Population-average mediation analysis for zero-inflated count outcomes.

Mediation analysis is an increasingly popular statistical method for explaining causal pathways to inform intervention. While methods have increased, there is still a dearth of robust mediation methods for count outcomes with excess zeroes. Current mediation methods addressing this issue are computationally intensive, biased, or challenging to interpret. To overcome these limitations, we propose a new mediation methodology for zero-inflated count outcomes using the marginalized zero-inflated Poisson (MZIP) model and the counterfactual approach to mediation. This novel work gives population-average mediation effects whose variance can be estimated rapidly via delta method. This methodology is extended to cases with exposure-mediator interactions. We apply this novel methodology to explore if diabetes diagnosis can explain BMI differences in healthcare utilization and test model performance via simulations comparing the proposed MZIP method to existing zero-inflated and Poisson methods. We find that our proposed method minimizes bias and computation time compared to alternative approaches while allowing for straight-forward interpretations.

A Moderated Mediation Analysis of HIV and Intersectional Stigmas and Antiretroviral Adherence in People Living with HIV in the Dominican Republic.

Experiencing HIV and intersectional stigmas in healthcare settings may affect antiretroviral treatment (ART) adherence among people with HIV (PWH), given their need for frequent interactions with clinical settings and healthcare providers. Considering the importance of reducing stigmas to promote well-being and the need to elucidate how stigma influences health across various settings, we examined how experienced HIV stigma in Dominican Republic healthcare settings impacts ART adherence through internalized HIV stigma and whether race or sexual orientation stigma moderates this relationship. Participants were 471 PWH (aged 17-71) who were recruited from two HIV clinics in the Dominican Republic in 2021-2022. Results revealed a significant mediation effect (B=-0.10, SE = 0.05, CI [-0.234, - 0.014]) after adjusting for effect of age and time since HIV diagnosis, suggesting that experienced HIV stigma in healthcare settings was associated with more internalized HIV stigma (B = 0.39, SE = 0.11, p = .001), subsequently linked to lower ART adherence (B=-0.26, SE = 0.11, p = .016). The indirect effect was significant at low levels of race stigma (B=-0.16, SE = 0.09, CI [-0.369, - 0.001]) but not at high levels of race stigma (B=-0.06, SE = 0.05, CI [-0.175, 0.038]). This indirect effect was also significant at low levels of sexual orientation stigma (B=-0.19, SE = 0.10, CI [-0.401, - 0.023]) but not at high levels of sexual orientation stigma (B=-0.04, SE = 0.06, CI [-0.160, 0.074]). These findings suggest that addressing experienced HIV stigma in Dominican Republic healthcare settings, along with various dimensions of HIV-related stigma (e.g., internalized stigma) and intersecting stigmas (e.g., race, sexual orientation), is vital for improving health outcomes, such as optimal ART adherence.

RESUMEN: Experimentar estigmas relacionados con el VIH e interseccionales en entornos de atención médica puede afectar la adherencia al tratamiento antirretroviral (TAR) entre las personas que viven con VIH (PVVIH), dado que necesitan interacciones frecuentes con entornos clínicos y proveedores de atención médica. Considerando la importancia de reducir los estigmas para promover el bienestar y la necesidad de esclarecer cómo el estigma influye en la salud en diversos contextos, examinamos cómo el estigma del VIH experimentado en entornos de atención médica en la República Dominicana impacta la adherencia al TAR a través del estigma internalizado del VIH y si el estigma racial o de orientación sexual modera esta relación. Los participantes fueron 471 PVVIH (de 17 a 71 años) que fueron reclutados de dos clínicas de VIH en la República Dominicana en 2021­2022. Los resultados revelaron un efecto de mediación significativo (B=-0.10, SE = 0.05, CI [-0.234, − 0.014]) después de ajustar por el efecto de la edad y el tiempo desde el diagnóstico de VIH, sugiriendo que el estigma del VIH experimentado en entornos de atención médica estaba asociado con un mayor estigma internalizado del VIH (B = 0.39, SE = 0.11, p = .001), vinculado posteriormente a una menor adherencia al TAR (B=-0.26, SE = 0.11, p = .016). El efecto indirecto fue significativo en niveles bajos de estigma racial (B=-0.16, SE = 0.09, CI [-0.369, − 0.001]) pero no en niveles altos de estigma racial (B=-0.06, SE = 0.05, CI [-0.175, 0.038]). Este efecto indirecto también fue significativo en niveles bajos de estigma por orientación sexual (B=-0.19, SE = 0.10, CI [-0.401, − 0.023]) pero no en niveles altos de estigma por orientación sexual (B=-0.04, SE = 0.06, CI [-0.160, 0.074]). Estos hallazgos sugieren que abordar el estigma del VIH experimentado en entornos de atención médica en la República Dominicana, junto con diversas dimensiones del estigma relacionado con el VIH (por ejemplo, estigma internalizado) y estigmas interseccionales (por ejemplo, raza, orientación sexual), es vital para mejorar los resultados de salud, como la adherencia óptima al TAR.

