More than 30% of adults in the United States are estimated to drink alcohol at risky levels (7 standard drinks per week or 3 drinks per occasion for women and individuals older than 65 years, and 14 standard drinks per week or 4 drinks per occasion for men younger than 65 years) or meet the criteria for harmful or dependent alcohol use (Saitz, 2005). The U.S. Preventive Services Task Force recommends both screening and brief intervention (SBI) for alcohol misuse in primary care settings (Moyer, 2013), partly because risky alcohol use may be asymptomatic (Saitz, 2005). Therefore, the use of validated adult screening tools, such as the Alcohol Use Disorders Identification Test (AUDIT; Babor, Biddle-Higgins, Saunders, & Monteiro, 2001), the abbreviated AUDIT-C, or single-question screening test (Smith, Schmidt, Allensworth-Davies, & Saitz, 2009), may lead to the identification of potential misuse and the need for an intervention (Moyer, 2013). Potential negative consequences of excessive alcohol use are both chronic and acute and include alcoholic liver disease, gastroesophageal hemorrhage, liver cirrhosis, fall injuries, and motor vehicle crashes, among numerous others (Centers for Disease Control and Prevention, 2015). The total estimated combined cost of binge drinking, heavy drinking, underage drinking, and alcohol consumption during pregnancy in the United States was $249 billion in 2010 (Sacks, Gonzales, Bouchery, Tomedi, & Brewer, 2015). Encouraging primary care providers to incorporate SBI into practice is congruent with nurse practitioner competencies (National Organization of Nurse Practitioner Faculties, 2013) and with the Institute of Medicine report on the Future of Nursing (2010), which recommend that advanced practice nurses screen for chronic conditions. To help in the development of SBI skills, the Substance Abuse and Mental Health Services Administration (SAMHSA) expanded training grant opportunities in 2013 for screening, brief intervention, and referral to treatment (SBIRT) education programs to multiple groups of medical professionals, including nurses (SAMHSA, 2015b).
Scholars have assessed preliminary evidence related to nurse-delivered SBI and SBIRT implementation models and concluded that nurse-delivered SBIRT services are feasible (Finnell, 2012). The International Nurses Society on Addictions and the Emergency Nurses Association recommend that “nurses in all specialties and practice settings be prepared to deliver SBIRT to identify and effectively respond to alcohol use and related disorders across the lifespan” (Strobbe, Perhats, & Broyles, 2013, p. 204). Other scholars and health care professionals have concluded that nurse practitioners (NPs) are well positioned to implement SBIRT in their clinical practice (Mitchell et al., 2015) and that nurses have important existing skill sets in the areas used by SBIRT (Fischer, Pace, Reimann, & Swenson, 2013). NPs are educated in health promotion and disease prevention strategies in their professional programs. Therefore, they are well positioned to identify alcohol misuse that may potentially lead to adverse problems and offer interventions that may alter a chronic alcohol-related disease trajectory.
Although SBIRT training curricula for nurses supported by SAMHSA have demonstrated positive outcomes (Gotham, Knopf-Amelung, Krom, Stilen, & Kohnle, 2015), the factors that facilitate or inhibit SBIRT training, as well as the implications of training for clinical practice, are not yet fully understood (Broyles et al., 2013). SBIRT training needs to impart desired procedural and clinical skills such as motivational interviewing (MI), as well as address and mitigate barriers that may affect clinical practice, such as perceptions of insufficient time, lack of knowledge, negative attitudes toward individuals who use substances (Puskar et al., 2013), low self-efficacy, and perceived financial limitations (Holland, Pringle, & Barbetti, 2009). These barriers may affect a variety of different components of SBIRT, including screening, providing appropriate services to patients based on screening outcomes, referral practices, and, in some cases, appropriate documentation of these efforts (Agley, Gassman, Vannerson, & Crabb, 2014). Although several studies have examined the barriers and facilitators to the implementation of SBIRT by nurses (Broyles et al., 2012; Puskar et al., 2013), little research has linked these constructs to specific components of SBIRT performance. For each distinct component of SBIRT (e.g., screening), there is utility for both curriculum development and training evaluation in identifying the variables that predict completion and noncompletion of that component in clinical practice. In 2003, Gassman identified several factors (each assessed using multiquestion scales) that predicted nurses' and physicians' stated intentions to perform SBIRT for alcohol abuse in primary care. Building on that research, the current study examined whether those factors assessing knowledge, attitudes, and beliefs about SBIRT, which were measured following a mixed-methods SBIRT/MI training series, predicted whether 21 family nurse practitioner (FNP) students screened for alcohol use during actual clinical patient encounters. The purpose of doing so was to foster the further development of a valid means of assessing whether FNP students will screen for alcohol use when they begin performing clinical encounters with real patients.
