Advertisement

Implementing Evidence-Based Screening and Counseling for Unhealthy Alcohol Use with Epic-Based Electronic Health Record Tools

      Background

      Multiple national organizations recommend screening and counseling adults for unhealthy alcohol use.

      Methods

      An evidence-based approach to screening and counseling using Epic electronic health record (EHR) tools was implemented in a general medicine clinic. A dissemination package with actionable steps for clinics and systems wishing to implement similar processes was then produced. To evaluate the initial implementation and quality improvement project, run charts were created to track patients screened, patients counseled, and fidelity to protocols, and members of the original project team were interviewed to assess facilitators and barriers. The draft dissemination package was revised after feedback from health system representatives (key informants).

      Results

      More than 9,000 patients (73.9% of those eligible) were screened in 20 months. Sixty-four percent of patients with positive initial screens had documented screening-related assessment; 39.7% (141/355) were offered counseling when indicated. Initial project team members identified EHR tools, clinic leadership, quality improvement culture, a multidisciplinary team, and training for providers and nurses as facilitators; and competing demands, patient population size, and nursing staff/resident turnover as barriers. Six key informants evaluated the dissemination package. Most rated 10 of the 12 sections as very useful; all rated components specific to implementing alcohol screening and counseling as very useful. Ratings for general guidance on implementing evidence-based services in primary care were more mixed.

      Conclusion

      Evidence-based screening and counseling for unhealthy alcohol use can be implemented with EHR tools. A dissemination guide was viewed favorably by key informants and can serve as a guide for other clinics and systems.
      The World Health Organization estimates that 1 in 20 deaths globally is due to alcohol.
      • World Health Organization
      Global Status Report on Alcohol and Health 2018.
      Unhealthy alcohol use is one of the leading causes of preventable deaths in the United States and is associated with many societal and health problems.
      • Centers for Disease Control and Prevention
      • National Center for Health Statistics
      FastStats: Alcohol Use.
      • Saitz R.
      Clinical practice. Unhealthy alcohol use.
      Unhealthy alcohol use is an overarching term that includes risky drinking (consumption of alcohol in excess of the recommended amounts) as well as alcohol use disorder (AUD), a pattern of alcohol use that involves problems controlling drinking, preoccupation with alcohol, continuing to use alcohol despite associated problems, drinking more to get the same effect, or withdrawal symptoms on rapidly cutting back on or stopping alcohol use.
      • American Psychiatric Association (APA)
      Diagnostic and Statistical Manual of Mental Disorders.
      The National Institute on Alcohol Abuse and Alcoholism (NIAAA) recommends no more than 4 drinks per day and 14 drinks per week for men under the age of 65. For all women, and for men 65 and older, the recommended limits are 3 drinks or fewer per day and no more than 7 drinks per week.

      National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

      More than 20% of primary care patients in the United States drink alcoholic beverages in excess of the recommended limits.
      • Vinson D.C.
      • et al.
      Alcohol and sleep problems in primary care patients: a report from the AAFP National Research Network.
      Based on comprehensive systematic review findings, the US Preventive Services Task Force (USPSTF) recommends that clinicians screen adults for unhealthy alcohol use and provide persons engaged in risky drinking with brief counseling interventions.
      • Curry S.J.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: US Preventive Services Task Force Recommendation Statement.

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • Moyer V.A.
      • et al.
      Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement.
      • O’Connor E.A.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
      Despite such recommendations and a substantial burden of illness, less than a third of patients visiting primary care providers in the United States are asked about alcohol use or ever discuss alcohol use with a health professional.
      • D’Amico E.J.
      • et al.
      Identification of and guidance for problem drinking by general medical providers: results from a national survey.
      • McKnight-Eily L.R.
      • et al.
      Vital signs: communication between health professionals and their patients about alcohol use—44 states and the District of Columbia, 2011.
      Practices often lack a formal process for screening and subsequent delivery of appropriate interventions for primary care patients. Barriers to screening and counseling include competing priorities, lack of provider training, misconceptions about patient comfort with discussing alcohol, and lack of appropriate infrastructure and protocols. Implementing screening and counseling for unhealthy alcohol use requires formal protocols, staffing (for example, multidisciplinary team-based care), support systems, and additional provider and staff training.