Mendelian randomization in the multivariate general linear model framework.

Mendelian randomization (MR) is an application of instrumental variable (IV) methods to observational data in which the IV is a genetic variant. MR methods applicable to the general exponential family of distributions are currently not well characterized. We adapt a general linear model framework to the IV setting and propose a general MR method applicable to any full-rank distribution from the exponential family. Empirical bias and coverage are estimated via simulations. The proposed method is compared to several existing MR methods. Real data analyses are performed using data from the REGARDS study to estimate the potential causal effect of smoking frequency on stroke risk in African Americans. In simulations with binary variates and very weak instruments the proposed method had the lowest median [Q1 , Q3 ] bias (0.10 [-3.68 to 3.62]); compared with 2SPS (0.27 [-3.74 to 4.26]) and the Wald method (-0.69 [-1.72 to 0.35]). Low bias was observed throughout other simulation scenarios; as well as more than 90% coverage for the proposed method. In simulations with count variates, the proposed method performed comparably to 2SPS; the Wald method maintained the most consistent low bias; and 2SRI was biased towards the null. Real data analyses find no evidence for a causal effect of smoking frequency on stroke risk. The proposed MR method has low bias and acceptable coverage across a wide range of distributional scenarios and instrument strengths; and provides a more parsimonious framework for asymptotic hypothesis testing compared to existing two-stage procedures.

Partnership status and retention in care among cisgender heterosexual newly diagnosed people with HIV: a cohort study.

This study examined the role of partnership status (married, unmarried-partnered, and unpartnered) on retention in care among newly diagnosed (2013-2017), cisgender heterosexual people with HIV in Birmingham, Alabama (n = 152). This study evaluated all scheduled HIV primary care provider visits for two years following diagnosis date. A kept-visit measure was calculated such that, if an individual attended ≥1 visit in each of the four 6-month intervals, they were considered to have high visit constancy. A missed-visit measure was categorized as ≥1 no-show vs. 0 no-show for first and second year after diagnosis. Odds ratios (OR) and 95% confidence intervals (CI) were calculated using logistic regression models. Models were adjusted for confounding sociodemographic and clinical characteristics. The study population was 76% Black, 57% male, median age of 37 years. Overall, 65% had high visit consistency and 34.5% had ≥1 no-show in both years. Compared to unpartnered, married individuals had higher visit constancy [AOR (95% CI): 2.88 (1.02, 8.16)]; no differences were observed among unmarried-partnered individuals. No differences in having ≥1 no-show among partnership status groups were observed for either year. These findings suggest potential success of interventions involving a social confidant in optimizing retention in care among newly diagnosed, heterosexual PWH.

Development and implementation of a distributed data network between an academic institution and state health departments to investigate variation in time to HIV viral suppression in the Deep South.

