John Weiser

Service Delivery and Patient Outcomes in Ryan White HIV/AIDS Program–Funded and –Nonfunded Health Care Facilities in the United States

IMPORTANCE Outpatient human immunodeficiency virus (HIV) health care facilities receive funding from the Ryan White HIV/AIDS Program (RWHAP) to provide medical care and essential support services that help patients remain in care and adhere to treatment. Increased access to Medicaid and private insurance for HIV-infected persons may provide coverage for medical care but not all needed support services and may not supplant the need for RWHAP funding. OBJECTIVE To examine differences between RWHAP-funded and non-RWHAP-funded facilities and in patient outcomes between the 2 systems. DESIGN, SETTING, AND PARTICIPANTS The study was conducted from June 1, 2009, to May 31, 2012, using data from the 2009 and 2011 cycles of the Medical Monitoring Project, a national probability sample of 8038 HIV-infected adults receiving medical care at 989 outpatient health care facilities providing HIV medical care. MAIN OUTCOMES AND MEASURES Data were used to compare patient characteristics, service needs, and access to services at RWHAP-funded vs non-RWHAP-funded facilities. Differences in prescribed antiretroviral treatment and viral suppression were assessed. Data analysis was performed between February 2012 and June 2015. RESULTS Overall, 34.4% of facilities received RWHAP funding and 72.8% of patients received care at RWHAP-funded facilities. With results reported as percentage (95% CI), patients attending RWHAP-funded facilities were more likely to be aged 18 to 29 years (8.5% [7.4%-9.5%] vs 5.0% [3.9%-6.2%]), female (29.2% [27.2%-31.2%] vs 20.1% [17.0%-23.1%]), black (47.5% [41.5%-53.5%] vs 25.8% [20.6%-31.0%]) or Hispanic (22.5% [16.4%-28.6%] vs 12.9% [10.6%-15.2%]), have less than a high school education (26.1% [24.0%-28.3%] vs 10.9% [8.7%-13.1%]), income at or below the poverty level (53.6% [50.3%-56.9%] vs 23.9% [19.7%-28.0%]), and lack health care coverage (25.0% [21.9%-28.1%] vs 6.1% [4.1%-8.0%]). The RWHAP-funded facilities were more likely to provide case management (76.1% [69.9%-82.2%] vs 15.4% [10.4%-20.4%]) as well as mental health (64.0% [57.0%-71.0%] vs 18.0% [14.0%-21.9%]), substance abuse (33.6% [27.0%-40.2%] vs 12.0% [8.0%-16.0%]), and other support services; patients attending RWHAP-funded facilities were more likely to receive these services. After adjusting for patient characteristics, the percentage prescribed ART antiretroviral therapy, reported as adjusted prevalence ratio (95% CI), was similar between RWHAP-funded and non-RWHAP-funded facilities (1.01 [0.99-1.03]), but among poor patients, those attending RWHAP-funded facilities were more likely to be virally suppressed (1.09 [1.02-1.16]). CONCLUSIONS AND RELEVANCE A total of 72.8% of HIV-positive patients received care at RWHAP-funded facilities. Many had multiple social determinants of poor health and used services at RWHAP-funded facilities associated with improved outcomes. Without facilities supported by the RWHAP, these patients may have had reduced access to services elsewhere. Poor patients were more likely to achieve viral suppression if they received care at a RWHAP-funded facility.

Clinician Perspectives on Delaying Initiation of Antiretroviral Therapy for Clinically Eligible HIV-Infected Patients

Objectives: Guidelines for antiretroviral therapy (ART) initiation have evolved, but consistently note that adherence problems should be considered and addressed. Little is known regarding the reasons providers delay ART initiation in clinically eligible patients. Methods: In 2009, we surveyed a probability sample of HIV care providers in 582 outpatient facilities in the United States and Puerto Rico with an open-ended question about nonclinical reasons for delaying ART initiation in otherwise clinically eligible patients. Results: Very few providers (2%) reported never delaying ART. Reasons for delaying ART were concerns about patient adherence (68%), patient acceptance (60%), and structural barriers (33%). Provider and practice characteristics were associated with reasons for delaying ART. Conclusion: Reasons for delaying ART were consistent with clinical guidelines and were both patient level and structural. Providers may benefit from training and access to referrals for ancillary services to enhance their ability to monitor and address these issues with their patients.