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • Jonas D.E.
      • et al.
      Behavioral counseling after screening for alcohol misuse in primary care: a systematic review and meta-analysis for the U.S Preventive Services Task Force.
      This article, the first in a special series of articles highlighting key projects from the Agency for Healthcare Research and Quality’s (AHRQ) Evidence-based Practice Center (EPC) initiative to improve translation of high-quality evidence into practice by learning health systems, as introduced by Fiordalisi et al,
      • Fiordalisi C.
      • et al.
      Introduction to AHRQ Series on Improving Translation of Evidence into Practice for the Learning Health System.
      describes how we implemented screening and counseling in primary care for unhealthy alcohol use based on national recommendations
      • Moyer V.A.
      • et al.
      Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement.
      and findings from a systematic evidence review,

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      and provides results from our evaluation of the implementation. The article is based on an implementation guide
      • Agency for Healthcare Research and Quality. Implementing Evidence-Based Screening and Counseling for Unhealthy Alcohol Use with Epic-Based Electronic Health Record Tools
      A Guide for Clinics and Health Systems, Developed as Part of a Pilot Dissemination Project. Barclay C, et al. AHRQ Publication No.
      that, drawing on lessons learned and challenges faced in the implementation process, is intended to offer a practical road map to the process of integrating these evidence-based services into a clinic’s work flow. A report on the package’s development that provided material for this article is available on the Agency for Healthcare Research and Quality (AHRQ) Effective Health Care Program website.

      Agency for Healthcare Research and Quality. Development of a Primary Care Guide for Implementing Evidence-Based Screening and Counseling for Unhealthy Alcohol Use with Epic-Based Electronic Health Record Tools: A Pilot Dissemination Project. Barclay C, Viswanathan M, Jones DE. Methods Research Report. AHRQ Publication No. 18-EHC020-EF. Sep 2018. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/methods-research-report-unhealthy-alcohol-use.pdf.

      The material presented in this article is based on the experience of one team at a single general internal medicine clinic, in an academic health care system, using Epic (Epic Systems Corporation, Verona, Wisconsin) electronic health care records to implement a service of interest. Components of the implementation process are described and illustrated with details from our clinic’s implementation of screening and counseling for unhealthy alcohol use. The general steps can be applied to implementation of many other evidence-based services. In addition, we summarize feedback from representatives of our own and other health systems we engaged as key informants (KIs) during the development of the dissemination package on which this article is based.

      Methods

      Implementation Project: Preliminary Steps

      Rationale for Implementation

      With the proliferation of recommended preventive services in primary care, health care systems must prioritize those for which (1) the burden of disease in the system’s patient population is substantial, (2) the evidence base clearly shows that benefits outweigh harms, and (3) resources are available for implementation.
      • Curry S.J.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: US Preventive Services Task Force Recommendation Statement.

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • Moyer V.A.
      • et al.
      Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement.
      • O’Connor E.A.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
      • Jonas D.E.
      • et al.
      Behavioral counseling after screening for alcohol misuse in primary care: a systematic review and meta-analysis for the U.S Preventive Services Task Force.
      • Mokdad A.H.
      • et al.
      Actual causes of death in the United States, 2000.
      Screening and counseling for unhealthy alcohol use met these three criteria at our site.

      Team Members

      Recruiting a strong, multidisciplinary team for the implementation process required staffing flexibility—identifying individuals whose skill sets correspond to one or more roles and who can commit the needed effort, particularly during the initial period in. The key roles on our team were filled by individuals who combined quality improvement (QI) training, expertise in unhealthy alcohol use, and/or experience in clinical or project management. We included existing staff in the clinic and stakeholders from all parts of the screening, counseling, and treatment process, including patient representation. The team included the following:
      • Project lead: experienced primary care clinician and researcher
      • Clinic medical director: primary care clinician with QI and practice innovation expertise; facilitated implementation in the clinic
      • Project coordinator: carried out administrative management of the project
      • Clinic project assistant: assisted with data collection, tracking, and documenting new processes
      • Nurse manager: provided nursing perspective on protocols, training, and work flow planning
      • Social workers/counselors: advised on role of social work, counselors, motivational interviewing, and available services for those with AUD
      • Patient/family advisor: member of our existing formal Patient and Family Advisory Council; provided feedback on processes and materials from patient perspective

      Process Map

      Early adoption of a process flow diagram clarified the impact that staffing and competing demands would have on implementation. Following the screening approach endorsed by the NIAAA,

      National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

      we developed a map of the work flow (Figure 1). We modeled the alcohol screening process after an existing successful approach to depression screening used in our clinic. Familiarity with the depression screening process (which involves a brief initial screening followed by a longer questionnaire if the initial screen is positive) facilitated the learning curve for staff and providers during implementation.
      Figure 1
      Figure 1This work flow for alcohol screening and interventions was used at our general internal medicine clinic. AUDIT, Alcohol Use Disorders Identification Test; AUD, alcohol use disorder.