BACKGROUND: Achieving early and sustained viral suppression (VS) following diagnosis of HIV infection is critical to improving outcomes for persons with HIV (PWH). The Deep South of the United States (US) is a region that is disproportionately impacted by the domestic HIV epidemic. Time to VS, defined as time from diagnosis to initial VS, is substantially longer in the South than other regions of the US. We describe the development and implementation of a distributed data network between an academic institution and state health departments to investigate variation in time to VS in the Deep South. METHODS: Representatives of state health departments, the Centers for Disease Control and Prevention (CDC), and the academic partner met to establish core objectives and procedures at the beginning of the project. Importantly, this project used the CDC-developed Enhanced HIV/AIDS Reporting System (eHARS) through a distributed data network model that maintained the confidentiality and integrity of the data. Software programs to build datasets and calculate time to VS were written by the academic partner and shared with each public health partner. To develop spatial elements of the eHARS data, health departments geocoded residential addresses of each newly diagnosed individual in eHARS between 2012-2019, supported by the academic partner. Health departments conducted all analyses within their own systems. Aggregate results were combined across states using meta-analysis techniques. Additionally, we created a synthetic eHARS data set for code development and testing. RESULTS: The collaborative structure and distributed data network have allowed us to refine the study questions and analytic plans to conduct investigations into variation in time to VS for both research and public health practice. Additionally, a synthetic eHARS data set has been created and is publicly available for researchers and public health practitioners. CONCLUSIONS: These efforts have leveraged the practice expertise and surveillance data within state health departments and the analytic and methodologic expertise of the academic partner. This study could serve as an illustrative example of effective collaboration between academic institutions and public health agencies and provides resources to facilitate future use of the US HIV surveillance system for research and public health practice.

A novel Mendelian randomization method with binary risk factor and outcome.

BACKGROUND: Mendelian randomization (MR) applies instrumental variable (IV) methods to observational data using a genetic variant as an IV. Several Monte-Carlo studies investigate the performance of MR methods with binary outcomes, but few consider them in conjunction with binary risk factors. OBJECTIVE: To develop a novel MR estimator for scenarios with a binary risk factor and outcome; and compare to existing MR estimators via simulations and real data analysis. METHODS: A bivariate Bernoulli distribution is adapted to the IV setting. Empirical bias and asymptotic coverage probabilities are estimated via simulations. The proposed method is compared to the Wald method, two-stage predictor substitution (2SPS), two-stage residual inclusion (2SRI), and the generalized method of moments (GMM). An analysis is performed using existing data from the CLEAR study to estimate the potential causal effect of smoking on rheumatoid arthritis risk in African Americans. RESULTS: Bias was low for the proposed method and comparable to 2SPS. The Wald method was often biased towards the null. Coverage was adequate for the proposed method, 2SPS, and 2SRI. Coverage for the Wald and GMM methods was poor in several scenarios. The causal effect of ever smoking on rheumatoid arthritis risk was not statistically significant using a variety of genetic instruments. CONCLUSIONS: Simulations suggest the proposed MR method is sound with binary risk factors and outcomes, and comparable to 2SPS and 2SRI in terms of bias. The proposed method also provides more natural framework for hypothesis testing compared to 2SPS or 2SRI, which require ad-hoc variance adjustments.

Prevalence and impact of comorbid chronic pain and cigarette smoking among people living with HIV.

Rates of chronic pain and cigarette smoking are each substantially higher among people living with HIV (PLWH) than in the general population. The goal of these analyses was to examine the prevalence and impact of comorbid chronic pain and cigarette smoking among PLWH. Participants included 3289 PLWH (83% male) who were recruited from five HIV clinics. As expected, the prevalence of smoking was higher among PLWH with chronic pain (41.9%), than PLWH without chronic pain (26.6%, p < .0001), and the prevalence of chronic pain was higher among current smokers (32.9%), than among former (23.6%) or never (17%) smokers (ps < .0001). PLWH who endorsed comorbid chronic pain and smoking (vs. nonsmokers without chronic pain) were more likely to report cocaine/crack and cannabis use, be prescribed long-term opioid therapy, and have virologic failure, even after controlling for relevant sociodemographic and substance-related variables (ps < .05). These results contribute to a growing empirical literature indicating that chronic pain and cigarette smoking frequently co-occur, and extend this work to a large sample of PLWH. Indeed, PLWH may benefit from interventions that are tailored to address bidirectional pain-smoking effects in the context of HIV.

An Investigation of Selection Bias in Estimating Racial Disparity in Stroke Risk Factors.

