Computer-delivered Intervention Research Articles

Does self-affirmation augment the effects of a mandated personalized feedback intervention? A randomized controlled trial with heavy drinking college students.

Theory and evidence indicate that affirming the value of the self before exposure to a threatening message fosters more open-minded appraisal of message content. We predicted that college students mandated to a computer-delivered personalized feedback intervention (PFI) and who engaged in a self-affirmation (SA) exercise would demonstrate reduce drinking and consequences relative to those who received an attention control. Participants were 484 undergraduates (age 18-24, 56% male, 78% White) mandated to participate in an alcohol intervention following a first-time alcohol policy violation. After a baseline assessment, each was randomized to SA (n = 256) or attention control (n = 227) prior to a computer-delivered PFI intervention. Posttest measures included an affirmation manipulation check; primary outcomes (past month weekly quantity, peak drinks, binge frequency, consequences) were assessed at 1-, 3-, 6-, 9-, and 12-month follow-ups. We used latent growth curve modeling to test study hypotheses. The SA exercise increased positive self-evaluation at posttest (p < .001). Overall reductions in drinking and consequences were observed at early follow-up (all p < .05), but the SA manipulation was not associated with intercept or slope factor differences in the models. Engaging in assessments during COVID restrictions was generally associated with reduced drinking and consequences. The SA exercise did not differentially affect trajectories of alcohol use and consequences, despite evidence that the exercise had the predicted effect on participant's self-evaluations. The lack of SA effects could be attributed to a nonthreatening PFI intervention that is generally accepted among mandated students. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Students’ Perceptions on the Effectiveness of a Computer-Delivered Alcohol Intervention

Objective: Social adjustment, including alcohol use, directly affects the success of college students. Due to an increased reliance on computer-delivered alcohol interventions (CDIs) a need has emerged to further investigate alcohol use and web-based interventions. Methods: In-depth focus group interviews were conducted with 51 undergraduate students to elicit information from students on the shared experience of participating in a CDI. Results: Participants identified the influence of gender, culture, parents, and family on alcohol use behavior. A difference in personal factors, previous exposure, and experiences can affect the attitudes, behaviors, and outcomes of a CDI. Conclusion: Multiple approaches geared towards a wide variety of students from different backgrounds and environments are needed to be truly successful in preventing alcohol misuse.

Bridging the gap between genetic epidemiological research and prevention: A randomized control trial of a novel personalized feedback program for alcohol and cannabis use

BackgroundRisky substance use among college students is widespread and associated with numerous negative consequences. We created an online Personalized Feedback Program (PFP) for college students that targets genetically influenced risk pathways for substance use and provides feedback on four risk domains (Sensation Seeking, Impulsivity, Extraversion, and Neuroticism) along with individualized recommendations and campus resources. MethodsA pilot randomized controlled trial was conducted to evaluate the effects of the PFP on alcohol and cannabis use. First-year college students were randomized to one of four groups: (1) control, (2) PFP, (3) computer-delivered brief motivational intervention (BMI), and (4) combined group that included both the PFP and BMI (PFP+BMI). Students completed a baseline survey (n=251) that assessed alcohol and cannabis use and program satisfaction. Two follow-up surveys were administered at 30-days and 3-months post-intervention to evaluate longitudinal effects on substance use. ResultsParticipants reported high satisfaction with the PFP. There were no significant effects of intervention group on alcohol use at the follow-up timepoints, though trends were in the expected direction with participants in the PFP group showing decreased odds of alcohol use. There were significant reductions in cannabis use in the PFP group as compared to other groups. ConclusionsThe PFP was met with high satisfaction and had a positive impact on reducing cannabis use. With cannabis use at a historic high among college-aged adults, further research evaluating the effects of the PFP is warranted.

