- Email: [email protected]
Screening for At-Risk Alcohol Use and Drug Use in an Emergency Department: Integration of Screening Questions Into Electronic Triage Forms Achieves High Screening Rates
GENERAL MEDICINE/ORIGINAL RESEARCH
Screening for At-Risk Alcohol Use and Drug Use in an Emergency Department: Integration of Screening Questions Into Electronic Triage Forms Achieves High Screening Rates J. Aaron Johnson, PhD; Alexandra Woychek, MPH; Darlene Vaughan, RN; J. Paul Seale, MD
Study objective: Previous studies have shown that brief interventions for at-risk alcohol and drug use are significantly more likely to occur if patients are screened with a standardized, validated instrument, but high screening rates have traditionally been difficult to attain. Use of very brief screens can enable brief intervention specialists to focus their efforts on assessing and assisting patients most likely to need a brief intervention or more intensive treatment. This study describes the results of integrating brief substance abuse screens into an urban emergency department’s (ED’s) triage process. Methods: As part of a comprehensive initiative to increase alcohol and drug screening, brief intervention, and referral to treatment (SBIRT), 3 single-item screening questions were programmed into the electronic triage tool used in the ED to detect tobacco use, at-risk alcohol use, illicit drug use, or prescription drug misuse. Project staff conducted training sessions with nurses to ensure the questions were asked properly and ED supervisors provided ongoing performance feedback. Names of patients with positive responses to the alcohol or drug questions automatically populated a list forwarded to health education specialists, who provided assessments, brief interventions, and referrals. Results: Screening was conducted with 145,394 of 151,597 eligible patients, a 96% screening rate. Electronic reports revealed an 89% screening rate 30 days postimplementation and gradually increasing and stabilizing at approximately 97%. The overall percentage of patients screening positive for alcohol or drug use was similar to that of other ED-based studies (22%) but varied substantially by patient demographics. Conclusion: High rates of screening can be achieved if properly integrated into a clinical setting’s existing patient care processes with well-planned information technology support. [Ann Emerg Med. 2013;62:262-266.] Please see page 263 for the Editor’s Capsule Summary of this article. A feedback survey is available with each research article published on the Web at www.annemergmed.com. A podcast for this article is available at www.annemergmed.com. 0196-0644/$-see front matter Copyright © 2013 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2013.04.011
INTRODUCTION Clinician knowledge of patients’ unhealthy alcohol and drug use can prevent medical and medication errors and provide opportunity for early intervention, which can reduce alcohol and drug misuse, decrease health care use, and interrupt the trajectory to further illness, injury, and substance use disorders.1 In emergency departments (EDs) specifically, studies have shown that screening and brief intervention can decrease alcohol consumption, reduce driving after drinking, reduce injury recurrence, and decrease recurrent ED visits.2,3 Nonetheless, clinician screening and brief intervention is infrequently performed.4-7 One potential means of increasing clinician brief interventions is to have a brief initial screen administered at intake/triage to identify patients who report at-risk alcohol or drug use. Very brief single-item screens for unhealthy alcohol and drug use have been validated in primary care and 262 Annals of Emergency Medicine
demonstrate high levels of sensitivity and specificity.8-10 A previous primary care study found that patients with at-risk alcohol use were 13 times more likely to receive a brief intervention if nurses performed single-question alcohol screening during the measurement of patients’ vital signs.11 To date, however, there are no published studies of the effect of alcohol and drug screening administered by ED nurses during triage and patients’ likelihood of receiving a brief intervention. High screening rates are critical to the successful implementation of any screening, brief intervention, and referral to treatment (SBIRT) program, but previous studies have found that this goal is often difficult to attain. Previous research on screening in EDs report screening rates from 8% to 68%.7,12 Furthermore, screening rates often decline without consistent and ongoing feedback by SBIRT project staff.11,13,14 To date, ongoing screening rates of greater than 90% have been reported Volume , . : September
Johnson et al
SBIRT Screening Using Electronic Health Records
Editor’s Capsule Summary
What is already known on this topic Though contradictory, there is some evidence that screening, brief intervention, and referral to treatment (SBIRT) can be effective in modifying patients’ use of street drugs and alcohol. What question this study addressed Whether embedding brief screening questions in an electronic triage system would increase identification of patients who might benefit from SBIRT. What this study adds to our knowledge Brief screening was achieved in 97% of roughly 150,000 patients treated during the 3-year study; 22% screened positive and 60% of them received SBIRT. How this is relevant to clinical practice This study provides evidence that embedding brief screens in an electronic system can achieve nearuniversal screening. We await outcomes data on the effect of the SBIRT on the patients’ behavior.
only in the Veterans Health Administration primary care system, which has implemented universal annual alcohol screening with the 3-question Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) and reported a screening rate of 93% after adoption of a mandatory performance measure for alcohol screening in 2003.15 This article reports on SBIRT screening rates attained in the ED of a nonprofit, Level I trauma hospital after integrating brief alcohol and drug screening questions into the electronic triage system and the nurse triage process.
