Educational video module for increasing treatment rates for alcohol use disorders in inpatients

Abstract

A study by Twohig et al evaluated the impact of an educational video module (EVM) on the treatment of alcohol use disorder (AUD) in hospitalized patients with alcohol-related liver disease (ALD). This single-center prospective study involved 42 patients, and the results were compared with those of a retrospective control group. EVM increased the rates of pharmacological (50% vs 22%, P = 0.0008) and psychosocial (73.8% vs 44%, P = 0.001) treatments within 30 days post-treatment. The rate of alcohol relapse decreased significantly (7.9% vs 35.6%, P = 0.003) after the intervention. All the participants recommended the EVM. These findings suggest that standardized educational interventions can address knowledge gaps and improve treatment engagement for AUD in patients with ALD.

Key Words: Alcohol use disorder; Alcohol-related liver disease; Educational video module; Addiction treatment; Patient education; Hospital-based intervention

Core Tip: This study demonstrates the effectiveness of a 15-minute educational video module in increasing alcohol use disorder treatment rates in hospitalized patients with alcoholic liver disease. This standardized intervention addresses knowledge gaps among providers and patients, resulting in significant increases in the initiation of pharmacological and psychosocial treatment and a dramatic decrease in alcohol relapse rates. Its scalability and high acceptance make the educational video module a promising strategy for widespread implementation in healthcare systems.

TO THE EDITOR

We read with great interest the article published by Twohig et al[1] on the development of an educational video module (EVM) to improve treatment rates for alcohol use disorder (AUD) in hospitalized patients with alcoholic liver disease (ALD). This single-center prospective study provides convincing empirical evidence of the effectiveness of standardized educational interventions in addressing AUD treatment gaps.

AUD TREATMENT GAPS AND THE URGENCY OF INTERVENTION

Alcohol use disorder is the seventh leading risk factor for premature death and disability worldwide, with ALD accounting for 50% of all liver-related deaths[2,3]. Data from the Twohig et al’s study[1] revealed that in the retrospective control group (2018-2020), only 22% of hospitalized patients with ALD received pharmacological treatment, and 44% received psychosocial therapy within 30 days post-treatment. These baseline figures confirm the existence of a substantial treatment gap that requires systematic interventions. Although multimodal AUD treatment has been shown to reduce alcohol consumption and hepatic decompensation by increasing abstinence rates and survival, the level of treatment utilization in clinical practice remains low[4,5]. A survey of hepatology providers identified systemic barriers, with 71% having never prescribed pharmacotherapy for AUD, 50% being unfamiliar with Food and Drug Administration-approved medications, and 90% wanting additional training on AUD treatment. This gap creates a cycle in which patients do not receive adequate treatment information, thereby lowering treatment initiation rates, even when patients are motivated to change[6]. This knowledge gap creates systemic barriers to effective AUD treatment in different regions.