Single-Session Computer-Delivered Intervention for Parental Accommodation and Overcontrolling Behaviors: A Pilot Study

ABSTRACT Despite repeated calls regarding the need to target parental accommodation and overcontrol in child anxiety treatment, brief protocols specifically targeting these behaviors in parents of clinically anxious youth are currently limited in number. The present investigation piloted and refined a theoretically informed, internet-delivered, single-session intervention (SSI) specifically aimed at reducing parental accommodation and overcontrolling behaviors. Socioeconomically and racially diverse parents of clinically anxious children (N = 12; M age =10.08 years, SD = 1.74, range = 7–14 years) were recruited in two phases. In Phase One, three online focus groups with two parents each (n = 6) received the intervention and provided qualitative and quantitative feedback on the feasibility and acceptability of the intervention program. Parental feedback from the focus groups was used to revise the program. In Phase Two, the revised program was administered to six new parents to examine the program’s acceptability and feasibility for increasing psychoeducational knowledge regarding child anxiety, parental accommodation, and overcontrol. Across both phases, the intervention was well-received by parents in terms of relevance, acceptability, and feasibility. Parents demonstrated significant increases in knowledge of parental accommodation, overcontrol, and child anxiety from pre- to post-intervention. Qualitative data indicated that parents benefited from the unique online format as well as the content. Findings support the feasibility and acceptability of an accessible and time-efficient SSI designed to target problematic parenting behaviors among parents of clinically anxious children.

Computer-delivered intervention for individuals with obesity and elevated anxiety sensitivity: feasibility, acceptability, and initial test

ABSTRACT Obesity is one the fastest growing public health problems related to numerous chronic diseases and frequently co-occurs with negative mood states. One promising approach to address the heterogeneity of affective vulnerabilities and obesity is to focus on processes (e.g. anxiety sensitivity [AS]) that underpin both conditions. Thus, this study aimed to evaluate the feasibility, acceptability, and initial efficacy of a brief, computer-delivered AS reduction program (AST) for adults with obesity and elevated AS. Participants were 131 individuals with obesity and elevated AS who received either AST or a health information control (HC). Latent growth curve modeling was utilized to examine the effect of treatment condition on change in AS, emotional eating, expectancies of eating to manage mood, and exercise self-efficacy from baseline to 1-week, 2-weeks, and 1-month follow-up. Results indicated it was feasible to recruit and retain individuals with obesity and elevated AS through the 1-month follow-up. Positive qualitative feedback was also provided for both AST and HC. AST was a statistically significant predictor of change in expectancies of eating to manage negative affect. No other statistically significant treatment effects were indicated. Current data provide initial empirical support for future work focused on persons with obesity and elevated AS.

Development and Cultural Adaptation of a Computer-Delivered and Multi-Component Alcohol Reduction Intervention for Russian Women Living with HIV and HCV.

Background:There is elevated prevalence of problem drinking among Russian women living with HIV and HCV co-infection. This paper describes the development and cultural adaptation of a multi-component alcohol reduction intervention incorporating a brief, computer-delivered module for Russian women living with HIV and HCV co-infection.Methods:The format and content of the intervention were adapted to be linguistic-, cultural-, and gender-appropriate using the ADAPT-ITT framework. A computer-delivered module and brief clinician-delivered individual and telephone sessions were developed.Results:We describe the theoretical foundations of the intervention, the cultural adaptation of the intervention, and overview the content of the intervention’s multiple components.Discussion:Interventions to reduce alcohol use that can be integrated within Russian HIV treatment centers are urgently needed. If efficacious, the culturally-adapted intervention offers the promise of a cost-effective, easily disseminated intervention approach for Russian women living with HIV/HCV co-infection engaging in problematic alcohol use.

Computer-delivered brief alcohol intervention for patients with liver disease: a randomized controlled trial.