MATERIALS AND METHODS Integration of alcohol and drug screening questions into the electronic triage system was part of a larger effort to implement a comprehensive SBIRT program into the ED. In addition to electronic screening, the program included the introduction of continuous ED coverage by health education specialists who would provide additional screening and appropriate SBIRT services to patients screening positive. Two members of the SBIRT administrative team met with the ED nurse director and medical director to plan project implementation during a period in which many ED procedures were being reorganized to improve efficiency and decrease patients’ waiting times. After an initial meeting with ED administration, an SBIRT implementation team was formed that included representatives from the emergency physician group, nursing, and information technology (informatics). The SBIRT implementation team met Volume , . : September
1. Have you used any tobacco products in the past 12 months? ○ Yes ○ No 2. (a) WOMEN: How many times in the past 12 months have you had 4 or more drinks in a day? ○ 25 or more times ○ 13-24 times ○ 6-12 times ○ 1-5 times ○ None (b) MEN: How many times in the past 12 months have you had 5 or more drinks in a day? ○ 25 or more times ○ 13-24 times ○ 6-12 times ○ 1-5 times ○ None 3. In the past twelve months, did you smoke pot (marijuana), use another street drug, or use a prescription painkiller, stimulant, or sedative for a non-medical reason? ○ No ○ Yes If yes, Which ones?
Figure 1. Screening questions integrated into Electronic Health Record.
3 times during a 2-month period (December 2008 through January 2009) to discuss program logistics. Before initiation of the delivery of SBIRT services in February 2009, 3 single-item screening questions (Figure 1) were programmed into the electronic triage tool in the ED to detect tobacco use, at-risk alcohol use, and illicit drug use or prescription drug misuse. Integration and testing of these items in the electronic triage system required approximately 10 hours of programming by information technology staff assigned to the ED. Both the single-item alcohol and drug screening questions had been previously validated and shown to have good sensitivity and specificity in primary care.8-10 The ED electronic health record system was programmed to inform SBIRT health education specialists in 2 ways when patients gave a positive response to the alcohol or drug questions. Electronic tracking screens are located throughout the ED, listing relevant patient information, including location, length of time in the ED, and pending orders (laboratory tests, discharge, etc). Like other pending orders, an icon (a white cross in a blue box) was deployed beside the patient’s name on the ED’s electronic tracking screens to alert health education specialists that the patient required SBIRT services. Though health education specialists and other ED staff recognized the icon, it was intentionally nondescript to protect patient privacy. In addition to the icon, the patient’s name was automatically added to an electronic SBIRT patient list. The electronic health record was also programmed to deploy a red “electronic flag” if SBIRT questions were skipped, indicating an incomplete step in the triage process. Immediately before the start date for screening (February 2009), 1 member of the SBIRT administrative team conducted 15-minute training sessions with nurses during 6 consecutive nursing report sessions (3 days at 6:45 AM and 6:45 PM) to reach all nurses. These training sessions briefly introduced SBIRT, demonstrated the location of the new screening questions in the triage system, and stressed the importance of asking the questions as written. Subsequent training of new nurses was conducted periodically by nurse supervisors in the ED, and nursing supervisors provided triage nurses with ongoing performance feedback, including individual feedback to nurses who skipped or reworded SBIRT questions. Data on screening rates were collected through weekly reports automatically generated by the electronic health record and e-mailed to the Annals of Emergency Medicine 263
SBIRT Screening Using Electronic Health Records
Johnson et al 100
100 98 96 94 92 90 88 86 84 82
80 60 40 20 Mar 2009
Sep 2009
Mar 2010
Sep 2010
Mar 2011
Sep 2011
Mar 2012
Figure 2. Screening rates between March 2009 and March 2012.
SBIRT project director. The SBIRT director monitored weekly reports for consistency and notified the ED administrator of significant changes in screening rates or screen-positive rates. All adult patients aged 18 years and older were eligible for screening questions at triage and, if the screening result was positive, additional screening and appropriate services (eg, brief interventions, referral to treatment) were delivered by health education specialists. Weekly screening rates were calculated by dividing the number of patients asked the 3-question screen by the total number of adult patients presenting to the ED. Patients were considered screen positive if they answered yes to either the alcohol or drug screening question (or both). Screen positive rates were the number of screen positives divided by the total number of patients screened. The tobacco screening question was not included in screen-positive calculations.