      Implementation Project: Implementation Steps

      Process Development and Testing

      Clinics preparing to implement a new process will vary in their current state of readiness. Our implementation site’s history of ongoing improvement initiatives and receptivity to innovation favored the use of established QI methods to quickly develop, test, and modify components of the process. We conducted repeated, rapid Plan-Do-Study-Act (PDSA) cycles
      • Institute for Healthcare Improvement
      How to Improve: Science of Improvement: Testing Changes.
      with a limited number of providers and staff to optimize protocols, tools, and training procedures before implementing and testing them more widely across the clinic. In addition, we streamlined development and built on familiar processes by modeling our approach after existing clinic protocols, such as depression screening and treatment. Key features of our development process included the following:
      • Testing low-tech tools and protocols before launching permanent electronic health record (EHR)–based clinical support tools
      • At the outset, manually flagging a limited number of eligible patients in the clinic schedule before expanding to all eligible patients
      • Starting with a few providers and nurses before expanding to all providers
      • Developing a temporary user-editable tool to facilitate screening by allowing nurses to easily insert text into a chart. In Epic EHR, a nurse can document screening responses by typing a SmartPhrase name preceded by a period to link directly to a data entry Flowsheet.
      • Developing paper-based decision support tools and user-editable tools (SmartPhrases in Epic) for providers to guide assessment, counseling, or referral of patients with positive screens
      • Developing visit-based reminders and final EHR tools using what we learned from the previous steps
      • Collecting data and tracking progress

      Use of Validated Screening Instruments

      Recommendations for evidence-based practices usually stipulate the use of well-validated instruments for screening, which facilitates provider and staff buy-in and optimizes the benefits of the clinical service. We used the initial screening questions (shown in Figure 1) recommended by the NIAAA,

      National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

      which have demonstrated sensitivity and specificity roughly comparable to those reported for longer questionnaires,

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • O’Connor E.A.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
      followed by the 10-question Alcohol Use Disorders Identification Test

      National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

      (AUDIT) (Appendix 1, available in online article) for those with positive screens. The AUDIT can be completed in five minutes or less and has been tested extensively in primary care settings.

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • O’Connor E.A.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
      In initial testing cycles nurses administered a paper version, which was then incorporated into a nurse SmartPhrase and then a visit-based reminder.

      Assessment After Positive Initial Screen

      In our alcohol screening process, we stocked all clinic exam rooms with AUDIT forms printed on brightly colored paper. Nurses ask patients with positive initial screens to complete the AUDIT, which then serves as a visual cue for the provider (they would see the brightly colored paper on entering the room, indicating that the AUDIT was provided) to review it and calculate the score. The AUDIT helps to distinguish patients with AUD from those with risky drinking who do not have AUD. The back of our printed version of the AUDIT (Appendix 2) summarizes how AUDIT scores can be used to help with screening-related assessment and shows the scores that indicate whether an AUD is likely.

      Evidence-Based Interventions

      Effective evidence-based interventions must be available for a screening program to be recommended. Health care system decision makers and clinical directors should familiarize themselves with evidence reports and recommendation statements pertaining to topics of interest. The report on which we based our implementation supported the effectiveness of counseling interventions in reducing alcohol consumption, with numbers needed to treat (NNT) of less than 10 over one year for some drinking-related outcomes (for example, achieving drinking within recommended limits).

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • O’Connor E.A.
      • et al.
      Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
      • Jonas D.E.
      • et al.
      Behavioral counseling after screening for alcohol misuse in primary care: a systematic review and meta-analysis for the U.S Preventive Services Task Force.
      As described in the previous section, the provider reviews the completed AUDIT during clinical encounters and uses the scores to determine which patients have risky drinking behavior (but not an AUD) and are candidates for brief counseling interventions in primary care versus those whose scores indicate a likely AUD. The back of the printed AUDIT form summarizes the suggested intervention for risky drinking and provides a list of resources for patients with AUD (for example, 12-step programs, outpatient alcohol and substance abuse program).

      Interventions for People with Risky Drinking Without AUD

      Behavioral counseling interventions for risky drinkers aim to moderate a patient’s alcohol consumption to safer levels. The systematic review conducted for the USPSTF found that the best evidence for improving drinking outcomes was for brief (10–15 minutes) multicontact (≥ 2 visits) interventions.

      Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

      • Jonas D.E.
      • et al.
      Behavioral counseling after screening for alcohol misuse in primary care: a systematic review and meta-analysis for the U.S Preventive Services Task Force.
      Providers offer counseling using techniques from motivational interviewing, an evidence-based behavioral counseling approach that uses a patient-centered, guiding (rather than directing) style to elicit behavior change by helping patients to explore and resolve ambivalence and identify their personal motivations for change.
      • Miller W.R.
      • Motivational Interviewing Rollnick S.
      Preparing People to Change Addictive Behavior.
      • Rollnick S.
      • et al.
      Motivational interviewing.
      Printed pamphlets are available to support clinicians (Appendix 3), organized using a 5 A’s approach: Assess, Advise, Assist, Agree, and Arrange follow-up, with motivational interviewing techniques corresponding to each step. A paper pamphlet for patients titled Rethinking Drinking contains information about health risks, recommended drinking limits, definitions of standard drinks, a menu of options/goals for reducing risky drinking, and a diary to record alcohol consumption (Appendix 4). Both the Provider Guide and Rethinking Drinking pamphlet include portions of publicly available materials developed by the NIAAA.

      National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

      Interventions for People with AUD

      If providers identify an AUD, they conduct brief motivational interviewing to determine if the patient is willing to set a goal of abstinence, and then engage in shared decision making regarding options for more intensive treatment, referring to the list of available resources organized both by type of service (for example, detoxification, intensive outpatient programs, residential programs, individual therapy) and by county.

      Epic Tools to Support Delivery of Interventions

      Team members created and disseminated user-editable tools (Epic SmartPhrases) specifically for use by providers. Entering the appropriate SmartPhrase expedites documentation of the AUDIT score and provider response in the chart; guides the provider through an initial or follow-up counseling session; provides access to the current list of treatment and referral resources; or inserts the Rethinking Drinking content in a visit summary, which can be printed and given to the patient.

      Nurse and Provider Training

      Training aimed to convey the rationale for providing the new service, along with the relationship of the service to existing prevention efforts in the clinic and health system, the importance of validated instruments, specific protocols, and EHR tool use. Some of this material was imparted during presentations. Protocols and tool use were detailed in user guides and tip sheets distributed in the clinic and made available in the EHR.
      Among the nursing staff, our PDSA cycles revealed several points that required reinforcement in training sessions, materials, and one-on-one conversations, including informal “debriefs” after screening tool use, to obtain feedback in real time and improve the tools more rapidly. We emphasized the importance of using the correct, validated language. If the first question in the initial screen is not read as written (“Do you sometimes drink beer, wine, or other alcoholic beverages?”), patients may misinterpret the question. For example, if only asked whether they “. . . sometimes drink alcohol,” patients commonly think the question refers only to spirits (hard liquor) and does not include beer and wine. Correct completion of the SmartPhrase in the patient chart was revisited when necessary to ensure familiarity with pull-down options for age/gender groups, navigating through the phrase, remembering that the screening question is correctly answered as “number of times” not “number of drinks,” and being sure to select an answer (even when it is N/A) to all questions. Nurses were instructed in appropriate use of the AUDIT. Nurse feedback was solicited as part of the weekly PDSA cycle and in two 15- to 30-minute structured training sessions, with food and coffee provided, that were incorporated into the weekly nurse meeting. Our team also sought to increase buy-in and to provide positive reinforcement with acknowledgment of the nurses with the highest fidelity to the screening process (based on percentage of eligible patients screened) on the clinic’s visual management/nurse appreciation board.
      Provider training took place during the fall and spring pre-clinic conferences for residents, for which we developed content on rationale for screening, motivational interviewing, and hypothetical cases (Appendix 5). We refreshed existing knowledge and demonstrated new protocols and tool use in presentations during Grand Rounds, General Internal Medicine Division Meetings, and resident conferences, and reinforced these with e-mails from the medical director and chief resident.

      EHR Tools for Sustainability

      Streamlining and automating the screening and intervention process enhanced sustainability when staffing capacity for manual flagging of eligible visits or data entry of screening results diminished after the implementation project ended. We worked with the institution’s Epic development team to build and launch visit-based reminders that triggered the process and encouraged its completion. Prior EHR tool builds for other services with similar work flow and EHR functionality helped the development team efficiently produce our project’s tools. Other health systems’ EHR functionality that includes electronic reminders is likely to allow development of tools similar to those described in this article. The features and functionality of our EHR tools are summarized in Table 1.
      Table 1Best Practice Advisories (BPAs): Screening and Interventions for Unhealthy Alcohol Use
      BPA NameBPA TriggerUsersInstructionsLinksActions/Buttons
      Initial Alcohol ScreenNo initial alcohol screen or AUDIT documented in the past yearNursesClick DocFlowsheet to administer screen and document results• DocFlowsheet: Initial Alcohol Screening