Qualifications, Demographics, Satisfaction, and Future Capacity of the HIV Care Provider Workforce in the United States, 2013–2014

BACKGROUND The human immunodeficiency virus (HIV)-infected population in the United States is increasing by about 30 000 annually (new infections minus deaths). With improvements in diagnosis and engagement in care, additional qualified HIV care providers may be needed. METHODS We surveyed a probability sample of 2023 US HIV care providers in 2013-2014, including those at Ryan White HIV/AIDS Program (RWHAP)-funded facilities and in private practices. We estimated future patient care capacity by comparing counts of providers entering and planning to leave practice within 5 years, and the number of patients under their care. RESULTS Of surveyed providers, 1234 responded (adjusted response rate, 64%): 63% were white, 11% black, 11% Hispanic, and 16% other race/ethnicity; 37% were satisfied/very satisfied with salary/reimbursement, and 33% were satisfied/very satisfied with administrative time. Compared with providers in private practice, more providers at RWHAP-funded facilities were HIV specialists (71% vs 43%; P < .0001) and planned to leave HIV practice within 5 years (11% vs 4%; P = .0004). An estimated 190 more full-time equivalent providers (defined as 40 HIV clinical care hours per week) entered practice in the past 5 years than are expected to leave in the next 5 years. If these rates continue, by 2019 patient care capacity will increase by 65 000, compared with an increased requirement of at least 100 000. CONCLUSIONS Projected workforce growth by 2019 will not accommodate the increased number of HIV-infected persons requiring care. RWHAP-funded facilities may face attrition of highly qualified providers. Dissatisfaction with salary/reimbursement and administrative burden is substantial, and black and Hispanic providers are underrepresented relative to HIV patients.

Trends in cigarette smoking among adults with HIV compared with the general adult population, United States - 2009–2014

Smoking increases HIV-related and non-HIV-related morbidity and mortality for persons with HIV infection. We estimated changes in cigarette smoking among adults with HIV and adults in the general U.S. population from 2009 to 2014 to inform HIV smoking cessation programs. Among HIV-positive adults, rates of current smoking declined from 37.6% (confidence interval [CI]: 34.7-40.6) in 2009 to 33.6% (CI: 29.8-37.8) in 2014. Current smoking among U.S. adults declined from 20.6% (CI: 19.9-21.3) in 2009 to 16.8% (CI: 16.2-17.4) in 2014. HIV-positive adults in care were significantly more likely to be current smokers compared with the general U.S. population; they were also less likely to quit smoking. For both HIV-positive adults in care and the general population, disparities were noted by racial/ethnic, educational level, and poverty-level subgroups. For most years, non-Hispanic blacks, those with less than high school education, and those living below poverty level were more likely to be current smokers and less likely to quit smoking compared with non-Hispanic whites, those with greater than high school education, and those living above poverty level, respectively. To decrease smoking-related causes of illness and death and to decrease HIV-related disparities, smoking cessation interventions are vital as part of routine care with HIV-positive persons. Clinicians who care for HIV-positive persons who smoke should utilize opportunities to discuss and implement smoking cessation strategies during routine clinical visits.

Reproductive health counseling delivered to women living with HIV in the United States

ABSTRACT Advances in antiretroviral therapy (ART) and reproductive technologies have made transmission of HIV to partners and infants almost completely preventable. Comprehensive reproductive health counseling (CRHC) is an important component of care for women living with HIV, but few women report discussing reproductive health with an HIV care provider. We surveyed a probability sample of U.S. HIV care providers during 2013–2014. Of 2023 eligible providers, 1234 responded (64% adjusted provider response rate). We estimated the percentage delivering CRHC to their female patients. CRHC was defined as delivering each of five components of reproductive health care to most or all female patients. We assessed associations between provider characteristics and delivering CRHC using chi-squared tests and multivariable logistic regression. Of all providers, 49% (95% confidence interval [CI], 42–55) reported delivering all components of CRHC: 71% assessed reproductive intentions of reproductive-aged women, 78% explained perinatal transmission risk, 87% discussed ART for preventing perinatal transmission, 76% provided contraception as appropriate, and 64% provided referrals for preconception care. Among providers who offered primary care (83% of sample), 52% (CI: 44–60) delivered CRHC compared to 33% (CI: 22–44) of providers who did not offer primary care (P = .01). More female providers (46% of sample) compared to male providers delivered CRHC (57% [CI: 48–65] vs. 40% [CI: 31–50], P < .01). Delivery of CRHC by providers did not differ by patient caseload. After adjusting for gender, years of HIV experience, and patient caseload, providing primary care to HIV-infected patients remained associated with delivering CRHC (adjusted prevalence ratio [aPR] 1.48, 95% CI 1.02–2.16). Provider delivery of CRHC is not consistent with current guidelines that recommend discussing reproductive health with all reproductive-aged women who are living with HIV, even among providers offering primary care to their HIV patients.

Prevention counseling practices of HIV care providers with patients new to HIV medical care: medical monitoring project provider survey, 2009.