0
March Sept March Sept March Sept March 2009 2009 2010 2010 2011 2011 2012
Figure 3. Percentage of patients screening positive March 2009 to March 2012. Table. Comparison of screen-positive rates by patient demographics. Demographic Sex Male Female Race Black White Other Age, y 18–29 30–44 45–54 ⱖ55
Prescreen Positive (%) 30.2 13.6 22.0 20.0 21.6 25.3 28.1 28.4 10.6
RESULTS During the 3-year period reported here (March 2009 to March 2012), 151,597 adults patients were eligible for screening, with 145,394 (96%) receiving the 3-question screen. Patients presenting to the ED were 55% women, with an equal distribution of black and white patients (47% each) and a small percentage of other races and ethnicities (6%). More than one third of patients were aged 55 years or older (36.8%), with relatively equal numbers of patients aged 18 to 29 years (21.1%), 30 to 44 years (22.6%), and 45 to 54 years (19.4%). During the first 30 days after implementation of SBIRT services, 89% of patients were screened for alcohol and drugs (Figure 2). Screening rates gradually increased during the next 18 months to approximately 97%. Since reaching a plateau in September 2010, weekly rates have remained consistent. There is little information about the patients who are not screened. There were no previous exclusions instructing triage nurses to skip certain subpopulations. Those missed were likely critically ill patients or patients otherwise incapacitated when they presented to the ED. The percentage of patients screening positive for either drug use or at-risk alcohol use (tobacco users are excluded from this rate) ranged from 19.5% to 26% during the 3-year period. Of 264 Annals of Emergency Medicine
the 145,394 patients screened, 31,971 (22%) screened positive for at-risk alcohol or drug use, a rate that is consistent with that of previous ED-based studies.7,16,17 As with screening rates, the screen-positive rates stabilized in September 2010 as nurses became comfortable with the screening process and the need to ask questions in a standardized manner (Figure 3). The Table compares the screen-positive rates across patient demographics, with women, whites, and persons aged 55 years and older having lower screen-positive rates compared with men, blacks, and younger individuals, respectively. With such a large sample, even small variations in rates (eg, the difference between blacks and whites) are statistically, if not substantively, significant. With approximately 200 patients per week screening positive, health education specialists staffing the ED were not able to consult with all patients eligible for SBIRT services, despite their best efforts. Often, patients with minor illnesses or injuries would be treated by ED staff and discharged before health education specialists were able to administer SBIRT services. Health education specialists completed services with approximately 60% of screen-positive patients during this period. Volume , . : September
Johnson et al
LIMITATIONS Findings of this single-site study may not be applicable to all ED settings. Documentation of answers to alcohol screening questions does not guarantee that patients were asked validated alcohol screening questions exactly as they were written. All patients with positive screen results did not receive brief interventions because of limited SBIRT staffing and the fast pace of many ED visits. The medicolegal risk of positive brief screens that are not addressed is unknown, although positive screen results only indicate a need for further assessment, not imminent danger or risk to patients.
DISCUSSION Following the work of Bradley et al15 in the Veterans Health Administration, this is the second published report of a screening and intervention program that has attained and sustained high rates (⬎90%) of substance use screening during an extended period and the first to accomplish this in an ED and to report such high rates of screening for both alcohol and drugs, to our knowledge. Factors contributing to program success came from a variety of areas mentioned in conceptual models of implementation of health care innovations, including aspects of the “inner setting” (the organization of this particular ED), characteristics of the intervention itself, the process of implementation, and a key aspect of the project’s “outer setting” (the larger context of the project, in particular the availability of grant funding).18 In this study, external funding served as the impetus to introduce a simple, adaptable intervention into a busy urban ED with strong staff communication and training processes during a period of ongoing change focused on quality improvement. The SBIRT administrative team engaged opinion leaders and internal implementation personnel and collaborated with ED staff in execution and ongoing evaluation. Rather than adding an entirely new procedure to ED patient flow, project directors integrated a brief 3question screening process, similar in length to the 3question AUDIT-C used in the Veterans Health Administration alcohol screening process,15 into the existing electronic triage system. Information technology personnel played a key part in implementation by integrating screening questions into the electronic triage form, flagging patients with incomplete screens, identifying patients with positive screen results by adding SBIRT icons to the ED’s electronic triage system, and automatically generating a list of screenpositive patients for health education specialists. The electronic integration of this process offered the additional advantage of being able to monitor staff performance and provide timely feedback to both nurses and SBIRT specialists. The 24-hour presence of SBIRT specialists who followed up on positive screen results institutionalized the SBIRT process. Although the larger SBIRT program (eg, continuous coverage by health education specialists) may change with the end of grant funding, the initial screening conducted by nurses at triage has become a standard of care Volume , . : September
SBIRT Screening Using Electronic Health Records within this ED. During 18 months, universal substance use screening became a standard of care, with only a small percentage (3%) of patients with life-threatening emergencies (eg, severe trauma, chest pain) or altered mental status remaining unscreened. It is possible to achieve long-term high rates of nurseadministered screening for alcohol, tobacco, and other drugs. Aspects of this program believed to be important in achieving and sustaining high screening rates were the use of very brief screening tools; integration of substance use screening into an existing step of the ED clinical care system; key input from information technology in creating simple, well-placed electronic protocols that discouraged incomplete records and easily identified patients needing SBIRT services; ongoing performance feedback; and a constant SBIRT “presence” within the setting. Further studies in other ED settings should broaden knowledge of factors that are critical to achieving high screening rates. Supervising editor: David L. Schriger, MD, MPH Author affiliations: From the Mercer University School of Medicine, Macon, GA (Johnson, Woychek, Seale); and the Department of Family Medicine (Johnson, Woychek, Seale) and Department of Emergency Medicine (Vaughan), Medical Center of Central Georgia, Macon, GA. Author contributions: JAJ and JPS were responsible for study design. AW was responsible for data analysis. DV was responsible for study execution. All authors were responsible for drafting the article. JAJ takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist. This research was funded by grant T1019545 from the Substance Abuse and Mental Health Services Administration. Publication dates: Received for publication September 14, 2012. Revisions received February 16, 2013, and March 22, 2013. Accepted for publication April 12, 2013. Available online May 18, 2013. Address for correspondence: J. Aaron Johnson, PhD, E-mail [email protected].