Selection due to survival or attrition might bias estimates of racial disparities in health, but few studies quantify the likely magnitude of such bias. In a large national cohort with moderate loss to follow-up, we contrasted racial differences in 2 stroke risk factors, incident hypertension and incident left ventricular hypertrophy, estimated by complete-case analyses, inverse probability of attrition weighting, and the survivor average causal effect. We used data on 12,497 black and 17,660 white participants enrolled in the United States (2003-2007) and collected incident risk factor data approximately 10 years after baseline. At follow-up, 21.0% of white participants and 23.0% of black participants had died; additionally 22.0% of white participants and 28.4% of black participants had withdrawn. Individual probabilities of completing the follow-up visit were estimated using baseline demographic and health characteristics. Adjusted risk ratio estimates of racial disparities from complete-case analyses in both incident hypertension (1.11, 95% confidence interval: 1.02, 1.21) and incident left ventricular hypertrophy (1.02, 95% confidence interval: 0.84, 1.24) were virtually identical to estimates from inverse probability of attrition weighting and survivor average causal effect. Despite racial differences in mortality and attrition, we found little evidence of selection bias in the estimation of racial differences for these incident risk factors.

HIV Status and Other Risk Factors for Prevalent and Incident Sexually Transmitted Infection during Pregnancy (2000-2014).

Background: Sexually transmitted infections (STIs) are associated with adverse birth outcomes. Current prenatal STI screening guidelines define "risk" without explicit consideration of HIV status. Our objective was to test the hypothesis that HIV status is associated with bacterial STI in pregnant women. Methods: We designed a retrospective cohort study to identify pregnant women with HIV who delivered at our facility during 2000-2014. HIV+ women were compared to HIV- women with matching by year of delivery. Logistic regression was used to model adjusted odds of prevalent and incident STI. Prevalent STI was defined as chlamydia (CT), gonorrhea (GC), syphilis, or trichomoniasis detected on an initial prenatal screening test and incident STI as a newly positive result following a negative prenatal test. Results: The cohort included 432 women, 210 HIV+ and 222 HIV-. Most pregnant women were screened for STI (92% of HIV+ women and 74% of HIV- women). STI rates were high and particularly elevated in HIV+ women: 29% vs 18% (p=0.02), for prevalent STI and 11% vs 2% (p<0.001) for incident STI. Risk factors for prevalent STI were as follows: HIV status (aOR 3.0, CI: 1.4-6.4), Black race (aOR 2.7, 95% CI: 1.1-6.6), and more recent delivery (2007-2014 compared to 2000-2006) (aOR 2.3, CI: 1.1-4.7). HIV status was an independent risk factor for incident STI (aOR 7.2, CI: 2.1-25.0). Conclusion: Pregnant women who delivered in our center had high STI rates. Since HIV infection was independently associated with prevalent and incident STI, prenatal screening guidelines may need to incorporate HIV status as a high-risk group for repeat testing.

Early detection of cardiac dysfunction in the type 1 diabetic heart using speckle-tracking based strain imaging.

Enhanced sensitivity in echocardiographic analyses may allow for early detection of changes in cardiac function beyond the detection limits of conventional echocardiographic analyses, particularly in a small animal model. The goal of this study was to compare conventional echocardiographic measurements and speckle-tracking based strain imaging analyses in a small animal model of type 1 diabetes mellitus. Conventional analyses revealed differences in ejection fraction, fractional shortening, cardiac output, and stroke volume in diabetic animals relative to controls at 6-weeks post-diabetic onset. In contrast, when assessing short- and long-axis speckle-tracking based strain analyses, diabetic mice showed changes in average systolic radial strain, radial strain rate, radial displacement, and radial velocity, as well as decreased circumferential and longitudinal strain rate, as early as 1-week post-diabetic onset and persisting throughout the diabetic study. Further, we performed regional analyses for the LV and found that the free wall region was affected in both the short- and long-axis when assessing radial dimension parameters. These changes began 1-week post-diabetic onset and remained throughout the progression of the disease. These findings demonstrate the use of speckle-tracking based strain as an approach to elucidate cardiac dysfunction from a global perspective, identifying left ventricular cardiac regions affected during the progression of type 1 diabetes mellitus earlier than contractile changes detected by conventional echocardiographic measurements.

Nationwide secondary overtriage in level 3 and level 4 trauma centers: are these transfers necessary?

