Faculty Bibliography

BACKGROUND: An Audio Computer-assisted Self Interview (ACASI) version of the Alcohol, Smoking, and Substance Involvement Screening Test (ASSIST) could reduce barriers to substance use screening and assessment in primary care settings. This study evaluated the diagnostic accuracy of an ACASI ASSIST for identification of unhealthy substance use and substance use disorders (SUD). METHODS: 399 adult patients were consecutively recruited from an urban safety-net primary care clinic. ACASI ASSIST scores for tobacco, alcohol, marijuana, and cocaine were compared against reference standard measures to assess the instrument's diagnostic accuracy for identifying unhealthy use and SUD, first using empirically-derived optimal cutoffs, and second using the currently recommended ASSIST cutoffs. RESULTS: For identifying any unhealthy use, at the empirically-derived cutoffs the ACASI ASSIST had 93.6% sensitivity and 85.8% specificity (AUC=0.90) for tobacco, 85.9% sensitivity and 60.3% specificity (AUC=0.73), for alcohol in men, 100% sensitivity and 62.4% specificity (AUC=0.81) for alcohol in women, 94.6% sensitivity and 81.6% specificity (AUC=0.88) for marijuana, and 86.1% sensitivity, 84.0% specificity (AUC=0.85) for cocaine. For SUD, sensitivity ranged from 79% (for alcohol in males), to 100% (for tobacco), and specificity was 83% or higher (AUCs ranged 0.83-0.91). For substances other than tobacco, empirically-derived cutoff scores were lower than the standard cutoffs, and resulted in higher sensitivity and lower specificity for identifying unhealthy substance use. CONCLUSIONS: The ACASI ASSIST is a valid measure of unhealthy use and SUD for substances that are commonly used by primary care patients, and could facilitate effective and efficient screening for substance use in medical settings.

Recruitment of people who use drugs (PWUD) for HIV-related research has been undertaken since early in the epidemic. In early studies, recruitment was often performed by outreach workers with familiarity with the target population, who distributed risk reduction materials, and administered the surveys being conducted on drug use and risk behaviors. The evolution of effective treatments for HIV has provided opportunities for PWUD to participate in biobehavioral studies testing the efficacy of medical treatment advances and exploring the underlying biomedical basis for prevention and treatment efforts. Recruitment for these studies has led to new challenges for outreach workers and institutions conducting this research. PWUD, particularly those from race/ethnic minority populations, have had lower rates of engagement in HIV care and have been underrepresented in HIV/AIDS medical studies. To address these health disparities, enhanced efforts are needed to increase their participation in biomedical studies. This article examines the challenges identified by experienced outreach workers in recruiting PWUD for HIV-related biomedical studies, including individual (participant)-, institutional-, and recruiter-level challenges, and provides recommendations for addressing them.

BACKGROUND: Human immunodeficiency virus (HIV) disease in the United States disproportionately affects minorities, including Latinos. Barriers including language are associated with lower antiretroviral therapy (ART) adherence seen among Latinos, yet ART and interventions for clinic visit adherence are rarely developed or delivered in Spanish. OBJECTIVE: The aim was to adapt a computer-based counseling tool, demonstrated to reduce HIV-1 viral load and sexual risk transmission in a population of English-speaking adults, for use during routine clinical visits for an HIV-positive Spanish-speaking population (CARE+ Spanish); the Technology Acceptance Model (TAM) was the theoretical framework guiding program development. METHODS: A longitudinal randomized controlled trial was conducted from June 4, 2010 to March 29, 2012. Participants were recruited from a comprehensive HIV treatment center comprising three clinics in New York City. Eligibility criteria were (1) adults (age >/=18 years), (2) Latino birth or ancestry, (3) speaks Spanish (mono- or multilingual), and (4) on antiretrovirals. Linear and generalized mixed linear effects models were used to analyze primary outcomes, which included ART adherence, sexual transmission risk behaviors, and HIV-1 viral loads. Exit interviews were offered to purposively selected intervention participants to explore cultural acceptability of the tool among participants, and focus groups explored the acceptability and system efficiency issues among clinic providers, using the TAM framework. RESULTS: A total of 494 Spanish-speaking HIV clinic attendees were enrolled and randomly assigned to the intervention (arm A: n=253) or risk assessment-only control (arm B, n=241) group and followed up at 3-month intervals for one year. Gender distribution was 296 (68.4%) male, 110 (25.4%) female, and 10 (2.3%) transgender. By study end, 433 of 494 (87.7%) participants were retained. Although intervention participants had reduced viral loads, increased ART adherence and decreased sexual transmission risk behaviors over time, these findings were not statistically significant. We also conducted 61 qualitative exit interviews with participants and two focus groups with a total of 16 providers. CONCLUSIONS: A computer-based counseling tool grounded in the TAM theoretical model and delivered in Spanish was acceptable and feasible to implement in a high-volume HIV clinic setting. It was able to provide evidence-based, linguistically appropriate ART adherence support without requiring additional staff time, bilingual status, or translation services. We found that language preferences and cultural acceptability of a computer-based counseling tool exist on a continuum in our urban Spanish-speaking population. Theoretical frameworks of technology's usefulness for behavioral modification need further exploration in other languages and cultures. TRIAL REGISTRATION: ClinicalTrials.gov NCT01013935; https://clinicaltrials.gov/ct2/show/NCT01013935 (Archived by WebCite at http://www.webcitation.org/6ikaD3MT7).