REFERENCES 1. Madras BK, Compton WM, Avula D, et al. Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: comparison at intake and 6 months later. Drug Alcohol Depend. 2009;99:280-295. 2. Desy PM, Howard PK, Perhats C, et al. Alcohol screening, brief intervention, and referral to treatment conducted by emergency nurses: an impact evaluation. J Emerg Nurs. 2010;36:538-545.
Annals of Emergency Medicine 265
SBIRT Screening Using Electronic Health Records
Johnson et al
3. D’Onofrio G, Fiellin DA, Pantalon MV, et al. A brief intervention reduces hazardous and harmful drinking in emergency department patients. Ann Emerg Med. 2012;60:181-192. 4. Cunningham RM, Harrison SR, McKay MP, et al. National survey of emergency department alcohol screening and intervention practices. Ann Emerg Med. 2010;55:556-562. 5. Schermer CR, Gentilello LM, Hoyt DB, et al. National survey of trauma surgeons’ use of alcohol screening and brief intervention. J Trauma. 2003;55:849-856. 6. Babor TE, Higgins-Biddle J, Dauser D, et al. Alcohol screening and brief intervention in primary care settings: implementation models and predictors. J Stud Alcohol. 2005;66:361-368. 7. Désy PM, Perhats C. Alcohol screening, brief intervention, and referral in the emergency department: an implementation study. J Emerg Nurs.2008;34:11-19. 8. Smith PC, Schmidt SM, Allensworth-Davies D, et al. Primary care validation of a single-question alcohol screening test. J Gen Intern Med. 2009;24:783-788. 9. Smith PC, Schmidt SM, Allensworth-Davies D, et al. A singlequestion screening test for drug use in primary care. Arch Intern Med. 2010;170:1155-1160. 10. Dawson DA, Pulay AJ, Grant BF. A comparison of two single-item screeners for hazardous drinking and alcohol use disorder. Alcohol Clin Exp Res. 2010;34:364-374. 11. Seale JP, Shellenberger S, Velasquez MM, et al. Impact of vital signs screening and clinician prompting on alcohol and tobacco screening and intervention rates: a pre-post intervention comparison. BMC Fam Pract. 2010;11:18.
12. Cherpitel CJ, Moskalewicz J, Swiatkiewicz G, et al. Screening, brief intervention, and referral to treatment (SBIRT) in a Polish emergency department: three-month outcomes of a randomized, controlled clinical trial. J Stud Alcohol Drugs. 2009;70:982-990. 13. Seale JP, Shellenberger S, Boltri JM, et al. Effects of screening and brief intervention training on resident and faculty alcohol intervention behaviours: a pre- post-intervention assessment. BMC Fam Pract. 2005;6:46. 14. Johnson JA, Seale JP, Shellenberger S, et al. Impact of systemlevel changes and training on alcohol screening and brief intervention in a family medicine residency clinic: a pilot study. Subst Abuse Treat Prev Policy. 2013;8:9. 15. Bradley KA, Williams EC, Achtmeyer CE, et al. Implementation of evidence-based alcohol screening in the Veterans Health Administration. Am J Manag Care. 2006;12:596-606. 16. D’Onofrio G, Degutis LC. Integrating Project ASSERT: a screening, intervention, and referral to treatment program for unhealthy alcohol and drug use into an urban emergency department. Acad Emerg Med. 2010;17:903-911. 17. Hankin A, Daugherty M, Bethea A, et al. The emergency department as a prevention site: a demographic analysis of substance use among ED patients. Drug Alcohol Depend. http:// dx.doi.org/10.1016/j.drugalcdep.2012.10.027. 18. Damschroder LJ, Aron DC, Keith RE, et al. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4:50.
Did you know? Annals accepts audio and video files as ancillaries to the main article.
Visit http://www.annemergmed.com/content/instauth/ for more details!