Reducing alcohol consumption by liver disease patients can reduce morbidity and mortality. This study compared a computer-delivered brief alcohol intervention (cBAI) with standard care in a sample of US military veterans with liver disease. Multi-site, randomized controlled trial of a cBAI plus standard care (n=67) versus standard care only (n=71). Participants were assessed at baseline and 3- and 6-month follow-up. US Veterans Health Administration liver clinics. Participants were mostly male and diagnosed with hepatitis C. A cBAI tailored to veterans with liver disease and consisting of assessment and personalized feedback. Standard care was brief education and advice about alcohol and liver disease. Primary outcomes were self-reported number of drinking days and unhealthy drinking days (defined as more than two drinks for men and more than one for women) in the past 30days at 6-month follow-up. Secondary outcomes were these two variables at 3-month follow-up, and drinks consumed per drinking day, depression and overall health at 3- and 6-month follow-ups. Missing data were imputed using multiple imputation. Compared with standard care, cBAI participants reported significantly fewer drinking days at 6-month follow-up and fewer unhealthy drinking days at both 3- and 6-month follow-ups. Least square means (LS-means) for number of drinking days were 3.78 for the cBAI condition and 6.89 for the standard care condition at 6months [LS-mean ratio=3.78/6.89=0.55, 95% confidence interval (CI)=0.34, 0.89]. LS-means for number of unhealthy drinking days were 1.04 for the cBAI condition and 2.57 for the standard care condition at 3-month follow-up (LS-mean ratio=1.04/2.57=0.41, 95% CI=0.19, 0.85). At 6-months follow-up, LS-means were 1.18 for the cBAI condition and 2.75 for the standard care condition (LS-mean ratio=1.18/2.75=0.43, 95% CI=0.20, 0.91). A computer-delivered brief alcohol intervention reduced drinking days and unhealthy drinking days at 6-month follow up in military veterans with liver disease compared with brief education and advice to reduce consumption.

Participant Satisfaction with Computer-Delivered Intervention Components and Its Relation to Alcohol Outcomes

Background Few computer-delivered brief intervention (CDBI) studies have evaluated participant satisfaction with individual elements of the intervention, or whether participant satisfaction impacts intervention outcomes. Purpose: This factorial trial examined whether subjective reactions to a CDBI for heavy drinking (1) varied depending on the presence versus absence of an animated narrator, a spoken voice, empathic reflections, and motivational interviewing (MI) strategies and (2) were associated with drinking outcomes at 3-month follow-up. Methods: Participants were 352 heavy drinking university students. All participants were randomly assigned to one of 16 versions of a CDBI. After finishing the CDBI, participants completed measures of intervention likability and perceived empathy. Alcohol use outcomes were assessed at 3-month follow-up. Results: CDBI characteristics had minimal effects on participant ratings of likeability and perceived empathy. However, higher likeability ratings were associated with decreases in alcohol use outcomes over the 3-month assessment period. Conclusions: Results indicate that subjective reactions to CDBIs can have important effects on alcohol use outcomes.

Predictors of receptivity to an alcohol intervention among mandated students

ObjectiveStudents in violation of campus alcohol policy often must participate in an alcohol risk reduction intervention, and some mandated students express overt resistance to change. This study tested a priori hypotheses about psychological variables that might predict receptivity to an online brief alcohol intervention. MethodParticipants were 193 students (56% male, 88% underclassmen, 70% white) who were referred to the campus Office of Alcohol and Other Drugs after being cited for alcohol policy violations. Before their mandated online alcohol intervention, they completed a baseline assessment consisting of demographics and questionnaires about their alcohol use, attitudes toward moderate and heavy drinking, beliefs about the centrality of alcohol in college life, perceived descriptive and injunctive drinking norms, trait defensiveness, social desirability and response to the event leading to their sanction. They also responded to a set of questions about how interested in, open-minded, and likely to benefit they were from the computer-delivered alcohol intervention they were about to take. ResultsControlling for covariates and all other predictors, greater receptivity to intervention was predicted by heavier alcohol consumption, positive attitude towards moderate drinking, perceptions that peers were less approving of heavy drinking, and taking personal responsibility for the incident. ConclusionsAmong students mandated to participate in an alcohol intervention, openness to benefitting from the intervention varies in predictable ways. Because intervention receptiveness is associated with post-intervention change, we offer suggestions to tailor mandated interventions in ways that might improve outcomes.