Training and Data Collection
In fall 2014, 21 FNP students at a midwestern, urban, public university enrolled in a mixed-methods SBIRT/MI training program. The study was approved by the university institutional review board. The sample was primarily female and White non-Hispanic (95.5% for both categories). The curriculum components included a 1-hour online SBIRT module, a 1-hour interactive online MI module, and a 4-hour interactive face-to-face training that covered SBIRT and MI using real play, role-play, and lecture. Each student completed a paper survey immediately following the face-to-face training (which occurred subsequent to completion of the online modules). The survey instrument consisted of 51 items designed to inform educational and programmatic improvement (Gassman, 2003) and facilitate reporting to the funding agency according to the federal Government Performance and Results Modernization Act of 2010 (SAMHSA, 2015a). The survey was completed by all 21 FNP students.
Following completion of the program in its entirety, students implemented SBIRT during their clinical practice-specific management courses, which required approximately 500 clinical hours (encounters occurred from September 11, 2014, through July 20, 2015, hereafter called the study period). Students also completed eight questions regarding their use of SBIRT/MI following each encounter in the Typhon Group's Nurse Practitioner Student Tracking (NPST™) System, an electronic student tracking system that logs students' clinical encounters throughout their program.
In total, 21 students completed 2,164 clinical encounter SBIRT/MI evaluations during the study period. Multiple clinical encounters were removed from the data set. Cases were eliminated if open-ended clinical notes revealed that the patient was younger than age 12 years (n = 72), the patient was being seen for extremely serious physical diagnoses requiring immediate action (n = 11), the nurse was present only as an observer to a physician visit (n = 29), the patient was developmentally disabled to the extent he or she could not communicate or exhibited other issues with communication (n = 6), or the evaluation was a duplicate that was mistakenly entered (n = 4). The final set of clinical encounters comprised 2,042 responses. The mean number of clinical encounters per student was 92.5, and the median number of encounters per student was 29.5.
Item Selection and Scale Development
Because the purpose of this study was to determine which knowledge, attitude, and belief scales, if any, predicted nurses' performance of alcohol screening in clinical settings, items were selected from the survey instrument to match that purpose. In prior research, 21 of the survey items (across six scales) were investigated for their potential to predict the likelihood of performing SBI among physicians and nurses (Gassman, 2003). Preliminary evidence also suggests that four items (one scale) from the Government Performance and Results Act item set (SAMHSA, 2015a) covering trainee satisfaction may have the ability to discriminate between affective responses to differing training structures (Agley, Gassman, DeSalle, Vannerson, Carlson, & Crabb, 2014), so they were included as well. After assessing the interrater reliability of each of the seven scales, two scales were removed due to Cronbach's alpha levels lower than .600 (Henson, 2001). The remaining 17 items comprised five scales, which assessed satisfaction with the training (alpha = .935), self-reported competence with SBI behaviors (alpha = .822), perceived role legitimacy (alpha = .766), skepticism of behavioral health care (alpha = .845), and time utilization and compensation (alpha = .611). The final scale was retained despite marginally adequate internal reliability because it was composed of two questions covering differing concepts, and heterogeneous constructs and short scale length may artificially deflate alpha levels (Tavakol & Dennick, 2011). Each scale and its component items is provided in the Table.
Scales and Component Items Used in Study Analyses
Computation of the Dependent Variable
To assess the clinical performance of alcohol screening by the FNP students, the researchers assessed their responses to the question “Was the client/patient screened for alcohol use?” Students were able to indicate whether they screened (yes/no) using the AUDIT-10 for adults or the CRAFFT screening test for adolescents, and whether the screening was positive or negative. All cases from each student were aggregated and all instances of “no” responses were calculated as a ratio of the total clinical encounters. For example, one student completed 171 clinical encounters and failed to screen during 19 of them so was assigned a noncompletion ratio of 0.111 (11.1%).
