Search

Motivational interviewing (MI) is a patient-centered counseling technique used for behavior change. MI has been evaluated in hundreds of clinical trials across contexts, focusing on mental health and addictions. Despite promise as a technique in primary care, questions still remain around how MI might work, how best to measure clinician competence, and the extent to which this technique is transferable to different cultures. This dissertation seeks to understand how MI can be used in clinical settings. The first is a systematic review of MI for chronic disease. The second study presents a tool to measure clinician competence in Motivational interviewing and validates this against the current gold standard. Finally, the third study is a qualitative analysis of Singaporean clinicians and their views on incorporating MI into clinical practice. The meta-analysis shows that MI seems to be a useful tool in the prevention and management of chronic diseases. With a predominantly female sample, this technique indicates statistically significant effect sizes ranging between d=0.11 for physical activity, and d=1.01 for blood pressure. Subgroup analyses indicated stronger effects for US vs. international studies, in particular within US communities where there are high proportions of minority groups. In the second study, the validation of the OnePass tool for measuring MI competence, interrater reliability was good to excellent (between 0.44 <kappa < 0.84) for all compulsory items, and a strong correlation between overall scores on the MITI and on the OnePass, indicating validity. The final study of 37 Singaporean clinicians showed that Motivational Interviewing is appropriate in real world contexts where multiple languages may be spoken during the course of an average day. Despite minimal institutional support, the clinicians in the sample all described using some aspects of MI with their patients, and perceived benefits to both their own practice and the health outcomes of their patients. Overall conclusions from this dissertation indicate that MI is an appropriate intervention for behavior change in a range of clinical contexts and settings. Future research should relate clinician competence to patient outcomes, and start to examine the nuances of culture, values and language on the use of MI.

BACKGROUND: Adolescent hazardous alcohol use is prevalent and has serious short- and long-term consequences. The trial 'Our Choice' examines efficacy, feasibility and acceptability of prevention interventions targeting school, parent, and student levels at Danish high schools. We hypothesize that students in a structural intervention (school and parent levels) reduce hazardous alcohol use and related health behaviors compared to students in an assessment only control group 12 months post baseline; and that adding group-based Motivational Interviewing (group MI) yields further improvements. The study examines the efficacy of interventions targeting multiple levels with the aim of providing novel insights into prevention of adolescent hazardous alcohol use and related health outcomes. METHOD: The study employs a parallel group cluster randomized controlled trial design with three conditions: (1) structural condition targeting school and parent levels, (2) structural condition combined with group MI which also targets the student level, and (3) assessment-only control condition. A participatory approach is used to adapt and develop interventions. Sixteen high schools in Denmark and about N = 3100 first-year students (15-18 years) enrolled in high school in August 2023 will be recruited. Data will be collected via online questionnaires pre-interventions (baseline), 2, 6, 9 and 12 month post baseline and analyzed with generalized linear mixed models. The primary outcome is past month high intensity drinking; secondary outcomes are alcohol use, alcohol-related consequences, well-being, tobacco, and illegal substance use. Feasibility and acceptability will be assessed via surveys (students) and interviews (high school staff) to inform future implementation. DISCUSSION: 'Our Choice' is the first trial to compare the efficacy of a structural intervention targeting school- and parent levels to an intervention targeting these levels and the student level via group MI - on hazardous drinking and related health outcomes among students. Preventing and reducing hazardous alcohol use during adolescence is crucial due to the short- and long-term negative consequences. The tested interventions can be implemented at low cost. The study has significant implications for adolescent health and well-being and has potential to inform evidence-based decisions on alcohol prevention policy, education, and health professions. TRIAL REGISTRATION NUMBER: The trial was retrospectively registered at ClinicalTrials.gov on August 24th, 2023. TRIAL REGISTRATION NUMBER: ID NCT06018389.

Because adolescents are unlikely to seek, receive, or complete treatment for alcohol and/or cannabis misuse, it is important to enhance the lasting impact of clinical contacts when they do occur. Adolescents (N = 506; 72.5% Hispanic) were randomized to motivational interviewing (MI) versus alcohol and cannabis education (ACE). Latent growth models estimated change over time. Significant reductions in alcohol use were observed, with slightly greater reductions by 12-month follow-up for MI. Both interventions significantly reduced cannabis use, with no treatment group differences. When outcomes were examined comparing Hispanic to non-Hispanic participants, there were no significant differences in intervention efficacy by group. MI's inherently client-centered and culturally adaptive approach may contribute to its equitable degree of behavior change for youth across race/ethnic backgrounds.