Budget Impact Analysis of a Computer-Delivered Brief Alcohol Intervention in Veterans Affairs (VA) Liver Clinics: A Randomized Controlled Trial

ABSTRACT A recent clinical trial found that a computer-delivered BAI (cBAI) was effective in reducing alcohol use among patients, in the Department of Veterans Affairs (VA), with liver disease. This is encouraging clinically, but whether the intervention can be rolled on within VA nationally depends on its economic impact. We conducted a randomized clinical trial in 110 patients with liver disease in two VA clinics. We used VA national administrative data to obtain and evaluate health utilization and costs 12 months before and after implementing the cBAI, using a difference-in-differences (DID) approach. The cBAI and usual care group did not differ on inpatient, outpatient, and pharmacy health care utilization and costs at baseline. The results of the DID analysis of the total health services expenses demonstrated similar costs between the cBAI and usual care group, 12 months before and after BAI intervention. Implementing a cBAI in VA liver clinics reduces alcohol consumption while not having any long-term impact on health care utilization and costs.

Reducing sexual health risks and substance use in the prenatal setting: A study protocol for a randomized controlled trial

Sexually transmitted infections (STIs) are at a record high in the United States, and STI risk is a critical and costly public health concern for childbearing women. STIs can lead to a number of serious health risks including premature birth, low birth weight, ectopic pregnancy, and fetal death. Similarly, there has been a dramatic increase in substance use during pregnancy, leading to complications during pregnancy and poorer birth outcomes. Women who misuse substances are disproportionately more likely to engage in risky sexual behaviors that can result in STIs. The proposed study will test whether the Health Check-Up for Expectant Moms (HCEM), a computer-delivered brief intervention that simultaneously targets STI risk and alcohol/illicit drug use during pregnancy, reduces antenatal and postpartum risk more than an attention, time, and information matched control condition among pregnant women seeking prenatal care. The study is a two-group, randomized controlled trial in which a diverse sample of 250 pregnant women will be recruited from prenatal care clinics and assigned to either (a) a computer-delivered, single-session brief intervention plus two booster sessions); or (b) a computer-delivered control condition. Follow-up assessments will occur at 2 and 6 months from baseline, and at 6 weeks postpartum. Our objective measures include STI incidence and birth outcomes. The results of this trial will fill a critical gap and provide much-needed data on the efficacy, costs, and resource utilization of a practical computer-delivered, brief motivational intervention tailored to reach high-risk women during pregnancy and extending impact to postpartum.

Computer-Delivered Intervention for Obese Sedentary Individuals With Elevated Anxiety Sensitivity

Computer-Delivered Intervention for Obese Sedentary Individuals With Elevated Anxiety Sensitivity

Exploring the components of an efficacious computer brief intervention for reducing marijuana use among adults in the emergency department

ObjectiveTo examine the efficacious components of a computer-delivered brief intervention (CBI) for reducing marijuana use among adults presenting to a low-income urban emergency department (ED), which a prior report found to decrease marijuana use at a 6-month follow-up. MethodParticipants were 237 ED patients reporting recent drug use (46% male; 54% African-American; mean age, 30.7) who were randomized to receive a CBI consisting of an interactive program guided by a virtual health counselor. The primary outcome was past 30-day marijuana use at 6-month follow-up assessed using the Timeline Follow-Back (TLFB). Intervention components related to change in marijuana use at 6 month follow-up examined in the current study included participant responses to items within five CBI domains that were rooted in motivational interviewing: goals for change, strengths, evoking-change (concerns about use and benefits of change), challenges, and tools for change. ResultsThe evoking-change domain was related to significant reductions in marijuana use at 6 months (B = −2.91, SE = 1.10, p < .01). Within this domain, items focused on concerns about family and friends were related to reductions in marijuana use of up to 5.5 fewer days of marijuana use in the past month (B = −5.49, SE = 1.63, p < .01). ConclusionsAn ED-based brief intervention, delivered by computer, was effective in reducing marijuana use. Intervention components focused on perceived concerns about use and benefits of change in relation to family and friends were critical domains within a CBI associated with reductions in marijuana use at 6-month follow-up.