Background: Persistent immune activation is associated with a variety of adverse clinical outcomes. People who inject drugs (PWID) have high levels of immune activation in blood and mucosal tissues; however, the relative contributions of chronic HCV infection, highly prevalent among PWIDs, and the non-sterile injection of illicit drugs have remained obscure. Methods: We recruited (N=48 for each group): 1) active injectors of heroin 2) individuals who ceased injecting heroin for 1-2 months 3) individuals who ceased injecting heroin for 3-4 months 4) healthy non-injecting volunteers. Soluble (including sCD14, hs-CRP, TNF-a, IFN-g, IL-10, MIP-1a) and cell associated (CD38+HLA-DR+ CD4 and CD8+ T cells) markers of immune activation were quantified. Mixed-effects regression models with random intercepts to account for participation in more than one group were used to compare groups on markers of immune activation. Results: Participant characteristics are shown in Table 1 below. Levels of IL-12p70, IL-15, IL-1b, IL-2, IL-4, and IL-6 determined by multiplex ELISA were at or below the level of detection in 50% or more of the active injectors and were not analyzed. Mean levels of selected markers of systemic and cellular immune activation are shown in Table 2 below. Participants in Groups 2 and 3 had statistically significantly lower levels of TNF-a and % CD4+ and CD8+ CD38+/HLA-DR+ T cells compared to actively injecting Group 1 subjects only if HCV infection was spontaneously controlled or if subjects were HCV uninfected (HCV-aviremic). sCD14 levels in HCV-aviremic Group 3 subjects were significantly lower than in aviremic Group 1 subjects and comparable to Group 4. Additionally, hs-CRP levels were significantly lower in Group 2 but not in Group 3 compared to Group 1 subjects. In contrast, in HCV-viremic subjects, the above parameters were not significantly different between the groups and were significantly higher than in the healthy non-injecting volunteers. Levels of IFN-g, IL-10, and MIP-1a were comparable across Groups 1, 2, and 3 independent of the presence/absence of viremia. Conclusions: Active IDU and HCV viremia are associated with persistent immune activation. Select markers of immune activation are significantly lower among the HCV-aviremic who cease injecting but not in those who are HCV viremic. These findings may have public health consequences. Aggressive treatment of HCV infection as well as enhanced harm reduction efforts should converge to optimize long-term outcomes. (Table Presented)

BACKGROUND: Patients undergoing hematopoietic cell transplantation (HCT) for hematological malignancies experience a number of challenges during treatment. There is growing evidence that malglycemia (hyperglycemia, hypoglycemia, and/or increased glycemic variability) contributes to HCT-related complications, even in patients without preexisting diabetes. The purpose of this pilot study was to investigate factors influencing glycemic status and associated infection occurrences in nondiabetic autologous HCT recipients. METHODS: Oncology patients without preexisting diabetes treated with autologous HCT at a National Cancer Institute-designated cancer center were followed from admission through discharge or 28 days post-HCT. Patients had morning fasting laboratory tests. Descriptive statistics and Cox proportional hazards models were used to examine associations between BG levels and risk for infection while adjusting for baseline covariates including age, body mass index (BMI), cumulative glucocorticoid dose, and diagnosis. RESULTS: The sample included 28 female and 25 male predominately non-Hispanic White patients (mean age 55.7 years, SD = 11.32). Blood glucose (BG) range was 35-325 mg/dl. Twenty-three patients incurred at least one infection. BMI >/= 25 kg/m2 was associated with high BG and infections. In the multivariate Cox model, an increase of 1 interquartile range in BG 2 days before infection was associated with a moderately increased risk of infection (hazard ratio = 1.44, p = .008). CONCLUSIONS: Understanding the contributors to and consequences of malglycemic events can lead to better protocols for identifying patients at greater risk for infection. Further investigation is warranted for interventions to mitigate BG events for improved outcomes.

