Quyen M. Ngo

Background: Sexually transmitted infections (STIs) are at a record high in the United States and are a significant health problem for childbearing women. Rates of substance use, particularly cannabis and opioid use, have increased in recent years and are linked to negative health consequences for pregnant women and their infants. Addressing these health concerns together during this vulnerable time is a priority. Objective: This study aims to test whether the Health Check-Up for Expectant Moms (HCEM), an innovative, theory-driven, technology-delivered, and fully automated brief intervention, reduced condomless sex and STI risk, alcohol, or drug use, compared to a control condition in pregnant women seeking prenatal care. Methods: We recruited a sample of 176 pregnant women (all were at risk for alcohol or drug use or STIs during pregnancy) from clinics and using social media campaigns (Facebook and Instagram) in the state of Michigan and randomized them to the motivational interviewing-consistent HCEM intervention or to an attention-, time-, and information-matched control condition delivered using the same technology platform. We followed these women at 2 and 6 months after the initial intervention visit. Primary outcomes included self-report assessment of alcohol, drug, or cannabis use and unprotected sexual occasions during pregnancy. Results: A total of 88 women were randomized to the intervention, and 88 to the control condition. Cannabis use was the most prevalent substance reported during pregnancy; a total of 35.2% (62/176) reported recent use (within the last 90 days) at baseline, with 10.2% (18/176) reporting use in the month prior to baseline. There were significant reductions in alcohol and cannabis use over time during pregnancy (at 2 and 6 months compared to baseline) in both HCEM and control groups; however, these reductions were not significantly different between conditions (time-by-arm interaction), and most were sustained from spontaneous reductions reported in the month before study enrollment. Moreover, there were no statistically significant differences in the change pattern of condomless sex across the groups at either follow-up. Conclusions: There are many potential benefits of a technology-delivered approach to support the behavioral health of pregnant women in a private and convenient way. Our sample was largely low-risk, and as such, an intervention effect may have been impossible to observe given substantial self-change. Future trials are needed to examine efficacy in other samples of pregnant women with a higher risk of current alcohol or drug use. Trial Registration: ClinicalTrials.gov NCT03826342; https://clinicaltrials.gov/study/NCT03826342 International Registered Report Identifier (IRRID): RR2-10.2196/30367

BACKGROUND Sexually transmitted infections (STIs) are at a record high in the United States and are a significant health problem for childbearing women. Rates of substance use, particularly cannabis and opioid use, have increased in recent years and are linked to negative health consequences for pregnant women and their infants. Addressing these health concerns together during this vulnerable time is a priority. OBJECTIVE This study aims to test whether the Health Check-Up for Expectant Moms (HCEM), an innovative, theory-driven, technology-delivered, and fully automated brief intervention, reduced condomless sex and STI risk, alcohol, or drug use, compared to a control condition in pregnant women seeking prenatal care. METHODS We recruited a sample of 176 pregnant women (all were at risk for alcohol or drug use or STIs during pregnancy) from clinics and using social media campaigns (Facebook and Instagram) in the state of Michigan and randomized them to the motivational interviewing-consistent HCEM intervention or to an attention-, time-, and information-matched control condition delivered using the same technology platform. We followed these women at 2 and 6 months after the initial intervention visit. Primary outcomes included self-report assessment of alcohol, drug, or cannabis use and unprotected sexual occasions during pregnancy. RESULTS A total of 88 women were randomized to the intervention, and 88 to the control condition. Cannabis use was the most prevalent substance reported during pregnancy; a total of 35.2% (62/176) reported recent use (within the last 90 days) at baseline, with 10.2% (18/176) reporting use in the month prior to baseline. There were significant reductions in alcohol and cannabis use over time during pregnancy (at 2 and 6 months compared to baseline) in both HCEM and control groups; however, these reductions were not significantly different between conditions (time-by-arm interaction), and most were sustained from spontaneous reductions reported in the month before study enrollment. Moreover, there were no statistically significant differences in the change pattern of condomless sex across the groups at either follow-up. CONCLUSIONS There are many potential benefits of a technology-delivered approach to support the behavioral health of pregnant women in a private and convenient way. Our sample was largely low-risk, and as such, an intervention effect may have been impossible to observe given substantial self-change. Future trials are needed to examine efficacy in other samples of pregnant women with a higher risk of current alcohol or drug use. CLINICALTRIAL ClinicalTrials.gov NCT03826342; https://clinicaltrials.gov/study/NCT03826342 INTERNATIONAL REGISTERED REPORT RR2-10.2196/30367