A Randomized Pilot Study Comparing the Efficacy of a Therapist-Delivered Motivational Interview to a Brief Computer-Delivered Intervention

Aims: The purpose of this study was to compare the preliminary efficacy of a computer-assisted intervention (CAI), in which a computer-delivered intervention was immediately followed up with a brief therapist review session, to a therapist-delivered intervention (TDI) for adolescent substance use. Design, Setting, and Participants: Both conditions were examined in a pilot randomized clinical trial. All participants were recruited from a family court in the northeast United States. The sample included a total of 36 adolescents court-referred for an adolescent substance-related offense.Measures: Measures included adolescent alcohol and marijuana-use frequency, quantity, and problems as well as self-efficacy to resist the urge to use.Findings: While no significant time-by-condition differences were noted between the CAI and TDI conditions, significant time effects were found for both the TDI and CAI indicating a decrease in the total number of alcohol- or marijuana-use days over the six-month follow-up period.Conclusions: Given that CAIs are inexpensive, require minimal training, can be implemented with a high degree of fidelity, and are portable when compared to some TDIs, their use for decreasing substance use and related problems, particularly among adolescents with low access to substance-use interventions, seems promising. A fully powered trial of CAI efficacy is indicated.

Perceived Facilitators and Barriers in Response to a Walking Intervention in Rural Cancer Survivors: A Qualitative Exploration.

Physical activity has numerous associated benefits for cancer survivors. Compared to their urban counterparts, rural and remote Australians experience a health disadvantage, including poorer survival rate after the diagnosis of cancer. The purpose of this qualitative study was to (a) investigate factors that motivated or inhibited walking in rural participants during a 12-week intervention and (b) to investigate factors that motivated or inhibited physical activity behavior change three months post-intervention. Ten cancer survivors living in rural areas of South Australia participated in a 12-week computer-delivered walking-based intervention during which they reported daily steps, daily affect, and ratings of perceived exertion. Based on this information, individualized daily step goals were sent to them to increase walking. Following the intervention, participants engaged in face-to-face semi-structured interviews. Interviews were recorded, transcribed and coded using thematic analysis. Participants identified a range of physical, psychological, social, environmental, and organizational motivators and barriers. Participants appreciated the monitoring and support from the research team, but some voiced a need for better transition to post-program and many desired ongoing support to maintain their motivation. Future studies should incorporate strategies to help walking behavior to become more intrinsically motivated and therefore sustained.

Adapting Coordinated Anxiety Learning and Management for Veterans Affairs Community-Based Outpatient Clinics: Iterative Approach.

BackgroundA national priority at the US Department of Veterans Affairs (VA) is to increase the availability and accessibility of evidence-based psychotherapies (EBPs) across all VA medical facilities. Yet many veterans, particularly those who use remote outpatient VA clinics, still do not receive much needed evidence-based treatment. Strategies are needed for supporting mental health providers at rural VA community-based outpatient clinics (CBOCs) as they translate their clinical training to routine practice. The Coordinated Anxiety Learning Management (CALM) program is a computer-delivered program that supports the delivery of cognitive behavioral therapy (CBT) by providers in outpatient settings to patients with depression and anxiety, including posttraumatic stress disorder.ObjectiveThe objectives of our study were to (1) adapt an existing computer-based program to rural VA CBOCs through feedback from key stakeholder focus groups; (2) develop a prototype of the adapted program; and (3) determine the adapted program’s acceptability and feasibility. Mental health stakeholders included VA leaders (n=4) in the implementation of EBPs, VA experts (n=4) in CBT, VA CBOC mental health providers (n=8), and veterans (n=8) diagnosed with a mental health condition treated using the CALM program and receiving treatment in a VA CBOC.MethodsAn iterative approach comprising 3 waves of focus group discussions was used to develop a modified prototype of CALM. Following each wave of focus group discussions, template analysis was used to rapidly communicate stakeholder recommendations and feedback to the design team. The original program was first adapted through a process of data collection, design modification, and product development. Next, a prototype was developed. Finally, the redesigned program was tested for acceptability and feasibility through a live demonstration.ResultsKey stakeholders suggested modifications to the original CALM program that altered its modules’ appearance by incorporating veteran-centric content. These modifications likely have no impact on the integrity of the original CALM program, but have altered its content to reflect better the demographic characteristics and experiences of rural veterans. Feedback from stakeholder groups indicates that changes will help VA patients identify with the program content, potentially enhancing their treatment engagement.ConclusionsThe development model was effective for economically gathering actionable recommendations from stakeholders to adapt a computer-based program, and it can result in the development of an acceptable and feasible computer-delivered intervention. Results have implications for developing computer-based programs targeting behavior change more broadly and enhancing engagement in EBP.