      • Printable Initial Alcohol Screen
      • Delay—Other clinical priorities (suppresses for 6 weeks)

      • Patient declines (suppresses for 1 year)

      • Initial Alcohol Screen Complete (suppresses for 72 hours for reporting purposes if responses entered in flowsheet)
      Alcohol Incomplete ScreenNo AUDIT results documented (entered in AUDIT Flowsheet) after patient screened positive on initial screenNursesProvide paper AUDIT and document results• DocFlowsheet: AUDIT

      • Printable AUDIT
      • Delay—Other clinical priorities (suppresses for 6 weeks)

      • Patient declines (suppresses for 1 year)

      • AUDIT completed
      Alcohol Screen PositiveNo documentation (in visit note or with appropriate BPA tools) of brief intervention, counseling, or referral after positive initial screenProvidersProvide appropriate intervention and documentation by using system dotphraseNone• Open SmartSet
      The SmartSet contains single-click options for adding documentation to the visit note, making an ambulatory referral, ordering medications to treat AUD and vitamins, and choosing a follow-up time frame.
      (Alcohol Screen Positive)

      • Add Problem (Excessive drinking of alcohol [without AUD])

      • Add Problem (Alcohol consumption binge drinking)

      • Add Problem (Alcohol use disorder [RAF-HCC])

      • Delay—Other clinical priorities (suppresses for 6 weeks)

      • Patient declines (suppresses for 1 year)

      • Shared decision (suppresses for 1 year)
      AUDIT, Alcohol Use Disorders Identification Test; AUD, alcohol use disorder; RAF-HCC, Risk Adjustment Factor–Hierarchical Condition Category.
      * The SmartSet contains single-click options for adding documentation to the visit note, making an ambulatory referral, ordering medications to treat AUD and vitamins, and choosing a follow-up time frame.

      Data Collection and Tracking

      We formulated a measurement approach to assess the intervention’s site-specific reach and effectiveness, aiming to capture data on numbers and characteristics of patients receiving the new services and nursing staff/provider completion of protocols. Table 2 shows the measures collected during alcohol screening implementation in the University of North Carolina General Internal Medicine Clinic.
      Table 2Measures: Implementation of Screening and Interventions for Unhealthy Alcohol Use
      Measure NameMeasure CalculationData Source
      Patients screenedNumerator: Number of patients completing the single-question screen

      Denominator: All patients seen in GIM clinic who are eligible for screening
      Electronic health records
      Proportion of eligible visits in which initial screening was completedNumerator: Number of visits in which screening was completed

      Denominator: Visits in which patients were eligible for screening
      Electronic health records
      Patients with positive screens who complete the AUDITNumerator: Patients with AUDIT documented in EHR

      Denominator: Patients with positive single-question screens
      Electronic health records
      Documentation of whether patients likely have an AUDNumerator: Number with documentation of whether screening-related assessment indicates an AUD

      Denominator: Patients completing the AUDIT
      Electronic health records; chart review
      Patients appropriately offered counseling for risky drinkingNumerator: Patients offered counseling in primary care

      Denominator: Patients who engage in risky drinking but do not have an AUD
      Electronic health records; chart review
      Patients appropriately offered referral or treatment for an AUDNumerator: Patients offered referral or treatment for AUD

      Denominator: Patients with newly identified AUDs based on AUDIT and screening-related assessment
      Electronic health records; chart review
      Best Practice Advisories (BPA) UseNumerator: BPAs opened and completed by clinical team

      Denominator: Visits for screening eligible patients in which BPA is deployed
      Electronic health records
      GIM, general internal medicine; AUDIT, Alcohol Use Disorders Identification Test; AUD, alcohol use disorder.
      Run charts are a tool for tracking progress during a QI initiative or implementation effort. They allow a quick visual assessment of whether goals (for example, for number of patients screened) are being met and whether specific changes to protocols have led to improvements.
      • Perla R.J.
      • Provost L.P.
      • Murray S.K.
      The run chart: a simple analytical tool for learning from variation in healthcare processes.
      We created run charts of selected measures on a weekly basis during early PDSA cycles, and monthly later in the project period, In addition to collection of these data during the project period, the clinic project assistant or the project coordinator regularly recorded factors that facilitated or acted as barriers to implementation. Post-project, we also conducted semistructured interviews with original team members to incorporate their perceptions of facilitators and barriers.