RESEARCH FINDINGS EVALUATION

The study by Twohig et al[1] used a prospective single-center design (n = 42) compared with a retrospective control design (n = 109) from 2018-2020. The specific timeframe is methodologically significant because the 2018-2020 control period preceded major changes in clinical practice guidelines and institutional protocols for AUD management. This temporal separation between the control and intervention groups introduces a systematic bias that threatens internal validity[7,8]. The intervention consisted of three inseparable components: (1) Brief intervention using motivational interviewing for all 52 enrolled patients; (2) A 15-minute EVM based on the Articulate StoryLine™ in English and Spanish; and (3) An electronic order set to facilitate AUD medication prescriptions. This multicomponent design prevents the determination of which specific element motivational interviewing, video education, or electronic prescribing support contributed to the observed increase in the treatment rate. Without a factorial design or component isolation, attributing effects to the EVM alone represents a fundamental limitation in causal inference[9]. The use of historical controls from the 2018-2020 period introduced substantial temporal bias. Several specific confounding factors emerged during the intervals between the control and intervention periods. The 2024 publication of the American College of Gastroenterology Clinical Guidelines on Alcohol-Associated Liver Disease dramatically increased provider awareness of AUD treatment integration in hepatology care[3]. This guideline includes strong recommendations (grade 1A evidence) for universal AUD screening and pharmacotherapy initiation in patients with ALD, fundamentally altering the standard of care. Institutional protocols for AUD screening evolved between 2020 and 2024, with many academic medical centers implementing mandatory electronic screening tools triggered by elevated liver enzymes or alcohol-related admissions[10-12]. Medication accessibility has changed substantially, as insurance coverage for naltrexone, acamprosate, and disulfiram has expanded following the 2021 American Psychiatric Association Practice Guideline updates[10,11]. The coronavirus disease 2019 pandemic (2020-2023) fundamentally altered healthcare delivery patterns, patient engagement with digital health interventions, and hospital admission thresholds, creating additional unmeasured confounders[13,14]. During the interval between the control and intervention periods, several fundamental changes occurred in the AUD treatment landscape: (1) The 2024 publication of the American College of Gastroenterology Clinical Guideline on Alcohol-Associated Liver Disease, which increased provider awareness of the importance of AUD treatment; (2) The evolution of institutional protocols for universal AUD screening in many academic medical centers; and (3) Changes in medication availability and insurance coverage for AUD pharmacotherapy, including the wider adoption of extended-release naltrexone. These temporal confounders indicate that the control and intervention groups differed not only in exposure to EVM but also in the baseline standard of care, medication formulary access, insurance authorization procedures, provider training levels, and institutional culture regarding addiction treatment. The magnitude of these secular trends cannot be quantified without concurrent control[8,15]. The control and intervention groups were not comparable in terms of exposure to EVM, standard care, medication accessibility, or institutional culture regarding AUD treatment. Concurrent randomization with contemporary controls eliminates temporal confounding. Concurrent randomization with contemporary controls would eliminate temporal confounding by ensuring that both the intervention and control groups are exposed to identical secular trends, institutional protocols, medication availability, and standard-of-care practices. Randomization does not assume protocol stability; rather, it distributes unmeasured confounders equally between groups through random allocation[8,16-18]. When patients are randomized during the same time period, any changes in clinical guidelines, institutional policies, or medication access affect both groups equally, thus isolating the specific effect of the intervention[4]. In contrast, historical controls introduce systematic bias because the control group operates under fundamentally different conditions than the intervention group does, making the observed differences attributable to temporal trends rather than the intervention itself[7,8]. For EVM evaluation, a contemporary randomized design would involve enrolling patients simultaneously and randomly assigning them to either EVM plus standard care or standard care alone, thereby controlling for all time-varying factors and enabling valid causal inferences regarding EVM effectiveness[15]. The study findings revealed that after the intervention, the treatment rate increased significantly, ranging from 50% to 22% in the control group (P = 0.0008) and from 73.8% to 44% in the control group (P = 0.001). The rate of return to alcohol use decreased from 35.6% to 7.9% (P = 0.003). This decrease was accompanied by a reduction in the mean alcohol consumption (12.6 drinks/day vs 2.1 drinks/day), serum ethanol (118 g/dL vs 76.7 g/dL), and phosphatidylethanol (467.6 g/dL vs 18.7 g/dL) levels. However, the interpretation of these results is limited by several methodological weaknesses, including: (1) The absence of component isolation, as all patients received brief intervention plus EVM plus Mediterranean diet support, making it impossible to determine the specific contribution of EVM. The study did not have a group that received only EVM without brief intervention; (2) Sample size and statistical power: With n = 42 vs 109 controls (imbalanced 1:2.6), this study was underpowered to detect differences in many secondary outcomes; (3) Outcome validation: The reduction in alcohol use was verified only through self-reports via telephone (timeline follow-back method) without consistent biomarker confirmation. Ethanol and phosphatidylethanol biomarkers were available for only a small proportion of patients; (4) Follow-up duration: The 30-day period was too short for the evaluation of AUD, a chronic relapsing-remitting condition that requires a minimum of 6-12 months of follow-up to assess sustained behavior change; and (5) Selection bias: The study excluded 12 patients who refused participation after a brief intervention, indicating that 19% of the eligible population was unresponsive to the educational approach. The final population may represent a subgroup with higher baseline motivation levels.