Evaluative conditioning: A brief computer-delivered intervention to reduce college student drinking

BackgroundRecent research suggests that a brief computer-delivered intervention based on evaluative conditioning (EC) can change the implicit evaluation of alcohol and reduce drinking behaviors among college students. We tested whether we could obtain similar findings in a high-powered preregistered study and whether hazardous drinking moderates these effects. MethodBefore the intervention, 122 French college students were screened for hazardous drinking using the Alcohol Use Disorder Identification Test (AUDIT). Implicit evaluation of alcohol was assessed before and immediately after the intervention using an Implicit Association Test (IAT). Drinking behavior was assessed before the intervention and approximately two weeks after using the TimeLine Follow Back (TLFB) method. The EC consisted of 120 trials of words (related to alcoholic beverages, soft drinks or neutral) paired with pictures (neutral, positive or negative). In the EC condition, alcohol-related words were systematically paired with negative pictures. In the control condition, alcohol-related words were systematically paired with neutral pictures. ResultsThe EC did not change the implicit evaluation of alcohol, Cohen's d = 0.01, 95CI [−0.35, 0.35]. However, the EC reduced drinking behavior, Cohen's d = 0.37, 95CI [0.01, 0.72]. This effect was independent of hazardous drinking behavior, but it was especially pronounced among participants with the most positive implicit evaluation of alcohol before the intervention. ConclusionThis preregistered study suggests that evaluative conditioning can successfully reduce drinking behavior among college students by 31% (compared to 4% in the control condition) without causing an immediate change in the implicit evaluation of alcohol.

Sexual HIV risk behavior outcomes of brief interventions for drug use in an inner-city emergency department: Secondary outcomes from a randomized controlled trial

BackgroundDrug use is an established risk factor for HIV. Brief Interventions (BIs) targeting reductions in both drug use and HIV risk behaviors may help curtail these related epidemics. The present study evaluates the impact of BIs for drug use and HIV risk reduction on sexual HIV risk behaviors among a primarily marijuana-using sample during a 12-month post-intervention follow-up period. MethodsWe conducted a randomized controlled trial of 780 adult patients in an Emergency Department (ED) with past 3-month drug use (primarily non-injecting). This study used a 3 × 2 factorial design (3 ED-based conditions: computer-delivered brief intervention [Computer BI], therapist-delivered, computer-guided BI [Therapist BI], or enhanced usual care (EUC-ED) for drug-using adults; 2 follow-up conditions at 3 months: booster or control). This analysis examines the outcomes of the BIs on sexual HIV risk behaviors at 3-, 6-, and 12-months. ResultsCompared to the enhanced usual care control, the combined Therapist BI with booster resulted in significant reductions in scores on the sexual risk subscale of the HIV Risk Taking Behaviour Scale over 12-months, when controlling for baseline sexual risk, gender, and drug dependency status. The baseline interventions alone, booster alone, and Computer BI plus booster did not differ from the comparison group (EUC plus control). ConclusionsA therapist-delivered BI for drug use and HIV risk behaviors, combined with a follow-up therapist-delivered booster, shows promise for reducing sexual HIV risk behaviors among a primarily marijuana using, non-injecting sample.

A Pilot Randomized Controlled Trial of a Computer-Delivered Brief Intervention for Substance Use and Risky Sex During Pregnancy.