      Dissemination Package Evaluation

      During a subsequent AHRQ–funded methods pilot project, we developed a dissemination package based on our implementation project experience and intended to offer a practical road map to the process of integrating these evidence-based services into a clinic’s work flow. To evaluate the package and strengthen its utility, we identified and solicited feedback from KIs. We sought representatives of health care systems that use Epic EHR (both academic and nonacademic, of various sizes, and from various regions of the United States), such as clinic medical directors, in a position to make decisions about implementation of services into primary care. We used a snowball sampling strategy,
      • Sampling Snowball
      contacting General Internal Medicine division heads, asking the task order officer on this project to suggest colleagues, and contacting directors at other Evidence-based Practice Centers with a brief explanation of the project and a request to recommend appropriate individuals who might serve as KIs.
      Individuals who agreed to serve as KIs were asked to review a draft and provide input via individualized links to a Qualtrics questionnaire focusing on the degree of usefulness of each component, as well as general feedback on the package, additional information that could help other clinics or health systems with uptake, and factors that may predict successful dissemination and implementation.

      Results

      Implementation Project

      Because a systematic process to screen for unhealthy alcohol use was not in place at our site at the outset of this implementation project, preimplementation data were not available. After implementation began, more 9,000 patients (73.9% of those eligible) were screened during the period from July 2016 to February 2018 (Figure 2); 39.7% (141/355) were offered counseling when indicated (Figure 3). Appendix 6 contains additional run charts showing the proportion of eligible patient visits in which initial screening for unhealthy alcohol use was completed, patients with positive initial screen who had AUDIT results documented, patients with documented AUDIT results who had likely AUD, and patients with likely AUD who were offered referral or treatment. Best Practice Advisory (BPA) use was quite stable during the 11 months between its launching and the end of the project period. The mean percentage of initial screening BPA firings in which the nurse opened the screening activity link was 62.6%.
      Here we note the facilitators and barriers encountered during alcohol screening and implementation at our clinic (recorded in PDSA cycle reports and/or identified in subsequent interviews with original team members). Conditions at other sites, as well as characteristics of the service being implemented, will generate their own set of facilitators and barriers, but in general they will be related to the organization’s financial and leadership resources, staff and provider familiarity and agreement with the rationale and tools for implementation, the presence of competing priorities, and patient factors.
      Figure 2
      Figure 2This chart shows the total number of patients who were screened (from July 2016 to February 2018) for unhealthy alcohol use.
      Figure 3
      Figure 3This graph shows the percentage of patients offered counseling for risky drinking when indicated on the day of screening. (Counseling is indicated for male patients with AUDIT [Alcohol Use Disorders Identification Test] scores 5–14 and for female patients with AUDIT scores 4–12.)
      Facilitators: Factors That Contributed to the Success of Our Implementation Project
      • Funding provided by the university’s Institute for Healthcare Quality Improvement permitted us to assemble a team with dedicated time for the implementation and QI project.
      • The Division of General Internal Medicine, where the project took place, has a history of hosting and supporting QI initiatives.
      • All team members received training in QI methods, which facilitated development and testing.
      • We modeled our screening and assessment protocols on the clinic’s existing depression screening process. The similarity to a familiar process facilitated nurse and provider uptake.
      • Developing a comprehensive list of resources for people with AUD increased willingness to identify and treat AUD. The list of available local resources for patients with AUD has received positive feedback from both patients and providers. Access to a useful clinic resource for all our patients with known AUD, not just those newly identified by our screening process, was an unintended positive consequence.
      Barriers: Factors That Presented Challenges in Our Implementation Project
      • Staffing issues, particularly a high proportion of float nurses (substitute nurses who do not routinely work in our clinic) on occasion, created gaps in training and compliance. Float nurses in particular sometimes gave patients the AUDIT first instead of starting with the initial screening questions.
      • Although paper forms may serve as useful prompts in the screening, assessment, and intervention cascade, collecting them and entering the responses in the EHR can pose a challenge. Our team put a labeled box for completed AUDITs in each provider’s room and sent reminders to providers when they did not document anything demonstrating follow-up on a positive initial screen.
      • Ensuring that resources were available for the various possible outcomes required time and effort from our multidisciplinary team, although it saved time and enhanced value after it was launched.
      • Competing priorities and lack of time pose the most substantial barrier. The initial screening itself can be quickly done, but providers required an estimated 5 to 10 minutes to perform the screening-related assessment when a patient had positive screening results. Additional time and visits are required for delivery of the behavioral counseling interventions to those with risky drinking behaviors.