BACKGROUND: Secondary overtriage (SO) refers to the interfacility transfer of trauma patients who are rapidly discharged home without surgical intervention by the receiving institution. SO imposes a financial hardship on patients and strains trauma center resources. Most studies on SO have been conducted from the perspective of the receiving hospital, which is usually a level 1 trauma center. Having previously studied SO from the referring rural hospital's perspective, we sought to identify variables contributing to SO at the national level. METHODS: Using data from the 2008-2012 National Trauma Data Bank, we isolated patients transferred to level 1 trauma centers who were: (1) discharged home within 48 h and (2) did not undergo any surgical procedure. This population was subsequently compared with similar patients treated at and discharged directly from level 3 and 4 centers. Multivariate logistic regression analysis was used to isolate variables that independently influenced a patient's risk of undergoing SO. Injury patterns were characterized by use of subspecialty consultants. RESULTS: A total of 99,114 patients met inclusion criteria, of which 13.2% were discharged directly from level 3 or 4 trauma centers, and 86.8% of them were transferred to a level 1 trauma center before discharge. The mean Injury Severity Score of the nontransfer and transfer groups was 5.4 ± 4.5 and 7.3 ± 5.7, respectively. Multivariate regression analysis showed that Injury Severity Score > 15, alcoholism, smoking, drug use, and certain injury patterns involving the head, vertebra, and face were associated with being transferred. In this minimally injured population, factors protective against transfers were: age > 65 y, female gender, systolic blood pressure <80, a head computed tomography scan and orthopedic injuries. CONCLUSIONS: SO results from the complex interplay of variables including patient demographics, facility characteristics, and injury type. The inability to exclude a potentially devastating neurologic injury seems to drive SO.

Surrogates of protection in repeated low-dose challenge experiments.

A critical step toward developing a successful vaccine to control the human immunodeficiency virus pandemic entails evaluation of vaccine candidates in non-human primates (NHPs). Historically, these studies have usually entailed challenges (i.e., exposures) with very high doses of a simian version of human immunodeficiency virus, resulting in infection of all NHPs in the experiment after a single challenge. More recently, researchers have begun to conduct repeated low-dose challenge (RLC) studies in NHPs that are believed to more closely mimic typical exposure in natural human transmission settings. One objective of RLC studies is to assess whether measured immune responses to vaccination can serve as surrogate endpoints for the primary endpoint of interest, namely infection. In this paper, different designs of RLC studies for assessing a binary surrogate of protection are considered.

Functional deficiencies of subsarcolemmal mitochondria in the type 2 diabetic human heart.

The mitochondrion has been implicated in the development of diabetic cardiomyopathy. Examination of cardiac mitochondria is complicated by the existence of spatially distinct subpopulations including subsarcolemmal (SSM) and interfibrillar (IFM). Dysfunction to cardiac SSM has been reported in murine models of type 2 diabetes mellitus; however, subpopulation-based mitochondrial analyses have not been explored in type 2 diabetic human heart. The goal of this study was to determine the impact of type 2 diabetes mellitus on cardiac mitochondrial function in the human patient. Mitochondrial subpopulations from atrial appendages of patients with and without type 2 diabetes were examined. Complex I- and fatty acid-mediated mitochondrial respiration rates were decreased in diabetic SSM compared with nondiabetic (P ≤ 0.05 for both), with no change in IFM. Electron transport chain (ETC) complexes I and IV activities were decreased in diabetic SSM compared with nondiabetic (P ≤ 0.05 for both), with a concomitant decline in their levels (P ≤ 0.05 for both). Regression analyses comparing comorbidities determined that diabetes mellitus was the primary factor accounting for mitochondrial dysfunction. Linear spline models examining correlative risk for mitochondrial dysfunction indicated that patients with diabetes display the same degree of state 3 and electron transport chain complex I dysfunction in SSM regardless of the extent of glycated hemoglobin (HbA1c) and hyperglycemia. Overall, the results suggest that independent of other pathologies, mitochondrial dysfunction is present in cardiac SSM of patients with type 2 diabetes and the degree of dysfunction is consistent regardless of the extent of elevated HbA1c or blood glucose levels.

Falls from ladders: age matters more than height.