Sexually transmitted infections (STIs) are an increasingly critical and costly health problem for American childbearing women. Pregnant women who misuse substances are more likely to engage in risky sexual behavior that leads to STIs. Substance use and risky sex during pregnancy are both associated with numerous negative consequences for the woman and the developing fetus. A two-group, randomized controlled trial. Recruitment of 50 pregnant women (30% Latina; 24.4 years old [SD = 5.31]) with an average of 13 weeks gestation (SD = 4.5 weeks) was conducted at a prenatal clinic in a large inner-city hospital. Recruitment took place between 2015 and 2016, and data analysis took place in 2016. A computer-delivered, single-session brief motivational intervention plus booster session addressing both substance use and STI risk. To assess participants' perceptions of the intervention and to examine the preliminary efficacy in reduction of substance use and risky sex at 4-month follow-up assessment. There were consistently very high ratings of acceptability of the intervention, ranging between 6.3 and 6.8 on a 1-7 scale. At the 4-month follow-up, participants in the intervention arm reported a significantly larger reduction (54%) in any marijuana or alcohol use compared with participants in the control group (16%) (p = 0.015) based on two-group clustered logistic regression using a generalized estimating equations approach. There was a higher reduction in condomless vaginal sex at follow-up in the health checkup for expectant moms (HCEM) arm than control (27% vs. 5%), although this was not significant (p = 0.127). The results of this pilot study are encouraging with respect to the acceptability and preliminary efficacy of an intervention in reducing alcohol/marijuana use and condomless sex during pregnancy, supporting the next step of testing the intervention in a larger sample.

Initial Progress Toward Development of a Voice-Based Computer-Delivered Motivational Intervention for Heavy Drinking College Students: An Experimental Study.

BackgroundComputer-delivered interventions have been shown to be effective in reducing alcohol consumption in heavy drinking college students. However, these computer-delivered interventions rely on mouse, keyboard, or touchscreen responses for interactions between the users and the computer-delivered intervention. The principles of motivational interviewing suggest that in-person interventions may be effective, in part, because they encourage individuals to think through and speak aloud their motivations for changing a health behavior, which current computer-delivered interventions do not allow.ObjectiveThe objective of this study was to take the initial steps toward development of a voice-based computer-delivered intervention that can ask open-ended questions and respond appropriately to users’ verbal responses, more closely mirroring a human-delivered motivational intervention.MethodsWe developed (1) a voice-based computer-delivered intervention that was run by a human controller and that allowed participants to speak their responses to scripted prompts delivered by speech generation software and (2) a text-based computer-delivered intervention that relied on the mouse, keyboard, and computer screen for all interactions. We randomized 60 heavy drinking college students to interact with the voice-based computer-delivered intervention and 30 to interact with the text-based computer-delivered intervention and compared their ratings of the systems as well as their motivation to change drinking and their drinking behavior at 1-month follow-up.ResultsParticipants reported that the voice-based computer-delivered intervention engaged positively with them in the session and delivered content in a manner consistent with motivational interviewing principles. At 1-month follow-up, participants in the voice-based computer-delivered intervention condition reported significant decreases in quantity, frequency, and problems associated with drinking, and increased perceived importance of changing drinking behaviors. In comparison to the text-based computer-delivered intervention condition, those assigned to voice-based computer-delivered intervention reported significantly fewer alcohol-related problems at the 1-month follow-up (incident rate ratio 0.60, 95% CI 0.44-0.83, P=.002). The conditions did not differ significantly on perceived importance of changing drinking or on measures of drinking quantity and frequency of heavy drinking.ConclusionsResults indicate that it is feasible to construct a series of open-ended questions and a bank of responses and follow-up prompts that can be used in a future fully automated voice-based computer-delivered intervention that may mirror more closely human-delivered motivational interventions to reduce drinking. Such efforts will require using advanced speech recognition capabilities and machine-learning approaches to train a program to mirror the decisions made by human controllers in the voice-based computer-delivered intervention used in this study. In addition, future studies should examine enhancements that can increase the perceived warmth and empathy of voice-based computer-delivered intervention, possibly through greater personalization, improvements in the speech generation software, and embodying the computer-delivered intervention in a physical form.