266 Annals of Emergency Medicine
Volume , . : September
Screening for At-Risk Alcohol Use and Drug Use in an Emergency Department: Integration of Screening Questions Into Electronic Triage Forms Achieves High Screening Rates J. Aaron Johnson, PhD; Alexandra Woychek, MPH; Darlene Vaughan, RN; J. Paul Seale, MD
Study objective: Previous studies have shown that brief interventions for at-risk alcohol and drug use are significantly more likely to occur if patients are screened with a standardized, validated instrument, but high screening rates have traditionally been difficult to attain. Use of very brief screens can enable brief intervention specialists to focus their efforts on assessing and assisting patients most likely to need a brief intervention or more intensive treatment. This study describes the results of integrating brief substance abuse screens into an urban emergency department’s (ED’s) triage process. Methods: As part of a comprehensive initiative to increase alcohol and drug screening, brief intervention, and referral to treatment (SBIRT), 3 single-item screening questions were programmed into the electronic triage tool used in the ED to detect tobacco use, at-risk alcohol use, illicit drug use, or prescription drug misuse. Project staff conducted training sessions with nurses to ensure the questions were asked properly and ED supervisors provided ongoing performance feedback. Names of patients with positive responses to the alcohol or drug questions automatically populated a list forwarded to health education specialists, who provided assessments, brief interventions, and referrals. Results: Screening was conducted with 145,394 of 151,597 eligible patients, a 96% screening rate. Electronic reports revealed an 89% screening rate 30 days postimplementation and gradually increasing and stabilizing at approximately 97%. The overall percentage of patients screening positive for alcohol or drug use was similar to that of other ED-based studies (22%) but varied substantially by patient demographics. Conclusion: High rates of screening can be achieved if properly integrated into a clinical setting’s existing patient care processes with well-planned information technology support. [Ann Emerg Med. 2013;62:262-266.] Please see page 263 for the Editor’s Capsule Summary of this article. A feedback survey is available with each research article published on the Web at www.annemergmed.com. A podcast for this article is available at www.annemergmed.com. 0196-0644/$-see front matter Copyright © 2013 by the American College of Emergency Physicians. http://dx.doi.org/10.1016/j.annemergmed.2013.04.011
INTRODUCTION Clinician knowledge of patients’ unhealthy alcohol and drug use can prevent medical and medication errors and provide opportunity for early intervention, which can reduce alcohol and drug misuse, decrease health care use, and interrupt the trajectory to further illness, injury, and substance use disorders.1 In emergency departments (EDs) specifically, studies have shown that screening and brief intervention can decrease alcohol consumption, reduce driving after drinking, reduce injury recurrence, and decrease recurrent ED visits.2,3 Nonetheless, clinician screening and brief intervention is infrequently performed.4-7 One potential means of increasing clinician brief interventions is to have a brief initial screen administered at intake/triage to identify patients who report at-risk alcohol or drug use. Very brief single-item screens for unhealthy alcohol and drug use have been validated in primary care and 262 Annals of Emergency Medicine
demonstrate high levels of sensitivity and specificity.8-10 A previous primary care study found that patients with at-risk alcohol use were 13 times more likely to receive a brief intervention if nurses performed single-question alcohol screening during the measurement of patients’ vital signs.11 To date, however, there are no published studies of the effect of alcohol and drug screening administered by ED nurses during triage and patients’ likelihood of receiving a brief intervention. High screening rates are critical to the successful implementation of any screening, brief intervention, and referral to treatment (SBIRT) program, but previous studies have found that this goal is often difficult to attain. Previous research on screening in EDs report screening rates from 8% to 68%.7,12 Furthermore, screening rates often decline without consistent and ongoing feedback by SBIRT project staff.11,13,14 To date, ongoing screening rates of greater than 90% have been reported Volume , . : September
Johnson et al
SBIRT Screening Using Electronic Health Records
Editor’s Capsule Summary
What is already known on this topic Though contradictory, there is some evidence that screening, brief intervention, and referral to treatment (SBIRT) can be effective in modifying patients’ use of street drugs and alcohol. What question this study addressed Whether embedding brief screening questions in an electronic triage system would increase identification of patients who might benefit from SBIRT. What this study adds to our knowledge Brief screening was achieved in 97% of roughly 150,000 patients treated during the 3-year study; 22% screened positive and 60% of them received SBIRT. How this is relevant to clinical practice This study provides evidence that embedding brief screens in an electronic system can achieve nearuniversal screening. We await outcomes data on the effect of the SBIRT on the patients’ behavior.
only in the Veterans Health Administration primary care system, which has implemented universal annual alcohol screening with the 3-question Alcohol Use Disorders Identification Test - Consumption (AUDIT-C) and reported a screening rate of 93% after adoption of a mandatory performance measure for alcohol screening in 2003.15 This article reports on SBIRT screening rates attained in the ED of a nonprofit, Level I trauma hospital after integrating brief alcohol and drug screening questions into the electronic triage system and the nurse triage process.
