Computerized Alcohol Misuse Intervention System for Health Care (CAMI)

This project expands upon the proof-of-concept work done under the pilot project in order to get Interventionaire© poised for more widespread use. Interventionaire© is designed to deploy a computerized interactive questionnaire to patients in a physician waiting room. The information gathered is intended to be a conversation starter for primary care providers and patients (who may feel more at ease with the computer in revealing health risk behaviors, like alcohol abuse). Widespread use of Interventionaire© could result in a substantial cost savings by delivering a wide reaching and highly acceptable prevention strategy for integrating risky health behavior screening and brief intervention into mainstream medical practice.The investigators overall hypothesis is that participants receiving the computerized intervention will have greater improvement than those receiving the attention control condition.

Alcohol misuse continues to pose a major burden for public health. Screening and brief intervention for alcohol misuse in primary care settings has been recommended for decades, and the evidence base supporting it in primary care is robust. However, many primary care providers do not intervene in alcohol misuse, and those that do often fail to do so systematically or in accordance with best-evidence standards. Computerization of screening and brief intervention for alcohol misuse holds the promise of alleviating many of the barriers that have prevented its adoption into routine care, and also offers new opportunities for optimizing the effectiveness of brief interventions. This Phase 2 Small Business Technology Transfer (STTR) project follows up on Research Circle Associates' (RCA) successful proof-of-concept work done under Phase 1, which led to the development of a versatile software platform (Interventionaire©) to support computerized alcohol misuse screening and brief intervention using evidence based strategies including motivational interviewing and normative feedback. The investigators' computerized alcohol misuse intervention includes a normative feedback component that is unique in its micro-targeting of alcohol use feedback based on participant demographic characteristics (race/ethnicity, gender, and age), drawing population normative data from the National Surveys on Drug Use and Health. This represents a novel use of epidemiological data to inform clinical services. The Phase 2 project is designed to answer a number of important research questions and prepare the product for successful commercialization. Specifically, during Phase 2 the investigators intend to implement several refinements into the Interventionaire© platform that will increase its usability and streamline the creation of complex content with tailored feedback. The investigators will also upgrade the software to make it compatible with a major electronic health record system used by the investigators' clinical partner site. This will increase the reach of the product and ensure maximum readiness for deployment into the modern healthcare marketplace, a dynamic and rapidly evolving industry. The investigators will also conduct a study at the investigators' clinical partner site using qualitative key informant, focus group, and survey methodologies to investigate barriers to adopting computerized alcohol screening and brief intervention systems. A critical question is whether the computerized alcohol misuse intervention is effective in changing drinking behavior. To answer this question, the investigators will conduct a cross-over randomized controlled trial examining the short-term impact of the computerized alcohol misuse intervention with micro-targeted normative feedback and motivational messaging compared to an "attention-control" (placebo) generic health promotion message. This trial will help to establish the short-term efficacy of the intervention and set the stage for larger trials and adoption of the product in healthcare settings. Finally, the investigators will conduct a projection analysis of the economic and health impacts of the intervention under a variety of assumptions regarding intervention effect size, target population, and product penetration. Satisfying these Phase 2 objectives will leave RCA well-poised to deploy the product commercially, which will leverage the power of the market to improve clinical service delivery and promote public health.

Arm 1 is a cross-over randomized controlled trial (XRCT) comparing the short-term effects of the computerized alcohol misuse intervention (CAMI) vs. a generic health education attention-control (AC) condition in reducing past 30 day: (a) days of alcohol use, (b) days of heavy episodic drinking, (c) typical number of drinks consumed on a day of drinking, and (d) alcohol use in dangerous situations; (2) investigate changes in (a) patient motivation and (b) perceived norms as mediators of intervention effectiveness.

Arm 2 is not an active intervention but an attention control condition - it is not expected to exert any real intervention effect. Hence, any observed effects for the AC condition can be attributed to natural change via regression-to-the-mean, assessment reactivity, or a placebo effect. The effect size estimate of the intervention, both from an efficacy standpoint (i.e., above and beyond what change would have normally occurred) and an effectiveness standpoint (i.e., the magnitude of behavior change that can be expected in real-world implementation), can be readily calculated.

Interventionaire© is a computerized alcohol misuse intervention that includes a normative feedback component unique in its micro-targeting of alcohol use feedback based on participant demographic characteristics (race/ethnicity, gender, and age), drawing population normative data from the National Surveys on Drug Use and Health. This represents a novel use of epidemiological data to inform clinical services.

These products are designed for entertainment and educational purposes only. No medical claims are made for them and, the participant's interaction with them should be conceptually unrelated to alcohol use behavior.In order to avoid any carryover effect from comparing two active treatments, we have chosen to use a simple attention control condition that is not expected to exert any real intervention effect. In that way, the effect size estimate of the intervention can be readily calculated. No research data will be collected as part of the AC, beyond the amount of time spent on the program.

Change in drinking behavior as measured by self-reported alcohol use after completing the computerized alcohol misuse intervention (CAMI)

At one month, the investigators expect the attention control group (A) to reduce their alcohol use somewhat, but the intervention group (B) is expected to show a much steeper reduction. Thus, the investigators first planned contrast tests A vs. B, and can be considered similar to a parallel group trial examining 1-month outcomes for intervention vs. attention control. After the cross-over at 1 month, the magnitude of the reduction from Month 1 to Month 2 for those receiving the intervention (C) is expected to be similar to that experienced by those who received the intervention initially.

Changes in participant motivation to stop or cut down alcohol use as measured by self-reported interest in reduction of alcohol use after completing the computerized alcohol misuse intervention (CAMI)

It is expected that the effect of CAMI will be carried by changes in motivation and changes in perceived norms.The investigators will use advanced modern mediation analysis techniques to investigate the mechanisms that may drive reductions in drinking.

Inclusion Criteria: Adult age 18 and older; Positive AUDIT screening score; and Willing to provide contact information for themselves and two friends or family members (to assist with tracking for follow-up). Exclusion Criteria: Inability to give informed consent (e.g., due to cognitive impairment); Acute medical or psychiatric problem that precludes ability to complete the study interview or intervention; Inability to comprehend or read English.

Neighbors C, Larimer ME, Lewis MA. Targeting misperceptions of descriptive drinking norms: efficacy of a computer-delivered personalized normative feedback intervention. J Consult Clin Psychol. 2004 Jun;72(3):434-47. doi: 10.1037/0022-006X.72.3.434.