BACKGROUND: Primary care settings like federally qualified health centers (FQHC) are optimal locations to identify individuals with substance use disorders (SUD) and link them to SUD treatment, yet successful linkage has proven difficult. Recovery management checkups for primary care (RMC-PC) is a promising method for increasing linkage to care, engagement in treatment, and reducing substance use. METHODS: Participants (n = 266) who received screening, brief intervention, and referral to treatment (SBIRT) at four FQHC sites and needed SUD treatment were randomized to receive SBIRT only or SBIRT+RMC-PC. All participants received SBIRT prior to randomization as part of usual care while those in the experimental group also received quarterly checkups. All participants completed research interviews at enrollment and 3, 6, 9, and 12 months post-enrollment. The primary outcome was whether participants received any days of SUD treatment. Key secondary outcomes were days of SUD treatment (total and by SUD level of care), days of alcohol or drug abstinence, and a reduction in days of specific substance use, all based on self-report. RESULTS: Relative to participants receiving SBIRT only, participants assigned to SBIRT+RMC-PC were significantly more likely to have received any SUD treatment over 12 months (adjusted odds ratio [AOR] = 3.85) and more days of SUD treatment over 12 months (Cohen's effect size d = +0.41). The SBIRT+RMC-PC group also reported significantly more days of abstinence over 12 months (d = +0.30), fewer days of alcohol use (d = -0.20) and cannabis use (d = -0.20), and lower combined substance use frequency (d = -0.25). Days of treatment were found to positively mediate the direct effect of SBIRT+RMC-PC on days of abstinence. CONCLUSION: This study provides further evidence of the effectiveness of the "referral to treatment" component of SBIRT when combined with RMC for patients in primary care settings, including those with drug use problems. Moreover, results demonstrate the value of repeated checkups on longer-term treatment and substance use outcomes.

Recent years have witnessed remarkable progress made in large language models (LLMs). Such advancements, while garnering significant attention, have concurrently elicited various concerns. The potential of these models is undeniably vast; however, they may yield texts that are imprecise, misleading, or even detrimental. Consequently, it becomes paramount to employ alignment techniques to ensure these models to exhibit behaviors consistent with human values. This survey endeavors to furnish an extensive exploration of alignment methodologies designed for LLMs, in conjunction with the extant capability research in this domain. Adopting the lens of AI alignment, we categorize the prevailing methods and emergent proposals for the alignment of LLMs into outer and inner alignment. We also probe into salient issues including the models' interpretability, and potential vulnerabilities to adversarial attacks. To assess LLM alignment, we present a wide variety of benchmarks and evaluation methodologies. After discussing the state of alignment research for LLMs, we finally cast a vision toward the future, contemplating the promising avenues of research that lie ahead. Our aspiration for this survey extends beyond merely spurring research interests in this realm. We also envision bridging the gap between the AI alignment research community and the researchers engrossed in the capability exploration of LLMs for both capable and safe LLMs.

BACKGROUND: Studies have found that reductions in World Health Organization (WHO) drinking risk levels may be a stable outcome of treatment for alcohol use disorder (AUD) and associated with functional improvements. The aim of this study was to investigate whether posttreatment reductions in WHO drinking risk levels are stable over time among older adults and associated with a decrease in consequences of drinking and AUD symptoms and improved quality of life. METHODS: Participants. Individuals 60+ years old, suffering from DSM-5 AUD (n = 693), and seeking outpatient treatment. MEASUREMENTS: WHO drinking risk levels, prior to treatment and at all follow-up points up to 1 year after treatment start, were assessed with Form 90. Outcomes at follow-up included consequences of drinking (Drinker Inventory of Consequences), quality of life (WHOQOL-BREF), and DSM-5 AUD symptoms (Mini International Neuropsychiatric Interview). Logistic regression and linear mixed models were used to examine the probability of maintaining risk-level reductions at follow-up and the association between risk-level reductions and outcomes, respectively. RESULTS: Reductions in risk levels were maintained over time (at least 1 level: OR 5.39, 95% CI 3.43, 8.47; at least 2 levels: OR 9.30, 95% CI 6.14, 14.07). Reductions were associated with reduced consequences of drinking and number of AUD symptoms, and minor, but statistically significant, improvements in quality of life. CONCLUSIONS: Maintaining reductions in WHO risk levels appears achievable for older adults seeking treatment for AUD. The small reduction of AUD symptoms and improvement of quality of life indicates that these reductions may not be adequate as the only treatment goal.

BACKGROUND: Smoking, poor diet, and physical inactivity account for as much as 60% of cancer risk. Latinos experience profound disparities in health behaviors, as well as the cancers associated with them. Currently, there is a dearth of controlled trials addressing these health behaviors among Latinos. Further, to the best of our knowledge, no studies address all three behaviors simultaneously, are culturally sensitive, and are guided by formative work with the target population. Latinos represent 14% of the U.S. population and are the fastest growing minority group in the country. Efforts to intervene on these important lifestyle factors among Latinos may accelerate the elimination of cancer-related health disparities. METHODS/DESIGN: The proposed study will evaluate the efficacy of an evidence-based and theoretically-driven Motivation And Problem Solving (MAPS) intervention, adapted and culturally-tailored for reducing cancer risk related to smoking, poor diet, and physical inactivity among high-risk Mexican-origin smokers who are overweight/obese (n = 400). Participants will be randomly assigned to one of two groups: Health Education (HE) or MAPS (HE + up to 18 MAPS counseling calls over 18 months). Primary outcomes are smoking status, servings of fruits and vegetables, and both self-reported and objectively measured physical activity. Outcome assessments will occur at baseline, 6 months, 12 months, and 18 months. DISCUSSION: The current study will contribute to a very limited evidence base on multiple risk factor intervention studies on Mexican-origin individuals and has the potential to inform both future research and practice related to reducing cancer risk disparities. An effective program targeting multiple cancer risk behaviors modeled after chronic care programs has the potential to make a large public health impact because of the dearth of evidence-based interventions for Latinos and the extended period of support that is provided in such a program. TRIAL REGISTRATION: National Institutes of Health Clinical Trials Registry # NCT01504919.