MATERIALS AND METHODS Integration of alcohol and drug screening questions into the electronic triage system was part of a larger effort to implement a comprehensive SBIRT program into the ED. In addition to electronic screening, the program included the introduction of continuous ED coverage by health education specialists who would provide additional screening and appropriate SBIRT services to patients screening positive. Two members of the SBIRT administrative team met with the ED nurse director and medical director to plan project implementation during a period in which many ED procedures were being reorganized to improve efficiency and decrease patients’ waiting times. After an initial meeting with ED administration, an SBIRT implementation team was formed that included representatives from the emergency physician group, nursing, and information technology (informatics). The SBIRT implementation team met Volume , . : September
1. Have you used any tobacco products in the past 12 months? ○ Yes ○ No 2. (a) WOMEN: How many times in the past 12 months have you had 4 or more drinks in a day? ○ 25 or more times ○ 13-24 times ○ 6-12 times ○ 1-5 times ○ None (b) MEN: How many times in the past 12 months have you had 5 or more drinks in a day? ○ 25 or more times ○ 13-24 times ○ 6-12 times ○ 1-5 times ○ None 3. In the past twelve months, did you smoke pot (marijuana), use another street drug, or use a prescription painkiller, stimulant, or sedative for a non-medical reason? ○ No ○ Yes If yes, Which ones?
Figure 1. Screening questions integrated into Electronic Health Record.
3 times during a 2-month period (December 2008 through January 2009) to discuss program logistics. Before initiation of the delivery of SBIRT services in February 2009, 3 single-item screening questions (Figure 1) were programmed into the electronic triage tool in the ED to detect tobacco use, at-risk alcohol use, and illicit drug use or prescription drug misuse. Integration and testing of these items in the electronic triage system required approximately 10 hours of programming by information technology staff assigned to the ED. Both the single-item alcohol and drug screening questions had been previously validated and shown to have good sensitivity and specificity in primary care.8-10 The ED electronic health record system was programmed to inform SBIRT health education specialists in 2 ways when patients gave a positive response to the alcohol or drug questions. Electronic tracking screens are located throughout the ED, listing relevant patient information, including location, length of time in the ED, and pending orders (laboratory tests, discharge, etc). Like other pending orders, an icon (a white cross in a blue box) was deployed beside the patient’s name on the ED’s electronic tracking screens to alert health education specialists that the patient required SBIRT services. Though health education specialists and other ED staff recognized the icon, it was intentionally nondescript to protect patient privacy. In addition to the icon, the patient’s name was automatically added to an electronic SBIRT patient list. The electronic health record was also programmed to deploy a red “electronic flag” if SBIRT questions were skipped, indicating an incomplete step in the triage process. Immediately before the start date for screening (February 2009), 1 member of the SBIRT administrative team conducted 15-minute training sessions with nurses during 6 consecutive nursing report sessions (3 days at 6:45 AM and 6:45 PM) to reach all nurses. These training sessions briefly introduced SBIRT, demonstrated the location of the new screening questions in the triage system, and stressed the importance of asking the questions as written. Subsequent training of new nurses was conducted periodically by nurse supervisors in the ED, and nursing supervisors provided triage nurses with ongoing performance feedback, including individual feedback to nurses who skipped or reworded SBIRT questions. Data on screening rates were collected through weekly reports automatically generated by the electronic health record and e-mailed to the Annals of Emergency Medicine 263
SBIRT Screening Using Electronic Health Records
Johnson et al 100
100 98 96 94 92 90 88 86 84 82
80 60 40 20 Mar 2009
Sep 2009
Mar 2010
Sep 2010
Mar 2011
Sep 2011
Mar 2012
Figure 2. Screening rates between March 2009 and March 2012.
SBIRT project director. The SBIRT director monitored weekly reports for consistency and notified the ED administrator of significant changes in screening rates or screen-positive rates. All adult patients aged 18 years and older were eligible for screening questions at triage and, if the screening result was positive, additional screening and appropriate services (eg, brief interventions, referral to treatment) were delivered by health education specialists. Weekly screening rates were calculated by dividing the number of patients asked the 3-question screen by the total number of adult patients presenting to the ED. Patients were considered screen positive if they answered yes to either the alcohol or drug screening question (or both). Screen positive rates were the number of screen positives divided by the total number of patients screened. The tobacco screening question was not included in screen-positive calculations.