BACKGROUND: Falls from ladders account for a significant number of hospital visits. However, the epidemiology, injury pattern, and how age affects such falls are poorly described in the literature. MATERIALS AND METHODS: Patients ≥18 y who suffered falls from ladders over a 5½-y period were identified in our trauma registry. Dividing patients into three age groups (18-45, 46-65, and >66 y), we compared demographic characteristics, clinical data, and outcomes including injury pattern and mortality. The odds ratios (ORs) were calculated with the group 18-45 y as reference; group means were compared with one-way analysis of variance. RESULTS: Of 27,155 trauma patients, 340 (1.3%) had suffered falls from ladders. The average age was 55 y, with a male predominance of 89.3%. Average fall height was 9.8 ft, and mean Injury Severity Score was 10.6. Increasing age was associated with a decrease in the mean fall height (P < 0.001), an increase in the mean Injury Severity Score (P < 0.05), and higher likelihood of admission (>66 y: OR, 5.3; confidence interval [CI], 2.5-11.5). In univariate analysis, patients in the >66-y age group were more likely to sustain traumatic brain injuries (OR, 3.4; CI, 1.5-7.8) and truncal injuries (OR, 3.6; CI, 1.9-7.0) and less likely to sustain hand and/or forearm fractures (OR, 0.3; CI, 0.1-0.9). CONCLUSIONS: Older people are particularly vulnerable after falling from ladders. Although they fell from lower heights, the elderly sustained different and more severe injury patterns. Ladder safety education should be particularly tailored at the elderly.

Viral load and sexually transmitted infection testing among youth with HIV in a southern United States clinic.

Background: As compared to their older peers, youth with HIV (YWH) are less likely to attain viral suppression and have higher rates of sexually transmitted infections (STI). In this exploratory study, we examine the relationship between HIV viral suppression, STI testing, and STI diagnosis among YWH receiving care at a clinic in the southern United States.Methods: Data from 933 clinical visits (2017-2020) were aggregated into singular patient records for YWH aged 10-24 years in Alabama (N = 139). Analyses included univariate generalized linear mixed models performed with the PROC GLIMMIX procedure approximating the marginal likelihood by using Laplace's method.Results: Sample median age was 22 years at the index visit. Most YWH were 20-24 years old (69.1%), male (67.6%), and identified as Black (77%); 58.3% were virally unsuppressed at index visit. YWH who identified as White or of other races had 4.79 times higher odds of being virally suppressed as compared to Black YWH (p < .01); STI testing behavior and STI positive diagnosis were associated with lower odds of being virally suppression.Conclusions: Findings suggest that among YWH, receiving STI testing and having an STI diagnosis is associated with a lack of viral suppression, suggesting that extra efforts may be necessary to support YWH who have an STI to attain suppression. Research is needed to examine individual behaviors, structural forces, and clinic features that could impact STI care engagement, specifically among unsuppressed YWH.

The Characteristics of Youth With Missed HIV Visits in Alabama.

Gaps in knowledge remain related to understanding missed human immunodeficiency virus (HIV) visits and youth with HIV (YWH). This study examined data from an Alabama academic HIV clinic with clients aged 16 to 24 years old and found that non virally suppressed and older YWH were associated with missed visits among YWH.

Re-evaluating the relationship between youth with HIV and BMI in an age of increasing rates of overweight and obese youth.

BACKGROUND: Newer antiretrivirals (ART) have shifted the metabolic experiences of people with HIV (PWH) from those of wasting syndrome to increases in body mass index (BMI). This study sought to examine the relationship between BMI and ART use among youth with HIV (YWH). METHODS: Charts from YWH ages 10-24 with at least two documented BMIs at least 6 months apart between 2017 and 2020 were included (N = 44). Statistical analyses were conducted in SAS 9.4. RESULTS: Clients were predominately African American (66%) males (73%) aged 19-24 years (64%), with men having sex with men (48%) being the most common mode of transmission. YWH on non-integrase inhibitor (INSTI) regimens had greater absolute increases in BMI compared to those on INSTI regimens (p = 0.03). Fourteen percent of clients using INSTI experienced an increase in BMI class from normal to overweight or overweight to obese; no non-INSTI users changed BMI class. Time since diagnosis and BMI change due to weight gain were positively associated (p = 0.03) among behaviorally-acquired YWH. CONCLUSIONS: Increasing BMI and changing BMI classes may be more likely among YWH using INSTI. More longitudinal studies inclusive of diet and exercise profiles are needed to understand the relationship between INSTI and YWH BMI.

Sharpening bounds on principal effects with covariates.