Fairness of machine learning models in healthcare has drawn increasing attention from clinicians, researchers, and even at the highest level of government. On the other hand, the importance of developing and deploying interpretable or explainable models has been demonstrated, and is essential to increasing the trustworthiness and likelihood of adoption of these models. The objective of this study was to develop and implement a framework for addressing both these issues - fairness and explainability. We propose an explainable fairness framework, first developing a model with optimized performance, and then using an in-processing approach to mitigate model biases relative to the sensitive attributes of race and sex. We then explore and visualize explanations of the model changes that lead to the fairness enhancement process through exploring the changes in importance of features. Our resulting-fairness enhanced models retain high sensitivity with improved fairness and explanations of the fairness-enhancement that may provide helpful insights for healthcare providers to guide clinical decision-making and resource allocation.

Background: The use of virtual treatment services increased dramatically during the COVID-19 pandemic. Unfortunately, large-scale research on virtual treatment for substance use disorder (SUD), including factors that may influence outcomes, has not advanced with the rapidly changing landscape. Objective: This study aims to evaluate the link between clinician-level factors and patient outcomes in populations receiving virtual and in-person intensive outpatient services. Methods: Data came from patients (n=1410) treated in a virtual intensive outpatient program (VIOP) and an in-person intensive outpatient program (IOP), who were discharged between January 2020 and March 2021 from a national treatment organization. Patient data were nested by treatment providers (n=58) examining associations with no-shows and discharge with staff approval. Empathy, comfort with technology, perceived stress, resistance to change, and demographic covariates were examined at the clinician level. Results: The VIOP (β=–5.71; P=.03) and the personal distress subscale measure (β=–6.31; P=.003) were negatively associated with the percentage of no-shows. The VIOP was positively associated with discharges with staff approval (odds ratio [OR] 2.38, 95% CI 1.50-3.76). Clinician scores on perspective taking (β=–9.22; P=.02), personal distress (β=–9.44; P=.02), and male clinician gender (β=–6.43; P=.04) were negatively associated with in-person no-shows. Patient load was positively associated with discharge with staff approval (OR 1.04, 95% CI 1.02-1.06). Conclusions: Overall, patients in the VIOP had fewer no-shows and a higher rate of successful discharge. Few clinician-level characteristics were significantly associated with patient outcomes. Further research is necessary to understand the relationships among factors such as clinician gender, patient load, personal distress, and patient retention.

BACKGROUND The use of virtual treatment services increased dramatically during the COVID-19 pandemic. Unfortunately, large-scale research on virtual treatment for substance use disorder (SUD), including factors that may influence outcomes, has not advanced with the rapidly changing landscape. OBJECTIVE This study aims to evaluate the link between clinician-level factors and patient outcomes in populations receiving virtual and in-person intensive outpatient services. METHODS Data came from patients (n=1410) treated in a virtual intensive outpatient program (VIOP) and an in-person intensive outpatient program (IOP), who were discharged between January 2020 and March 2021 from a national treatment organization. Patient data were nested by treatment providers (n=58) examining associations with no-shows and discharge with staff approval. Empathy, comfort with technology, perceived stress, resistance to change, and demographic covariates were examined at the clinician level. RESULTS The VIOP (β=–5.71; P=.03) and the personal distress subscale measure (β=–6.31; P=.003) were negatively associated with the percentage of no-shows. The VIOP was positively associated with discharges with staff approval (odds ratio [OR] 2.38, 95% CI 1.50-3.76). Clinician scores on perspective taking (β=–9.22; P=.02), personal distress (β=–9.44; P=.02), and male clinician gender (β=–6.43; P=.04) were negatively associated with in-person no-shows. Patient load was positively associated with discharge with staff approval (OR 1.04, 95% CI 1.02-1.06). CONCLUSIONS Overall, patients in the VIOP had fewer no-shows and a higher rate of successful discharge. Few clinician-level characteristics were significantly associated with patient outcomes. Further research is necessary to understand the relationships among factors such as clinician gender, patient load, personal distress, and patient retention. CLINICALTRIAL