0
March Sept March Sept March Sept March 2009 2009 2010 2010 2011 2011 2012
Figure 3. Percentage of patients screening positive March 2009 to March 2012. Table. Comparison of screen-positive rates by patient demographics. Demographic Sex Male Female Race Black White Other Age, y 18–29 30–44 45–54 ⱖ55
Prescreen Positive (%) 30.2 13.6 22.0 20.0 21.6 25.3 28.1 28.4 10.6
RESULTS During the 3-year period reported here (March 2009 to March 2012), 151,597 adults patients were eligible for screening, with 145,394 (96%) receiving the 3-question screen. Patients presenting to the ED were 55% women, with an equal distribution of black and white patients (47% each) and a small percentage of other races and ethnicities (6%). More than one third of patients were aged 55 years or older (36.8%), with relatively equal numbers of patients aged 18 to 29 years (21.1%), 30 to 44 years (22.6%), and 45 to 54 years (19.4%). During the first 30 days after implementation of SBIRT services, 89% of patients were screened for alcohol and drugs (Figure 2). Screening rates gradually increased during the next 18 months to approximately 97%. Since reaching a plateau in September 2010, weekly rates have remained consistent. There is little information about the patients who are not screened. There were no previous exclusions instructing triage nurses to skip certain subpopulations. Those missed were likely critically ill patients or patients otherwise incapacitated when they presented to the ED. The percentage of patients screening positive for either drug use or at-risk alcohol use (tobacco users are excluded from this rate) ranged from 19.5% to 26% during the 3-year period. Of 264 Annals of Emergency Medicine
the 145,394 patients screened, 31,971 (22%) screened positive for at-risk alcohol or drug use, a rate that is consistent with that of previous ED-based studies.7,16,17 As with screening rates, the screen-positive rates stabilized in September 2010 as nurses became comfortable with the screening process and the need to ask questions in a standardized manner (Figure 3). The Table compares the screen-positive rates across patient demographics, with women, whites, and persons aged 55 years and older having lower screen-positive rates compared with men, blacks, and younger individuals, respectively. With such a large sample, even small variations in rates (eg, the difference between blacks and whites) are statistically, if not substantively, significant. With approximately 200 patients per week screening positive, health education specialists staffing the ED were not able to consult with all patients eligible for SBIRT services, despite their best efforts. Often, patients with minor illnesses or injuries would be treated by ED staff and discharged before health education specialists were able to administer SBIRT services. Health education specialists completed services with approximately 60% of screen-positive patients during this period. Volume , . : September
Johnson et al
LIMITATIONS Findings of this single-site study may not be applicable to all ED settings. Documentation of answers to alcohol screening questions does not guarantee that patients were asked validated alcohol screening questions exactly as they were written. All patients with positive screen results did not receive brief interventions because of limited SBIRT staffing and the fast pace of many ED visits. The medicolegal risk of positive brief screens that are not addressed is unknown, although positive screen results only indicate a need for further assessment, not imminent danger or risk to patients.
DISCUSSION Following the work of Bradley et al15 in the Veterans Health Administration, this is the second published report of a screening and intervention program that has attained and sustained high rates (⬎90%) of substance use screening during an extended period and the first to accomplish this in an ED and to report such high rates of screening for both alcohol and drugs, to our knowledge. Factors contributing to program success came from a variety of areas mentioned in conceptual models of implementation of health care innovations, including aspects of the “inner setting” (the organization of this particular ED), characteristics of the intervention itself, the process of implementation, and a key aspect of the project’s “outer setting” (the larger context of the project, in particular the availability of grant funding).18 In this study, external funding served as the impetus to introduce a simple, adaptable intervention into a busy urban ED with strong staff communication and training processes during a period of ongoing change focused on quality improvement. The SBIRT administrative team engaged opinion leaders and internal implementation personnel and collaborated with ED staff in execution and ongoing evaluation. Rather than adding an entirely new procedure to ED patient flow, project directors integrated a brief 3question screening process, similar in length to the 3question AUDIT-C used in the Veterans Health Administration alcohol screening process,15 into the existing electronic triage system. Information technology personnel played a key part in implementation by integrating screening questions into the electronic triage form, flagging patients with incomplete screens, identifying patients with positive screen results by adding SBIRT icons to the ED’s electronic triage system, and automatically generating a list of screenpositive patients for health education specialists. The electronic integration of this process offered the additional advantage of being able to monitor staff performance and provide timely feedback to both nurses and SBIRT specialists. The 24-hour presence of SBIRT specialists who followed up on positive screen results institutionalized the SBIRT process. Although the larger SBIRT program (eg, continuous coverage by health education specialists) may change with the end of grant funding, the initial screening conducted by nurses at triage has become a standard of care Volume , . : September
SBIRT Screening Using Electronic Health Records within this ED. During 18 months, universal substance use screening became a standard of care, with only a small percentage (3%) of patients with life-threatening emergencies (eg, severe trauma, chest pain) or altered mental status remaining unscreened. It is possible to achieve long-term high rates of nurseadministered screening for alcohol, tobacco, and other drugs. Aspects of this program believed to be important in achieving and sustaining high screening rates were the use of very brief screening tools; integration of substance use screening into an existing step of the ED clinical care system; key input from information technology in creating simple, well-placed electronic protocols that discouraged incomplete records and easily identified patients needing SBIRT services; ongoing performance feedback; and a constant SBIRT “presence” within the setting. Further studies in other ED settings should broaden knowledge of factors that are critical to achieving high screening rates. Supervising editor: David L. Schriger, MD, MPH Author affiliations: From the Mercer University School of Medicine, Macon, GA (Johnson, Woychek, Seale); and the Department of Family Medicine (Johnson, Woychek, Seale) and Department of Emergency Medicine (Vaughan), Medical Center of Central Georgia, Macon, GA. Author contributions: JAJ and JPS were responsible for study design. AW was responsible for data analysis. DV was responsible for study execution. All authors were responsible for drafting the article. JAJ takes responsibility for the paper as a whole. Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist. This research was funded by grant T1019545 from the Substance Abuse and Mental Health Services Administration. Publication dates: Received for publication September 14, 2012. Revisions received February 16, 2013, and March 22, 2013. Accepted for publication April 12, 2013. Available online May 18, 2013. Address for correspondence: J. Aaron Johnson, PhD, E-mail [email protected].