Estimation of treatment effects in randomized studies is often hampered by possible selection bias induced by conditioning on or adjusting for a variable measured post-randomization. One approach to obviate such selection bias is to consider inference about treatment effects within principal strata, that is, principal effects. A challenge with this approach is that without strong assumptions principal effects are not identifiable from the observable data. In settings where such assumptions are dubious, identifiable large sample bounds may be the preferred target of inference. In practice these bounds may be wide and not particularly informative. In this work we consider whether bounds on principal effects can be improved by adjusting for a categorical baseline covariate. Adjusted bounds are considered which are shown to never be wider than the unadjusted bounds. Necessary and sufficient conditions are given for which the adjusted bounds will be sharper (i.e., narrower) than the unadjusted bounds. The methods are illustrated using data from a recent, large study of interventions to prevent mother-to-child transmission of HIV through breastfeeding. Using a baseline covariate indicating low birth weight, the estimated adjusted bounds for the principal effect of interest are 63% narrower than the estimated unadjusted bounds.

Relationship of ethnicity and CD4 Count with glucose metabolism among HIV patients on Highly-Active Antiretroviral Therapy (HAART).

BACKGROUND: HIV patients on HAART are prone to metabolic abnormalities, including insulin resistance, lipodystrophy and diabetes. This study purports to investigate the relationship of ethnicity and CD4+ T cell count attained after stable highly-active antiretroviral treatment (HAART) with glucose metabolism in hyperrtriglyceridemic HIV patients without a history of diabetes. METHODS: Demographic, anthropometric, clinical, endocrinologic, energy expenditure and metabolic measures were obtained in 199 multiethnic, healthy but hypertriglyceridemic HIV-infected patients [46% Hispanic, 17% African-American, 37% Non-Hispanic White (NHW)] on stable HAART without a history of diabetes. The relationship of glucose and insulin responses to ethnicity, CD4 strata (low (<300/cc) or moderate-to-high (≥ 300/cc)), and their interaction was determined. RESULTS: African-Americans had significantly greater impairment of glucose tolerance (P < 0.05) and HbA1c levels (P < .001) than either Hispanics or NHWs. In multivariate models, after adjusting for confounders (age, sex, HIV/HAART duration, smoking, obesity, glucose, insulin and lipids), African-Americans and Hispanics had significantly higher HbA1c and 2-hour glucose levels than NHW's. Demonstrating a significant interaction between ethnicity and CD4 count (P = 0.023), African Americans with CD4 <300/cc and Hispanics with CD4 ≥300/cc had the most impaired glucose response following oral glucose challenge. CONCLUSIONS: Among hypertriglyceridemic HIV patients on HAART, African-Americans and Hispanics are at increased risk of developing diabetes. Ethnicity also interacts with CD4+ T cell count attained on stable HAART to affect post-challenge glycemic response.

Geospatial Analysis of Time to Human Immunodeficiency Virus (HIV) Diagnosis and Adult HIV Testing Coverage Highlights Areas for Intervention in the US Southeast.

Background: In the United States (US), 44% of people with human immunodeficiency virus (PWH) live in the Southeastern census region; many PWH remain undiagnosed. Novel strategies to inform testing outreach in rural states with dispersed HIV epidemics are needed. Methods: Alabama state public health HIV testing surveillance data from 2013 to 2017 were used to estimate time from infection to HIV diagnosis using CD4 T-cell depletion modeling, mapped to county. Diagnostic HIV tests performed during 2013-2021 by commercial testing entities were used to estimate HIV tests per 100 000 adults (aged 15-65 years), mapped to client ZIP Code Tabulation Area (ZCTA). We then defined testing "cold spots": those with <10% adults tested plus either (1) within or bordering 1 of the 13 counties with HIV prevalence >400 cases per 100 000 population or (2) within a county with average time to diagnosis greater than the state average to inform testing outreach. Results: Time to HIV diagnosis was a median of 3.7 (interquartile range [IQR], 0-9.2) years across Alabama, with a range of 0.06-12.25 years. Approximately 63% of counties (n = 42) had a longer time to diagnosis compared to national US estimates. Six hundred forty-three ZCTAs tested 17.3% (IQR, 10.3%-25.0%) of the adult population from 2013 to 2017. To prioritize areas for testing outreach, we generated maps to describe 47 areas of HIV-testing cold spots at the ZCTA level. Conclusions: Combining public health surveillance with commercial testing data provides a more nuanced understanding of HIV testing gaps in a state with a rural HIV epidemic and identifies areas to prioritize for testing outreach.