estimation (Table A; available as a supplement to the online version of this article at http://www.ajph.org). In both models, the intercept became more negative and significant, suggesting that the probability of using adequate risk reduction was being overstated for the type of provider represented by zeroes on all of the covariates (which may not be entirely meaningful in all models). For drug risk reduction, the coefficient for delivering care in a language other than English became nonsignificant. For the sexual risk reduction outcome, the male provider coefficient became significant, and the Black provider, nurse practitioner, and integrated team effects became even stronger. Finally, the estimated variability of the random facility intercepts was clearly being overstated when ignoring the weights, and the weighted models explained more of the variance in the outcomes at each level.
The weights at each level were clearly informative about the parameters defining these models, and ignoring them in analysis would have led to erroneous inferences with respect to the sample design used. Notably, these results held despite the inclusion of available covariates related to the sampling weights in the models. In practice, covariates used to compute the weights or the weights at each level of the data hierarchy may not be available to the public, making appropriate design-adjusted estimation of multilevel models difficult or impossible. We encourage analysts fitting multilevel models to survey data to carefully examine the variables available for weighted estimation in these data sets, make use of the powerful software1–3,11 that has been developed in this area, and (when possible) examine whether weighted estimation or adjustment for covariates related to the weights affects their inferences.
The United States spends more per capita on prescription drugs than do other high-income countries (1). In 2017, patients paid 14% of this cost out of pocket (2). Prescription drug cost-saving strategies, including nonadherence to medications due to cost concerns, have been documented among U.S. adults (3) and can negatively affect morbidity and, in the case of persons with human immunodeficiency virus (HIV) infection, can increase transmission risk (4,5). However, population-based data on prescription drug cost-saving strategies among U.S. persons with HIV are lacking. CDC's Medical Monitoring Project* analyzed cross-sectional, nationally representative, surveillance data on behaviors, medical care, and clinical outcomes among adults with HIV infection. During 2016-2017, 14% of persons with HIV infection used a prescription drug cost-saving strategy for any prescribed medication, and 7% had cost saving-related nonadherence. Nonadherence due to prescription drug costs was associated with reporting an unmet need for medications from the Ryan White AIDS Drug Assistance Program (ADAP), not having Medicaid coverage, and having private insurance. Persons who were nonadherent because of cost concerns were more likely to have visited an emergency department, have been hospitalized, and not be virally suppressed. Reducing barriers to ADAP and Medicaid coverage, in addition to reducing medication costs for persons with private insurance, might help to decrease nonadherence due to cost concerns and, thus contribute to improved viral suppression rates and other health outcomes among persons with HIV infection.
Background: During 2008–2015, the estimated annual HIV incidence rate in the United States decreased for each transmission risk category, except for men who have sex with men (MSM). Racial/ethnic disparities exist, with higher incidence rates for Black and Hispanic/Latino MSM. Setting: This analysis examines changes, 2010–2015, in disparities of HIV incidence among Black, Hispanic/Latino and White MSM. Methods: We compared results from the rate ratio, rate difference, weighted and unweighted index of disparity, and population attributable proportion. We calculated incidence rates for MSM using HIV surveillance data and published estimates of the MSM population in the United States. We generated 95% confidence intervals for each measure and used the Z statistic and associated P values to assess statistical significance. Findings: Results from all but one measure, Black-to-White rate difference, indicate that racial/ethnic disparities increased during 2010–2015; not all results were statistically significant. There were statistically significant increases in the Hispanic/Latino-to-White MSM incidence rate ratio (29%, P < 0.05), weighted index of disparity with the rate for White MSM as the referent group (9%, P < 0.05), and the population attributable proportion index (10%, P < 0.05). If racial/ethnic disparities among MSM had been eliminated, a range of 55%–61% decrease in overall MSM HIV incidence would have been achieved during 2010–2015. Conclusions: A large reduction in overall annual HIV incidence among MSM can be achieved by eliminating racial/ethnic disparities among MSM. Removing social and structural causes of racial/ethnic disparities among MSM can be effective in reducing overall annual HIV incidence among MSM.