To determine the extent to which each of the scales predicted alcohol screening in clinical practice, the researchers used a general linear model with “family(binomial)” and “link(logit),” with “robust” standard error parameters in STATA® version 13 software. A general linear model with those parameters is considered the best model estimate for data when the dependent variable is a proportion that includes instances of zero that are not structural (McDowell & Cox, 2015), both of which were the case with this data set. The two-tailed alpha level was set a priori at .05.
Of the five scales that were included in the model, only one significantly predicted alcohol screening in clinical encounters during the study period. FNP students who reported higher levels of perceived competence in their posttraining surveys were more likely to screen for alcohol in the clinical setting (adjusted odds ratio = 4.49, Z = 3.46, p = .001, 95% confidence interval [1.92, 10.53]). Specifically, each of the four questions composing the Competence Scale had response options ranging from 1 (strongly agree) to 5 (strongly disagree), where 1 was structured as the response that endorsed competence. Thus, the continuous value represented by the Competence Scale ranged from 1 = high perceived competence to 5 = low perceived competence. The Margins command in the STATA software provided predicted probabilities of the dependent variable being 1 (in this case, 1 would represent 100% noncompletion of screening for alcohol abuse across all clinical encounters during the study period) for each level of competence, with all other values in the model being held at their mean. Students reporting a 1 for competence (the optimal response) would have had a 6% probability of failing to administer an alcohol screening 100% of the time, whereas students reporting a 5 for competence (the least optimal response) would have had a 96% chance of failing to administer an alcohol screening 100% of the time.
Results indicate that students' perception of their competence with SBIRT significantly predicted whether they would screen patients for alcohol use during clinical encounters. As described in the Method section, competence was assessed following a three-part, mixed-methods training. The scale was composed of items assessing endorsement of four concepts: knowing what questions to ask patients to obtain information on their alcohol consumption, comfort level when asking about a patient's drinking patterns, knowing how to effectively help patients reduce their drinking, and ease of making such statements. This result is consistent with a prior finding that self-assessed competence predicts stated likelihood of performing SBI in clinical practice by physicians and nurses (Gassman, 2003). The current study adds to that conclusion by testing live clinical practice, rather than cross-sectional estimates of future behavior. If larger future studies validate that finding, then a training or curriculum evaluation report indicating that a particular program increased nurses' perceived competence with SBIRT may suggest that the program will be effective in promoting the clinical practice of preventive screening behaviors for alcohol abuse. In addition, if this finding is further validated, then SBIRT training programs that intend to promote clinical practice might specifically emphasize activities that increase perceived competence with SBIRT, especially knowledge about and comfort with using valid screening tools.
The current study was limited by a number of factors. First, although the number of clinical encounters assessed was large, the number of participants was small (n = 21), limiting statistical power. Second, all of the nurses were part of the same cohort of graduate students and thus were expected by their faculty to perform SBIRT during clinical practice. This may have affected their frequency of screening. Third, one or more unmeasured factors that may affect preventive care for alcohol, such as clinical experience (Savage, Johnson, Finnell, & Seale, 2015), may have altered the results, although this possibility was mitigated by enrolling students who were all part of the same graduate cohort. Finally, it is possible that several clinical encounters were marked as incomplete screenings without being noted as cases where the client was under the recommended age to use the CRAFFT screening tool for adolescents.
National competencies for NPs focus on collecting and documenting relevant health histories for patients of all ages, including behavioral screening and identification of potential risk factors affecting health and health promotion (National Organization of Nurse Practitioner Faculties, 2013). Screening for alcohol misuse and identifying patients engaged in hazardous drinking meet these essential competencies.
Of importance, the study results pertain only to students' screening behavior for alcohol and not to related SBIRT practices, such as identifying appropriate levels of intervention, providing a brief intervention, referring to treatment, or appropriately using MI and other relevant clinical skills. This work should be replicated using a prospective design with a larger number and a greater diversity (i.e., pediatric and adult NPs) of students to validate the results. However, these preliminary findings present a promising means of quickly assessing whether an SBIRT program or curriculum for nurses is likely to produce clinical practitioners who will actively screen for alcohol use.