Objectives: To determine the prevalence of prevention counseling discussions between HIV care providers and their patients who are newly linked to care and to assess factors that facilitate such discussions. Methods: In 2009, a probability sample of HIV care providers in 582 outpatient settings in the United States and Puerto Rico was surveyed regarding provider’s HIV prevention discussions with HIV-infected patients newly linked to HIV medical care. Results: A majority of providers reported consistently discussing HIV transmission risk reduction (76%), sexually transmitted disease risk (66%), and adherence to antiretroviral regimens (87%). Only 35% of providers reported consistently discussing partner counseling services. Conclusion: The proportion of providers engaged in HIV prevention counseling with patients newly linked to HIV care is generally high, but more work is needed to encourage providers to fully participate as partners in prevention, which is central to preventing onward transmission of HIV.

Prescribing of Human Immunodeficiency Virus (HIV) Pre-exposure Prophylaxis by HIV Medical Providers in the United States, 2013–2014

Abstract Background. Clinical trials have demonstrated the effectiveness of human immunodeficiency virus (HIV) pre-exposure prophylaxis (PrEP) for reducing HIV acquisition. Understanding how HIV care providers are prescribing PrEP is necessary to ensure success of this prevention strategy. Methods. During 2013–2014, we surveyed US HIV care providers who also provided care to HIV-negative patients. We estimated percentages who had prescribed PrEP and assessed associations between provider characteristics and PrEP prescribing. Results. An estimated 26% (95% confidence interval [CI], 20–31) had ever prescribed PrEP. Of these, 74% (95% CI, 61–87) prescribed for men who have sex with men (MSM), 30% (95% CI, 21–39) for women who have sex with men, 23% (95% CI, 9–37) for men who have sex with women, 23% (95% CI, 15–30) for uninfected partners in HIV-discordant couples trying to conceive, and 1% (95% CI, 0–2) for persons who inject drugs. The following provider characteristics were significantly associated with having prescribed PrEP: male vs female (32% vs 16%; adjusted prevalence ratio [aPR], 1.5; 95% CI, 1.0–2.2), lesbian/gay/bisexual vs heterosexual orientation (50% vs 21%; aPR, 2.0; 95% CI, 1.3–2.9), and HIV caseload (>200, 51–200, and ≤50 patients, 39%, 29%, and 14%, respectively; >200 vs ≤50 patients, aPR 2.4, 95% CI 1.1–5.2, and 51–200 vs ≤50 patients, aPR 2.2, 95% CI 1.2–4.0). Conclusions. In 2013–2014, one quarter of HIV care providers reported having prescribed PrEP, most commonly for MSM and rarely for persons who inject drugs. Lesbian/gay/bisexual providers and male providers were more likely than others to have prescribed PrEP. Additional efforts may enable more providers to prescribe PrEP to underserved clients needing the service.

Delivery of HIV Transmission Risk-Reduction Services by HIV Care Providers in the United States-2013.

Objectives: Evidence-based guidelines have long recommended that HIV care providers deliver HIV transmission risk-reduction (RR) services, but recent data are needed to assess their adoption. Methods: The authors surveyed a probability sample of 1234 US HIV care providers on delivery of 9 sexual behavior– and 7 substance use–related HIV transmission RR services and created an indicator of “adequate” delivery of services in each area, defined as performing approximately 70% or more of applicable services. Results: Providers were most likely to encourage patients to disclose HIV status to all partners since HIV diagnosis (81%) and least likely to ask about disclosure to new sex and drug injection partners at follow-up visits (both 41%). Adequate delivery of sexual behavior– and substance use–related RR services was low (37% and 43%, respectively). Conclusion: The majority of US HIV care providers may need additional support to improve delivery of comprehensive HIV transmission RR services.

Retention in Care Services Reported by HIV Care Providers in the United States, 2013 to 2014:

Objectives: Evidence-based guidelines recommend that HIV care providers offer retention-in-care services, but data are needed to assess service provision. Methods: We surveyed a probability sample of 1234 HIV care providers to estimate the percentage of providers whose practices offered 5 recommended retention services and describe providers’ perceptions of barriers to care among patients. Results: An estimated 21% of providers’ practices offered all 5 retention services. Providers at smaller (<50 versus >400 patients), private, and non-Ryan White HIV/AIDS Program (RWHAP)-funded practices, and practices without on-site case management were significantly less likely to provide patient navigation services or do systematic monitoring of retention. Providers’ most commonly perceived barriers to care among patients were mental health (40%), substance abuse (36%), and transportation (34%) issues. Conclusion: Deficiencies in the provision of key retention services are substantial. New strategies may be needed to increase the delivery of recommended retention services, especially among private, non-RWHAP-funded, and smaller facilities.

Weighted Multilevel Models: A Case Study

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.