CLINICAL IMPLICATIONS AND MECHANISM OF ACTION

The findings show a positive association, and claims regarding the mechanism of action of the EVM require empirical verification. This study did not measure mediator variables, such as knowledge retention, self-efficacy, or patient empowerment, via validated instruments. The hypothesis that ‘the video format provides consistent, standardized information’ cannot be tested without data on the variation in counseling received by the control group or without measuring knowledge acquisition before and after EVM. The high acceptance rate (100% of patients recommended EVM, 92.5% found it helpful) must be interpreted in the context of the absence of adherence monitoring. The study did not report: (1) The percentage of patients who completed the EVM; (2) Whether patients could recall key messages from the EVM at follow-up; or (3) Whether the EVM adhered to the prescribed medication regimen. Positive acceptance of education does not automatically translate into behavior change, especially in the context of addiction, where the insight-action gap is well documented. Although the increase in the treatment rate reached statistical significance, the absolute risk reductions need to be evaluated from the perspective of clinical meaningfulness and feasibility of implementation. For pharmacologic treatment, an absolute increase of 28% (from 22% to 50%) resulted in a number needed to treat (NNT) of 3.6, meaning that 3-4 patients need to receive the intervention for one additional patient to receive pharmacologic treatment. For psychosocial treatment, an absolute increase of 29.8% resulted in an NNT of 3.4. These figures are superficially impressive, but their interpretation requires consideration of the intervention and opportunity costs. The development of the EVM using the Articulate Storyline™ requires expertise in instructional design, content validation by a multidisciplinary team (addiction psychiatrists, hepatologists, and internists), translation and cultural adaptation of the Spanish version, and integration with the hospital IT infrastructure. The implementation requires bedside tablets in each unit, technical support for troubleshooting, and ongoing maintenance for content updates. The bedside tablet infrastructure involves more than just simple video display devices. These tablets must be configured with hospital-grade security protocols to protect patient privacy, integrated with electronic health record systems to track completion and document educational interventions, and equipped with multilingual interfaces to serve diverse patient populations[19,20]. The tablets can be programmed with specialized applications designed for longitudinal patient engagement beyond the initial educational interventions. Potential applications include: (1) Automated medication adherence tracking through daily prompted self-reports; (2) Symptom monitoring for alcohol withdrawal or medication side effects with automated alerts to care teams; (3) Access to telemedicine consultations with addiction psychiatry services for follow-up support; (4) Integration with mobile health applications on patients’ personal devices to continue monitoring after discharge; and (5) Delivery of booster educational content at scheduled intervals to reinforce key messages[21]. However, implementing these extended functionalities requires substantial additional infrastructure, including secure data storage compliant with Health Insurance Portability and Accountability Act regulations, interoperability with community addiction treatment programs for continuity of care, technical support available 24/7 to address device malfunctions, and patient training to ensure the effective use of digital tools[20,22]. Without a formal cost-effectiveness analysis comparing the costs of EVM development and implementation to incremental health benefits (measured in health-adjusted life years gained or liver-related complications prevented), it cannot be determined whether the resource investment is justified compared with alternative strategies, such as provider education, system-level interventions (electronic clinical decision support), or enhanced access to addiction psychiatry consultation. Future implementation research should incorporate economic evaluations via frameworks such as the Consolidated Health Economic Evaluation Reporting Standards to inform scalability decisions[23].

IMPLEMENTATION CONSIDERATIONS

The generalizability of the findings is limited by the characteristics of the study population and setting. The sample, which included 74% females (vs 63% males in the control group), was not representative of the gender distribution of AUD in the general population, where the prevalence among males was greater. This single-center study at an academic medical center with advanced technological infrastructure (bedside iPads in every unit) cannot be directly applied to community hospitals or rural settings with limited resources. In addition, the EVM is available only in English and Spanish, limiting its implementation in populations with greater linguistic diversity. The contribution of brief interventions to outcomes cannot be ignored. All 52 consenting patients underwent a motivational interviewing session before watching the EVM. The literature shows that brief interventions alone can increase treatment engagement in patients with AUD. This study did not include a group that received only brief interventions (without EVM) to control for this effect. Consequently, the claim that ‘EVM increases treatment rates’ may overestimate the actual contribution of the video module compared with human interaction through motivational interviewing. The study also implemented an Epic^® order set to facilitate AUD medication prescription, representing a system-level intervention independent of the EVM. Order sets have been shown to increase medication prescription rates by reducing the cognitive load and standardizing practices. Without a control group receiving the order set without the EVM, it is impossible to disaggregate the effects of technology-facilitated prescribing from patient education.