- Agley, J., Gassman, R.A., DeSalle, M., Vannerson, J., Carlson, J. & Crabb, D. (2014). Screening, brief intervention, referral to treatment (SBIRT) and motivational interviewing for PGY-1 medical residents. Journal of Graduate Medical Education, 6, 765–769. doi:10.4300/JGME-D-14-00288.1 [CrossRef]
- Agley, J., Gassman, R.A., Vannerson, J. & Crabb, D. (2014). Assessing the relationship between medical residents' perceived barriers to SBIRT implementation and their documentation of SBIRT in clinical practice. Public Health, 128, 755–758. doi:10.1016/j.puhe.2014.05.011 [CrossRef]
- Babor, T.F., Biddle-Higgins, J.C., Saunders, J.B. & Monteiro, M.G. (2001). AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for use in primary health care. Geneva, Switzerland: World Health Organization.
- Broyles, L.M., Gordon, A.J., Rodriguez, K.L., Hanusa, B.H., Kengor, C. & Kraemer, K.L. (2013). Evaluation of a pilot training program in alcohol screening, brief intervention, and referral to treatment for nurses in in-patient settings. Journal of Addictions Nursing, 24, 8–19. doi:10.1097/JAN.0b013e31828767ef [CrossRef]
- Broyles, L.M., Rodriguez, K.L., Kraemer, K.L., Sevick, M.A., Price, P.A. & Gordon, A.J. (2012). A qualitative study of the anticipated barriers and facilitators to the implementation of nurse-delivered alcohol screening, brief intervention, and referral to treatment for hospitalized patients in a Veterans Affairs medical center. Alcohol Science & Clinical Practice, 7(7), 1–20.
- Centers for Disease Control and Prevention. (2015). Alcohol and public health: Alcohol-Related Disease Impact (ARDI). Retrieved from http://nccd.cdc.gov/DPH_ARDI/default/default.aspx
- Finnell, D.S. (2012). A clarion call for nurse-led SBIRT across the continuum of care. Alcoholism: Clinical and Experimental Research, 36, 1134–1138. doi:10.1111/j.1530-0277.2012.01870.x [CrossRef]
- Fischer, L., Pace, E., Reimann, B. & Swenson, C. (2013). Innovative approaches to integrating screening, brief intervention, and referral to treatment training into nursing education and practice in Colorado. Addiction Science & Clinical Practice, 8(Suppl. 1), A25.
- Gassman, R.A. (2003). Medical specialization, profession, and mediating beliefs that predict stated likelihood of alcohol screening and brief intervention: Targeting educational interventions. Substance Abuse, 24, 141–156. doi:10.1080/08897070309511544 [CrossRef]
- Gotham, H.J., Knopf-Amelung, S., Krom, L., Stilen, P. & Kohnle, K. (2015). Competency-based SBIRT training for health-care professionals: Nursing and social work students. Addiction Science & Clinical Practice, 10(Suppl. 1), A14. doi:10.1186/1940-0640-10-S1-A14 [CrossRef]
- Henson, R.K. (2001). Understanding internal consistency reliability estimates: A conceptual primer on coefficient alpha. Measurement and Evaluation in Counseling and Development, 34, 177–189.
- Holland, C.L., Pringle, J.L. & Barbetti, V. (2009). Identification of physician barriers to the application of screening and brief intervention for problem alcohol and drug use. Alcoholism Treatment Quarterly, 27, 174–183. doi:10.1080/07347320902784890 [CrossRef]
- Institute of Medicine. (2010). The future of nursing: Leading change, advancing health. Retrieved from http://books.nap.edu/openbook.php?record_id=12956&page=R1
- McDowell, A. & Cox, N.J. (2015). How do you fit a model when the dependent variable is a proportion?. Retrieved from http://www.stata.com/support/faqs/statistics/logit-transformation/
- Mitchell, A.M., Hagle, H., Puskar, K., Kane, I., Lindsay, D., Talcott, K. & Goplerud, E. (2015). Alcohol and other drug use screenings by nurse practitioners: Clinical issues and costs. The Journal for Nurse Practitioners, 11, 347–351. doi:10.1016/j.nurpra.2014.12.007 [CrossRef]
- Moyer, V.A. (2013). Screening and behavioral counseling interventions in primary care to reduce alcohol misuse: U.S. Preventive Services Task Force recommendation statement. Annals of Internal Medicine, 159, 210–218.