REFERENCES 1. Madras BK, Compton WM, Avula D, et al. Screening, brief interventions, referral to treatment (SBIRT) for illicit drug and alcohol use at multiple healthcare sites: comparison at intake and 6 months later. Drug Alcohol Depend. 2009;99:280-295. 2. Desy PM, Howard PK, Perhats C, et al. Alcohol screening, brief intervention, and referral to treatment conducted by emergency nurses: an impact evaluation. J Emerg Nurs. 2010;36:538-545.
Annals of Emergency Medicine 265
SBIRT Screening Using Electronic Health Records
Johnson et al
3. D’Onofrio G, Fiellin DA, Pantalon MV, et al. A brief intervention reduces hazardous and harmful drinking in emergency department patients. Ann Emerg Med. 2012;60:181-192. 4. Cunningham RM, Harrison SR, McKay MP, et al. National survey of emergency department alcohol screening and intervention practices. Ann Emerg Med. 2010;55:556-562. 5. Schermer CR, Gentilello LM, Hoyt DB, et al. National survey of trauma surgeons’ use of alcohol screening and brief intervention. J Trauma. 2003;55:849-856. 6. Babor TE, Higgins-Biddle J, Dauser D, et al. Alcohol screening and brief intervention in primary care settings: implementation models and predictors. J Stud Alcohol. 2005;66:361-368. 7. Désy PM, Perhats C. Alcohol screening, brief intervention, and referral in the emergency department: an implementation study. J Emerg Nurs.2008;34:11-19. 8. Smith PC, Schmidt SM, Allensworth-Davies D, et al. Primary care validation of a single-question alcohol screening test. J Gen Intern Med. 2009;24:783-788. 9. Smith PC, Schmidt SM, Allensworth-Davies D, et al. A singlequestion screening test for drug use in primary care. Arch Intern Med. 2010;170:1155-1160. 10. Dawson DA, Pulay AJ, Grant BF. A comparison of two single-item screeners for hazardous drinking and alcohol use disorder. Alcohol Clin Exp Res. 2010;34:364-374. 11. Seale JP, Shellenberger S, Velasquez MM, et al. Impact of vital signs screening and clinician prompting on alcohol and tobacco screening and intervention rates: a pre-post intervention comparison. BMC Fam Pract. 2010;11:18.
12. Cherpitel CJ, Moskalewicz J, Swiatkiewicz G, et al. Screening, brief intervention, and referral to treatment (SBIRT) in a Polish emergency department: three-month outcomes of a randomized, controlled clinical trial. J Stud Alcohol Drugs. 2009;70:982-990. 13. Seale JP, Shellenberger S, Boltri JM, et al. Effects of screening and brief intervention training on resident and faculty alcohol intervention behaviours: a pre- post-intervention assessment. BMC Fam Pract. 2005;6:46. 14. Johnson JA, Seale JP, Shellenberger S, et al. Impact of systemlevel changes and training on alcohol screening and brief intervention in a family medicine residency clinic: a pilot study. Subst Abuse Treat Prev Policy. 2013;8:9. 15. Bradley KA, Williams EC, Achtmeyer CE, et al. Implementation of evidence-based alcohol screening in the Veterans Health Administration. Am J Manag Care. 2006;12:596-606. 16. D’Onofrio G, Degutis LC. Integrating Project ASSERT: a screening, intervention, and referral to treatment program for unhealthy alcohol and drug use into an urban emergency department. Acad Emerg Med. 2010;17:903-911. 17. Hankin A, Daugherty M, Bethea A, et al. The emergency department as a prevention site: a demographic analysis of substance use among ED patients. Drug Alcohol Depend. http:// dx.doi.org/10.1016/j.drugalcdep.2012.10.027. 18. Damschroder LJ, Aron DC, Keith RE, et al. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science. Implement Sci. 2009;4:50.
Did you know? Annals accepts audio and video files as ancillaries to the main article.
Visit http://www.annemergmed.com/content/instauth/ for more details!
266 Annals of Emergency Medicine
Volume , . : September