RECOMMENDATIONS FOR FUTURE RESEARCH AND PRACTICE

On the basis of the findings of Twohig et al[1], several steps that could advance this field include the need for multicenter studies with more diverse patient populations to strengthen the generalizability of the findings (Table 1). Evaluating EVM in outpatient settings could expand the reach of the intervention beyond inpatient care. Longer follow-up periods will clarify the durability of the treatment effects and their impact on long-term clinical outcomes, such as liver disease progression and mortality[12,17]. The integration of EVM into the standard care pathway for patients with ALD should be considered. Hospitals can implement protocols for universal AUD screening and the automatic delivery of EVM to identified patients. Parallel training of service providers in AUD treatment will complement patient education and create a more supportive care environment. The 30-day return to alcohol use rate of 7.9% in the intervention group needs to be contextualized against the existing evidence. A systematic review by Kaner et al[24] analyzing 69 randomized controlled trials (RCTs) revealed that brief interventions reduced alcohol consumption by a mean difference of -20 g/week (95%CI: -28 to -12), with a modest but consistent effect size across settings. Chi et al[25] reported a sustained reduction at the 12-month follow-up after a brief intervention in primary care, with 42% of patients maintaining a clinically meaningful reduction compared with 34% in usual care (P = 0.03). The magnitude of the effect reported by Twohig et al[1] (77.8% relative reduction in return to drinking) substantially exceeds that reported in controlled trials of brief interventions, raising questions about whether short-term results are sustainable or primarily reflect the Hawthorne effect and selection bias. Without extended follow-up beyond 30 days, it is unknown whether this impressive reduction represents a durable behavior change or a transient response to intensive attention and novel interventions. Before widespread implementation can be recommended, validation through adequately powered multicenter RCTs that address the fundamental limitations of the current study is needed[26]. An ideal phase III study would involve 300-400 patients to ensure sufficient statistical power across subgroups and enable the detection of clinically meaningful differences in long-term outcomes, such as hospitalizations, liver-related mortality, and sustained abstinence rates. Additionally, for achieving AUD treatment, we recommend: (1) Standardizing EVM content and developing a core curriculum for AUD education that can be adapted by various institutions; (2) Creating an open-access repository of EVMs that should be made freely available under a creative commons license to enable wider implementation and replication studies; and (3) Developing an implementation science framework; future studies should use frameworks such as reach, effectiveness, adoption, implementation, maintenance to comprehensively evaluate scalability.

Table 1 Evaluation of methodological limitations and recommendations.

Domain	Limitations of the current study	Recommendations for improvement
Trial design	Two-arm (EVM + brief intervention vs retrospective control), unable to isolate the effect of EVM	Three-arm RCT: (1) EVM + brief intervention + order set; (2) Brief intervention + order set only; and (3) Usual care. Minimum follow-up of 12 months
Sample size	n = 42 intervention, no published power calculation	A priori sample size calculation to detect a 15% difference in the primary outcome with 80% power, P value = 0.05
Outcome measures	Self-reported alcohol use without consistent biomarker validation; no histology	Primary outcome: Composite of validated biomarkers (CDT, PEth) + structured clinical assessment (AUDIT score). Secondary: Liver stiffness measurement via FibroScan to track disease progression
Adherence monitoring	No data on EVM completion rate or medication adherence	(1) Video analytics to track completion; (2) Pill counts or pharmacy refill data; and (3) Knowledge assessment pre-post EVM using validated instrument
Population	74% female, single center, United States/Spain only	Balanced sex distribution, multicenter design including community hospitals, multilingual EVM
Statistical analysis	Retrospective control with demographic imbalance	Concurrent randomization with stratification by disease severity (MELD score) and prior treatment history

AUDIT: Alcohol use disorders identification test; RCT: Randomized controlled trial; CDT: Carbohydrate-deficient transferrin; EVM: Educational video module; MELD: Model for end-stage liver disease; PEth: Phosphatidylethanol.

CONCLUSION

Twohig et al[1] presented preliminary evidence that multicomponent interventions (brief motivational interviewing + EVM + provider order set) are associated with increased AUD treatment initiation rates and reduced short-term alcohol use. However, substantial methodological limitations-particularly the inability to isolate the specific effects of EVM, small sample size, reliance on self-reported outcomes without consistent biomarker validation, and inadequate follow-up duration-limit the ability to draw causal conclusions regarding EVM effectiveness as a standalone intervention. Future research with a three-arm randomized design, validated objective outcome measures, adequate statistical power, and extended follow-up (minimum 12 months) is needed before implementation guidelines can be formulated. Given the potential scalability and low cost of video-based education, investment in rigorous evaluation through well-designed RCTs is justified to identify whether the EVM provides added value beyond standard brief interventions and provider system changes.