In the current paper, we review existing tools for solving variable selection problems in psychology. Modern regularization methods such as lasso regression have recently been introduced in the field and are incorporated into popular methodologies, such as network analysis. However, several recognized limitations of lasso regularization may limit its suitability for psychological research. In this paper, we compare the properties of lasso approaches used for variable selection to Bayesian variable selection approaches. In particular we highlight advantages of stochastic search variable selection (SSVS), that make it well suited for variable selection applications in psychology. We demonstrate these advantages and contrast SSVS with lasso type penalization in an application to predict depression symptoms in a large sample and an accompanying simulation study. We investigate the effects of sample size, effect size, and patterns of correlation among predictors on rates of correct and false inclusion and bias in the estimates. SSVS as investigated here is reasonably computationally efficient and powerful to detect moderate effects in small sample sizes (or small effects in moderate sample sizes), while protecting against false inclusion and without over-penalizing true effects. We recommend SSVS as a flexible framework that is well-suited for the field, discuss limitations, and suggest directions for future development.

Engagement with HIV-prevention services varies among sexual minority men (SMM). 183 HIV-negative SMM completed a baseline assessment including sociodemographic, psychosocial, identity and stigma, and sexuality measures, as well as HIV-prevention behaviors: (1) date of most recent HIV test, (2) whether they discussed HIV or (3) sexual behavior with their provider, and (4) disclosure to provider about engaging in condomless anal sex (CAS). Factor analysis of these four items yielded an HIV-prevention engagement factor score. Stochastic search variable selection (SSVS) followed by multiple linear regression identified variables associated with HIV-prevention engagement. SSVS identified three variables for inclusion in a multiple linear regression model. Not disclosing sexual orientation to one's provider (p < 0.001), discomfort discussing sex with provider (p < 0.001) and lower education (p = 0.007) were associated with less HIV-prevention engagement. Findings suggest the importance of training providers in culturally competent care to mitigate observed barriers.

Objective

Neuropsychological instruments are often developed in English and translated to other languages to facilitate the clinical evaluation of diverse populations or to utilize in research environments. However, the psychometric equivalence of these assessments across language must be demonstrated before populations can validly be compared.

Method

To test this equivalence, we applied measurement invariance procedures to a subsample (N = 1,708) of the Hispanic Community Health Survey/Study of Latinos (HCHS/SOL) across English and Spanish versions of a neurocognitive battery. Using cardinality matching, 854 English-speaking and 854 Spanish-speaking subsamples were matched on age, education, sex, immigration status (U.S. born, including territories, or foreign-born), and Hispanic/Latino heritage background. Neurocognitive measures included the Six-Item Screener (SIS), Brief-Spanish English Verbal Learning Test (B-SEVLT), Word Fluency (WF), and Digit Symbol Substitution (DSS). Confirmatory factor analysis was utilized to test item-level invariance of the SIS, B-SEVLT, and WF, as well as factor-level invariance of a higher-order neurocognitive functioning latent variable.

Results

One item of both the SIS and WF were more difficult in Spanish than English, as was the DSS test. After accounting for partial invariance, Spanish-speakers performed worse on each of the subtests and the second-order neurocognitive functioning latent variable.

Conclusions

We found some evidence of bias at both item and factor levels, contributing to the poorer neurocognitive performance of Spanish test-takers. While these results explain the underperformance of Spanish-speakers to some extent, more work is needed to determine whether such bias is reflective of true cognitive differences or additional variables unaccounted for in this study. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Introduction

Little is known about onward HIV transmissions from people living with HIV (PLWH) in care. Antiretroviral therapy (ART) has increased in potency, and treatment as prevention (TasP) is an important component of ending the epidemic. Syndemic theory has informed modelling of HIV risk but has yet to inform modelling of HIV transmissions.

Methods

Data were from 61,198 primary HIV care visits for 14,261 PLWH receiving care through the Centers for AIDS Research (CFAR) Network of Integrated Clinical Systems (CNICS) at seven United States (U.S.) sites from 2007 to 2017. Patient-reported outcomes and measures (PROs) of syndemic conditions - depressive symptoms, anxiety symptoms, drug use (opiates, amphetamines, crack/cocaine) and alcohol use - were collected approximately four to six months apart along with sexual behaviours (mean = 4.3 observations). Counts of syndemic conditions, HIV sexual risk group and time in care were modelled to predict estimated HIV transmissions resulting from sexual behaviour and viral suppression status (HIV RNA < 400/mL) using hierarchical linear modelling.

Results

Patients averaged 0.38 estimated HIV transmissions/100 patients/year for all visits with syndemic conditions measured (down from 0.83, first visit). The final multivariate model showed that per 100 patients, each care visit predicted 0.05 fewer estimated transmissions annually (95% confidence interval (CI): 0.03 to 0.06; p < 0.0005). Cisgender women, cisgender heterosexual men and cisgender men of undisclosed sexual orientation had, respectively, 0.47 (95% CI: 0.35 to 0.59; p < 0.0005), 0.34 (95% CI: 0.20 to 0.49; p < 0.0005) and 0.22 (95% CI: 0.09 to 0.35; p < 0.005) fewer estimated HIV transmissions/100 patients/year than cisgender men who have sex with men (MSM). Each within-patient syndemic condition predicted 0.18 estimated transmissions/100 patients/year (95% CI: 0.12 to 0.24; p < 0.0005). Each between-syndemic condition predicted 0.23 estimated HIV transmissions/100 patients/year (95% CI: 0.17 to 0.28; p < 0.0005).

Conclusions

Estimated HIV transmissions among PLWH receiving care in well-resourced U.S. clinical settings varied by HIV sexual risk group and decreased with time in care, highlighting the importance of TasP efforts. Syndemic conditions remained a significant predictor of estimated HIV transmissions notwithstanding the effects of HIV sexual risk group and time in care.

Psychosocial syndemic conditions have received more attention regarding their deleterious effects on HIV acquisition risk than for their potential impact on HIV treatment and viral suppression. To examine syndemic conditions' impact on the HIV care continuum, we analyzed data collected from people living with HIV (N = 14,261) receiving care through The Centers for AIDS Research Network of Integrated Clinical Systems at seven sites from 2007 to 2017 who provided patient-reported outcomes ∼4-6 months apart. Syndemic condition count (depression, anxiety, substance use, and hazardous drinking), sexual risk group, and time in care were modeled to predict antiretroviral therapy (ART) adherence and viral suppression (HIV RNA <400 copies/mL) using multilevel logistic regression. Comparing patients with each other, odds of ART adherence were 61.6% lower per between-patient syndemic condition [adjusted odds ratio (AOR) = 0.384; 95% confidence interval (CI), 0.362-0.408]; comparing patients with themselves, odds of ART adherence were 36.4% lower per within-patient syndemic condition (AOR = 0.636 95% CI, 0.606-0.667). Odds of viral suppression were 29.3% lower per between-patient syndemic condition (AOR = 0.707; 95% CI, 0.644-0.778) and 27.7% lower per within-patient syndemic condition (AOR = 0.723; 95% CI, 0.671-0.780). Controlling for the effects of adherence (AOR = 5.522; 95% CI, 4.67-6.53), each additional clinic visit was associated with 1.296 times higher odds of viral suppression (AOR = 1.296; 95% CI, 1.22-1.38), but syndemic conditions were not significant. Deploying effective interventions within clinics to identify and treat syndemic conditions and bolster ART adherence and continued engagement in care can help control the HIV epidemic, even within academic medical settings in the era of increasingly potent ART.