      Dissemination Package Evaluation

      Six KIs reviewed the dissemination package and completed the questionnaire. The 12 sections of the package, each corresponding to an important component of the implementation process we developed and tested in our QI project, were rated as “very useful” by a majority of the KIs. All 6 KIs considered the sections on validated instruments and data collection to be “very useful,” while 5/6 found 3 other sections (on rationale for implementation, evidence-based interventions, and EHR tools for sustainability) to be “very useful.” No KI rated any section as “not useful.” Where feedback from the KIs’ evaluation of the revised draft clearly indicated ways the package could be improved, we modified the document according to their suggestions.
      KIs were also asked what factors they considered most predictive of successful dissemination and implementation efforts. Responses included a strong evidence base, confidence in the practice that implementation will not require too many additional resources, the presence of practice champions, early engagement of all stakeholders and frequent feedback from them, leveraging power of story as well as data, frequent and early small tests of change, transparency of data, building in sustainability from the start, buy-in from the team, organizational culture, and leadership support for implementation.

      Discussion

      The screening rates achieved during our implementation project period exceed national estimates, in which less than 25% of US adults report ever discussing alcohol use with a health professional.
      • McKnight-Eily L.R.
      • et al.
      Vital signs: communication between health professionals and their patients about alcohol use—44 states and the District of Columbia, 2011.
      We successfully screened the vast majority of our clinic population. Further, the proportion of our patients with positive screens for unhealthy alcohol use who were offered counseling (39.7%) compares favorably with national rates showing that less than 25% of those with significant problems from alcohol use received a recommendation to stop drinking,
      • D’Amico E.J.
      • et al.
      Identification of and guidance for problem drinking by general medical providers: results from a national survey.
      though it falls short of our admittedly ambitious target of 80%. We will continue to assess opportunities for improvement in uptake of counseling. In a recent prospective observational cohort study of quality of alcohol care for the population of patients screening positive for unhealthy alcohol use in a large US Department of Veterans Affairs health system, just over 30% of patients with high-risk alcohol use were offered a psychosocial intervention.
      • Hepner K.A.
      • et al.
      Rates and impact of adherence to recommended care for unhealthy alcohol use.
      Our data may underestimate the frequency of counseling, as it was collected for day of screening and thus does not capture counseling offered on subsequent visits. In addition to the barriers identified in team member interviews (for example, competing demands during visits, turnover among resident providers, undertraining of providers, data collection limited to the day of screening), another influence on rates of counseling is worth noting. Individuals with unhealthy drinking patterns who are screened for the first time are often in the precontemplation stage of change, as described in the Transtheoretical Model of behavior change,
      • Prochaska J.O.
      • Velicer W.F.
      The Transtheoretical Model of health behavior change.
      and therefore may not be likely to accept counseling at the time of the screening. Although our measures specified intervention offered rather than completed, rates of documentation may be lower when a patient declines counseling.
      The KIs involved in evaluation of our dissemination package identified information with a direct and practical application to the topic of alcohol screening and intervention, and to appropriate metrics for it, as most valuable for other health systems wishing to undertake similar implementation efforts. Their responses suggest that end users may want a concise document with a “just tell me what to do” perspective. During our original implementation we received this feedback from many providers. This type of document is also desirable because the level of effort involved in our process (conducted with dedicated funding and a multidisciplinary team) may not be replicable in many systems interested in implementing an evidence-based service.

      Conclusion

      Evidence-based screening and counseling for unhealthy alcohol use can be successfully implemented with EHR tools; our initial QI project, which benefited from dedicated funding and a multidisciplinary team, resulted in screening and counseling rates greater than those reported in national data. A dissemination package describing the process, barriers, and facilitators was viewed favorably by KIs and can serve as a guide for other clinics and systems.

      Funding

      Colleen Barclay, Shana Ratner, and Daniel Jonas received funding from the Institute for Healthcare Quality Improvement, University of North Carolina at Chapel Hill, for the implementation/quality improvement project on which this article is based. The article also draws upon materials prepared by the RTI International–University of North Carolina at Chapel Hill Evidence-based Practice Center (EPC) under contract to the Agency for Healthcare Research and Quality (AHRQ), Rockville, Maryland.

      Conflicts of Interest

      All authors report no conflicts of interest.