Objective: Although previous studies have discussed the promise of deep brain stimulation (DBS) as a possible treatment for substance use disorders (SUDs) and collected researcher perspectives on possible ethical issues surrounding it, none have consulted people with SUDs themselves. We addressed this gap by interviewing people with SUDs. Methods: Participants viewed a short video introducing DBS, followed by a 1.5-hour semistructured interview on their experiences with SUDs and their perspective on DBS as a possible treatment option. Interviews were analyzed by multiple coders who iteratively identified salient themes. Results: We interviewed 20 people in 12-step–based, inpatient treatment programs (10 [50%] White/Caucasian, 7 Black/African American [35%], 2 Asian [10%], 1 Hispanic/Latino [5%], and 1 [5%] Alaska Native/American Indian; 9 women [45%], 11 men [55%]). Interviewees described a variety of barriers they currently faced through the course of their disease that mirrored barriers often associated with DBS (stigma, invasiveness, maintenance burdens, privacy risks) and thus made them more open to the possibility of DBS as a future treatment option. Conclusions: Individuals with SUDs gave relatively less weight to surgical risks and clinical burdens associated with DBS than previous surveys of provider attitudes anticipated. These differences derived largely from their experiences living with an often-fatal disease and encountering limitations of current treatment options. These findings support the study of DBS as a treatment option for SUDs, with extensive input from people with SUDs and advocates.

Background and Aims Deep brain stimulation (DBS) has shown promise as a treatment option for substance use disorders (SUDs), but may have unique ethical risks due to stigma and other factors. Previous studies have elicited researcher and clinician opinions on those risks, but none have studied perspectives of people living with SUDs. Methods Participants were recruited through a national inpatient treatment network employing purposive sampling for representation of minoritized groups and diversity of substances. Participants viewed a short video introducing DBS, followed by a 1.5 hour semi-structured interview on their experiences with SUDs and their perspectives on DBS as a future treatment option. Interviews were analyzed by multiple coders who iteratively identified salient themes. Results We interviewed 20 people in 12-step based, inpatient treatment programs (10 [50%] white/Caucasian, 7 Black/African American [35%], 2 Asian [10%], 1 Hispanic/Latino [5%], and 1 [5%] Alaska Native/American Indian; 11 [45%] women). Interviewees described a variety of barriers they currently faced through the course of their disease that mirrored barriers often associated with DBS (stigma, invasiveness, maintenance burdens, privacy risks). The majority of respondents expressed interest in DBS as a future treatment option, emphasizing the importance of exploring novel treatment options and keeping individual treatment goals as key guiding considerations. Conclusions The perspectives of people with lived experience of SUDs contrasted with previous surveys of provider attitudes on DBS for SUDs. Individuals with SUDs gave relatively less weight to surgical risks and clinical burdens associated with DBS. These differences derived largely from their experiences living with an often-fatal disease, encountering limitations of current treatment options, and their familiarity with 12-step treatment paradigms that prioritize having multiple therapeutic “tools.” These findings support the study of DBS as a treatment option for SUDs, with extensive input from people with SUDs and advocates.

Background: The COVID-19 pandemic has profoundly transformed substance use disorder (SUD) treatment in the United States, with many web-based treatment services being used for this purpose. However, little is known about the long-term treatment effectiveness of SUD interventions delivered through digital technologies compared with in-person treatment, and even less is known about how patients, clinicians, and clinical characteristics may predict treatment outcomes. Objective: This study aims to analyze baseline differences in patient demographics and clinical characteristics across traditional and telehealth settings in a sample of participants (N=3642) who received intensive outpatient program (IOP) substance use treatment from January 2020 to March 2021. Methods: The virtual IOP (VIOP) study is a prospective longitudinal cohort design that follows adult (aged ≥18 years) patients who were discharged from IOP care for alcohol and substance use–related treatment at a large national SUD treatment provider between January 2020 and March 2021. Data were collected at baseline and up to 1 year after discharge from both in-person and VIOP services through phone- and web-based surveys to assess recent substance use and general functioning across several domains. Results: Initial baseline descriptive data were collected on patient demographics and clinical inventories. No differences in IOP setting were detected by race (χ22=0.1; P=.96), ethnicity (χ22=0.8; P=.66), employment status (χ22=2.5; P=.29), education level (χ24=7.9; P=.10), or whether participants presented with multiple SUDs (χ28=11.4; P=.18). Significant differences emerged for biological sex (χ22=8.5; P=.05), age (χ26=26.8; P<.001), marital status (χ24=20.5; P<.001), length of stay (F2,3639=148.67; P<.001), and discharge against staff advice (χ22=10.6; P<.01). More differences emerged by developmental stage, with emerging adults more likely to be women (χ23=40.5; P<.001), non-White (χ23=15.8; P<.001), have multiple SUDs (χ23=453.6; P<.001), have longer lengths of stay (F3,3638=13.51; P<.001), and more likely to be discharged against staff advice (χ23=13.3; P<.01). Conclusions: The findings aim to deepen our understanding of SUD treatment efficacy across traditional and telehealth settings and its associated correlates and predictors of patient-centered outcomes. The results of this study will inform the effective development of data-driven benchmarks and protocols for routine outcome data practices in treatment settings.