- National Organization of Nurse Practitioner Faculties. (2013). Population-focused nurse practitioner competencies. Retrieved from http://c.ymcdn.com/sites/www.nonpf.org/resource/resmgr/Competencies/CompilationPopFocusComps2013.pdf
- Puskar, K., Gotham, H.J., Terhorst, L., Hagle, H., Mitchell, A.M., Braxter, B. & Burns, H.K. (2013). Effects of screening, brief intervention, and referral to treatment (SBIRT) education and training on nursing students' attitudes toward working with patients who use alcohol and drugs. Substance Abuse, 34, 122–128. doi:10.1080/08897077.2012.715621 [CrossRef]
- Sacks, J.J., Gonzales, K.R., Bouchery, E.E., Tomedi, L.E. & Brewer, R.D. (2015). 2010 national and state costs of excessive alcohol consumption. American Journal of Preventive Medicine, 49(5), e73–e79. doi:10.1016/j.amepre.2015.05.031 [CrossRef]
- Saitz, R. (2005). Unhealthy alcohol use. The New England Journal of Medicine, 352, 596–607. doi:10.1056/NEJMcp042262 [CrossRef]
- Savage, C.L., Johnson, J.A., Finnell, D.S. & Seale, J.P. (2015). Baseline measures of importance, confidence, and current practice in an SBIRT training program: The role of experience for advance practice RNs. Retrieved from http://augusta.openrepository.com/augusta/bitstream/10675.2/559355/1/savage.pdf
- Smith, P.C., Schmidt, S.M., Allensworth-Davies, D. & Saitz, R. (2009). Primary care validation of a single-question alcohol screening test. Journal of General Internal Medicine, 24, 783–788. doi:10.1007/s11606-009-0928-6 [CrossRef]
- Strobbe, S., Perhats, C. & Broyles, L.M. (2013). Expanded roles and responsibilities for nurses in screening, brief intervention, and referral to treatment (SBIRT) for alcohol use. Journal of Addictions Nursing, 24, 203–204. doi:10.1097/JAN.0b013e3182a6914f [CrossRef]
- Substance Abuse and Mental Health Services Administration. (2015a). GPRA Modernization Act of 2010 tools. Retrieved from http://www.samhsa.gov/grants/gpra-measurement-tools
- Substance Abuse and Mental Health Services Administration. (2015b). Screening, brief intervention, and referral to treatment (SBIRT) grantees. Retrieved from http://www.samhsa.gov/sbirt/grantees
- Tavakol, M. & Dennick, R. (2011). Making sense of Cronbach's alpha. International Journal of Medical Education, 2, 53–55. doi:10.5116/ijme.4dfb.8dfd [CrossRef]
Scales and Component Items Used in Study Analyses
|Scale||Study Component Item|
|Scale 1: Participant Satisfaction (alpha = .935)a||How satisfied are you with the overall quality of this training?|
|How satisfied are you with the quality of the instruction?|
|How satisfied are you with the quality of the training materials?|
|Overall, how satisfied are you with your training experience?|
|Scale 2: Self-Reported Competence (alpha = .822)b||I know what questions to ask patients to obtain information on their alcohol consumption.|
|I am comfortable asking about a patient's drinking patterns.|
|I know how I would effectively help patients to reduce their drinking.|
|I am at ease making these statements.|
|Scale 3: Perceived Role Legitimacy (alpha = .766)c||How often do you think colleagues in your field screen patients for drinking problems?|
|How do you think colleagues in your field would feel about your screening patients for drinking problems?|
|How do you think your colleagues would feel about your stating your concerns about a patient's drinking patterns and health risks?|
|How often do you think your colleagues state their concerns about patients' drinking patterns and related health risks?|
|Scale 4: Skepticism of Behavioral Health Care (alpha = .845)b||I am not aware of a single problem drinker who ever cut back on his or her drinking upon advice from his or her care provider (e.g., physicians, nurses, social workers).|
|Advising patients with alcohol-related problems to seek assistance from an outside agency is the best I can do for them.|
|In general, I am somewhat skeptical about the efficacy of behavioral medicine.|
|Scale 5: Time Utilization and Compensation (alpha = .611)b||There is not enough time to advise patients about drinking.|
|Patients would not be willing to pay a fee for alcohol counseling.|