BACKGROUND: Ecstasy (MDMA) in the US is commonly adulterated with other drugs, but research has not focused on purity of ecstasy since the phenomenon of "Molly" (ecstasy marketed as pure MDMA) arose in the US. METHODS: We piloted a rapid electronic survey in 2015 to assess use of novel psychoactive substances (NPS) and other drugs among 679 nightclub/festival-attending young adults (age 18-25) in New York City. A quarter (26.1%) of the sample provided a hair sample to be analyzed for the presence of select synthetic cathinones ("bath salts") and some other NPS. Samples were analyzed using fully validated UHPLC-MS/MS methods. To examine consistency of self-report, analyses focused on the 48 participants with an analyzable hair sample who reported lifetime ecstasy/MDMA/Molly use. RESULTS: Half (50.0%) of the hair samples contained MDMA, 47.9% contained butylone, and 10.4% contained methylone. Of those who reported no lifetime use of "bath salts", stimulant NPS, or unknown pills or powders, about four out of ten (41.2%) tested positive for butylone, methylone, alpha-PVP, 5/6-APB, or 4-FA. Racial minorities were more likely to test positive for butylone or test positive for NPS after reporting no lifetime use. Frequent nightclub/festival attendance was the strongest predictor of testing positive for MDMA, butylone, or methylone. DISCUSSION: Results suggest that many ecstasy-using nightclub/festival attendees may be unintentionally using "bath salts" or other NPS. Prevention and harm reduction education is needed for this population and "drug checking" (e.g., pill testing) may be beneficial for those rejecting abstinence.

OBJECTIVE: We investigated the relationships between self-reported discrimination (SRD) and mental and physical health (self-reported physical health conditions and direct, physiologic measures [BMI, waist-to-hip ratio, and blood pressure]) among Sikh Asian Indians (AI), a group that may be particularly discriminated against because of physical manifestations of their faith, including a tendency to wear turbans or ethnic clothing. METHODS: Sikh AIs (N = 196) were recruited from Sikh gurdwaras in Queens, New York. Data were collected on SRD, social support, and self-reported health, along with multiple direct physiological measures for cardiovascular health. RESULTS: Participants who wore turbans/scarves reported higher levels of discrimination than those who did not wear turbans/scarves. As hypothesized, multiple regression analysis supported that discrimination is significantly associated with poorer self-reported mental (B = -.53, p < .001) and physical health (B = -.16, p = .04) while controlling for socioeconomic, acculturation, and social support factors. The study did not support an association between SRD and physiologic measures (elevated BMI, waist-to-hip ratio, and blood pressure). CONCLUSION: Consistent with previous discrimination and health reports, this study demonstrated an inverse relationship between discrimination and health among Sikh AIs, an understudied yet high-risk minority population. Community-based efforts are also needed to reduce the occurrence or buffer the effects of discrimination experienced by Sikh AIs. (PsycINFO Database Record

We evaluated performance, accuracy, and acceptability parameters of unsupervised oral fluid (OF) HIV self-testing (HIVST) in a general population in western Kenya. In a prospective validation design, we enrolled 240 adults to perform rapid OF HIVST and compared results to staff administered OF and rapid fingerstick tests. All reactive, discrepant, and a proportion of negative results were confirmed with lab ELISA. Twenty participants were video-recorded conducting self-testing. All participants completed a staff administered survey before and after HIVST to assess attitudes towards OF HIVST acceptability. HIV prevalence was 14.6 %. Thirty-six of the 239 HIVSTs were invalid (15.1 %; 95 % CI 11.1-20.1 %), with males twice as likely to have invalid results as females. HIVST sensitivity was 89.7 % (95 % CI 73-98 %) and specificity was 98 % (95 % CI 89-99 %). Although sensitivity was somewhat lower than expected, there is clear interest in, and high acceptability (94 %) of OF HIV self-testing.