      Supplementary data

      References

        • World Health Organization
        Global Status Report on Alcohol and Health 2018.
        • Centers for Disease Control and Prevention
        • National Center for Health Statistics
        FastStats: Alcohol Use.
        • Saitz R.
        Clinical practice. Unhealthy alcohol use.
        N Engl J Med. 2005 Feb 10; 352: 596-607
        • American Psychiatric Association (APA)
        Diagnostic and Statistical Manual of Mental Disorders.
        5th ed. APA, Arlington, VA2013
      1. National Institute on Alcohol Abuse and Alcoholism. Helping Patients Who Drink Too Much: A Clinician’s Guide. NIH Publication No. 07-3769. 2005. (Updated: Jul 2016.) Accessed Jun 10, 2019. http://pubs.niaaa.nih.gov/publications/Practitioner/CliniciansGuide2005/guide.pdf.

        • Vinson D.C.
        • et al.
        Alcohol and sleep problems in primary care patients: a report from the AAFP National Research Network.
        Ann Fam Med. 2010; 8: 484-492
        • Curry S.J.
        • et al.
        Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: US Preventive Services Task Force Recommendation Statement.
        JAMA. 2018 Nov 13; 320: 1899-1909
      2. Agency for Healthcare Research and Quality. Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse. Jonas DE, et al. Comparative Effectiveness Review No. 64. AHRQ Publication No. 12-EHC055-EF. Jul 2012. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/alcohol-misuse_research.pdf.

        • Moyer V.A.
        • et al.
        Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement.
        Ann Intern Med. 2013 Aug 6; 159: 210-218
        • O’Connor E.A.
        • et al.
        Screening and behavioral counseling interventions to reduce unhealthy alcohol use in adolescents and adults: updated evidence report and systematic review for the U.S. Preventive Services Task Force.
        JAMA. 2018 Nov 13; 320: 1910-1928
        • D’Amico E.J.
        • et al.
        Identification of and guidance for problem drinking by general medical providers: results from a national survey.
        Med Care. 2005; 43: 229-236
        • McKnight-Eily L.R.
        • et al.
        Vital signs: communication between health professionals and their patients about alcohol use—44 states and the District of Columbia, 2011.
        MMWR Morb Mortal Wkly Rep. 2014 Jan 10; 63: 16-22
        • Jonas D.E.
        • et al.
        Behavioral counseling after screening for alcohol misuse in primary care: a systematic review and meta-analysis for the U.S Preventive Services Task Force.
        Ann Intern Med. 2012 Nov 6; 157: 645-654
        • Fiordalisi C.
        • et al.
        Introduction to AHRQ Series on Improving Translation of Evidence into Practice for the Learning Health System.
        Jt Comm J Qual Patient Saf. 2019; 45: 558-565
        • Agency for Healthcare Research and Quality. Implementing Evidence-Based Screening and Counseling for Unhealthy Alcohol Use with Epic-Based Electronic Health Record Tools
        A Guide for Clinics and Health Systems, Developed as Part of a Pilot Dissemination Project. Barclay C, et al. AHRQ Publication No.
        in: 18-EHC020-1-EF. Sep 2018
      3. Agency for Healthcare Research and Quality. Development of a Primary Care Guide for Implementing Evidence-Based Screening and Counseling for Unhealthy Alcohol Use with Epic-Based Electronic Health Record Tools: A Pilot Dissemination Project. Barclay C, Viswanathan M, Jones DE. Methods Research Report. AHRQ Publication No. 18-EHC020-EF. Sep 2018. Accessed Jun 10, 2019. https://effectivehealthcare.ahrq.gov/sites/default/files/pdf/methods-research-report-unhealthy-alcohol-use.pdf.

        • Mokdad A.H.
        • et al.
        Actual causes of death in the United States, 2000.
        JAMA. 2004 Mar; 291: 1238-1245
        • Institute for Healthcare Improvement
        How to Improve: Science of Improvement: Testing Changes.
        • Miller W.R.
        • Motivational Interviewing Rollnick S.
        Preparing People to Change Addictive Behavior.
        Guilford Press, New York1991
        • Rollnick S.
        • et al.
        Motivational interviewing.
        BMJ. 2010 Apr 27; 340: c1900
        • Perla R.J.
        • Provost L.P.
        • Murray S.K.
        The run chart: a simple analytical tool for learning from variation in healthcare processes.
        BMJ Qual Saf. 2011; 20: 46-51
        • Sampling Snowball
        Lewis-Beck M.S. Bryman A. Liao T.F. The SAGE Encyclopedia of Social Science Research Methods. Sage, Thousand Oaks, CA2004: 1043-1044
        • Hepner K.A.
        • et al.
        Rates and impact of adherence to recommended care for unhealthy alcohol use.
        J Gen Intern Med. 2019; 34: 256-263
        • Prochaska J.O.
        • Velicer W.F.
        The Transtheoretical Model of health behavior change.
        Am J Health Promot. 1997; 12: 38-48